Hydrocarbon Processing October 2017

HARNESS THE PO POWER WER OF MANUFACTURING INNOVATION

RENTECH engineers build unmatched power and performance into every boiler we deliver. Our 80-acre manufacturing facility—the industry’s most technologically advanced—includes heavy bay and light bay areas with direct access to rail, cross-country cross-country trucking routes and shipping facilities. We master every detail to deliver elemental element al power for clients worldwide. Take Take an expanded tour of our facilities faciliti es today at www.rentechboilers.com/facilities HARNESS THE POWER WITH RENTECH.

HEAT RECOVERY STEAM GEN ERAT ERATORS ORS WASTE HEAT BOILERS FIRED PACKAGED WATERTUBE WATERTUBE BOILERS SPECIALTY BOILERS

Select 53 at www.HydrocarbonProcessing.com/RS

WWW.RENTECHBOILERS.COM

OCTOBER 2017 | HydrocarbonProcessing.com

PROCESS CONTROL AND AN D INSTRU I NSTRUME MENT NTA ATI TION ON The next generation of advanced process control control Optimizing ethylene production with laser technology Conceptual troubleshooting training for refinery opera op erators tors PROCESS OPTIMIZATION

HF alkylation conversion is finally within reach—Part 2 Optimization of a refinery’s H 2 network REGIONAL REPORT

Diversification: The catalyst for the Middle East’s downstream transformation BUSINESS TRENDS

Smart refineries: Le Leveraging veraging data for operational intelligence


PROCESS CONTROL AND AN D INSTRU I NSTRUME MENT NTA ATI TION ON The next generation of advanced process control control Optimizing ethylene production with laser technology Conceptual troubleshooting training for refinery opera op erators tors PROCESS OPTIMIZATION

HF alkylation conversion is finally within reach—Part 2 Optimization of a refinery’s H 2 network REGIONAL REPORT

Diversification: The catalyst for the Middle East’s downstream transformation BUSINESS TRENDS

Smart refineries: Le Leveraging veraging data for operational intelligence

K O B E L C O

C E N T R I F U G A L

C O M P R E S S O R S

multiple centrifugal compressor solutions for your most challenging environments High-efficiency performance in the largest refineries and power applications around the world, Kobelco’s API compliant Integrally Geared and Single Shaft Overhang (DH) compressors work in the harshest flammable, explosive, corrosive and toxic environments. Thousands of Kobelco centrifugal compressors have been in service worldwide in air separation, refining, petrochemical, energy services and many other applications. Kobelco’s integrally geared compressor is a perfect fit for cryogenic services and air separation services while keeping a small footprint. Kobelco’s DH compressor offers an open back-end design to eliminate powder build up and simplify maintenance. Let Kobelco take you where others don’t. Call today and ask how. SCREW

RECIPROCATING

CENTRIFUGAL

Kobelco Compressors America, Inc. Houston Office: [email protected] 713.655.0015 www.kobelcocompressors.com

Kobelco DH compressor allows for easy maintenance because of its simple design.

To k y o , J a p a n • H o u s t o n , Te x a s • M u n i c h , G e r m a n y • J u r o n g , S i n g a p o r e • D u b a i , U . A . E .


OCTOBER 2017

| Volume 96 Number 10

HydrocarbonProcessing.com

36

59 SPECIAL FOCUS: PROCESS CONTROL AND INSTRUMENTATION

37

Optimizing ethylene production with laser technology A. Gogates and J. Gunnell

41

The next generation of advanced process control A. Kern and B. Andrew

43

DEPARTMENTS

4 8

15 17 93 95 96 97 98

Conceptual troubleshooting training for refinery operators W. P. Averill

PROCESS OPTIM IZATION

47

Optimize a refinery H2 network using spreadsheet tools H. Salgado

53

HF alkylation conversion is finally within reach—Part 2 J. Nunez, S. Presley and D. Rana

7

67

19

Diversification: The catalyst for the Middle East’s downstream transformation L. Nichols

74

21 23 25

Turn to open-path laser detectors for HF gas safety K. Dean

SULFUR MANAGEMENT—SUPPLEMENT

S-85

COSO case study: 22 years of reliable sulfur removal M. Kolar, W. Echt and S. Osgood

GAS PROCESSING SUPPLEM ENT

GP-1

29

Technology and Business Information for the Global Gas Processing Industry

Editorial Comment Project Awards 2017

Reliability

Automation Strategies

Engineering Case Histories

Supply Chain

Roundtable The reality of process safety risk

35

New burner structure targets significant NOx reduction J. Colannino

83

People

Integration: The next frontier in supply chain and production management

Ensure uniformity of air preheater flow to obtain maximum heat transfer efficiency M. Badshahwala, R. C. Patil and A. Gupta

Design atmospheric aboveground storage tanks for accident prevention S. Zardynezhad

Advertiser Index Events

Case 98: Why performing calculations is important to an engineer’s career

Minimize engineering errors with competency and proper engineering review J. Tharakan

77

Marketplace

Expanding scope of automation events and alerts

Troubleshooting the repetitive failure and shaft seizure of a hot oil circulation pump W. K. Allah, A. Al-Abood, F. Ballard and K. Brashler

ENVIRONMENT AND SAFETY

Show Preview: ADIPEC

Technology transfer training is most effective for small groups

HEAT TRANSFER

72

Global Project Data

HP Top

MAINTENANCE AND RELIABILITY

63

Industry Metrics

COLUMNS

REGIONAL REPORT

59

Industry Perspectives Business Trends

Viewpoint Finding value in the age of smart control

DIGITAL EXCLUSIVES

Petrochemicals Innovations

Cover Image: ExxonMobil uses advanced control systems, like this one at its Baton Rouge refinery, to efficiently operate and optimize its plants. Photo courtesy of ExxonMobil.

www.HydrocarbonProcessing.com

P. O. Box 2608 Houston, Texas 77252-2608, USA Phone: +1 (713) 529-4301 Fax: +1 (713) 520-4433 [email protected]

PUBLISHER

Industry Perspectives

Catherine Watkins [email protected]

EDITOR/ASSOCIATE PUBLISHE R

Lee Nichols [email protected]

EDITORIAL

What is the future of the global HPI? Over the past year, the world has witnessed significant downstream capacity growth in all sectors of the hydrocarbon processing industry (HPI). The majority of this growth took place in Asia-Pacific, the Middle East and the US. These three regions will continue to invest billions of dollars to increase oil and gas processing capacity into the early 2020s. Although the Asia-Pacific, Middle East and US regions will be the leaders in the development of downstream processing capacity, other regions are investing in additional distillation and secondary units construction, petrochemical capacity and LNG liquefaction and regasification terminals. These investments are ensuring that global demand for refined fuels, petrochemical products and natural gas are met in the future. However, the present and future of the HPI can vary immensely by region. What are the major trends within each sector of the global HPI? Which regions are investing in new processing capacity to meet demand? How has supply and demand of oil and gas products shifted over the past several years, and what will the future of the HPI look like? These questions are a major focus of Hydrocarbon Processing’s HPI Market Data 2018. HPI Market Data 2018 provides top-level insight and transparency into market trends, spending and activity in the global petrochemical, refining and natural gas/LNG industries. The publication provides downstream oil and gas professionals with detailed insight into the activities and operations of the global HPI. HPI Market Data 2018 features: • Global spending in the refining, petrochemical and gas processing sectors • Forecasts of the HPI’s capital, maintenance and operating budgets for 2018 • Short-term and long-term implications of today’s low crude oil prices • An exploration of changing markets and demand within the global HPI, with discussion of emerging markets • More than 35 tables and 70 figures • Editorial analysis of worldwide economic, social and political trends driving HPI activity across all sectors. Additional highlights include: • The surge in petrochemical capacity investments • How crude oil supply and demand is affecting the downstream HPI, as well as global and regional trade • A detailed overview of LNG import/export terminal construction • A detailed overview on new sulfur and clean-fuels regulations around the world • A detailed overview on the IMO 2020 Global Sulfur cap on marine fuels. For more information, visit GulfPub.com/mdb2018 4


Executive Editor Managing Editor Technical Editor Digital Editor Reliability/Equipment Editor Contributing Editor Contributing Editor Contributing Editor MAGAZINE PRODUCTION /

Adrienne Blume Mike Rhodes Bob Andrew Ashley Smith Heinz P. Bloch Alissa Leeton ARC Advisory Group Anthony Sofronas

+1 (713) 525-4633

Vice President, Production Manager, Advertising Production Manager, Editorial Production Assistant Manager, Editorial Production Artist/Illustrator Graphic Designer

Sheryl Stone Cheryl Willis Angela Bathe Dietrich Lindsey Craun David Weeks Andreina Keller

ADVERTISING SALES

See Sales Offices, page 96. CIRCULATION

/ +1 (713) 520-4498 / [email protected]

Manager, Circulation

Suzanne McGehee

SUBSCRIPTIONS

Subscription price (includes both print and digital versions): One year $399, two years $679, three years $897. Airmail rate outside North America $175 additional a year. Single copies $35, prepaid. Hydrocarbon Processing’s Full Data Access subscription plan is priced at $1,695. This plan provides full access to all information and data Hydrocarbon Processing

has to offer. It includes a print or digital version of the magazine, as well as full access to all posted articles (current and archived), process handbooks, the HPI Market Data book, Construction Boxscore Database project updates and mo re. Because Hydrocarbon Processing is edited specifically to be of greatest value to people working in this specialized business, subscriptions are restricted to those engaged in the hydrocarbon processing industry, or service and supply company personnel connected thereto. Hydrocarbon Processing is indexed by Applied Science & Technology Index, by

Chemical Abstracts and by Engineering Index Inc. Microfilm copies available through University Microfilms, International, Ann Arbor, Mich. The full text of Hydrocarbon Processing is also available in electronic versions of the Business Periodicals Index. DISTRIBUTION OF ARTICLES

Published articles are available for distribution in a PDF format or as professionally printed handouts. Contact Foster Printing at Mossberg & Co. for a price quote and details about how you can customize with company logo and contact information.

For more information, contact Jill Kaletha with Foster Printing at Mossberg & Co. at +1 (800) 428-3340 x 149 or [email protected]. Hydrocarbon Processing (ISSN 0018-8190) is published monthly by Gulf Publishing

Company, 2 Greenway Plaza, Suite 1020, Houston, Texas 77046. Periodicals postage paid at Houston, Texas, and at additional mailing office. POSTMASTER: Send address changes to Hydrocarbon Processing, P.O. Box 2608, Houston, Texas 77252. Copyright © 2017 by Gulf Publishing Compan y. All rights reserved. Permission is granted by the copyright owner to libraries and others registered with the Copyright Clearance Center (CCC) to photocopy any articles herein for the base fee of $3 per copy per page. Payment should be sent directly to the CCC, 21 Congress St., Salem, Mass. 01970. Copying for other than personal or internal reference use without express permission is prohibited. Requests for special permission or bulk orders should be addressed to the Editor. ISSN 0018-8190/01.

President/CEO CFO Vice President Vice President Vice President, Finance and Operations Vice President, Production

John Royall Alan Millis Andy McDowell Ron Higgins Pamela Harvey Sheryl Stone

Publication Agreement Number 40034765

Printed in USA TM

Other Gulf Publishing Company titles include: Gas Processing , Petroleum Economist ©, World Oil ® , Pipeline & Gas Journal and Underground Construction.

™

INNOVATING CAUSTIC TREATING TECHNOLOGY ®

Introducing FFC Plus™, the next generation of FIBER FILM Contactor technology. Merichem Company’s advanced FIBER FILM technology with an integrated coalescer and treater pushes the boundaries of extractive technologies. With this latest innovation, hydrocarbon treating rates can be increased by up to 150% through increased mercaptan extraction efficiency. The integrated coalescing and treating device offers a simple installation, reduced chemical use, and reduced service and maintenance complexity. FFC Plus truly is mercaptan treating made better. Select 84 at www.HydrocarbonProcessing.com/RS

www.merichem.com

At the end of the day, you want a technology supplier who works with you. You’re committed to progress and success. We’re committed to you. And we demonstrate our commitment through licensing world-class refining, gas, chemical technologies and specialty catalysts that drive exceptional performance. You can count on our proven technology and long-term collaboration to help you keep pace with the increasingly complex challenges of today’s evolving marketplace. From initial consultation through plant startup and beyond, our global team offers you practical guidance based on years of real-world operating experience. Our goal is your success. Learn more about how we can work together to grow your business. www.catalysts-licensing.com . n o i t a r o p r o C l i b o M n o x x E f o s k r a m e d a r t e r a n i e r e h s e m a n t c u d o r p l l a d n a e c i v e d ” X “ g n i k c o l r e t n i e h t , o g o l l i b o M n o x x E e h t , l i b o M n o x x E . d e v r e s e r s t h g i r l l A . n o i t a r o p r o C l i b o M n o x x E 6 1 0 2 ©


Editorial Comment

HP Top

LEE NICHOLS, EDITOR/ASSOCIATE PUBLISHER [email protected]

Project Awards 2017

It’s that time again! Hydrocarbon Processing has announced the nominees for its annual Top Project awards. The editors of Hydrocarbon Processing have identified nine projects that will contribute significantly to the expansion of the hydrocarbon processing industry, whether through contributing capital expenditures, satisfying or increasing domestic or regional demand, diversifying product offerings or adding to the resurgence in refining and/or petrochemical processing capacity. These nine projects span the globe and represent more than $114 B in total capital expenditures. The winners of this prestigious award over the last two years include: • Refining

o 2014�Saudi Aramco and Total Refining and Petrochemcial Co.’s (SATORP) Jubail Refinery o 2015�SOCAR’s Turkey Aegean Refinery (STAR)

o 2016�KNPC’s Clean Fuels Project • Petrochemicals

o 2014�Saudi Aramco and Dow Chemical’s SADARA Petrochemical Complex o 2015�Sasol’s Ethane Cracker and Derivatives Complex o 2016�Dow Chemical’s Oyster Creek PDH unit project. This year’s refining nominees (TABLE 1) represent nearly 2 MMbpd of new refining capacity by the early 2020s. These projects represent a total capital investment of more than $54 B. The five petrochemical nominees (TABLE 2) have a total cost of nearly $60 B and represent more than 36 MMtpy of additional petrochemicals production by the early 2020s. The choice is now up to you! Beginning October 1, readers of Hydrocarbon Processing can make their voices heard in an exclusive online poll. The winners will be revealed in Hydrocarbon Processing’s December issue.

TABLE 1. Top refining project nominees

Operator

Project

Location

Capacity

KNPC

Al-Zour refinery

Al-Zour, Kuwait

615 Mbpd

PDVSA

Puerto La Cruz Deep Conversion project

Puerto la Cruz, Venezuela

210 Mbpd

Zhejiang Petrochemical

Dayushan Island Refinery and Dayushan Island, Petrochemical Integrated Complex China

NIORDC

Persian Gulf Star refinery

Bandar Abbas, Iran

800 Mbpd (Phases 1 and 2) 360 Mbpd

TABLE 2. Top petrochemical project nominees

Operator

Project

Location

Capacity

ORPIC

LIWA Plastics Industries Complex

Sohar, Oman

1 MMtpy+

Sibur

ZapSibNeftekhim Petrochemical Complex (ZapSib-2)

Near Tobolsk, Russia

1.5 MMtpy+

Negin Mokran Development Petrochemical Co.

Mokran Petrochemical Complex

Chabahar, Iran

25 MMtpy (Phases 1–3)

Petronas

Refining and Petrochemical Integrated Development (RAPID)

Pengerang, Johor, Malaysia

7.7 MMtpy

S-Oil

Residue Upgrading and Olefin Ulsan, South Korea Downstream Complex (RUC/ODC)

705 Mtpy

INSIDE THIS ISSUE

29

Roundtable. In this first

36

Special Focus. As the

47

Process Optimization.

installment of Hydrocarbon Processing’s new roundtable discussions, four respected oil and gas industry professionals discuss how the industry thinks it manages risk, how it is actually managed and how it can be improved.

downstream processing industry moves toward a more digital world, advanced process control solutions provide hydrocarbon processing companies with inventive ways to manage difficult and complex operational and reporting applications. This month’s Special Focus section showcases innovations in control devices, hardware and software to increase plant efficiency.

Articles focus on optimizing a refinery’s hydrogen network, and the continuing discussion on converting alkylation units from hydrofluoric acid to sulfuric acid.

S-86

Sulfur Supplement.

Around the world, legislation mandating decreased emissions and lower levels of airborne pollutants is coming into effect. In response, refiners are implementing operational and processing changes to reduce sulfur levels in transportation fuels. Hydrocarbon Processing’s Sulfur Supplement includes companies that are taking on this sulfur challenge, as well as those that handle this element every day.

93

Show Preview.

Hydrocarbon Processing details the upcoming Abu Dhabi International Petroleum Exhibition and Conference (ADIPEC). The conference will feature more than 900 speakers at 200 technical and strategic conference sessions, as well as a new downstream program that will feature plenary panels on downstream investment, expansion, diversification and development. Hydrocarbon Processing | OCTOBER 2017

7

| Business Trends A convergence of information technology (IT) and real-time operational technology (OT) is raising the bar for managing downstream operations. The volume and subsequent values of the information generated in production operations are increasing, and the convergence of IT and OT is increasing the volumes and complexity of data and information. Embedding intelligence in daily refinery operations is one of the key steps toward operational intelligence (OI). OI monitors, identifies and detects situations related to inefficiencies, opportunities and threats, and provides operational solutions that help refineries address all aspects of plant performance in real-time. The level of OI achieved is almost entirely dependent on how well the plant, operational and business data are managed. This month’s Business Trends discusses the need for an operational intelligence framework, a methodology and the technology needed for implementation. Photo: The

control room is instrumental in the manual and digital monitoring of assets. Photo courtesy of ABB.

M. D. AGRAWAL, Tata Consultancy Services Ltd., Powai, Mumbai, India; and A. MORAWALA, Tata Consultancy Services Ltd., Andheri, Mumbai, India

Business Trends

Smart refineries: Leveraging data for operational intelligence In the highly competitive, capital- and asset-intensive oil and gas industry, demand is growing for the implementation of information technology (IT) practices covering the following functional requirements: • Informed decision-making capabilities� collaborative decision-making and intelligent decision automation • Analytic capabilities�descriptive diagnostics, predictive and prescriptive techniques • Information management capabilities�describe, organize, integrate, share, govern and implement. Executives do not want to look at past performance, but would rather use present and historical performance, as well as operational models, to navigate the road ahead, including how to satisfy near-term changes in demand. They must know which actions to take now to yield the most profitable outcomes tomorrow�or even later today. A convergence of IT and real-time operational technology (OT) is raising the bar for managing downstream operations. The volume and subsequent values of the information generated in production operations are increasing, and the convergence of IT and OT is increasing the volumes and complexity of data and information. The need for an operation intelligence framework, a methodology and the technology needed for implementation are discussed here, along with business outcomes. About the domain. Downstream refining is a potentially

risky business. Global oil and gas economics, local recessions and plant productivity are all factors in defining business profitability. A trend toward “smart manufacturing” is seen, through which the risks posed by debilitating factors can be mitigated, to a large extent. Smart refineries are the result of informed, near-real-time decision-making that is based on historical and present process and business data, providing the ability to predict near-future productivity. When built into operations, such intelligence results in operational intelligence (OI)�the ability to analyze information to predict future performance. Operational intelligence: An industry perspective. Em-

bedding intelligence in daily refinery operations is one of the key steps toward achieving OI, according to global best practices. OI monitors, identifies and detects situations related to inefficiencies, opportunities and threats, and provides opera-

tional solutions that help refineries address all aspects of plant performance in real time. Informed real-time decisions are achieved in key areas, including: • Improved feedstock flexibility • Growth in high-value product yield • Guaranteed availability • Increased logistics capabilities • Reduced risk. OI delivers visibility and insight into data, real-time events and business operations, and enables businesses to make decisions and act on these analytic insights through manual or automated actions. Data consolidation and the integration of operation and business systems offer a firm foundation for business transformation. Making sense of the surfeit of data is a major step toward OI, which: • Enables a common view of refinery key performance indicators (KPIs) • Provides a uniform enterprise monitoring system that encompasses all operations • Maximizes data visibility and asset availability • Delivers multidimensional analytics on critical business parameters • Improves response to change in operational parameters and takes proactive action through predictive analytics • Develops business and operations forecasting • Optimizes resource allocation and production capacity • Increases profitability. These performance indicators rely on a strong real-time data management strategy. Data is continuously generated through multiple interactions between refinery processes and functional areas. The goal is to aggregate such data, contextualize it, generate useful information through advanced analytical tools, and then disseminate this actionable information to its target audience. Role of data management in OI. In any production enter-

prise, data is generated by numerous systems and processes. These can be broadly classified into four sets: production data, process data, master data and business data ( FIG. 1). When data is generated from any base system, it contains no links to other processes; the raw data generated relates to only those parameters that create it. Multiple systems may make use of similar data, but only from their individual databases. Such data has limited “information value” because: Hydrocarbon Processing | OCTOBER 2017

9

Business Trends • These data clusters are islands of information • They vary with different systems at different locations • They are individualistic and lack comprehensive information of the entire system • The same data may be available at multiple locations (i.e., duplicity of data) • Data is generated from disparate systems and applications across the organization. Once all data is consolidated in one unified, structured data base, it can be accessed by all systems and processes to achieve close coordination through feedback loops.

For example, the “optimization” function occurs across the entire refinery process chain, from crude intake to final product selling. Depending on the process, optimization can happen in any form: technical (asset strategy, blending); business-oriented (crude cost, product cost); customer relation management (customer segment servicing, product pricing); and retail (risk management, improved payment schedule). Similarly, a process can derive information from multiple functional areas. Each of the functional parameters, optimization through “process control,” has information to offer to the “product management” process.

Downstream refining—data dependencies. Downstream

Managing data. Effectively controlling such complex opera-

refining comprises numerous processes and functions that are mutually related through information exchange. TABLE 1 pro vides insight into the interdependencies of processes and functional areas in a downstream refinery.

tions requires real-time insight into the various process cycles. Data management includes: • Gathering, aggregating and contextualizing data. Linking data to other devices or systems can increase its effectiveness. For example, data from a temperature sensor by itself is not as useful as data that is linked to various, temperature-dependent processes. • Analyzing data to identify actionable insights that create better outcomes. • Predicting refinery performance based on the analysis of past and present data. • Generating actionable information, which differs with respect to the perspective. For business users, this information can be in the form of charts, dashboards, reports, etc., to decipher the state of the refinery in real time. From an IT perspective, FIG. 2 depicts the basic architecture of a data management system.

Raw data from edge devices, instrumentation and control systems, data acquisition systems, process control, etc.

CRM, ERP, finance, sales, etc.

Process data

Business data

Production data

Master data

AN APPROACH TO DATA INTELLIGENCE Processed data from production systems, LIMS, energy, yield accounting, etc.

Governance information, HSE, planning and scheduling, asset data, customer services, etc.

FIG. 1. Sources of data in a downstream refinery.

Shown in FIG. 3, a rule-based approachan essential abstract tool to generate actionable intelligence from dataenables decision-making. The major containments within this approach are briefly described here.

TABLE 1. Downstream refinery processes and functional areas

Downstream processes r Functional areas d Optimization

Planning

Business deals

Process control

Feedstock management

Refining management

Product management

Marketing and sales

Purchase and HSE

Trade and cost

Asset strategy

Product supply and cost

Customer segments

Risk management

Feedstock selection

Refinery optimization

Grade and blending

Price

Enhanced payment schedule

Crude fingerprinting

Predictive maintenance

Supply and demand forecasting

Retail network management

Supply schedule and price

Production planning and scheduling

Transportation cost

Distribution optimization

Feedstock price

Blending

Feedstock inspection

Quality management

Product inspection

B2B contract management

Order delivery and invoicing

Feedstock invoicing

Asset management

Product transport

Orders and pricing

Safety management

Feedstock inventory

Product inventory

Marketing campaigns

Pipeline DCS Fleet management

10OCTOBER 2017 | HydrocarbonProcessing.com

DCS/APC/MES/ Plant assets

Plant DCS

Instrumentation

The best just got better

And smarter. And safer. The AMETEK Model 888 Tail Gas Analyzer was designed with safety in mind. With a web-enabled interface, smart diagnostics and a flange temperature alarm that detects bad steam quality, service personnel are isolated safely away from the process. The Model 888’s exclusive auto flow control also prevents sulfur entrainment during process turndown, while its advanced electronics resist overheating, reducing downtime and maintenance. Plus, its easy-to-read color

VGA display can also be viewed remotely via Modbus and Ethernet connections. Our proven Model 880 has long been the best analyzer in the industry. The Model 888 Tail Gas Analyzer—the evolution of a well-proven formula.

sru.ametekpi.com © 2017 by AMETEK Inc. All rights reserved.


Business Trends

Aggregate. Oil and gas operations involve distributed data

sources in process, business and asset management systems that must be consolidated and aggregated. The realities of such data sources include: • Dispersed geography�covering multiple assets and plants, distribution stations, refineries, etc. • Data source variety�manual and digital sensors on pumps and distribution lines, refinery process lines, laboratories • Data mode variety�continuous data from DCS/ historians, discrete data from field operator manuals, mobile device data collection, laboratory sample data. Contextualize. Create and maintain functional/operational

relationships between data elements from disparate sources. In FIG. 3, contextualization is evident within each horizontal block. Production system information can be functionally related to overall performance metrics, so business systems information can be related to performance as demanded by operations. Propagate. Transfer relevant operational performance infor-

mation to the appropriate business-level systems. For example, inventory management and control information are available

across the refinery in real time, based on the produced items. This allows procurement departments to monitor low inventory stocks and initiate procurement without being informed by others. This may be considered as closing a feedback loop. Alternatively, refinery operations may redefine schedules or other indices based on up-to-date information. Analyze. Data analytics transform data into real-time perfor-

mance intelligence through the application of business rules. This part of the framework is responsible for advanced features of smart refineries, such as predictive maintenance, which utilizes historical data from multiple sources to build accurate, testable predictive models, allowing the generation of predictions and risk scores. This provides a more effective and efficient way to maintain and monitor critical assets, with high availability and reliability. Visualize. Graphical representations of KPIs support the con-

text- or role-based navigation of information based on persistent interrelationships. In some instances, this means enabling drill-down from multi-plan representations to individual facilities and systems. In the process unit overview and utility overview dashboard in FIG. 4, critical parameters are graphically adapted to suit operator intuition. Such visuals assist rapid decision-making. This framework is built using multiple IT tools and digital technologies. Platforms and management approaches. Platforms and

FIG. 2. Indicative architecture of a data management system,

integrating disparate data silos into a single source of enterprise information. Source: Gartner, 2009.

Contextualize s m e t s y s s s e n i s u B s e c r u o s a t a D

s m e t s y s s p O s m e t s y s s s e c o r P

Analyze

ERP, procurement, finance, sales, CRM, HSE, HR Production systems, LIMS, yield accounting, energy Edge devices, control systems

Visualize

Demand performance

e t a g a p o r P

Overall process performance metrics for the refinery

Large amount of data from multiple sources related to the process systems

Aggregate

FIG. 3. An approach to building intelligence from refinery data.

12


management approaches that provide a foundation to build ro bust OI solutions include: • Data modeling: The ISA 95 standard describes a framework for connecting plant floor operations to enterprise applications. Data models specific to refineries based on this standard provide a good starting point. However, proper configuration of the models is required to accurately reflect operations. • Data lifecycle management: Short-term storage of actionable data and longer-term archival and retrieval of selected OI data. Providing full resolution of archived and short-term data is necessary, but the dif ference is in the access rate. Short-term operational data must be immediately available for analysis. Retrieval of long-term archived data will be used for strategic decision-making. � Real-time KPI and extensive historian data access are part of data lifecycle management. • Business process management, data mining and discovery: These business process-centric approaches are helpful for developing a desired layer of data aggregation, analysis and virtualization. � A business process is a collection of related, structured activities that produce a service or product that meets the needs of the business. � Data mining techniques help in the aggregation of data from heterogeneous sources. � Business process management helps provide modeling tools, establish links of common objects and enable process modeling, an important requirement of OI and real-time visualization.

Business Trends USE CASES—PUTTING THEORY INTO PRACTICE

Some examples of how data consolidation helps the refinery are listed here. Major parameters depicted are energy consumption, production parameters and KPI variance. Other information related to production includes asset health (maintenance), HSE parameters and human resources availability. Production monitoring. Production managers looking at the

consolidated dashboard can see the whole picture at a glance, including: • Amount of crude processed and the yield • Available inventory • Basic quality parameters of the products • Energy consumed and other utility (water, gas, steam, etc.) information • Unit KPI variance (actual KPI vs. target KPI) • Asset information and human resources availability. Production managers are informed of core production parameters, asset availability, work permits issued and human resources availability. These outer circle parameters ensure awareness of production efficiencies and provide foresights on threats to normal production. Real-time information about related metrics, KPIs and performance highlights enable timely action and escalations. Asset maintenance and work order(s) issued for each major unit and its ancillary assets can be visualized, and it can be determined if multiple work orders have

been issued, which will hamper production. Engineering departments generally track scheduled and breakdown maintenance to monitor the mean time between failure (MTBF) of assets. This historical information can be used to effectively schedule preventive maintenance. Such historical information, combined with manufacturer design and operational parameters and rule-based algorithms, can be directed toward predictive maintenance. While preventive or scheduled maintenance must be carried out irrespective of asset functionality, predictive maintenance saves resources by generating alerts only when the asset’s functionality tends to exceed its usual parameters.

FIG. 4. An

indicative dashboard of overall utility functions.

D O Y O U H AV E W H A T I T TA K E S T O B E THE BEST? NAPCON is hosting the Operator World Cup 2017 to find the best operators. Think you

OPERATOR WORLD CUP

www.napconsuite.com have what it takes? Join the OWC, and play

NAPCON Games Distiller to find out. For more information, please visit

napcongames.com

Select 151 at www.HydrocarbonProcessing.com/RS Hydrocarbon Processing | OCTOBER 2017

13

Business Trends

Enterprise data warehouses (PLM/ERP/SCM/Bl) Manufacturing data warehouse (historian or DB associated with MES, EH&S, etc.)

Business leadership d e v r e s s e l o r f o m u u n i t n o C

Manufacturing management Site leadership Functional/ task workers Operators, engineers, technicians

DB

DB Operational data stores (MES, CMMS, etc.)

Data historians

DB Operational technology (OT) Real time (seconds, minutes)

z e d c i a l i s S p e t i c s D B l y a n a

Transactional Staged data (minutes, hours, shifts) (shifts, days, weeks, months) Time frame for decision support

CMMS—Computerized maintenance management system; DB—Database; EH&S—Environmental health and safety

FIG. 5. Platforms in an OI architecture. Source: Gartner, 2011.

A dashboard, like that shown in FIG. 4, is used by process engineers and provides a total graphic representation of the various resources consumed and the cost of production at the unit. Tracking the energy and resources consumed is necessary, as they represent a high cost in the production cycle. Unit efficiency is also measured by how resources are spent in the production cycle. In the dashboard in FIG. 4, the consumption of various resources, such as steam, water, fuel and nitrogen, is tracked to monitor unit performance. Key operating parameters, such as boiler efficiency and compressor running hours, indicate the efficiency of the unit’s production. The overall cost of resources during production is also shown. This cost is compared with budgeted estimates and is a measure of the effectiveness of the production process. The architecture of OI. OI is not a batch of applications that

churns out the resulting intelligence, but is rather an architecture or design involving multiple platforms and systems spanning across timeframes and roles. The architecture includes: historians that provide a robust foundation for an OI database, with the ability to aggregate, contextualize and analyze large volumes of data from various sources such as DCS, control systems and other OT platforms; MES systems with built-in intelligence capabilities that operate on data within the context of the application’s data; and business analytics that offer historical, present and predictive views of business operations. To date, no single vendor delivers a complete OI solution. Customers generally take a blended approach, and IT service providers interface and integrate multiple applications to suit a customer’s requirements. Business outcomes. OI enables an improved decision-mak-

ing capability based on enhanced information quality and availability extracted from a mass of data points. Such information produces alerts based on KPIs and key risk indicators. Often, the greatest opportunities for key improvements and increased efficiencies occur when silos of previously independent data are integrated or related in a new and meaningful way. This requires effectively spanning numerous disparate sources of information from tag-based systems (control systems) and relational data sources (e.g., a production database) to enterprise ap14


plications (e.g., an ERP system). Information from these diverse systems must be capable of being combined in expressions or calculations and related through structures, filtering and navigation. Diverse operation and business data converge as a single, virtual database that contains all information that is otherwise distributed in various systems, such as SAP, CMMS, LIMS, planning, production, inventory, historian, machine monitoring, etc. This integration allows multi-dimensional analytics that can be analyzed from various perspectives (production, supply chain, planning, maintenance, etc.). A proprietary scalable and reliable platform delivers manufacturing integration and intelligence. This flexible solutiona ensures full integration between shop floor systems, MES and SAP ERP to deliver benefits that include: • A comprehensive platform that enables composite applications • The creation of an environment that integrates disconnected environments • Role-based visibility • A technology agnostic platform that interoperates across multiple platforms. In addition to functionality, platform scalability and delivery expertise, reusable templates, data models and a configuration kit reduce the overall development and deployment time. Key takeaways. The level of OI achieved is almost entirely de-

pendent on how well the plant, operational and business data are managed. This involves complex data management frame works, coupled with process changes, to absorb the intelligence that is spelled out by the transformed data, as well as adherence to a disciplined, structured manufacturing system. While many tools can assist in OI, such tools must be selected carefully based on individual needs. Each enterprise�even within the same sector�is an individual entity with its own quirks and needs, and the tools must be tweaked to give optimal results. OI is a prequel to operational and business excellence, and an important milestone to achieve. NOTES

a TCS Global Operations Solution MURLI D. AGRAWAL is an Advisor and Director for Downstream COE at TCS’s

Oil and Gas Global Practice. He formerly served as General Manager of IT and Chief Information Officer (CIO) of downstream manufacturing for BPCL, India’s second-largest oil major. Mr. Agrawal was an a rchitect of an integrated, real-time refinery information model for end-to-end processes that was implemented at two BPCL refineries. He is now involved in designing strategies for downstream domain excellence studies for global customers and mentoring for comprehensive solution design proposals for global refineries. Mr. Agrawal also serves as VP and President-elect of the Bombay Management Association (BMA), is a member of the Apex Professional University body of senior managers, and is a visiting faculty member at the Shailesh J. Mehta School of Management (SJMSOM), which is part of the Indian Institute of Technology (IIT) in Mumbai. ARDISH M. MORAWALA, associate author, is a Sales Consultant for engineering

and industrial systems in the Asia-Pacific region for Tata Consultancy Services Ltd. He has 38 yr of technical experience in electronic systems engineering and the installation of motor controls, automation, DCS and embedded devices. Mr. Morawala specializes in field implementations of complex electronic and IT systems. Throughout his career, he has been actively engaged at site installations of electronic drives in large steel rolling mills and UPS systems in EDP centers. For a decade, he was responsible for the engineering of DCS systems, including RTUs and central computer systems. As part of a team, he installed DCS systems for railway traction control, ONGC offshore oil rigs, water treatment plants and power distribution utilities.

MIKE RHODES, MANAGING EDITOR [email protected]

Industry Metrics

US product crack spreads surged amid refinery outages as Hurricane Harvey affected US Gulf Coast (USGC) refinery operations. Product crack spreads in Europe rose—excluding fuel oil, which declined—and were affected by supply outages on the USGC and increasing arbitrage opportunities to the US and Latin America. Asian product crack spreads improved across all products except fuel oil.

Global refining margins, 2016–2017* 20 15

10 5

An expanded version of Industry Metrics can be found online at HydrocarbonProcessing.com.

0 6 1 . g u A

US gas production (Bcfd) and prices (US$/Mcf) 100

7

5 60

4

40

3 Monthly price (Henry Hub) 12-month price avg. Production

20 0

2

6 1 . c e D

7 1 . n a J

7 1 . b e F

7 1 . r a M

7 1 l i r p A

7 1 y a M

7 1 e n u J

7 1 y l u J

7 1 . g u A

US EU 16

Japan Singapore

60 6 1 . g u A

2017

Production equals U.S. marketed production, wet gas. Source: EIA.

6 1 . t p e S

6 1 . t c O

6 1 . v o N

6 1 . c e D

7 1 . n a J

7 1 . b e F

7 1 . r a M

7 1 l i r p A

7 1 y a M

7 1 e n u J

7 1 y l u J

7 1 . g u A

US Gulf cracking spread vs. WTI, 2016–2017 *

Selected world oil prices, US$/bbl 60

80

50

l b b / $ S U , d a e r p s g n i k c a r C

Dubai Fateh W. Texas Inter. Brent Blend

70 l b b / $ S U , s e c i r p l i O

6 1 . v o N

% , 90 s e t a r n 80 o i t a z i l i t U 70

J A S O N D J F M A M J J A S O N D J F M A M J J A 2016

6 1 . t c O

100 f c M / $ , s e c i r p s a G

1 0

2015

6 1 . t p e S

Global refining utilization rates, 2016–2017 *

6

80 d f c B , n o i t c u d o r P

WTI, US Gulf Brent, Rotterdam Oman, Singapore

l b b / $ S U , s n i g r a M

60

40 30

Source: DOE

Prem. gasoline Jet/kero Diesel Fuel oil

20

50

10

40

0

-10 -20

30

6 1 . g u A

20 A

S

O

N

D

J

F

M

A

2015

M

J

J

A

S

O

N

D

J

F

M

A M

2016

J

J

A

2017

2013-Q1

2014-Q1

6 d p 5 b M M 4 , e c 3 n a l a 2 b d 1 n a e 0 g n a -1 h c k c -2 t o S -3

Forecast

Stock change and balance World supply World demand

2015-Q1

2016-Q1

2017-Q1

30

7 1 . n a J

7 1 . b e F

7 1 . r a M

7 1 l i r p A

7 1 y a M

7 1 e n u J

7 1 y l u J

7 1 . g u A

7 1 . t p e S

Prem. gasoline Jet/kero

Gasoil Fuel oil

10 0

-10

-20 6 1 . g u A

6 1 . t p e S

6 1 . t c O

6 1 . v o N

6 1 . c e D

7 1 . n a J

7 1 . b e F

7 1 . r a M

7 1 l i r p A

7 1 y a M

7 1 e n u J

7 1 y l u J

7 1 . g u A

7 1 . t p e S

Singapore cracking spread vs. Oman, 2016–2017 *

Brent dated vs. sour grades (Urals and Dubai) spread, 2016–2017 *

30 l b b / $ S U , d a e r p s g n i k c a r C

20

8

Dubai Urals

Prem. gasoline Jet/kero

Gasoil Fuel oil

10 0

-10

-20

-4

6 1 . g u A

6 1 . c e D

20

2018-Q1

6 1 . v o N

40 l b b / $ S U , d a e r p s g n i k c a r C

Source: EIA Short-Term Energy Outlook , September 2017

r u l o 6 s b b / m $ u S i U4 d , e d m a2 / t e e r p e s 0 w s e t d h u r-2 g c i L

6 1 . t c O

Rotterdam cracking spread vs. Brent, 2016–2017*

World liquid fuel supply and demand, MMbpd 102 d100 p b M98 M , d 96 n a m94 e d 92 d n a 90 y l p 88 p u S 86 84 2012-Q1

6 1 . t p e S

6 1 . t p e S

6 1 . t c O

6 1 . v o N

6 1 . c e D

7 1 . n a J

7 1 . b e F

7 1 . r a M

7 1 l i r p A

7 1 y a M

7 1 e n u J

7 1 y l u J

7 1 . g u A

7 1 . t p e S

6 1 . g u A

6 1 . t p e S

6 1 . t c O

6 1 . v o N

6 1 . c e D

7 1 . n a J

7 1 . b e F

7 1 . r a M

7 1 l i r p A

7 1 y a M

7 1 e n u J

7 1 y l u J

7 1 . g u A

7 1 . t p e S

* Material published permission of the OPEC Secretariat; copyright 2017; all rights reserved; OPEC Monthly Oil Market Report, September 2017. Hydrocarbon Processing | OCTOBER 2017

15

TECHNOLOGY INNOVATION IS OUR DNA We, at Shell Global Solutions, constantly create new tailored solutions with our clients to exceed their expectations. To learn more, please visit www.shell.com/globalsolutions

Shell Global Solutions Select 92 at www.HydrocarbonProcessing.com/RS

LEE NICHOLS, EDITOR/ASSOCIATE PUBLISHER [email protected]

Global Project Data

Hydrocarbon Processing’s Construction Boxscore Database is

tracking more than 1,700 active downstream construction projects around the world. Over the past three years, the Asia-Pacific region and the US have led in new project announcements. Developing nations in the Asia-Pacific region continue to build downstream

processing capacity to meet increasing demand. The US refining industry is investing in new distillation capacity to process additional light shale oil, as well as in secondary units to adhere to new Tier 3 fuel regulations. The Asia-Pacific and Middle East regions account for half of total active project market share.

16%, 12%, 12%

7%, 3%, 4% 33%, 25%, 23%

8%, 9%, 12% 6%, 7%, 9% 19%, 34%, 33%

11%, 10%, 7%

Market share analysis of new project announcements by region, 2015–2017 32 29

6% Africa

29

16% US 20

20

17 12

12

13

12 10

10

11

12

22% Middle East 28% Asia-Pacific 4% Canada 15% Europe

July- Aug.- Sept.- Oct.- Nov.- Dec.- Jan.- Feb.- Mar.- April- May- June- July- Aug.16 16 16 16 16 16 17 17 17 17 17 17 17 17

Boxscore new project announcements, July 2016–present

9% Latin America Market share analysis of active downstream projects by region

Detailed and up-to-date information for active construction projects in the refining, gas processing and petrochemical industries across the globe |ConstructionBoxscore.com

Hydrocarbon Processing | OCTOBER 2017

17


Reliability

HEINZ P. BLOCH, RELIABILITY/EQUIPMENT EDITOR [email protected]

Technology transfer training is most effective for small groups Training can take many different forms. Naming just a few, we can: • Send employees to public courses, conferences and meetings. • Arrange for experienced experts to make in-plant presentations at our facility and ask the attendees to turn off their cellular phones for the duration. • Invite vendor representatives to do “lunch-and-learn” sessions at a nearby offsite location, or in an in-plant meeting room; the chosen location could be temporarily reconfigured to accommodate the lunch crowd. It can be demonstrated that each of these different approaches has its advantages and disadvantages. Well-thought-out training is indispensable for plants that wish to excel in achieving safety, reliability and profitability. However, for a variety of reasons, relatively few of these highly appropriate goals are being implemented with the urgency they deserve. To remedy the situation, an increasingly large number of HPI facilities are f inding ways to accelerate the transfer of cost-saving and reliability enhancing technologies. Pick the right employees to receive targeted training.

Technology transfer is best absorbed by small teams of advisors, mid-level managers and senior technicians. Exposing them to expert teachers with strong implementation know-how is the key to understanding, accepting and rapidly implementing innovative solutions. Several relatively new technologies have been adopted in the past two decades by best-of-class companies. Mechanical seals, dry gas seals, advantageous grease and liquid oil lubrication technology, coupling upgrades, low-leakage wiper rings, and piston rod packing and optimal compressor selection are a few that immediately come to mind.1 Duration of training and right size of groups. In two re-

cent instances, groups of four and six participants from petrochemical plants in Asia and the Middle East met, on separate occasions, for five days. One group brought along an interpreter, and the second group was multilingual. In both instances, the presenter had previously been made aware of each group’s specific areas of primary interest and had followed a sequence of topics to which all parties had agreed. Flash drives were handed out to the technology transfer participants on the first day of the meeting at a central major US location. They uploaded hundreds of PowerPoint slides to their personal laptops and supplemented these uploads with their comments and the presenter’s personalized answers to questions.

FIG. 1. A

small group of four advisors and mid-level managers attend five productive days of technology updates away from their overseas facility.

The end results were highly favorable. Suitably selected offsite venues remove attendee-participants from the unavoidable interruptions that exist in a plant environment. The direct interfacing between attendee and presenter allows for seamlessly catering to special plant needs. Desirable or explanatory add-on information is readily retrieved by the presenter for discussion the following day. The attendee-participants are encouraged to communicate and “network” with the presenter in the future; many follow up on this no-cost invitation. Feedback affirmed the value of “TTT”�technology transfer training�along the lines described here. Consider TTT as a valuable common-sense tool and examine its cost-to-benefit ratio. Its usefulness may pleasantly surprise you. LITERATURE CITED 1

Bloch, Heinz P., Petrochemical Machinery Insights, Elsevier Publishing Company, Oxford, UK and Cambridge, Massachusetts, 2016. HEINZ P. BLOCH’S professional career commenced

in 1962 and included long-term assignments as Exxon Chemical’s Regional Machinery Specialist for the US. He ha s authored or cowritten close to 700 publications, among them 20 comprehensive books on practical machinery management, failure analysis, failure avoidance, compressors, steam turbines, pumps, oil mist lubrication and practical lubrication for industry. Mr. Bloch holds BS and MS degrees (cum laude) in mechanical engineering. He is an ASME Life Fellow and was awarded lifetime registration as a Professional Engineer in New Jersey. Hydrocarbon Processing | OCTOBER 2017

19

ZymeFlow® tough. Refinery and petrochem process unit decontamination.

FASTER. SAFER. PREPARE PLANT EQUIPMENT FOR ENTRY IN

8-12 HOURS Shorten your turnaround using ZymeFlow® Decon. With 25 years of experience in refining and petrochemical applications, the ZymeFlow brand of specialty decontamination chemistries and services is the global leader for fast and safe vessel entry. The unique ZymeFlow process simultaneously eliminates benzene, LEL's, hydrogen sulfide, and pyrophoric iron sulfides, dramatically reducing the time spent to prepare vessels for entry and hot work. Furthermore, the process is safe for people, equipment, and wastewater plants. ZymeFlow Decon is active in over 50 countries, performing about 200 specialized decontamination projects each year.

Save Time. Increase Safety. Choose ZymeFlow®.

©2016. United Laboratories International, LLC. All Rights Reserved.

For more information, contact: Meghan Kidwell, Global Marketing Manager [email protected] +1.832.775.1548 | www.zymeflow.com


Automation Strategies

E. C. COSMAN, CONSULTANT ARC Advisory Group

Expanding scope of automation events and alerts Detecting, reporting and responding to abnormal situations or events are important functional elements of a complete process automation strategy. Traditional control systems present this information in the form of alarms directed to the operations staff on the control system console. In abnormal situations, the volume of these alarms can present a significant challenge. Process-related alarms are only one type of alert typically presented to the operations staff. For example, the growing importance of industrial control systems (ICS) cybersecurity means that security-related events must also be captured and managed. The ISA/IEC 62443 series of standards on industrial automation and control systems security explicitly state that security-related events and alerts must be collected and maintained for analysis. The standards implicitly assume that once this information is collected, someone will be available to interpret it and take the appropriate action. Automation solution providers and end users must focus on the nature of the information, how to express it and, most importantly, to whom it should be conveyed. End users must state clear requirements and expectations as to the types of information they need when describing anomalous behavior in their processes. Abnormal situations. Detection and response have been

topics of considerable interest in industrial automation for many years. Early work in this area included that of the Abnormal Situation Management (ASM) Consortium, which began with Honeywell’s Alarm Management Task Force to address alarm floods. More recently, international standards bodies have addressed the topic�most notably the ISA18 committee on instrument signals and alarms. Alarm management. Alarm management has been an impor-

tant element of industry response. The ISA18 committee was created to “…establish terminology and practices for alarm systems, including the definition, design, installation, operation, maintenance, modification and work processes recommended to effectively maintain an alarm system over time.” The original standard from this committee, ISA-18.1-1979 (R2004), Annunciator sequences and specifications, is intended primarily for use with electrical annunciators that call attention to abnormal process conditions using individual illuminated visual displays and audible devices. In 2009, the committee completed ANSI/ISA-18.2, Management of alarm systems in the process industries. Additional technical reports have been developed for several aspects of alarm management: • ISA-TR18.2.3-2015, Basic alarm design • ISA-TR18.2.4-2012, Enhanced and advanced alarm methods

• ISA-TR18.2.5-2012, Alarm system monitoring, assessment and auditing • ISA-TR18.2.6-2012, Alarm systems for batch and discrete processes. Additional sources of alerts. An abnormal situation may be

signaled by more than just traditional process-related alarms. The increased use of monitoring systems and other “smart” de vices results in a wide variety of alerts that may require a timely response. These additional sources include: • Cybersecurity: It is becoming more common to have some sort of network or security monitoring in place, with alerts generated for events such as unusual network traffic, authorization failures or unauthorized network traffic. • Network monitoring: With the increased sophistication of process networks, it may be necessary to detect and alert on unusual situations, such as node or switch failures. • Physical security: Monitoring devices, such as cameras or physical access sensors, may also generate alerts or requests for attention. • Business process alerts: Various information or workflow automation solutions may alert staff to unusual behavior. While a growing amount of information may be generated with an expectation of further follow-up for analysis and resolution, it is not always clear exactly who is accountable for this follow-up and associated response. In the absence of an explicit definition, certain assumptions can be made. For example, it is safe to assume that the operations staff wil l always be available to react to abnormal situations. Operators and plant engineers generally understand the implications and potential consequences of such events on the integrity of the system under control. However, they may not have the skills or experience to make the best decisions in response to these new sources of information. Unfortunately, it is much easier to configure devices and systems to generate alerts than to fully define the intended audience and expected actions. This has resulted in automation systems with far too many alarms for which the optimum response is not clear. Adding new sources of alerts, such as security and network management, will only exacerbate this situation. ERIC C. COSMAN has more than 35 yr

of experience in developing, delivering, managing and supporting operations information technology solutions in the process industries. During his career, his assignments and responsibilities have included process automation systems development, communications network design, functional and technical architecture design, and technology lifecycle management. He retired as an operations IT consulting engineer with Dow Chemical Co.


21

JUST SAY:

WHEN WHERE & HOW MANY. Solve. Respond. Deliver. It’s not just a slogan. It’s the bedrock we’ve built our business on since 1954 – delivering standard-setting cooling solutions for gas compression and processing, refineries, lube oil cooling and more. harscoAXC.com 918.619.8000 (USA)


Engineering Case Histories

A. SOFRONAS, CONSULTING ENGINEER http://mechanicalengineeringhelp.com

Case 98: Why performing calculations is important to an engineer s career ’

Engineering calculations can help an engineer’s career by controlling the risks taken. The author has followed this path. Most engineers realize that they can proceed through their careers doing very few engineering calculations. Managing projects or people; selecting and procuring equipment; working with manufacturers to design products; troubleshooting equipment, pressure vessels or structures; or using sophisticated computer programs may not require engineering calculations. These types of work can be performed by talented personnel capable of making sound decisions based on the data they obtain. However, by using engineering mathematical techniques, engineers can greatly enhance their abilities by generating needed data and not relying solely on others’ data. Early in my career, I realized that my managers did not want to know the details of how I arrived at my results. What my managers noticed was that I was willing to make a difficult decision to solve the problem expediently, safely and cost-effectively. Usually, it meant that the system design or equipment was operating successfully. My managers’ lack of interest in my calculations did not bother me, since the calculations allowed me to do the following: • Perform a logical analysis and provide documentation that supported my decision, making me feel comfortable with the risk taken in the decision. • Use historical failure data to analyze new designs and prevent the same type of failure. An analysis can show why smaller units do not fail but larger ones do. • Verify the analysis by using other actual failures to see if the results are the same. Cracks and failure zones show where to center an analysis. • Feel confident about reviewing any type of unique equipment because I could simplify it, analyze it and understand it. I could understand the internal stresses, dynamics and other phenomena that may not have been known to others. Using equations, I could model and understand what was happening inside an operating system. • Calculate if and when something was likely to fail. Analysis can be like time travel in that it allows a look into a failure before it occurs.1 Performing calculations allowed me to develop a strong track record of successful risk-taking and decision-making, and become known as someone who could help resolve difficult problems. This distinction helped me advance and be recognized by management. A few simple examples serve to illustrate the point; more are provided in the references.

TABLE 1. Margins

for the twisted shaft

Hp (D = 4 in., rpm = 200)

σ shear

σ shear σyield

/

Margin

800

20,000

0.57

43%

1,200

30,000

0.86

14%

Example 1: Why did the shaft twist? A 20-ft-long extruder

shaft made of 316 stainless steel had twisted along its length. In a meeting to discuss the failure, speculation arose as to the cause. I asked if the shaft was original equipment, to which the answer was yes. After further questioning, I learned that the motor had been increased from 800 hp to 1,200 hp several years earlier to handle a tougher product. I wrote the following on a flip chart in the meeting:2 • Yield strength (σ yield ) for 316 steel = 35,000 lb/in 2 • Shear stress in shaft due to operating loads (σ shear) = 321,000 × hp ÷ (D3 × rpm) • TABLE 1, which shows margins for the shaft. Notice that the shaft had some margin from twisting at 800 hp, but much less at 1,200 ( TABLE 1)�this is why it did not fail at 800 hp. Enough margin was available to handle periodic ad verse operating conditions, such as cold product startups. A detailed analysis was performed, and the shaft was replaced with a higher-yield-strength 17-4 material. This quick calculation changed the direction and results of the meeting. A repeat failure would likely have occurred if the spare 316 shaft had been installed. Example 2: Can we start up the unit? A wrong answer

based on someone else’s opinion and a subsequent failure can jeopardize a career. This is where the ability to perform calculations can be important. In this case, a large, heated, rotating disk dryer operating at a plant3 had developed cracks at welds. Several similar, smaller units had not developed cracks. The problem was compounded since a new, much larger, scaled-up unit had been shipped overseas and was awaiting installation at a new facility. The question from management was, “What should we do next?” A simple finite element model was used to determine the stresses at the weld, along with a cyclic crack growth analysis on the cracked dryer. After performing these calculations on the dryers, I felt confident in saying, “Based on my analysis, if the new unit is put into service it will crack and fail in less than 1 yr.” I recommended that the unit be sent back to the US from Asia and strengthened by replacing the first three of 40 disks with thicker ones to expedite the repair. The analysis indicated that only these replacements were required, enhanced by better welding methods. After reviewing the calculations, project Hydrocarbon Processing | OCTOBER 2017

23

Engineering Case Histories management and the dryer manufacturer that had intended to install the unit agreed on my solution. The recommendation was a low-risk one, since I had used the same analysis on the smaller units and knew why they had not cracked. Good data from calculations makes for a strong position. In this case, mathematical analysis predicted the future, and the crack growth calculation proved to be a powerful tool for preventing equipment failure.

The calculations produced new data for a meeting on the process unit. It allowed the engineer to generate data to be compared with actual data, meaning torque on other cutter assemblies. Equipped with this information, the team decided to increase to the higher torque values as a low-risk solution. These types of calculations are well within the capability of most engineers who look for the opportunity to utilize them. LITERATURE CITED

Example 3: Not tight enough. An investigation team was re-

1

Sofronas, A., Survival Techniques for the Practicing Engineer, John Wiley & Sons, Hoboken, New Jersey, 2016. 2 Sofronas, A., Analytical Troubleshooting of Process Machinery and Pressure Vessels, John Wiley & Sons, Hoboken, New Jersey, 2006. 3 Sofronas, A., Case Histories in Vibration Analysis and Metal Fatigue, John Wiley & Sons, Hoboken, New Jersey, 2012.

viewing a failure. One of 20 cutter knives failed and destroyed a piece of processing equipment. A review of the repair indicated that someone unfamiliar with the equipment had replaced the resharpened knives several weeks earlier. The two studs holding the knives were 0.75 in. in diameter (Din ) and 1 in. free length (Lin ). Studs like these require enough torque (Tft-lb ) to stretch (δin ) more than 0.002 in. Any less torque, and the vibration could relax the threads and em bedded material under the stud/cutter assembly,2 and the stud would become loose. The stud stretch can be approximated as δ in ≈ 240 Tft-lb × Lin ÷ (πDin3E), where E = 30 × 10 6 lb/in2. The torque measured on the remaining intact cutter knife studs was 110 ft-lb–230 ft-lb. for these high-alloy studs. The stretch for the lower value was δin ≈ 0.002 in.; none of the higher-torque values were found to be loose. This measurement is too low, and the cutter most likely loosened in service due to improper tightening, causing the failure.

NOTE

Case 97 was published in HP in July. For past cases, please visit

HydrocarbonProcessing.com. TONY SOFRONAS, D. Eng, was the worldwide lead

mechanical engineer for ExxonMobil Chemicals before retiring. He now owns Engineered Products, which provides consulting and engineering seminars on machinery and pressure vessels. Dr. Sofronas has authored several engineering books and numerous technical articles on analytical methods.

IT ALL STARTS WITH

API PSSAP.

™

In the current economic cycle, it’s critical you remain compliant and competitive. API’s Process Safety Site Assessment Program (PSSAP) ensures that your site’s process safety systems conform to industry standards. PSSAP is focused on evaluating the quality of written programs and the effectiveness of field implementation. For assurance of your site’s compliance, start with API at www.API.org/PSSAP. It’s times like these you need people like us.®

877.562.5187 (Toll-free U.S. & Canada) +1.202.682.8041 (Local & International) [email protected] | www.api.org © 2017 – American Petroleum Institute, all rights reserved. API, API PSSAP, the API logo and the “It’s Times Like These” slogan are trademarks or registered trademarks of API in the United States and/or other countries.

24

Select 152 at www.HydrocarbonProcessing.com/RS OCTOBER 2017 | HydrocarbonProcessing.com

Supply Chain

D. MCDOWELL, Honeywell Connected Plant, Edmonton, Alberta, Canada; and S. DEY, Honeywell Connected Plant, Bangalore, India

Integration: The next frontier in supply chain and production management Hydrocarbon processing industry (HPI) companies have long used different applications to address their supply chain production management (SCPM) challenges. However, the next frontier in ad vancing supply chain excellence lies in inte grating SCPM functions to achieve the best financial outcome. This article reviews the steps organizations can take to achieve better functionality integration, and examines the benefits of this approach. The challenges. It is widely acknowl-

edged that supply chain excellence is a competitive advantage for any company. However, supply, production and distri bution networks are becoming increasingly global and complex ( FIG. 1), as are the planning and operation of these networks. At the same time, risks in the supply chain have increased, and improved transparency is critical to managing them. Optimizing the performance of supply chains within an organization requires tight cross-functional coordination, decisiveness and, occasionally, trade-offs. Most companies in the process industries are commodities businesses. Millions of dollars are tied up in purchasing, sales decisions and inventory�an investment exposed to fluctuating commodity prices and exchange rates. Supply chain processes typically span different functional domains, from sales and marketing through finance and accounting, as illustrated in FIG. 2. Decision steps are incorporated with different time horizons and granularity (e.g., from strategic to operational). Today, these processes are typically carried out in functional silos within an organization, with little coordination between them. It is this coordination across silos, and the information flows between them, with which process companies are struggling. Each function has its own view of the supply chain�and usually their own proprietary tools�so it is imperative to unite

these differing perspectives into a single comprehensive view. This will highlight conflicts and opportunities, and enable them to be more quickly and easily addressed. Even with the adoption of best-inclass SCPM applications, achieving a holistic overview of business functions remains a key challenge for many organizations. The solution. To achieve cross-function-

al SCPM excellence, organizations must create an integrated supply chain. It is impractical to rip and replace existing investments in supply chain tools, so the path forward should provide a framework that includes and leverages existing solutions. Implementing this framework essentially consists of three steps: 1. Collection: Developing seamless data integration between different

FIG. 1. Supply

FIG. 2. Supply

supply chain applications and a common data repository. 2. Contextualization: Providing the intelligence to understand the differences in the level of abstraction and time granularity required by various applications and transforming data as necessary. 3. Analysis: Exploiting the common intelligent data repository to enable business processes, such as plan vs. actual tracking, supply chain forecasting, retrospective analysis, what-if scenarios, and further reporting and advanced analytics. The first two steps can be collectively referenced as “integration,” and the third step as “analytics.” In the following sections, the value that supply chain integration and analytics (FIG. 3) provide is examined.

chain business challenges are becoming increasingly global and complex.

chain processes typically span different functional domains. Hydrocarbon Processing | OCTOBER 2017

25

Supply Chain supply chain functions, data and application resource availability (components, To understand the value that industry tions. Based on Suncor Sarnia analysis, 1 blender, product tank) must be checked, will get from a solution to this problem, closing the gap between “plan vs. actual” and the cost for the batch must be comthe performance areas in the context of in near-real time can result in up to an ad- puted. Each of these steps takes informathe supply chain operations reference ditional $0.50/bbl benefit over a medium- tion and communications (and time). (SCOR)a framework, shown in TABLE 1, term basis. Integrating the cross-functional information referenced above in a single consoli will be examined. This widely-used frame work assesses the performance of a supply Responsiveness. Responsiveness is the dated view can significantly decrease (by chain based on reliability, responsiveness, speed at which products are provided to 10%–30%) the time and effort taken to agility, cost and return on assets. customers. A typical scenario that tests a re- confirm the order, increasing customer finery’s responsiveness is the opportunity responsiveness and reducing overall order Reliability. An integrated supply chain, for a spot sale of product due to a competi- fulfillment cycle time. with functions sharing a clear and common tor’s unplanned refinery outage or logistics Integrated information and analytics understanding of production and inven- issue. This begins with a customer request, tools help supply chains collaboratively tory forecasts, allows for better financial which initiates a sequence of actions: con- make decisions to purchase and sell the tracking and for potential delivery issues firmation that the quantity of product is right grades, at the right price and at the to be more quickly spotted and resolved. available in inventory; a check of planned right time. This provides a significant comThis maintains customer satisfaction while shipments to verify that this inventory is petitive advantage: for a 200-Mbpd refineliminating unnecessary costs (e.g., penal- available for commitment and meets the ery, a reasonable assumption of one adties, demurrage, makeup imports). customer’s quality requirements; ensuring ditional opportunity cargo of 200 Mbbl/ Supply chain reliability is further im- that shipping capacity is free; and a poten- month at an incremental $0.50/bbl preproved by the ease and frequency with tial review of credit risk of the customer. mium would generate $1.2 MM/yr of in which a plan vs. actual analysis can be If the finished product must be made creased profit (or $0.02/bbl of crude run). conducted, due to the tight integration of to order (e.g., blended), then the producAgility. Supply chain agility means that any opportunities or disturbances that TABLE 1. SCOR framework arise in the market can be acted upon to Attribute Objective maximum advantage. An integrated supply Customer Supply chain reliability Consistently getting the orders right, chain system provides a common view for: product meets quality requirements • The supply chain time horizon, from long-term commitments Supply chain responsiveness The speed of providing products to customers through plan periods and scheduled Supply chain agility The ability to respond to changes in the events into the future market (external influences) • Visibility into crude and product Internal Supply chain costs Optimize costs associated with managing movements, inventories and operating the supply chain and forecasts Supply chain The effectiveness in managing the supply chain’s • Changing market conditions: asset management assets in support of fulfillment, measured by return on assets and return on working capital supply and demand, pricing trends • A common understanding of price incentive data across functions • The basis to perform a “What if?” exercise across functions to efficiently project the impact of alternatives being considered. Significant upstream market influences include: declining production and the decreasing quality of existing conventional crudes; the emergence of new crudes, conventional and synthetic; increasing pipeline flexibility; logistics rate and in ventory constraints; and resulting price differentials. All of these influences push refiners to consider increased amounts of opportunity crudes to maximize margin. At the same time, downstream influences, including the alteration of distribution patterns, ever-increasing product quality requirements, and volatile demands and FIG. 3. A conceptual view of SCPM integration and analytics. pricing, must be addressed. VALUE TO INDUSTRY

26


Supply Chain Consolidated market intelligence and forecasting significantly benefit agile operation in this scenario, as shown in FIG. 4. They also demand a planning model that accurately and robustly models refinery processing�real capacities, yields, parameters and, particularly, constraints� in the required operating environment. Rigorous LP model validation by analytical processes (e.g., technical back-casting) requires intensive data and preparation, and an integrated, practical information and analytics infrastructure. Without this validation process, a gap between the model predictions and operation capabilities frequently appears, with serious economic consequences for a refinery: running against incorrect constraints; overspending on crude, other feedstock or blendstocks; and producing excess or insufficient product (with resulting market penalties). Based on publicly available studies, a systematic process of updating model yields can result in a margin improvement of $0.10/bbl–$0.50/bbl, or more. 2

terials to flow back into the company af- reliability have been proposed, including ter finished goods have been delivered to a 1996 study 3 that noted record low levcustomers. A key contributor in refining els and suggested that even lower levels is work-in-progress, or inventory of crude might be optimal. A 2014 study 4 highand other feedstock, intermediates and lighted contributing factors and benefits finished product awaiting shipment. of inventory variance reduction. HowCrude price volatility and supply ever, many refiners find it difficult to concerns have, over time, changed tar- maintain target inventory levels due to get inventory strategies and numbers. high variability�inventory is allowed to Methodologies to optimize inventory float to absorb production/shipping imlevels based on margins and supply chain balances, so it differs greatly from target

Get accurate flare gas heat values fast, with the FlarePro™ Mass Spectrometer

Supply chain costs. Optimizing the

cost of the goods sold and the cost of operating the supply chain requires a process that evaluates performance to identify missed opportunities and improve planning and execution processes. This will utilize information on the projected and planned values and actual results for all factors that determine product costs. Analyzing this large quantity of data in a time-effective and rigorous manner requires a well-structured process. One analytical tool that has been used successfully is economic back-casting, in which the planning model for a previous period is run multiple times in sequence, while successively fixing prices, material inputs, process capacities and limits, and product qualities to the actual values for the period. The differences between planned and actual results (and objective function) for each step indicate the sensitivity to various factors; factors with a large impact should be analyzed to prevent reoccurrence of the issue. Benefits are highly variable, but a reduction of up to 1% of gross margin ($0.08/bbl) appears achievable. Supply chain asset management.

One of the most powerful supply-chain metrics is the cash-to-cash cycle time, or the time it takes for cash invested in ma-

With impending EPA rule changes for flare gas stack combustion, you’ll need to measure heat value at the stack quickly, accurately, and often. AMETEK Dycor®’s FlarePro mass spectrometer has the speed and specificity for the task, calculating make-up gas quantities 10X faster than chromatographs. It can speciate and quantify alkanes and alkenes up to C7 and convert ASTM calculations of concentration to yield flare gas BTU values, even with wildly changing flare gas streams. Features include: • 24” footprint (general or C1, D2 areas) • Low maintenance • 16-valve inlet system • Modbus/OPC communications • 1 to 100 amu mass range • Environmentally controlled to • Easy-to-service modular design accommodate a wide range of temperatures Learn more at www.ametekpi.com.

© 2017 by AMETEK Inc. All rights reserved.


27

Supply Chain TABLE 2. Supply

chain integration and analytics benefits Supply chain responsiveness

Supply chain agility

Supply chain costs

Supply chain asset management

Consistently getting orders right, product meets quality requirements

Speed of providing product to the customer

Ability to respond to changes in the market (external influences)

Costs associated with operating the supply chain

Maximizing return on fixed assets and working capital

Early warning indication of delivery issues

Improvements of cross-functional inputs to drive faster and more effective decisions

“What-if” scenarios to evaluate impact of opportunities

Economic back-casting to identify performance gaps/areas of improvement

Integrated Inventory forecast drives focus on inventory reduction strategies

Reduced operating expense and cost of product (increased margin)

Reduced working capital, one-day inventory reduced by reducing variation

< $0.08/bbl

$0.01/bbl

Attribute

Supply chain reliability

Definition

Supply chain integration and analytics

Timely, accurate plan vs. actual tracking

Quantified benefits

Technical back-casting identifies correct planning yields to deliver accurate plans

Reduced gap between plan and actual

Reduced overall order fulfillment cycle time

Improved opportunity capture (200 Mbbl/mos at $0.5/bbl)

Reduced demurrage and penalties

10%–30% cycle time improvement

Improved yield prediction and utilization of capacity and flexibility

< $0.50/bbl

$0.02/bbl

$0.1/bbl–$0.5/bbl

The petroleum refining industry has been acknowledged as a pacesetter in adopting technology to improve productivity and efficiency. Examples include the standardization of linear programming, the use of advanced scheduling algorithms and the implementation of statistical data reconciliation. The adoption of a consistent methodology for supply chain integration and analytics represents the next frontier to be conquered. NOTE

FIG. 4. A consolidated forecast view.

with extensive experience and knowllevels to provide a safety stock cushion. To achieve sustained cost reduction, edge of refinery operations. Integration inventory must be proactively managed. significantly reduces the effort spent on An integrated supply chain system offers data collection, collation and validation, significant improvement by consolidating and analytic tools can automate the proinformation from all activities impact- duction of complex reports. This allows ing inventory. It also provides different planners to spend more time on highfunctions with a common global view of value analysis and performance improveinventory forecasting�a uniform basis ment tasks. for improved control of inventory days Additionally, the adoption of such a of supply. Even with low crude prices solution brings an organization closer to (~$50/bbl), margins (~$8/bbl.) and the realization of the Industrial Internet interest rates (~4%), reducing average of Things (IIOT). A well-defined and crude and product inventory for a 200- managed integration can help to leverage Mbpd refinery by one day yields project- IIOT developments and apply them to ed returns of $864,000/yr (~$0.01/bbl). the supply chain. One of the smaller quantified benefits (with potentially the highest potential) Takeaway. A summary of the benefits of supply chain integration and analytics from supply chain integration and analytis the improved usage of personnel assets. ics against each of the SCOR attributes is Planning personnel are often resources shown in TABLE 2. 28


a

Supply Chain Operations Reference (SCOR®) is a cross-industry standard for supply chain management, developed and maintained by the Supply Chain Council, an independent, not-for-profit trade association. LITERATURE CITED

Complete literature cited available at HydrocarbonProcessing.com. DAVID MCDOWELL is Principal Consultant

for Honeywell Connected Plant in Edmonton, Alberta, Canada. His 43 yr of diverse experience in the process industry domain include: refining, petrochemical and mining operations; manufacturing execution and control systems; industry consulting; business process improvement; and large project leadership and execution. He holds BS and MS degrees in chemical engineering from the University of Alberta, Edmonton, Canada. SATRUJEET DEY is Product Marketing Manager

for Honeywell Connected Plant in Bangalore, India. Mr. Dey has more than 17 yr of diverse experience in the process industry domain, including refinery operations, software development, industry consulting and product management. He earned a BS degree in instrumentation engineering from the University of Mumbai, and an MBA degree from the Indian Institute of Management in Bangalore.

Petrochemicals

ALLIED MARKET RESEARCH Pune, Maharashtra, India

A global look at the chlorinated paraffins market With the rise in the polyvinyl chloride (PVC) industry, the chlorinated paraffins market has also been growing since it acts as secondary plasticizers in PVC. In addition, technological advancements made in the PVC market have propelled the growth of the chlorinated paraffins market. According to Allied Market Research’s “Chlorinated paraffin market�Global opportunity analysis and industry forecast, 2014–2022,” the global chlorinated paraffins market is expected to grow more than 27% from 2015 to 2022. The global chlorinated paraffins market size was $1.9 B in 2015, and is expected to reach more than $2.4 B by 2022. This increase in market size equates to a compound annual growth rate of 3.5% during the forecast period. Chlorinated paraffins are produced by the synthesis of chlorine gas unbranched paraffin fractions at a temperature range of approximately 80°C–100°C. They are colorless or yellowish , viscous, dense oils, except for long-carbon chain chlorinated paraffins with high chlorine content (approximately 70%), which are solid. Chlorinated paraffins are used as secondary plasticizers for PVC, and are expected to partially replace primary plasticizers, such as phthalates and phosphate esters, in the near future. Chlorinated paraffins offer advantages such as flame retardancy and low-temperature strength, as well as increases the flexibility of materials. Chlorinated paraffins are used as extreme pressure additives as metal-working lubricants or cutting oils. This is attributed to their compatibility with oils, viscous nature and property of emancipating hydrochloric acid at high temperatures. They improve the resistance to water and chemicals, which is most suitable when they are used in marine paints, as coatings for industrial flooring, vessels and swimming pools. In addition, technological advances coupled with escalating demand for products with PVC content, such as sheets, tubes, pipes, wires, and cablese, is expected to drive the demand for chlorinated paraffins during the forecast period. The global chlorinated paraffins market growth is characterized by vivid opportunities to market players, owing to disposable incomes and rapid growth in the global economy. The global chlorinated paraffins market is segmented based on product type, application and geography. Based on product type, the market is classified into short-chain, medium-chain and long-chain. Based on chlorine content, each product type is further segmented into less than 40%, 40%–70%, and more than 70%. By application, the chlorinated paraffins market is segmented into the following categories: • Lubricating additives • Plastics • Rubbers

h g i H h t w o r g t e k r a M

Paints

Adhesives

Lubricating addtives

Metal working Rulds

Rubber

Plastics

w o L

Low

Metal working fluids and lubricating additives segments are projected to be the top inventment pocket due to high-rising demand in industrial applications in Asia-Pacific

Market attractiveness

High

FIG. 1. Metal

working fluids is the top investment pocket in the chlorinated paraffin market. Source: Allied Market Research.

• Paints • Metal working fluids • Adhesives. In 2015, the lubricating additives and metal working fluids segments accounted for the major share in the market. In terms of value, these applications held approximately 54% market share in the global chlorinated parrafins market. Metal working fluids is the top investment pocket in the chlorinated paraffins market (FIG. 1). The reason for this is attributed to the increase in demand for extreme pressure additives in metalworking fluids, which allows the machining of difficult-tomachine materials. Moreover, the strong barrier to corrosion and thermal resistance property is expected to propel the stakeholders to capitalize on the underlying opportunities in this industry. Metal working fluids are widely used in industries, such as automobile, precision engineering and machinery construction. Thus, owing to the high growth rate and the significant revenue of these industries, the segment is anticipated to provide high stability, along with a significant return on investment for stakeholders. It has been estimated that consumption of the chlorinated paraffins market would increase in the Asia-Pacific region. The rise in consumption is due to a rise in population, development in the automotive and machine industries and stable economic growth in the region. The developing economies within this geographical boundary, especially India and China, will witness a tremendous growth in demand of the industry, along with a steady rise in demand of lubricating additives. The demand for long-chain chlorinated paraffins are expected to increase in North America due to a restriction in the use of short-chain chlorinated paraffins in certain applications. Moreover, the production cost of the chlorinated paraffin is low, making it a more affordable chemical in comparison to other chemicals. A full summary of the report can be found at Allied Market Research’s website. Hydrocarbon Processing | OCTOBER 2017

28–A

SAVE THE DATE WHEN

March 21–22, 2018

Gas Processing and World Oil are bringing the Eastern Mediterranean Gas Conference (EMGC) back to Nicosia, Cyprus. Now in its fifth year, EMGC is the region’s premier conference for companies operating in this rapidly evolving energy hub. The two-day conference will center around a high-level technical program featuring speakers from leading operating and service companies. Join the conversation and network with your peers at EMGC!

WHERE

Nicosia, Cyprus WEBSITE

EMGasConference.com

Sponsorship opportunities are available! For more information, please contact: Hortensia “Tish” Barroso, Business Development Manager at [email protected]

Organized by:

Roundtable

G. CLINE, Aberdeen Group, Boston, Massachusetts; K. KEIM, Seabrook, Texas; M. NEILL, Petrotechnics, Houston, Texas and J. THOMAS, Process Improvement Institute, Houston, Texas

The reality of process safety risk Unavoidably, process safety risks are such as startup, shutdown and other infreoften managed in different parts of an or- quent tasks. In some cases, companies in ganization. Consolidating these risks to countries without regulations have imple view their impact on the operational reali- mented excellent process safety managety of hydrocarbon assets is a real challenge. ment (PSM) programs; adding regula What the industry needs is to make tions may not always be the answer. GC: Incidents and accidents depend sure everyone assesses risk using the same on many things, including the regulacriteria, and has a practical understanding of how their decisions directly or indirect- tory environment and the overall level of ly influence risk and, ultimately, process safety awareness. Often, it is just human safety performance. By making process nature. People try to prepare and create a safety more operational�ensuring front- culture of safety, but slip-ups happen. JT: Also, people do not always unline personnel are aware of their roles and responsibilities, and are effectively and derstand all of the hazards or safeguards. consistently implementing processes and They get used to doing things a certain procedures�we can reduce incidents and way, and if nothing has happened, they improve sustainable production. feel it is okay to continue, even if it is not What is the reality of risk in the hydro- the safest thing to do. In addition, we do carbon sector? In this roundtable, senior not often identify all the hazards, espeindustry executives discuss what happens cially those related to infrequent modes of when process safety intent meets the re- operation�like startup and shutdown� ality of operations. The roundtable par- where a majority of incidents occur. Huticipants include Greg Cline (GC) of the man factors are not generally evaluated Aberdeen Group, Kelly Keim (KK) of and included in most PSM systems, so we ExxonMobil (retired), Mike Neill (MN) often “set the operators up to make errors.” KK: It is important to note, accident of Petrotechics and Jeff Thomas (JT) of the Process Improvement Institute. This rates for process safety incidents across roundtable will discuss how we think we the refining and petrochemical industries manage risk, how we actually manage it, are actually incredibly low. MN: I would say that most people in and how we can improve it practically and tangibly. the industry think, “I could almost guarantee we will have an accident,” rather than, “I can guarantee that we won’t.” But Industry regulation is at an all-time they do not know when, and they do not high. Every operator is committed know how big. The chances are that if you to safety and risk avoidance. Why are a big organization with a lot of operado you think incidents tions, you pretty much know that someand accidents still happen? JT: There are several reasons why acci- thing will happen eventually. KK: The good news is that the Ameridents still happen. First, not all countries have process safety regulations. Second, can Petroleum Institute (API), American even where good, detailed regulations Fuel and Petrochemical Manufacturers exist, it is hard to implement all of the (AFPM) and the American Chemistry processes and procedures they require Council (ACC) collect information on 100% correctly, all of the time. There are causes and causal factors on a consistent often conflicting priorities, particularly in basis. They are beginning to get a much the field, between safety, production and clearer picture of process safety related iscost. In addition, there are often no thor- sues. Traditionally, the industry looked at ough operating and maintenance proce- facility causes�equipment failure, corrodures that cover all modes of operations, sion, etc. Those risks are still big factors,

but the greatest proportion of incidents, based on industry evidence, is related to human performance, which is people failing to execute a procedure properly or missing an operating step. MN: There is a lot of focus on humans as the weak link in the chain. That is obviously part of it, but as much as we blame individuals when things go wrong, we need to credit them for reacting and recovering from problems. Where poor decisions are seen as the root cause of incidents, we need to examine whether competence was lacking, or if people just did not have the correct information with which to make a decision. KK: We are only just learning how to classify�let alone improve�human performance. There is still a lot to do. The industry does seek to get better by making reference to the nuclear and airline industries as shining examples. Those industries addressed things like equipment design, maintenance, management systems and people simultaneously, without preference. For example, in the oil and gas industry, the Mexico City and Bhopal disasters sparked PSM regulations in the US, focusing on systems and then eventually people. I think the nuclear and airline industries have been far more successful vs. the oil and gas industry’s phased approach. Is there a gap between what process safety key performance indicators (KPIs) and operational management systems are telling us, and the feeling on the front line? MN: I think there is. I have heard, an-

ecdotally, from operators that the KPIs say one thing and the reality at the plant is another. JT: A lot of people are still trying to figure out the process safety indicators they should focus on. We have had API standards in place for less than 10 yr. There is also a communications gap between field and office personnel, engineers and Hydrocarbon Processing | OCTOBER 2017

29

Roundtable

experienced one. There was a belief that these could happen, and people paid attention to avoid them. There was maybe JT: I would say most companies prob- a negative that people felt responsible for ably recognize their process safety per- putting their own lives at risk to minimize those events. Thankfully, we have almost eliminated this ‘cowboy approach.’ The indusExecutives must make sure the right people are involved try has the newest and rawest in the right processes and that they do the right process safety data. We have rethings. However, I would say operations are in control ally only been managing it for approximately 5 yr. With more of the plan. They are at the sharp end, so they should be time and data, we will be able to satisfied personally that the risk level is acceptable. say whether we are better than we think.

management who set up process safety indicators and processes. Generally, these indicators are not clearly communicated at an operator level in terms of what they are and their importance. I am not sure actions

Does the reality of risk management measure up to the intent of risk management?

are taken as a result of the process safety- formance is not where they want it to be. Do you think the relationship related data and the KPIs produced. One But we are doing a better job of under- between PSM and operational risk important KPI mentioned in the Center standing risk than we did when I started, management is close enough? GC: No! I think PSM is always aspifor Chemical Process Safety (CCPS) book 30 yr to 40 yr ago. MN: People are experienced enough rational, and the relationship between on incident investigation is near-miss data. It is critical to report both incidents to know that hazardous industries mean process safety and its impact on front-line and near-misses, and periodically analyze risky business. I do not think people operations can be better understood. JT: There are gaps in most cases. them to determine causal factors and root would publicly admit that risk is unprecauses to prevent future incidents. dictable. However, other industries�nu- There has been a lot of work focused on KK: I do not think we do a great job clear and airline�have managed to elim- developing PSM systems, improving risk on KPIs. I know very few sites that make inate some sources of unpredictable risk. related practices and developing PSM a big deal of reporting their process safety These sectors put a lot of emphasis on tools. However, there is often a “disconperformance to operations. They also do training, stop-work authority and redun- nect” between what the practices and not publicize their safety-critical equip- dancies in design so that if a system fails, processes intend and what actually hapment performance and inspections. If op- there is another that would take over. In pens at the grassroots operator level. Lots erations are not aware, then performance the process industries, we have become of companies are working on it, but I do starts to slip. somewhat normalized to risk, and we do not know of any that have a magic bullet. GC: There is always a gap, and there not come anywhere close to investing the MN: Process safety is in a different part should not be. We need to put capabili- same level of risk management resources. of the organization, so operations personties in place to minimize gaps and ensure There is a lot to gain from investing in nel struggle to understand some of the lanthat metrics are available enterprise- safety. Typically, with safety comes im- guage and how to apply it to their reality. wide. Also, it is important that peoples’ proved operational performance. Process safety people sit in a world of sceKK: I do think there is an undue con- narios and models in which it is easy to diperception of certain metrics match the reality of operations. fidence at both the executive and field verge from reality. It is a bit like your house MN: Major accidents are low-fre- levels that “those things just don’t happen being about to fall because it has wood terquency but high-consequence. If some- to us.” There is not that everyday sense of mites, but I am spending all my time paintthing happens, you cannot really make a caution that should be present in people ing it! I focus all my efforts on a process judgement on whether there is a trend, or who are one procedure away from a ma- for painting. It is a false sense of security. whether you are particularly vulnerable. jor catastrophic event. Most plant work- Operators need to know how to practically Some people try and extrapolate near- ers and managers have never experienced apply process safety in the plant. KK: Operators do not get a good misses and look at other performance a major process safety event, so they beindicators, but a lot of KPIs are based lieve it will not happen to them. We know picture of how change affects risk manon how well an organization implements that is not true. agement, or the aspects of the job where GC: Real safety happens on the they are the critical factor in managing safety processes. KK: Evaluating risk is always somewhat ground when people internalize it and do risk. Often, when investigating the failure subjective. For the most part, companies not view it as a burden on everyday busi- of an asset, the question to operations is have not been terribly transparent in the in- ness. That means risk exposure must be typically: Why weren’t you paying better formation they use for monitoring process made visible, prominent and available so attention? The challenge back: Pay better safety risk. Most people can point to their everyone can understand its impact on attention to what, and how? MN: Process safety designs safeguards. numbers for personnel injuries and behav- the operational reality. KK: When I first started in the indus- It does not really look at how risk is manioral safety observations, but catastrophic events are rare, so they are not front-of- try 40 yr ago, fires and explosions were aged in real time. Process safety teams are mind, even if the risk is always there. relatively common. Most workers had also not a strong voice in the organization. 30


Roundtable They do not have a significant budget, and are always vying for priority with plenty of other groups in the organization. KK: Our risk models rely on the operator for 99% execution. We do not often explain where operations teams really need to be at the top of their game. We do not explain that when facilities change, they are potentially operating in a higher-risk environment. The US Chemical Safety Board (CSB) report on the explosion of the electrostatic precipitator in the Torrance refinery pointed out that as operations became focused on the tasks required to complete the shutdown, they became unaware that the situation continued to change. They did not know the importance of the key process safety barriers they controlled. JT: It takes a lot of hard work and communication between the engineer, the management and the operators on what risk management is all about.

proach across business units to make it happen is another question. What critical process safety information do people who make the daily decisions about operating a plant need? GC: When we talk about making daily decisions, operational data must correlate with the management of process safety and vis-à-vis. Management needs

to analyze the plant and the processes that relate to PSM. Then, this needs to be incorporated into operational dash boards in an actionable way. MN: Operators need data that clearly shows if something unexpected is happening, what the impact could be, how that affects the program of work, the threats it creates and the effects of any remedial reaction. Their number-one priority is containment. They need data on the integrity

Who is responsible for managing risk? JT: From the CEO to an operator, me-

chanic, engineer, supervisorall levels of management and workers. Everyone has a key and different role to play, but risk management should permeate throughout the organization. KK: We are a long way from being able to take the operator out of risk management, particularly in refining and petrochemicals. Management is responsible for having systems in place to make operators aware of changing risk patterns. Ultimately, executives must recognize that this is part of managing process safety risk. MN: Ultimately, it lands at the top of the tree. Executives must make sure the right people are involved in the right processes and that they do the right things. However, I would say operations are in control of the plan. They are at the sharp end, so they should be satisfied personally that the risk level is acceptable. That said, where there are multiple levels of decision-making, it can be confusing when it comes to who owns risk. GC: In our most recent environmental, health and safety study, about one third of respondents have a formal risk management organization in place. That is presumably how they establish a frame work for risk management. Does it build a risk awareness culture across the organization? It can. Whether those companies also have the necessary collaborative apSelect 154 at www.HydrocarbonProcessing.com/RS

31

Roundtable of pipes and vessels and the condition of risk is of the tasks they are completing. the actual detection systems themselves. One of the key issues is that there JT: People need a lot of information is so much data; it is hard to figure out to make decisions. KPIs are needed at what is meaningful. You need to clearly the management level to help make deci- identify that type of information. The sions about operations, resources and pri- importance and the timing of activities orities. At the engineering level, they need are key, so operators can determine what inspection and test data to help determine is urgent and what can wait. You need a frequencies of maintenance and repairs. whole picture of ri sk based on data so deThe operator needs data to understand cisions are not isolated from everything the current state of a process and what the else that is going on in a facility.

AdvaSulf

How well-informed are front-line leaders and workers about the role of process safety barriers in preventing incidents?

TM

HySWEET® COSWEET

TM

SweetSulf

TM

Sprex® AdvAmine

TM

A UNIQUE TASTE OF SWEET FOR YOUR GAS 50 YEARS OF EXPERIENCE IN GAS SWEETENING AND SULPHUR RECOVERY PROCESSES

With its unique and complete proprietary technologies portfolio, PROSERNAT offers optimized solutions to bring on specs any type of gas contaminated with CO2, H2S, COS and organic sulphur species, while producing sulphur with the most stringent emissions standards. www.prosernat.com

32

KK: That consolidation of information is certainly vital to more rational decisiondecisionmaking. The trouble is we do not provide consolidated systems for operations to effectively assess if they can take one more step in their procedure. For example, the Deepwater Horizon event needed approximately 11 layers of protection in place to prevent the scenario that happened. One by one, those layers of protection were whittled away. The response was always, “Well that is okay because we have got this other ultimate layer of protection.” It shows that even a plant with multiple protection layers can experience a ma jor hazard because of an accumulation of relatively harmless decisions. The existing process safety barrier status must be visi ble to operations, but but also to managemen managementt so appropriate decisions can be made. MN: Ultimately, operators need data that shows whether it is safe to operate the plant.

Select 153 at www.Hydr www.HydrocarbonProcessing.com/RS ocarbonProcessing.com/RS

KK: I would say they are only barely aware of the layers of protection. In many cases, operationseven first-line engineersare not aware of the scenarios that could lead to a catastrophic event in their unit. The scenarios have never really been collated in a useful way for them. I think there is a general failure to really communicate, on a shift-by-shift basis, the status of key barriers on any given day. For example, I spoke to a team recently where there was something wrong with a detection device for a piece of safety-critical equipment. The company said that the operator is going to pay more attention, but no body translated translated that into what that meant for the operator, and how they would do it. That is the most important thing when operations are making daily decisions. JT: It varies by facility. To be frank, some do not have a clue, but some are doing a pretty good job. There is an opportunity for improvement to ensure operators, maintenance maintena nce technicians and the front line really understand key hazards, safeguards and the ideal state of the process safety barriers. That is critical. I have not seen a lot of facilities where they really have a good handle on the barriers and how they interrelate to prevent an incident.

Roundtable

GC: I think building a culture with the

ing and subsea, individuals need multiple right tools, right attitudes and right train- viewpoints. That is the source of informed informed ing can enhance the awareness of process decision-making: using technology to put safety barriers by making them part of the everyone in a much better position. standard operating procedures of frontHow can operators maintain their line leaders and workers. MN: I think that there is still a lack of safety and risk management information available. The further down standards over time? KK: For the most part, operators do the chain you go, the more abstract some of that information is. I am not sure people not get feedback on their risk levels, let really understand risk and what it means alone their risk management performance. to them. That can put them in a vulner- Even companies that are doing a good job able position to be exposed to risk they of tracking tier-three and tier-four process do not understand. If they did understand safety indicators are basing performance it, I think some of their decisions might on lagging data. They are certainly not be different. I think that is the industry’ industry’ss communicating this to operations. If you challenge. We need to give the front line do not get good feedback, you cannot the ability to be better informed about the improve. I am not aware of anyone using possible consequences of their actions, any process safety solution, other than the even when making minor decisions. most basic tools, such as work permit systems, to manage it. JT: I think there is merit to having a What are the obstacles to access this information in a timely manner, tool that shows an overall picture of hazards, operational risk, barriers and safeand how can they be eliminated? JT: There are a few. First, we have so guards�updated on a real-time basis. much data, particularly with things like Constant communication with operations digital process control systems (DCS), is key so that they know the impact of any safety instrumented systems (SIS), main- change�for example, management of tenance systems, etc. We get information change (MoC), and how best to adjust. GC: Safety and risk factors change overload, and it is not always clear what is most important. Second, there can be all the time, so best practices must be rea lag in the data. We do not always get sponsive to changing conditions. Creative it when we need it, and things can be solutions can help organizations maintain missed. Third, maintenance management and improve their safety performance and DCSs do not always make it easy to over time. MN: Safety standards define our risk extract data. And that is just the start! KK: The information is there, but it is tolerance, and risk tolerance is not an exact often in lots of different systems�some science. It is an interpretation of risk and of which may still be paper-based. Even whether certain outcomes are acceptable. acceptable. for a process safety engineer who has Managers would love to have a physical detell them they been on the site forever, it will take time vice with traffic signals that tell to pull all of that data together. If it is not need to do something or prioritize differconsolidated and condensed in useful ently; we all would. But it is more about forms, nobody actually uses it for making being sure that systems are effective. It is critical risk-based decisions. about an attitude of constant vigilance and GC: The Industrial Internet of Things questioning�giving people confidence in (IIoT) is enabling a new era where we each other and their data, and empowerhave the capability to monitor and im- ing them with systems they can rely on. prove processes to ensure that they are GREG CLINE is a Principal safe. Safety must be implicit. I think, to Market Analyst for the the extent that operators can connect Aberdeen Group. As the head of research for operations with the information needed, Aberdeen Group’s Product via IIo IIoT T or another framework, they can Innovation and overcome risk and help prevent incidents. Engineering (PIE) and MN: We need to connect the data we Manufacturing research practices, he covers topics have. We also need ways of assessing the related to developme development nt impact of doing something or�equally and manufacturing of important�NOT doing something. From products, ranging from maintenance and asset integrity to drill- new product development to embedded systems.

Previously, Greg spent 1 7 yr as a market intelligence Previously, manager and strategic product analyst at Intel Corporation, where he provided market insight to senior executives executives and managers. In addition, he spent 12 yr at globally respected market research firms, such as Yankee Group, IDC Government and In-Stat. Mr. Cline has an MS degree in computer and information systems from Dartmouth College. KELLY KEIM retired as

Chief Process Safety Engineer from ExxonMobil Research and Engineering after more than 33 yr. After 15 yr in several levels of management in operations and maintenance, Mr. Keim found his passion in process safety. safety. In his final years, Mr. Keim had the lead role in revamping ExxonMobil’s tools and methods for assessing and managing the risk of operating hazards with the highest potential consequences, such as BLEVE, toxic releases and vapor cloud explosions. Mr. Keim was part of a team receiving the Harry M. West Service Award for contributions made to the process safety center, as well as the Trevor Trevor Kletz Merit Award for contributions to the field of process safety. safety. MIKE NEILL is the

President of Petrotechnics USA. With more than 35 yr of experience, Mr. Neill has helped to improve safety and performance management for companies in hazardous industries around the world. Prior to joining Petrotech Petrotechnics, nics, he held roles in operations, drilling and petroleum engineering for BP in Scotland, Norway, the South of England and Egypt. Mr. Neill holds a BSc degree in mechanical engineering and an MSc in petroleum engineering from Imperial College of Science and Technology at the University of London, and an MBA degree in strategic management from the Peter F. Ducker Graduate Management Centre, Claremont Graduate School in California. He is an active member of the CCPS, AIChE, ASSE, GPA, and the Mary Kay O’Connor Process Safety Center. JEFF THOMAS is a

Senior Process Safety and Reliability Engineer for the Process Improvement Institute. He has more than 40 yr of experience in the upstream oil and gas industry, including positions in process/ facilitiess engineering, facilitie production operations and PSM. He has a BS degree in chemical engineering from Ohio State University, and is a registered Professional Engineer in the state of Texas. Mr. Thomas spent numerous years in roles as a process engineer and operations support advisor for gas plants, offshore platforms and other upstream operating facilities. He has significant experience in process safety, where he was the Leading Global Technical Authority for ExxonMobil’s production company. Hydrocarbon Processing | OCTOBER 2017

33

REAL-TIME LNG DATA Any project. Any time. Any place. The Energy Web Atlas is a comprehensive dataset covering the global LNG market. Updated in real-time, the web-based platform enables users to access worldwide facility and project data.

Real-time Data

Comprehensive Comprehen sive Functionality

The platform’s mapping, search and query tools allow users to interrogate real-time facility and project data and find the information needed for major decisions. Data points include:

Built for ease of use, the Energy Web Atlas facilitates access to the most relevant data available, with various functionality including:

• Owner/Operator

• LNG shipping/vessel information

• Custom annotation, reporting and graphing

• Shareholder data

• Interactive data filtering

• Process method.

• Layer/attribute access and modification

• Project status • Project storage and capacity

• Production quality maps, PDF exports • Interactive I nteractive charts/statistics. charts/statistics.

Questions and More Info

Additional Maps/Data US gas processing plant data is also available.

• Data exports

AMERICAS: J'Nette

Davis-Nichols, +1 (713) 520 4426 or [email protected] REST OF WORLD: Alastair Noakes, +44 (0) 203

409 2244

or [email protected]

D I S C O V E R M O R E A T E N E R G Y W E B AT AT L A S . C O M

Viewpoint

PETER MARTIN, VICE PRESIDENT Schneider Electric

Finding value in the age of smart control

PETER G. MARTIN, PhD, is the Vice President

of business value consulting at Schneider Electric. Dr. Martin has more than 35 yr in industrial control and automation. He has authored numerous articles, technical papers and books, and holds multiple patents. Dr. Martin was named as one of the 50 Most Influential Innovators of All Time by Intech, and received the Life Achievement Award from the International Society of Automation (ISA). In 2013, Dr. Martin was elected to the Process Automation Hall of Fame, and was selected as a Fellow of the International Society of Automation. He holds BA and MS degrees in mathematics and an MA degree in administration and management, a master of Biblical studies degree, a PhD in industrial engineering and a PhD in Biblical studies.

The Industrial Internet of Things (IIoT) is vastly transforming how manufacturing and process plants conduct their daily operations. While the term “IIoT” is being used left and right, whether the industry is really leveraging IIoT technology to find more value remains uncertain. Connectivity, networking, big data, predictive analytics, cloud computing, fog computing, wireless communication, cyber physical systems and edge computing are gaining acceptance every day, and when taken together, offer unprecedented promise. In fact, some experts believe the IIoT is eliminating many of the constraints tradi-

tional technology has imposed on automa- managing this would be to turn the pump tion systems, particularly the constraints off, which leads to profit loss. However, in that limited the ability of those systems to the age of smart control, the overheated improve operational profitability, which pump is registered and process controls are leveraged to communicate to the should be their inherent purpose. For example, traditional process con- pump that it needs to slow down. This altrol has focused almost exclusively on lows workers to safely control the operaimproving the efficiency of the operation. tion, but more importantly not diminish However, the rise of the IIoT is making it the profitability of the operation. This is possible for companies to begin control- the advantage of smart control, and it is ling other key business variables, as well. revolutionizing how companies drive realBecause they have more connectivity time operational profitability improveand computing power, today’s smarter, ments, safely. In the heyday of the IIoT age, managers connected products, such as pumps, are practically able to control, monitor and and plant operators have never had more secure themselves, essentially becoming opportunities to measure and improve the profitability of their operations in real time. autonomous assets. With strong IIoT technology and prac- Management is now able to understand the tices in place, these smarter assets can real-time variable constraints of the operaeasily work together to form asset sets. tion. At the same time, plant personnel can For example, equipment assets connect understand the impact their decisions have to form production unit asset sets, which on operational profitability in real time. While it may seem like autonomous are typically connected to form production area asset sets, which, in turn, are instruments are completely functional connected to form site asset sets. Sites are without the help of plant operators, the grouped into enterprises. Combining this systems do not have the knowledge to auequipment asset perspective with other tomatically control all business variables industrial assets, such as energy, materials, and process dynamics. Often, there is a production and human assets, opens the benefit to having manual control over the door to controlling the most critical busi- operation, as these self-service controllers ness variables in the same way operational can overlook key process dynamics that a human would be able to detect. Smart conefficiency is controlled. trol augments and empowers our workers This is called smart control. In the era of smart control, the industry to make better decisions. It is mind plus is able to control critical business variables machine for a better business outcome. The world is more connected, which in real time. This extended level of control enables presents the industry with new digitization the workforce to better communicate opportunities. The benefits this connec with the plant at a much faster rate, and at tivity provides mean that processing coma lower cost. An empowered workforce, panies and industrial manufacturers can aided by technology, drives total asset op- move from managing their businesses to timization and makes better business and actually controlling them. To be successful, risk factors like safety, reliability and profitoperating decisions at the right time. Today’s technology enables the indus- ability, all of which have traditionally been try to detect abnormal conditions and to managed daily, weekly or monthly, are now able to be controlled in real time too. The respond to them immediately. As an example of this, consider that a IIoT enables and pushes the boundaries to pump encasing harmful flammable chemi- find real, measurable and tangible value; cals begins to heat up. The previous way of that is its true promise. Hydrocarbon Processing | OCTOBER 2017

35

| Special Focus PROCESS CONTROL AND INSTRUMENTATION As the downstream processing industry moves toward a more digital world, advanced process control solutions provide hydrocarbon processing companies with inventive ways to manage difficult and complex operational and reporting applications. Plant personnel are living in an evolving world of wireless devices and software capabilities that allow management decisions in real time; all are supported by vast amounts of process data available instantaneously. This month’s Special Focus showcases innovations in control devices, hardware and software to increase plant efficiency. Photo: Air

Liquide control operations. Photo credit: Stéphane Rémael/La Company.

Special Focus

Process Control and Instrumentation A. GOGATES and J. GUNNELL, Emerson Automation Solutions

Optimizing ethylene production with laser technology Ethylene is one of the most valuable organic compound building blocks in the world. Global ethylene production is forecast to reach 200 MMtpy by 2020. Ethylene production is a highly competitive industry where purity is of utmost importance. Ethylene plants require highly responsive control technologies to keep the process stable. Traditionally, gas chromatographs (GCs) have been the measuring instrument of choice for all areas of ethylene production. From feed qualification, through the cracking furnaces and purification train, to the final product certification of ethylene delivered via ship or pipeline, a typical plant may have 40–50 GCs. In many of these areas, the GC remains the best analytic choice. However, in some critical segments of the plant, GC capabilities meet only basic requirements, specifically in ethylene fractionation, acetylene control and real-time product certification. The GC role. GC performance is more than adequate in eth-

ylene purity applications�unless the user has a better technology to choose, and that has been possible for only a relatively short period of time. Over the last few years, laser technology has quietly slipped into petrochemical manufacturing as an analytical option, and it has now advanced to a point where it pro vides a real improvement in terms of speed, precision, reliability and cost in critical areas of the purification train and product certification. Before plunging into the performance advantages of laser technology in the cold areas of ethylene production, it is important to emphasize the critical benefits of GCs in the “hot end,” which is the real moneymaker in an ethylene plant. Increased throughput achieved by using GCs for furnace optimization and control can mean big profits. In a plant producing 1 MMtpy of ethylene, a 1% change in throughput can produce an enormous improvement in the bottom line. Although laser technology does not have any application in this part of ethylene production, new developments in laser technology could change its role in this section of the ethylene process.

cess excursions. Precise and rapid control of the catalyst activity is vital to maximize the ethylene produced. If the catalyst is not active enough, then not all of the acetylene will be converted into ethylene. Conversely, if the catalyst is too active, then some of the ethylene could be converted back to ethane. The ability to measure carbon monoxide (CO) is a key to controlling catalyst activity. In addition, monitoring the concentration of acetylene is required to initiate the switch from the in-service to the standby unit. It is essential to look for acetylene breakthrough at the outlet of the converters to avoid process excursions downstream. Most importantly, measurement of the outlet must be done quickly and with a very low limit of detection. Ethylene fractionation. The goal of an ethylene plant is to

produce 99.99% pure product. This purity is ensured during the final purification step, which takes place in an ethylene fractionation tower (FIG. 1). An analysis is required for process control of the fractionator to ensure on-spec production. Ethane and ethylene have similar physical properties, which makes them difficult to separate. Process control of product purity requires a fine balance to maintain ethane close to the specification limit without going off-spec or recycling ethylene. Operating the tower efficiently offers considerable economic advantages in re-

Acetylene conversion. In the cracking process, some mol-

ecules are over-cracked into acetylene. An important step to maximize production is to convert acetylene back into ethylene. This process is done through the addition of hydrogen in catalytic beds called acetylene converters. Two acetylene converter units are used�one in service and one on standby. Analytical data are required for the inlet stream, mid-bed and outlet streams of acetylene converters to optimize conversion and avoid pro-

FIG. 1. Diagram

of a typical ethylene fractionator. Hydrocarbon Processing | OCTOBER 2017

37

Process Control and Instrumentation

ducing product giveaway, thereby minimizing energy usage and avoiding ethylene recycle. Measuring the C 1 and C2 molecules,

The novel feature is mounting the steering mirrors inside the cell. This setup might seem strange at first, but it is no different from the normal practice of allowing the cell mirrors to be in contact with the sample. For clean streams, such as those in high-purity ethylene appliHybrid QCL/TDL laser technologies can cations, no degradation of the optics is observed. The analyzer can combine up to six quantum casprovide a measureable impact on plant cade lasers (QCLs) and tunable diode lasers (TDLs) operations and financial outcomes. in a single system for multiple gas measurements in a zero-gap configuration. The lasers are housed in modules that are identical for QCLs and TDLs. The as well as CO and carbon dioxide (CO 2 ), allows the tower oplaser module has a gas-tight window situated directly in front eration to be fine-tuned for maximum efficiency and ensures of the laser package window, and the laser light is divergent as that ethylene production is on-spec. it enters the cell assembly. A collimation optic for each laser is situated within the gas path, and is mounted to the laser module to allow the laser to The purification demand. In the ethylene purification steps, recent advances in laser gas analyzer technology can have sub be aligned and collimated outside of the cell assembly. The collimated light from each of the six lasers is combined inside the stantial impacts. Historically, conducting high-sensitivity measurements of gas components, using laser absorption spectrosgas path using beam steering optics into a single coaxial beam, copy, has been compromised by the external path that the laser which is then directed into a section of the cell that houses either spherical or astigmatic mirrors to extend the length of beam travels in the optical system. This external light path can make a significant contribution to the spectrum, particularly the path the light travels. Two detectors can be installed in the where strong absorption lines are being targeted for high-sensisystem to allow detection of the mid- and near-infrared (IR) light, depending on the types of laser installed. The light is tivity measurements. A new approach is to completely eliminate the external light path by close-coupling the lasers and detector distributed between the detectors using wavelength selective to the measurement cell using a novel zero-gap design. beam splitters that are also located inside the gas path. The laser light’s total path length outside of the cell assembly is less than 1 mm in this configuration. This arrangement virtuHow zero-gap design is achieved. The external path spectral contribution is eliminated by reducing the portion of the ally eliminates any spectral absorption from the external light path, which enables parts-per-billion measurements of gases laser beam exposed to the air to near zero. A typical analyzer layout, where the laser beam passes through an external path, (such as H2O and CO2 ) for process gas streams. is shown in FIG. 2. A layout with a zero-gap laser and detector Multicomponent detection. QCL and TDL lasers are semiis shown in FIG. 3. conductor devices that produce light in the IR region. They are fabricated to emit light at a desired wavelength, and are made to scan a spectrum using a laser chirp technique. When the laser is pulsed with electrical energy to start the laser process, it heats up. As the temperature increases, the wavelength of the emitted light also increases. A laser chirp lasts about one microsecond. In this span, a spectrum of 1–3 wavenumbers is scanned. The raw detector signal is then proTABLE 1. Ethylene

purity measurements

FIG. 2. Standard analyzer layout with laser beam exposed to the

atmosphere.

Range1

Methane

0–1,000

5

ppmv

0–20

0.2

ppmv

0–500

5

ppmv

CO

0–5

0.05

ppmv

CO2

0–5

0.05

ppmv

Ammonia

0–20

0.2

ppmv

H2S

0–50

2

ppmv

H2O

0–10

0.1

ppmv

0–100

1

ppmv

Acetylene Process control

Adders for product certification

Ethane

Methanol FIG. 3. Zero-gap analyzer layout wherein the path of light through the

atmosphere is minimized.

38


1

Limit of detection (LOD)2 Units

Components

Components and ranges are indicative. Analyzer requirements will depend on complete gas list. Detailed specifications can be provided. 2 Repeatability ±1% of reading or LOD, whichever is greater.


cessed to convert it into a spectrum from which the concentration of analytes can be calculated. QCL and TDL lasers can be chirped at a frequency of up to 100 kHz, enabling many thousands of spectra to be gathered in a few seconds and processed to provide a strong signal with a good signal-to-noise ratio. The scanned wavelength region is selected to enable measurement of the desired analytes. By careful choice of lasers, compounds as diverse as hydrocarbons, sulfur species, sol vents and nitrogen compounds can be measured. In addition, each laser might be able to detect more than one compound. For example, between six and 12 highly varied measurements can be made in a single analyzer. When GCs are used, multiple instruments are required; therefore, the use of lasers can save capital expenditure costs. An advanced signal processing procedure enables real-time validation of measurements and greatly reduces the need for calibrations, reducing ongoing operational costs. Response time. Response time is a major issue in these ap-

plications, as well. In a QCL/TDL, the sample flows through a measurement cell where laser beams continuously analyze the gas. The response time is typically less than 10 sec to achieve 90% of a step change. The output is effectively continuous and in real time. Conversely, GCs work on the principle of injection followed by analysis. Cycle times for a GC vary from 1 min to more than 15 min, depending on the application; therefore, the concentration data is periodic rather than continuous.

This speed of the laser measurement is helpful in the process tower. However, it makes a huge difference in acetylene control, where it can quickly detect a process upset that might otherwise cost hundreds of thousands of dollars per hour to correct. Product certification. Before exporting to customers, eth-

ylene must be analyzed to ensure that it meets product specifications (TABLE 1). This step is critical to profitability. Traditionally, this analysis has been carried out using grab samples and laboratory analysis with GCs. However, components such as ammonia, methanol, nitrogen monoxide (NO), nitrogen dioxide (NO2 ) and hydrogen sulfide (H 2S) can now be measured online in one analyzer using the hybrid laser technology. This multi-component measurement is valuable in product certification, which would normally take 3–6 GCs to measure. Speed is critical at this stage of the delivery process, but precision of measurement should not be compromised for rapidity. The laser technology allows for both, and online, real-time product certification can be achieved. A significant leap. The use of GCs in ethylene production

is established and proven. Plant personnel are knowledgeable on the use of the technology. It requires a significant boost in performance improvement to suggest the need for a change. Hybrid QCL/TDL laser technologies can provide a measureable impact on plant operations and financial outcomes.

EFFICIENCY MATTERS RESTORING PLANTS QUICKLY AND SAFELY Managing turnarounds on time and on budget can present many challenges. Cudd Energy Services helps you meet these challenges head on. Our fleet of pumping, transport, and storage vessels accommodates a wide range of flow rates for HPHT, open-flame environments that get you back online safely and efficiently. With a 1.5 million scf capacity, the queen storage vessel tank reduces frequent deliveries that cause congestion, and can be safely replenished without interrupt ing pumping operations, saving you time and money. Equipped with emergency shut-down devices, the dual mode pump features a heat recovery system t hat reduces fuel costs, and its EPA Tier 2/C.A.R.B. emission rating helps reduce emissions. For more information about our industrial nitrogen solutions, visit us at www.cudd.com or call us at 832.452.2800.

WWW.CUDD.COM

Select 156 at www.HydrocarbonProcessing.com/RS Hydrocarbon Processing | OCTOBER 2017

39

The new WDG-V. Impressing even the world’s most demanding combustion managers. The new AMETEK Thermox WDG-V extractive combustion analyzer oers industry-leading safety support. First in its class to be third-party certied for SIL-2 implementation in safetyinstrumented systems, the WDG-V provides a complete solution for combustion process control and safety. Reliable detection of low-combustion oxygen and/or high CO in a red heater or boiler is critical to burner management system eectiveness. The WDG-V analyzer monitors hot, wet ue gas to minimize excess oxygen, lower NOx emissions, and improve operating eciency in power generation and petrochemical rening. It can also monitor methane levels to ensure safe burner start-up and shutdown. The all-new WDG-V. Combustion management and safety capabilities beyond expectations. Learn more at www.ametekpi.com.

© 2014 by AMETEK Inc. All rights reserved.


Special Focus

Process Control and Instrumentation A. KERN, APC Performance LLC, Houston, Texas; and B. ANDREW, Hydrocarbon Processing, Houston, Texas

The next generation of advanced process control In this article, Hydrocarbon Processing provides published references on model-based control, and contributing author Allan Kern presents his views on model-less control. Model predictive control (MPC) is well established for complex dynamic systems. However, a caveat to MPC’s definition says, “the additional complexity of the MPC control algorithm is not generally needed to provide adequate control of simple systems, which are often controlled well by generic PID controllers. Common dynamic characteristics that are difficult for PID controllers include large time delays and high-order dynamics.” 1 Hydrocarbon Processing has covered both model-based and model-less control for decades. In September 2002, R. B. Agnohitri and D. N. Jaguste authored, “Rationalizing PID tuning with a model-based approach.”2 Within the same issue, L.A. Robbins wrote, “Setpoint tuning method gives excellent response to load changes”.3 A. Kern authored, “The history, and possible future, of model-less multivariable control,” and “The lost art of proportional band, and the accidental art of performance monitoring,” in October 2016 and August 2017, respectively. 4,5 One final clarification is that the phrase “model-based design” can have a different meaning for process engineers vs. control engineers. For a process engineer, the term model can mean a full process simulation. WHAT IS APC 2.0?

For more than 30 yr, large matrix model-based multivariable control (i.e., MPC or APC 1.0) has been the dominant advanced process control (APC) technology used in industries such as oil refining, chemicals and petrochemicals. The virtues and successes of MPC aside, it has also earned a reputation as an ownership challenge in many areas, including cost, maintenance, support and performance. While MPC may always be the preferred solution in select applications, many end users have long been moving away from MPC as an everyday tool, even if the technology to backfill that gap (i.e., APC 2.0) has yet to fully emerge. Experience makes it increasingly clear that industry needs a multivariable control tool that is much more agile�more operation-friendly in every respect�to enable renewed progress in advanced process automation. Thirty years of the APC 1.0 paradigm have left many people thinking that large-matrix, model-based multivariable control and optimization must all go together�that you cannot have one part without all the others. However, experience shows that alternative solutions are actually quite possible, and even readily available, within existing modern control system capabilities.

Going forward, how will the emerging APC 2.0 paradigm differ in key respects from APC 1.0? A few examples include: • APC 2.0 will feature a matrix, but not necessarily models • Focus will shift to control-layer automation and business-layer optimization • Operational performance will become the new norm, after 75 yr of Zeigler-Nichols. Behold the matrix! One of the most important and endur-

ing contributions of the model-based era to process automation will prove to be the matrix (not models). The matrix has emerged as the natural way to define the essential aspects of the multivariable nature of almost any process. It captures the available handles, important constraints, useful interactions and optimization potentials in a useful intuitive format. It serves as a guide to operations and as a specification for automation. The matrix puts all members of the operating team� control engineers, process engineers and operations personnel�on the same page to align their efforts for more effective operation. Industry owes a large debt to APC 1.0 for bringing the concept of the matrix to the forefront of process operation and automation technology. Meanwhile, models have proven to be very unreliable and as much a vulnerability as a strength. Experience has shown that building an APC solution based on model-based control theory front-loads the solution with many ownership challenges. For more than 30 yr, it remains to be shown whether that burden is sustainable or ultimately outweighs the benefits of using modBusiness planning and optimization

Global

Operating team’s real-time knowledge

Process control and automation

Local

FIG. 1. The

APC 2.0 paradigm avails itself of business-domain planning and optimization, which eliminates redundancy and empowers each layer to focus on its core area of expertise and responsibility. Hydrocarbon Processing | OCTOBER 2017

41


TABLE 1. A

sample matrix for a crude oil distillation column. The principle elements of a matrix includes control handles, constraints, interactions and optimization objectives Constraints Handles

51A|110

51A|115

51A|116

51A|125

51PC105

51PD|104

NAP90

KERO90

KEROFLSH

DSL90

PRESSURE

TOP DP

51TC102

Feed heater

+1

51FC103

Bottom steam

+1

51FC128

Pumparound

-1

-1 OPT

51TC106

Top temperature

51FC112

Kerosine draw

51FC122

Diesel draw

51HC128A

Pumparound bypass

+1

+1

+1 OPT

+1 +1 OPT

+1 +1 +1 OPT

Target Error minimization Operational

FIG. 2. The

analogy of a passenger plane changing altitude illustrates why ‘operational performance’ criteria are more appropriate for industrial process operation than for traditional model-based error minimization.

els in the first place. Fortunately, experience and emerging technologies also suggest that models may actually be dispensable to closed-loop multivariable control performance (TABLE 1). Automation rules! In most industrial process operations, the

optimization solution (i.e., the target operating point), just like the matrix (i.e., the operating window), is well-known by the operating team. Process engineers are likely to make process economics and key process interactions the first priorities for new processes, and operators are unlikely to move up to the console or supervision arena without learning these aspects of operation. This knowledge is common among the operating team, so that building an optimizer into the APC solution to achieve a real-time basis is largely superfluous. It adds substantially to APC complexity and ownership burdens, without adding commensurate new value. A better paradigm is to allow optimization results to flow down from the business layer to the control layer, whether via the computer system or the chain of command (both are common today). The business layer has a greater global optimization horizon and more complete information, and results do not normally change in real time. The essential role of APC is to honor these optimization results in the live process en vironment, where the related process values (not the optimization results themselves) are subject to change in real time. This optimization paradigm is more streamlined, enables each player to focus on its essential role and eliminates one of the main sources (after models themselves) of MPC ownership challenges (FIG. 2). Operational control performance. Virtually every aspect of

modern process control technology is rooted in the concept of models. However, it turns out that control performance is 42

+1


one area where industry does not want a model-based solution. Industry wants operational performance. This has been another long, slow lesson of APC 1.0�and of single-loop PID control, for an even longer time. Operational performance means having preset process “speed limits.” This term means approaching targets and constraints in a steady first-order manner, avoiding overshoot and oscillation and taking extra precautions, where appropriate, to preserve and ensure process stability at all times. This definition both differs from, and conflicts with, traditional error-minimization control performance criteria, which is based on process models. Experience has shown that operators are willing to sacrifice automatic control�i.e., place loops in manual and circumvent (degrade or switch off) APCs�whenever operational performance is not forthcoming. To achieve greater success, APC 2.0�both single-loop and multivariable�must accommodate operational process control performance criteria (FIG. 2). Agility is the word. All of this information suggests a more ag-

ile APC 2.0 solution and paradigm. “Agile” means costs that fall within normal operating budgets; schedules that meet modern manufacturing precepts (days or weeks, not months or years); support requirements that fall within the purview of in-house control engineers; performance that meets operations’ expectations and approval; and technology that falls within existing control system capabilities (e.g., function blocks). To the extent that this paradigm has yet to fully emerge, industry would do well to make it a target paradigm. LITERATURE CITED

Complete literature cited available online at HydrocarbonProcessing.com. ALLAN KERN is the owner and President of APC Performance

LLC. He has 35 yr of industrial process control experience and has authored numerous papers on a wide range of practical process automation solutions. He earned a BS degree in chemical engineering from the University of Wyoming, and holds professional engineering licenses in chemical engineering and control systems engineering. Readers can provide feedback on APC 2.0 at APCperformance.com. BOB ANDREW is the Technical Editor for Hydrocarbon Processing. He has extensive experience in risk management,

having served as an HSE “best practices” coordinator for Mobil Global Professional Services in the mid-1990s, and more recently as an HSE or process lead on mega-projects in the LNG, oil and gas, and refining industries.

Special Focus

Process Control and Instrumentation W. P. AVERILL, Visual Performance Solutions Inc.,

Annapolis, Maryland

Conceptual troubleshooting training for refinery operators Throughout the refining industry, the average number of years of experience for console operators is decreasing, as large numbers of very experienced console operators are retiring. Many throughout the industry are calling this the “Great Crew Change.”1 Most operations managers and supervisors agree that the highest-value console operator competencies that are being lost in this crew change are associated with: • Understanding process and controls • The ability to troubleshoot abnormal process and control malfunctions. Closing this competency gap requires a new and better way of training troubleshooting skills and delivering the underlying prerequisite conceptual knowledge. As illustrated in FIG. 1, the goal of the program is “conceptual troubleshooting” competency, as it is the most challenging performance objective. The prerequisite knowledge and skills acquired in the journey toward that goal deliver significant added value beyond troubleshooting. Operators learn how to leverage that understanding when predicting (mentally simulating) process responses to setpoint adjustments, remaining mentally engaged when optimizing operations, and ensuring normal operation when performing critical procedures, such as mode changes, startups and shutdowns. This training is applicable for console operator and frontline engineer positions, which will be grouped together under the term “operators” here. High competency in these critical frontline positions will significantly contribute to the improvement of business performance.2 Mental simulation/understanding. The first stage of con-

ceptual-based training is to instill and develop an operator’s ability to mentally simulate (predict) process responses to control loop setpoint changes and instrument malfunctions (false high or low indications). This requires a deep understanding of concepts that include: • How changes to pressure and composition impact hydrocarbon boiling point temperatures. • How changes to heat and material balance in a distillation tower and other separation processes impact the flowrates and compositions of the products. • How changes in reactor chemistry factors, such as reaction severity, impact reactor product compositions.

• How changes in heat transfer associated with heat integration between liquid streams, liquid-to-vapor heat of vaporization, and vapor-to-liquid heat of condensation impact process chemistry. TROUBLESHOOTING

After developing a sound understanding of these concepts, operators are prepared to master conceptual troubleshooting, which includes detecting abnormal operation, iteratively and narrowly searching down to the root-cause component, and then implementing corrective action. Rule-based vs. conceptual troubleshooting. To imple-

ment and standardize a conceptual troubleshooting approach, a cultural transformation from rule-based to conceptual-based thinking must take place. The troubleshooting procedure in TABLE 1 is the simplest structure to illustrate a rule-based troubleshooting process. When operators detect an off-normal process condition, they: • Develop a list of observed symptoms that characterize the off-normal condition • Scan the list of symptom combinations in the troubleshooting table to locate the combination that exactly matches the observed symptoms list

y c n e t e p m o c g n i t o o h s e l b u o r t d n a s i s y l a n a l a u t p e c n o C

r ) f o U g C n C i t F ( o t o i h u n L4 troubleshooting l e s g b n u i o k r a c l t r r a c L3 cause and effects u t o t i c a t p n y e a l o c i t n a t a c C o d c r F i : L2 baseline control u 1 fl i c p T o L1 chemistry

Goal: Conceptual troubleshooting competency

r t o a r e n e r e g r t o a c R e : 2 i c p T o

Topic E3: Conceptual troubleshooting for basic distillation Lesson 4: Troubleshooting Lesson 3: Cause and effects Lesson 2: Baseline control Lesson 1: Chemistry Topic E2: Heat exchange concepts Topic E1: Baseline control concepts

l g a u n t i t p e l s o o t a n c h s o n C b l e e m u a r o n d t u f

g n i d n a t s r e d n U

Time

FIG. 1. Journey to conceptual troubleshooting competency. Hydrocarbon Processing | OCTOBER 2017

43


• Conclude that the corresponding root cause in the table is the problem • Perform the procedural steps for corrective action. Rule-based troubleshooting procedures are practical for individual pieces of equipment and for simple process systems, as operators need to recognize only a small number of malfunctions and a small number of symptoms that must be considered in the search for the root cause. Most competent troubleshooters in refineries have been implementing rule-based troubleshooting strategies on complex refinery processes throughout their careers. Over a long period of time (typically more than 10 yr), each of those trou bleshooters has accumulated the equivalent of a massive trou bleshooting procedure table in their long-term memory. The knowledge was acquired from many opportunities to interact with the control systems and work through a multitude of malfunction events on the job. For a complex refinery process, the knowledge to support a rule-based troubleshooting approach 1) Detect off-normal condition

2) Collect additional data

3) Develop hypotheses (possible root causes)

4) Mentally simulate each possible root cause

Yes

No (Repeat for another possible root cause)

1) Detect off-normal condition—This is based on knowledge of normal operating ranges and targets. 2) Collect additional data—Develop a list of observed symptoms that characterize the off-normal condition. 3) Develop hypotheses—Create a list of possible root cause components that, if malfunctioning, could be the root cause. 4) Mentally simulate each possible root cause—Predict how the system would respond to each component malfunction. 5) Compare the mental simulation to real-time data—Determine if the possible root cause component malfunction would result in all of the observed symptoms. 6) Root cause found?—If the comparison results in a match between predicted system response and the observed symptoms, then go to Step 7. If there is not a match, loop back to Step 3 and analyze the next possible root cause. 7) Develop and implement corrective action plan—The plan includes two steps: mitigate the malfunctioning component’s impact on the system; and determine and initiate actions required to repair or replace the malfunctioning component.

44

conceptual troubleshooting model.


complex processes, operators must recognize a large number of malfunctions and consider numerous symptoms in the search for the root cause. In these cases, conceptual troubleshooting is the most effective approach. The design basis for a performance-based training system should always begin with a performance vision. A universal performance vision (FIG. 2) is consistent with the analysis strategies employed by expert conceptual troubleshooters, regardless of the industries in which they work. 3 Conceptual troubleshooting is the most difficult to master, but it is also the most powerful. As a prerequisite, operators entering a conceptual troubleshooting program must have an aptitude for conceptual analysis. With a well-designed training program, operators can attain a high level of troubleshooting competency within months. Comparatively, reaching the same level of competency with a rule-based troubleshooting approach can require many years. While the conceptual knowledge that is prerequisite to conceptual troubleshooting is the most valuable, it is rarely addressed in traditional training and documentation systems. 4 the centerpiece for a conceptual knowledge training program, and it must convey a contextual understanding of the overall

7) Develop and implement corrective action plan

FIG. 2. Universal

Conceptual troubleshooting model and design. For

Create a system level control diagram. This diagram is

5) Compare mental simulation to collected data

6) Root cause found?

is massive, and efficient formal training strategies to develop that knowledge base simply do not exist. Additional disadvantages of rule-based troubleshooting include: • It does not prepare operators to troubleshoot malfunctions that they have never seen. • It is not transferrable to other processes or to new control strategies implemented on their processes. • Extensive and wasted effort by field operators can be required to check the functionality of controls and equipment that cannot possibly be the root cause.

TABLE 1. Troubleshooting

procedure tabular structure Then the

The corrective

If

And if

root cause is

action is

Symptom 1

Symptom 2

Malfunction 1

Procedure steps …

Symptom 1

Symptom 3

Malfunction 2

Procedure steps …

Lecture, 5% Reading, 10% s s e n e v i t c e ff e g n i s a e r c n I

Audio-visual, 20% Demonstration, 30% (examples) Exercises promote these three high-value learning activities

Discussion, 50% Practice, 75% Teach, 90%

FIG. 3. Effectiveness

of various training activities.

e e % i m % m 5 t 5 t i 2 7 g g n n g n n a i a n i h n n h i i t n i t i a s a r a r s t e r t r e f o t f g L o g M o n i n

i n i a r t

t o o h s e l

l a b n e u o i o t r m t i % i 5 d 2 t l e a a % r n g u m 5 i n t 7 T a i t p h n t i e n g a c a i n s r h n s t n t e i o L f o C e a

r r t o M f o


process system and relationships among the subsystems. 5 The diagram must also be user friendly and easy to interpret, pro vide a sufficient level of detail to facilitate conceptual analysis, and consolidate all system components into a single view on a single large sheet of paper. Create a large example and exercise library. Examples

and exercises are the most effective vehicles to translate the general vision of performance into explicit representations of the ultimate on-the-job performance objectives. As illustrated in FIG. 3, exercises are the primary vehicles to engage operators in discussions, practice and opportunities to mentor peers.6 This engagement strategy is the most effective way to promote conceptual knowledge acquisition, understanding, long-term memory retention and the ability to apply it on the job.

Topic E3: Final assessment g i n (provides three practice assessments) t o s o e Lesson 4: Conceptual troubleshooting h s y i s c c l e (intro to functional troubleshooting method e r t s n b x n u e p i o t customized for refinery processes) r o a n e n c e i e l t i l t Lesson 3: Cause and effects l a p t s c o t u p s t i e g m p (Interactions among DeC4 controls ) e c d i n o n c o t o c n c a i s Lesson 2: Baseline control o f c h s C o b l o : s r r (purpose of DeC4 controls) 3 a l e e b E o t b n i c f Lesson 1: Chemistry u m r o p e u o T T (hydrocarbon products boiling point ranges m d n a t s a n and heat and material balance for DeC4) t s d p t s e y p e p n t c i Topic E2: Final assessment x n u c e n c e f o n l o c o p (provides three practice assessments to prepare for final) g e c m g e y c g n n i c o Lesson 4: Fired heater a t s r h a n i t h g o c n m c i x (cause and effects among interacting heater controls) o a s h e t e x t s t e Lesson 3: Hydrocarbon condensers and reboilers h e C a r s a e p e e I n c H e l H c (pressure, composition, and boiling point temperature relationships) : n 2 b E u c o Lesson 2: Steam heaters and steam generators i c s o t p r c t s T o (pressure and boiling point temperature relationship) t p ff e c e d Lesson 1: Liquid-liquid heat exchangers n p n t s d e o n a (heat integration concepts) l c n c e s e a t r o i o s u s n Topic E1: Final assessment l c y C a c o l o (provides three practice assessments to prepare for final assessment) a t r n n i n e o l Lesson 4: Cause and effects a s e n e c l a (interactions among surge drum controls) a i B e l u t Lesson 3: Multi-level baseline control a s p B : e (operation of cascade, ratio, and split range controls) 1 c E n i c Lesson 2: Mental simulation statements (operation of single-loop controls: p o o C T temperature, flow, level, and pressure controls) Lesson 1: Manual and programmable control basics (fundamentals of control loop operation)

Design a simple-to-complex curriculum. The training

curriculum must progress in distinct stages and transition from simple to complex.7 FIG. 4 illustrates an example sequence for teaching operators conceptual troubleshooting fundamentals. This is the recommended first training module; it establishes a common foundation using generic materials. Subsequent training modules for specific refinery process units should employ the same sequence of topics as illustrated for Topic 3: chemistry, baseline control, cause and effects, and troubleshooting. Design a conceptual knowledge description. The pro-

cess description manual content must cover the relevant conceptual knowledge that supports the operator in the successful completion of the exercises. The organization of the content should be parallel to the structure of the curriculum. Separate chapters should be devoted to chemistry, baseline control, cause and effects, and troubleshooting. Design a standardized assessment strategy. Standardized assessments are needed to measure the degree that learning is taking place. The exercises in the assessments should be derived from the training exercises. Assessments are typically used as criteria for job qualification. Just as important, the analysis of assessment results provides insights into opportunities to improve training material. If multiple trainees score poorly on a specific assessment exercise, an in-depth analysis of the exercise might indicate that the exercise statement or layout is confusing, or that the supporting content in the process description manual may not adequately convey the prerequisite knowledge. Added value/benefits. One of the most important train-

ing objectives is to develop the operators’ ability to mentally simulate (predict) how refinery processes respond to change events. As illustrated in FIG. 4, this development is central to conceptual troubleshooting. An added value of this capability is that it can be applied in numerous other high-value activities. Establishing a standard language for process analysis increases engaged participation by multiple operators when troubleshooting and resolving operational issues. Monitoring of process responses to procedure steps.

When making a control setpoint adjustment, the console operator should mentally simulate the system response and then

Time

FIG. 4. Simple-to-complex conceptual troubleshooting fundamentals

learning curve.

verify normal operation. This is particularly important during unit startup, unit shutdown and when implementing operational mode changes. It is during these events when control malfunctions are most likely to be revealed and, if not recognized, produce serious consequences. Given an opportunity to increase profitability (increase throughput, increase quality, decrease costs, etc.), operators will think ahead and mentally simulate process responses to possible adjustments, and then determine possible desired benefits and undesired results. A less-structured optimization strategy such as, “Let’s try this…” can lead to undesired operational swings and possible upset conditions. LITERATURE CITED 1

Rhodes, M., “Overcoming the challenges of the ‘great crew change’,” Hydrocarbon Processing, June 2015. 2 Floyd, R. C., Liquid Lean: Developing Lean Culture in the Process Industries, Productivity Press, 2010. 3 Johnson, S. D., “Training technical troubleshooters,” Technical & Skills Training, University of Illinois periodical, October 1991. 4 Hall, E. P., S. P. Gott and R. A. Pokorny, A Procedural Guide to Cognitive Task Analysis: The PARI Methodology, Manpower and Personnel Division, Brooks AFB, Texas, November 1995. 5 Wallace, D. S., S. W. C. West, A. Ware and D. F. Dansereau, “The effect of knowledge maps that incorporate Gestalt principles on learning, The Journal of Education, September 1998. 6 Roffel, J. and M. Ross, “Leveling up: Progress from operator training to competency management,” AFPM Annual Meeting, San Francisco, California, March 2016. 7 Eisner, E. W., “A look into Benjamin Bloom’s life work,” Prospects: The quarterly review of comparative education, Vol. 30, No. 3, Unesco, Paris, September 2000. WARREN AVERILL has

been developing and perfecting a worldclass console operator training program, with an emphasis on conceptual troubleshooting skills, for almost three decades. Throughout his career, he has served as the lead on a long series of successful documentation and training projects for most major refining companies at more than 25 refineries and oil production plants. From 1990 to 2003, Mr. Averill was the Chief Engineer for the petrochemical training solutions department at RWD Technologies Inc. In 2003, he founded Visual Performance Solutions Inc. Mr. Averill holds a BS degree in electrical engineering from Oklahoma University, and an MS degree in electrical engineering from the Naval Post Graduate School. Hydrocarbon Processing | OCTOBER 2017

45

www.ConstructionBoxscore.com

Logon for a

FREE ONLINE DEMO!

, G N I N I F E R L A B O L G E H T R O F E C N E G I L L E T N I T MA R K E S E I R T S U D N I G N L / G N I S S E C O R P S A G D N A L A C I P E T R O C H E M Hydrocarbon Processing’s Construction Boxscore Database, the most reliable source to track active construction projects in the refining, petrochemical, gas processing, LNG and solids industries throughout the world, now reaches further and is more powerful than ever before!

• Project details on thousands of active projects and global construction contracts, including contact information for key personnel • Advanced search that filters the listings by project type, scope, region, investment and more • Daily updates for new and newly updated projects • The weekly Boxscore Update e-newsletter with new listings and trends analysis

For more information, contact: Laura Kane at [email protected] or +1 (713) 525-4626

Process Optimization H. SALGADO, Ecopetrol, Cartagena, Colombia

Optimize a refinery H2 network using spreadsheet tools One of the major goals in every refining complex is to minimize the utilization of its hydrogen (H 2) generation units by maximizing the H 2 recovery from residual streams. Achieving this goal can have several benefits, including savings in raw material for H2 generation, commonly natural gas, as well as more room to process opportunity crudes with high contaminants content. Conversely, increasing demand for H 2 for processing heavy crudes, and the need to meet fuel specifications, cause refining economics to be highly dependent on the availability and cost of H 2. This work provides a straightforward and systematic method to evaluate and optimize the H 2 distribution network in a refining complex by using common tools available to any process engineer, such as commercial spreadsheet software. As an illustration of this method, data from a high-conversion refinery is used to obtain a network redistribution proposal with optimal recovery of residual H2 streams. H2 and the modern refining industry. H 2 has become an

essential molecule to convert heavy oils into sellable, valuable fuel and products, such as diesel, jet fuel, naphtha, propylene and ethylene. Furthermore, refineries around the world are turning to heavier crude slates. These opportunity cr udes contain higher contents of contaminants, such as sulfur, nitrogen, oxygen and metals. Heavier crude slates demand more complex processing techniques to meet the desired conversion and product quality. One of the key processes that ensures both conversion and product quality is the hydrocracking of gas oils and other heavy cuts. However, this process has a high H 2 consumption�between 1,500 standard cubic feet per barrel (scfb) and 3,500 scfb. Another refining process that uses H 2 to meet the required quality, without providing significant conversion of the hydrocarbon molecules, is the hydrotreating of naphtha and distillates. This process consumes between 100 scfb and 800 scfb of H2 . The H2 consumption is driven by the quality of the processed crude slate, specifically its H 2 :C ratio and its contaminants content. These processes have an associated residual H2 production due to the gas purge used to eliminate undesirable compounds, such as methane and ethane, that are formed during side cracking reactions. Depending on the H 2 concentration, this resid-

ual gas is sent to purification units, such as pressure swing adsorption (PSA) units and H2 selective membranes, to recover the H2 present in the offgas. In other cases, this offgas is sent to the fuel gas network to be used in fired heaters and boilers. Other conversion processes, such as catalytic reforming of naphtha to produce high-octane gasoline and aromatics, and steam cracking to produce light olefins, have high residual H 2 production. This production can provide an alternative H2 source for the refinery H2 balance. Typical H2 consumption and production in several refining processes is shown in TABLE 1. In a typical refinery with several of these process units, the H2 is managed by a network that interconnects H 2 producing units (HPUs) with H2 consuming units, integrating H 2 purification units and compressors and delivering H2 at the required purity and pressure, according to the process needs of each consumer. Nevertheless, in the case of high-conversion refineries, or those that process heavy and contaminated crude slates, the demand for H 2 is high, which requires dedicated HPUs. Due to its flexibility in processing different hydrocarbon stocks, from natural gas to naphtha and heavy oils, steam reforming (SR) is the most common type of H2 production technology.3 Other technologies, such as autoreforming (AR) and partial oxidation (PO), are also used in several refineries, but to a lesser extent.4,5 TABLE 1. H2 consumption

and production in some refining

processes H2 consumption, scfb

Residual H2, scfb

H2 concentration in offgas, vol%

Naphtha hydrotreating

100–500

5–30

40–80

Distillate hydrotreating

300–800

30–150

50–90

Gasoil hydrotreating

700–2,000

100–400

50–90

Hydrocracking

1,500–3,500

210–700

50–90

Semi-regenerative catalytic reforming

670–1,090 (net)1,2

40–851

Continuous catalytic reforming

1,050–1,710 1,2

60–951

Catalytic cracking

70–200

10–301

Process


47

Process Optimization Regardless of the technology chosen, hydrocarbon stocks must be used for dedicated H 2 production, which increases the refinery’s operational costs. A primary goal of refinery operations is to maximize the H 2 recovery from the offgas coming from the process units, rather than producing H 2 from dedicated units. Since regulations for fuel quality have become more stringent with time ( FIG. 1), many older refineries have experienced H2 shortages; therefore, it is important to review the overall H2 balance and make the right investment to recover more H 2 to help meet new fuel specifications. TABLE 2. Methods

of optimizing H2 distribution networks

in refineries Method

Description

Graphic/pinch/linear programming

H2 surplus allocation using pinch methodology and final optimization by linear programming 2,6,7

Graphic/pinch/nonlinear programming

Final optimization done by mixed-integer nonlinear programming8,9 Introduction of variable pressure for producers and consumers in the equations 10 Introduction of compressors, purifiers and reformers in the equation 11,12,13,14

Classic methods to optimize H 2 distribution networks.

The previous statements justify the importance of setting the optimal configuration of an H2 distribution network to obtain the optimal recovery of H2. Although modern refineries are designed to recover most of the H 2 from the process units’ offgas, some H2 is left in the fuel gas system. Conversely, older refineries generally have large quantities of H2 being burned as fuel gas, instead of being used to produce high-quality fuels. Proposed methods in technical literature for H2 network optimization normally rely on pinch analysis, as well as on complex linear or nonlinear programing algorithms ( TABLE 2). Even though these methods are reliable and proven to provide applicable solutions, most of the refinery’s H2 networks are not very complex, and can be properly optimized by using any commercial spreadsheet software. The key factor to successful optimization is building a proper conceptual model, in a systematic way, with the right equations and constraints. To illustrate this approach, data from an actual H2 distribution network, in a high-conversion refinery, is used to obtain a redistribution proposal to minimize the utili zation of the dedicated HPUs; thereby improving the overall refining net margin. A three-step method is proposed to properly model and optimize an H 2 network in a conventional refinery. 10,000

Introduction of the effect of impurities, such as H2S and light hydrocarbons, in the equations15,16,17

TABLE 3. H2 sources

t w m p p , r u f l u S

with their initial values

Pressure, psig

H2 purity, vol%

Flow, MMscfd

PSA FG

110

43

4.25

NHT

250

76.1

5,06

H2 header

300

99.9 rated

104.53

118.13 MMscfd (max.)

HPU1/2

305

99.9 rated

94.19

100 MMscfd (max.)

PSA ROG

305

99.9 rated

10.34

18.13 MMscfd (max.)

Recycle NHT

330

84

10.44

Off HCK/DHT

350

85 rated

14.32 rated

DHT1

1,200

98

50.8

DHT2

1,200

92

53.58

MU DHT

1,350

99.9 rated

28.72

Tag

Recycle DHT1 Recycle DHT2

1,350 1,350

98 92

50.8 53.58

Comments

Japan US EU

10

1 1985

1990

1995

2000 Year

2005

2010

Recovered H2 H2 header 99.9%

10.34 MMscfd 0.22 MMscfd 76.1%

RE

4.38 MMscfd

71.46 MMscfd

32.65 MMscfd (max.) 58.93 MMscfd (max.) 58.93 MMscfd (max.)

NHT

HPU1/2

94.19 MMscfd

MU

442.24

MU HCK

2,200

99.9 rated

71.34

78.47 MMscfd (max.)

Recycle HCK

2,200

92.6

442.24

486.67 MMscfd (max.)

HCK RE

13.87 MMscfd MU

92.6

2015

FIG. 1. Evolution of fuel quality (road diesel) regulation in time.

RE

2,000


100

Fix due to process conditions

HCK

48

1,000

14.82 MMscfd

14.32 MMscfd 85%

DHT1 Residual H2 header 85%

RE DHT2

FIG. 2. Refinery H2 network under study.

PSA-ROG

4.26 MMscfd 43%

Fuel gas header

Process Optimization The proposed method. The following sections outline the

three steps to model and optimize the H 2 network. To facilitate this description, a case study from a high-conversion refinery is presented. The refinery H2 network scheme. The first step is to have a clear understanding of the H2 network, which can be depicted by the interconnections among producers and consumers. Since the optimization is to be carried out on the H2 network, the analysis on the hydrocarbon stream is not shown in the figures. Nevertheless, H2 chemical consumption in process units is a function of the hydrocarbon feed to such units. A process scheme of the H2network, with H2 flows before optimization, is shown in FIG. 2. Defining H2 sources and sinks. Unlike other reported methods, an alternative definition of sources and sinks is given in this study.8,9 Sources and sinks are not only defined for main process units, but also for main process equipment, such as makeup and recycle compressors, as well as H 2 and fuel gas headers. Conversely, certain units or equipment can be an H2 source and sink at the same time; therefore, the concept used in this work is related to the mass balance within the H 2 network (FIG. 3). Sources can vary in concentration and flow, while sinks are constrained by their process requirements and/or design parameters. Nevertheless, during the optimization of the H 2 net work, the flow and concentration of H2 from each source will be recalculated based on sink constraints and the network’s mass balance. In addition, some units can have multiple H2 sources.

The second step is the construction of the source/sink tables, where the H2 purity and pressure specification and/or constraints for each source and sink are included. The optimization constraints are defined in this step. It is important to mention that this step not only requires a deep level of knowledge about the refinery operational philosophy, but also knowledge about the design criteria of the H 2 network and the process units, such as the maximum capacity of process units and process equipment. This data is critical to properly select the possible sources for each sink, since the convergence will depend on such a selection. Additionally, the H 2 mass balance around process units and compressors must be taken into account. For this case study, the full list of sources with their respective initial values, and sinks with their respective constraints, are detailed in TABLES 3 and 4. H2 header SK

SC SC

HPU

SC

SK

SK

RE

SK

FIG. 3. H 2 source

MU

SC

SK

Process unit

PSA-ROG SC

SC SC SK

Fuel gas header

SC

and sink definition.

LIVE WEBCAST: Wednesday, November 1, 2017 11 a.m. EDT / 10 a.m. CDT / 3 p.m. UTC Christophe Romatier Business Leader Honeywell UOP Connected Performance Services

Katherine Jones Senior Product Line Manager Honeywell UOP Connected Performance Services

Leveraging IIoT to Drive Operational Excellence With reduced crude oil prices, the Oil & Gas industry faces a rapidly changing market place combined with lightening-speed advances in technology. The question facing refining and petrochemical producers is how to successfully adapt to the challenges by investing in technology that leverages insights from big data and drives business growth with operational improvements, increased reliability and higher margins. Join Honeywell UOP’s Connected Plant’s Business Leader, Christophe Romatier, and Sr. Product Line Manager, Katherine Jones, for an informative webinar. They will highlight a scenario around a new cloud-based service from Honeywell Connected Plant that targets these challenges, provides industrial users with around the clock monitoring of plant data and rigorous simulations, and offers on-going operational health checks and recommendations to close performance gaps and drive operational excellence.

Register For Free: HydrocarbonProcessing.com/Webcasts

Lee Nichols Editor/Associate Publisher Hydrocarbon Processing


49

Process Optimization Writing and solving the optimization model in the spreadsheet software. In this case, a proprietary spreadsheet

software was used. Any spreadsheet software can be used as long as it includes an optimization tool. As previously mentioned, the objective function selected was the amount of H 2 produced by the dedicated HPUs (HPU1/2), while the selected optimization method was the generalized reduced gradient (GRG) nonlinear method. All of the constraints defined in TABLES 3 and 4 can be easily written using the optimization tool button and parameters box, making reference to the cells in the spreadsheet where the variables are located. The best way to introduce the variables and data in the spreadsheet is with a matrix. The initial values in the spreadsheet correspond to the initial value of the H2 network before optimization. Since H2 source values depend on the objective function and the sink constraints satisfaction status, these values change after running the spreadsheet optimization tool; therefore, an iteration routine must be programmed in the spreadsheet tool so that the new values of the H2 source parameters match the previous calculation. Typically, only 2–4 iterations are needed. Results and analysis. After three iterations, the flow and con-

centration values for each H2 source were converged within a

relative maximum tolerance of 0.1%; therefore, the results from the last iteration are taken as solutions to minimize the utilization of the dedicated H2 production units. The optimized refinery H2 network scheme is depicted in FIG. 4. The minimum value obtained for the utilization of HPU1/2 was 93.24 MMscfd, which means a reduction of 0.95 MMscfd. Changes that must be implemented in the refinery H2 network to achieve this reduction include: • The H2 makeup to the naphtha hydrotreating (NHT) unit must be residual H 2 from the hydrocracking (HCK) and distillate hydrotreating units (DHT1/2) in a “once-through operation,” without H2 recycle. This process is possible due to the relatively low H2 chemical consumption and availability required for this unit. • Feed to the PSA-ROG will then be a combination of residual H2 from the HCK and DHT1/2 units, along with residual H2 from the NHT unit. • The feedrate to the PSA-ROG is reduced; therefore, recovered H2 to the header is also reduced. The NHT unit will not consume H 2 from the header, compensating the former effect. • Service of the NHT compressor must be changed from a recycle compressor to a residual H2 booster to feed the PSA-ROG.

TABLE 4. H2 sinks

Tag

Pressure, psig

H2 purity, vol%

Flow, MMscfd

FG

100

Notes2

PSA FG, NHT

Recycle NHT

200

Notes

10.44 (max.)

NHT3

290

84 (min.)

Notes4

3

MU-DHT

290

MU-HCK

2

Possible sources1

NHT, H2 header, Off HCK/CHT, DHT1/2, HCK Recycle NHT, H2 header, Off HCK/DHT, DHT1/2, HCK

2

32.65 (max.)

H2 header

2

Notes

290

Notes

78.47 (max.)

H2 header

H2 header5

300

99.9 rated

118.13 (max.)

HPU1/2, PSA ROG

PSA ROG

310

50 (min.)

23.7 (max.)

Recycle NHT, Off HCK/DHT, DHT1/2, HCK

Recycle DHT1

1,150

Notes

2

58.93 (max.)

DHT1

Recycle DHT2

1,150

Notes2

58.93 (max.)

DHT2

6

DHT1

7

4

1,300

98 (min.)

Notes

Recycle DHT1, MU DHT, HCK

DHT27

1,300

92 (min.)

Notes4

Recycle DHT2, MU DHT, HCK

Recycle HCK

1,915

Notes2

486.67 (max.)

HCK

2,100

92.6 (min.)

Notes

1

4

HCK Recycle HCK, MU HCK

The possible sources to be linked to certain sinks were limited to those with a higher pressu re, to ensure a “no new investment” solution, since no new compresso r will be required to implement the optimized H2 distribution. This constraint can be removed, if a given source with lower pressure than the corresponding sink is selected during the optimization; then, a new compressor must be studied as an alternative. 2 Calculated value during the optimization, based on the network mass balance. 3 Considering the lower operational pressure of the NHT unit, the optimization will be used to determine whether this compressor can use residual H2 from other units instead of H2 from the H2 header; therefore, other compressors and units are selected as possible sources. 4 Constraints for H2 flow to process units (NHT, DHT1/2 and HCK) are defined based on their H2 chemical consumption and H2 availability requirement (minimum total H / 2 hydrocarbon ratio at the reactor inlet). For this case study, chemical consumption and availability are defined as a function of factors and the hydrocarbon feedrate, as follows: • H2 chemical consumption: NHT: 271 scfb � 14,925 bpsd = 4.04 MMscfd (pure H2 ) DHT1: 334 scfb × 35,000 bpsd = 11.69 MMscfd (pure H2 ) DHT2: 360 scfb × 35,000 bpsd = 12.6 MMscfd (pure H2 ) HCK: 1,729 scfb × 35,000 bpsd = 60.52 MMscfd (pure H2 ) • H2 availability: NHT: 1.9 × chemical consumption DHT1: 5.4 × chemical consumption DHT2: 5.1 × chemical consumption HCK: 7.9 × chemical consumptio n. 5 The H2 header is a virtual unit that collects high-purity H2 from the HPU1/2 (dedicated H2 production units) and the H2 purification/concentration unit PSA ROG to supply to makeup compressors and the NHT unit. 6 It must be noted that the PSA ROG as the concentration unit has three concentrati on parameters: rated concentration s of 99.9 vol% and 43 vol% for the product and the residual gas, respectively, and a concentration constraint of a minimum of 50 vol% for the mixed offgas feed. 7 The optimization is used to determine whether DHT units can use residual H2 from the HCK unit.

50


Process Optimization Although the H2 reduction from the HPUs is marginal (1% reduction), the necessary changes are of minor scope, consisting mainly of piping modifications, since no new process equipment is needed. Therefore, after an economical assessment of the required modifications, it was concluded that the implementation of the proposal was economical. Regardless of the results, the proposed method was proven useful to easily evaluate a typical refinery’s H2 network by using common tools available to any process engineer. Due to the reduction in feedrate to the PSA ROG, an additional benefit that was identified was the potential recovery of even more H2 by using the spare capacity of the PSA to process residual streams from other units not considered in this study. This benefit is realized since only 44% of its 23.7 MMscfd of capacity would be used after optimization. Potential streams to be considered are shown in TABLE 5. Since the additional residual streams are at a low pressure, a new compressor would be needed to recover the H 2 from these streams. To make the PSA ROG suitable to process the new streams, an adsorbent change needs to be considered, along with a 5% increase in capacity. These modifications on the PSA ROG design are feasible and would increase the H 2 recovery by an additional 4 MMscfd. After implementing all of the proposed modifications, the utilization of the HPUs would be reduced by 4.95 MMscfd (5.2%). Takeaway. A refinery’s H2 distribution network can be opti-

mized by using common tools available to process engineers, such as commercial spreadsheet software, without using sophisticated optimization tools that are not often available in refineries. The proposed method was proven to be useful, and it obtained reliable outcomes that were in accordance with the actual refinery operation. The method was relatively easy to implement and did not require any investment in new process equipment. Further analysis of the results from the primary optimization project showed additional H2 recovery (less utilization of the dedicated HPUs). However, investment in new process equipment would be required. NOTE

Microsoft Excel and Solver were used for this study because of the availability of such tools at the time the study was carried out, and must be taken as only an example; therefore, this study does not recommend any spreadsheet software in particular. ACKNOWLEDGM ENT

The author acknowledges Sebastian Moreno, Ricardo Cortes and Luis C. Ballestas from Ecopetrol’s Cartagena refinery for contributing to the literature and data review at the beginning of this study. REFERENCES 1

Aitani, A. M., “Processes to enhance refinery hydrogen production,” Hydrogen Energy, Vol. 21, Iss. 4, 1996. 2 Alves, J., “Analysis and design of refinery hydrogen distribution systems,” PhD thesis, University of Manchester, UK, 1999. 3 Bhat, S. A. and J. Sadhukhan, “Process intensification aspects for steam methane reforming: An overview,” AIChE Journal, Vol. 55, Iss. 2, 2009. 4 Holladay, J. D., J. Hu, D. L. King and Y. Wang, “An overview of hydrogen production technologies,” Catalysis Today, Vol. 139, Iss. 4, 2009. 5 Kothari, R., D. Buddhi and R. L. Sawhney, “Comparison of environmental and economic aspects of various hydrogen production methods,” Renewable and Sustainable Energy Reviews, Vol. 12, Iss. 2, 2008. 6 Alves, J. J. and G. P. Towler, “Analysis of refinery hydrogen distribution systems,”

Recovered H2

6.91 MMscfd

H2 header 99.9%

Fuel gas header

RE NHT 5.37 MMscfd 71.7%

RE 71.46 MMscfd HPU1/2

93.24 MMscfd

MU

HCK

PSA ROG 5.03 MMscfd 85%

RE 13.87 MMscfd

DHT1 Residual H2 header 85%

RE

MU 14.82 MMscfd

2.83 MMscfd 43%

DHT2

FIG. 4. Refinery 2H network after optimization.

TABLE 5. Residual

streams for further H 2 recovery

Unit

Flow, MMscfd H2 concentration, vol%

Optimized feed to the PSA ROG

10.4

78.1

FCCU

12.1

25

HCK/DHT low-pressure offgas

2.4

86

24.9

53.1

Total

Industrial and Engineering Chemistry Research, Vol. 41, Iss. 23, 2002. Zhao, Z., G. Liu and X. Feng, “The integration of the hydrogen distribution system with multiple impurities,” Chemical Engineering Research and Design, Vol. 85, Iss. 9, 2007. 8 Hallale, N. and F. Liu, “Refinery hydrogen management for clean fuels production,” Advances in Environmental Research, Vol. 6, Iss. 1, 2001. 9 Liu, F. and N. Zhang, “Strategy of purifier selection and integration in hydrogen networks,” Chemical Engineering Research and Design, Vol. 82, Iss. 10, 2004. 10 Kumar, A., G. Gautami and S. Khanam, “Hydrogen distribution in the refinery using mathematical modeling,” Energy, Vol. 35, Iss. 9, 2010. 11 Jeon, C. and C. Han, “Byproduct hydrogen network design using pressure swing adsorption and recycling unit for the petrochemical complex,” Industrial and Engineering Chemistry Research, Vol. 50, Iss. 6, 2011. 12 Liao, Z., J. Wang, Y. Yang and G. Rong, “Integrating purifiers in refinery hydrogen networks: A retrofit case study,” Journal of Cleaner Production, Vol. 18, Iss. 3, 2010. 13 Sardashti-Birjandi, M. R. and F. Shahraki, “Off-gases optimization in hydrogen network refinery,” Chemical Engineering and Technology, Vol. 34, Iss. 12, 2011 14 Zhou, L., Z. Liao, J. Wang and D. Hui, “Optimal design of sustainable hydrogen networks,” International Journal of Hydrogen Energy, Vol. 38, Iss. 7, 2013. 15 Jia, N. and N. Zhang, “Multi-component optimization for refinery hydrogen net works,” Energy, Vol. 36, Iss. 8, 2011 16 Zhou, L., Z. Liao, J. Wang, B. Jiang and Y. Yang, “Hydrogen sulfide removal process embedded optimization of hydrogen network,” International Journal of Hydrogen Energy, Vol. 37, Iss. 23, 2012. 17 Umana, B., A. Shoaib, N. Zhang and R. Smith, “Integrating hydroprocessors in refinery hydrogen network optimization,” Applied Energy, Vol. 133, 2014. 7

HERNANDO SALGADO is a Senior Process Engineer in the

refining industry, specializing in the areas of fluid catalytic cracking, hydroprocessing, crude distillation and energy management. He earned a degree in chemical engineering from the Industrial University of Santander in Colombia, and a professional doctorate in engineering ( process design) from Delft University of Technology in the Netherlands. Dr. Salgado has experience in process analysis, troubleshooting, optimization and development, as well as in the economic evaluation of new facilities and modification of existing units. Hydrocarbon Processing | OCTOBER 2017

51

Are you taking full advantage of Hydrocarbon Processing?

Discover all the benefits of being a premium subscriber and gain full access to HydrocarbonProcessing.com DECE M BER 2016 | Hydro roc carbonPro roc cessi sin ng.c .co om

P LANT D ESI GN, E N GI NE E R I NG AND CON ST RUCTI O N Modula larriz iza ati tio on: Th The ke key y to su succ cce ess i in n t to oday’s mark rket et Eff ffiici cie ently tly a an nd sa safe felly rev evam amp p an RF RFC CC a and nd r re egenera rato torr

PRO PR O CESS O OP PTIMIZ IZA ATION Optim imiz ize e s ste team am u usa sage ge in in re refi fin nery fla lare res s

TOP P TOP PR ROJECT AWARDS Win inne ners rs of of th the e top c con onst strruc ucti tio on pro j jec ectt aw award rds s for 2 20 016, as chosen by HP rea eade ders rs

Subscriber Only Benefit Benefitss A subscription includes twelve monthly issues in print or digital format and premium access to HydrocarbonProcessing.com, where you will find: • All the latest issues and Process Handbooks • HP’s extensive archive archive containing containing 10 years of back issues • Receive each upcoming issue of in your Hydrocarbon Processing in choice of print or digital format digital format • HPI Market Data 2017 • Daily e-newsletters Published since 1922, Hydrocarbon provides operational and Processing provides technical information to improve plant reliability, profitability, safety and end-produc e nd-productt quality qual ity..

Subscribe Subscr ibe Today! Log on to HydrocarbonProcessing.com or call +1 (713) 520-4426.

Process Optimization J. NUNEZ, S. PRESLEY and D. RANA,

DuPont Clean Technologies, Overland Park, Kansas

HF alkylation conver conversion sion is finally within reach—Part 2 In Part 1 of this two-part series, recent technology developments to convert hydrofluoric acid (HF) alkylation to sulfuric acid alkylation were discussed. Not only can HF conversion be completed with a much lower capital investment than previously thought, but significant expansion is also possible with little additional cost due to some inherent differences between alkylation technologies that utilize HF and sulfuric acid as the catalyst. The two expansion case studies introduced in Part 1 will be explored in more detail in Part 2 of this series. Part 1 recap. HF conversion is finally within reach, thanks to

the development of proprietary, proprietar y, cost-effective, proven technologies. In conjunction with the conversion from HF to sulfuric acid alkylation, these technologies can offer a significant capacity increase with minimal additional cost. Two of these solutions include: • Gravity-flow HF alkylation conversion conversion technology�

This solution replaces the existing HF acid coolers with conventional reactors.1 The existing HF acid settler will be reused reused with the effluent routed routed back back to to the reactor reactor tube bundles. The light compone components nts in the effluent are flashed flashed across a pressure control valve prior to entering the tube bundle inlet inlet to remov removee the heat heat of reaction. reaction. As in a typical effluent refrigerated alkylation unit, the tube bundle outlet is routed to a new refrigeration section before proceeding to fractionation. The flashed vapors are subsequently compressed, condensed and returned to the reactor feed, providing additional cooling and contributing a significant amount of isobutane to the isobutane-to-olefin ratio. The optimization of acid consumption and alkylate quality can be achieved by staging the acid flow through the unit. The converted acid settler can be segmented such that one large acid settler can function like two or three individual settlers. In most cases, the entire fractionation section can be reused, reused, with little little or no no piping modification modificationss required. required. • Pumped-flow HF alkylation conversion technology�

This solution retrofits the existing horizontal acid settler(s) into a novel reactor(s), which utilizes proprietary mixing and separation equipment. The design of the patentpending novel reactor incorporates equipment that is familiar to refinery operators, has no moving parts and has

a robust design. As with the previous conversion solution, the entire fractionation section can be reused with little or no piping modifications. This solution is intended to be a lower lower-cost -cost solutio solutionn with a slight perform performance ance debit compared to the gravity-flow HF conversion technology. Both solutions offer significant capital cost savings by maximizing the reuse of existing HF alkylation equipment. The two case studies described in this article provide examples of this equipment reuse. Depending on the conversion solution selected, along with the configuration of the existing HF alkylation unit, the estimated cost of conversion is 40%–60% of the cost of a similar-sized grassroots sulfuric acid alkylation unit. Because sulfuric acid alkylation units operate o perate at lower reaction temperatures than HF alkylation units, a new refrigeration section is required for either conversion option. The addition of this refrigeration section, including a refrigerant compressor, is the most significant new construction required. If plot space is available near the existing unit, then the installation of this new equipment can take place during normal operation outside of the turnaround window. This would make it very feasible to complete the remainder of the work during a typical 30 d–45 d turnaround. One of the most attractive aspects of these conversion solutions is the ability to expand the alkylation unit simultaneously with the conve conversion, rsion, with only minim minimal al impa impact ct on cost. This benefit is possible by leveraging leveraging an inherent inherent difference difference between the HF and sulfuric acid alkylation technologies. In both technologies, a large recycle flow of isobutane to the reaction section is required to promote the desirable alkylation reactions, and to suppress polymerization reactions that can negatively impact acid consumption and alkylate quality. In an HF alkylation unit, all isobutane is recycled from the fractionation section, while in a sulfuric acid alkylation unit, only about one-half of the isobutane is recycled from the fractionation section, with the remainder recycled from the refrigeration section. Therefore, it is possible to double the effective capacity of the alkylation unit without requiring significant changes to the fractionation section equipment, thereby making the conversion more cost-effective. Metallurgical considerations. One consideration for any con-

version from HF to sulfuric acid alkylat alkylation ion is metallurgy metallurgy.. Both the HF and sulfuric acid alkylation technologies are similar in that Hydrocarbon Processing | OCTOBER 2017

53

Process Optimization carbon steel is widely used in the majority of the equipment and piping. However, in a few selected services in both technologies,

of the HF alkylation unit that contain this type of alloy are normally not reused. However, because some refiners have upgraded the metallurgy in other parts of the unit, a metallurgy evaluation is recommended. While this special alloy is not suitable for streams consisting of a large concenIt is important to note that there is no tration of high-strength sulfuric acid, it does provide “one-size-fits-all” solution, and optimization good corrosion resistance in hydrocarbon-continuous within each stage is unique for each streams with dispersed sulfuric acid due to the protection of the metal by the hydrocarbon phase. refiner depending on feed composition,

alkylation unit configuration configuration and the condition of the alkylation equipment. carbon steel does not provide adequate corrosion resistance, and upgraded alloys are required. For HF alkylation units, a special nickel-copper alloy is typically used in these cases.2 Although this special alloy can be used effectively in low-strength sulfuric acid services, there is a sharp increase in the corrosion rate for higher acid strengths. Since most sulfuric acid alkylation units operate with acid strengths strengths greater greater than 85 wt%, this special alloy is not compatible in services where large concentrations of sulfuric acid are present. Typically, this is not a concern when converting from HF to sulfuric acid alkylation because those portions

Combined settler effluent

Fine separation coalescing media

Benefits of acid staging. To make HF alkylation

conversion even more attractive to refiners, both pre viously mentione mentionedd solutions take full advan advantage tage of innovative and fully optimized sulfuric acid staging concepts (FIG. 1). The HF acid settler is retrofitted as follows: the bottom portion of the acid settler is segmented into up to three compartments, and nozzles are installed for the emulsion and acid flows. As in a traditional alkylation unit, 3 the sulfuric acid-hydrocarbon emulsion flows from the reactors to the acid settler, and acid is recycled back to the reactor. Bottom nozzles are provided for the acid transfer between stages to allow each reaction zone to operate at an independent acid strength, which minimizes acid consumption. Coalescing media is installed at two different heights to provide bulk and fine separation of sulfuric acid from the hydrocarbons. Combined settler effluent exits the top of the settler and flows through the reactor tube bundles to remove remove the heat of the alkylation alkylation reaction. reaction. Makeup fresh acid is fed to the first stage, with the flexibility to add to intermittent stages, if needed. The spent acid strength is 90 wt% for butylene buty lene alkylation units, with lower acid strengths achievable for other feedstocks to minimize acid consumption. Acid recontactor

Emulsion from reaction stage 3 Acid settler

Bulk separation coalescing media

Emulsion from reaction stage 2

Emulsion from reaction stage 1

Acid recycle to reaction stage 3


Acid cooler

Acid cooler


Depropanizer

Propane stripper

Defluorinators

Propane Feed dryers FC AI

FC

Fresh acid Intermediate 99.2 wt% ~95 wt%

M

Contactor reactor

F C AI

Spent acid ~ 90 wt%*

Intermediate ~92 wt%– 93 wt%

M

Contactor reactor

Olefin feed

AI

Contactor reactor

M

Reactor effluent

M/U isobutane

ASO Washer

ASO HC relief header

ASO surge

Alkylate Reuse Repurpose Abandon/demo Acid Hydrocarbon

Refrigerant recycle Olefin and isobutane

n-Butane

ASO

Acid relief header

Olefin and isobutane

KOH treater

Rerun

Olefin and isobutane

Isostripper

Acid relief neutralizer FIG. 1. Conventional

reactors integrated with an HF acid settler.1 *Lower spent acid strengths are achievable with other feedstocks.


FIG. 2. The

original HF alkylation unit from Case Study 1.

Process Optimization The target acid strength for each stage is determined to minimize the acid consumption and maximize alkylate quality. A patent-pending olefin and acid staging solution has also been developed to optimize variables such as reactor feed composition, reaction temperature and acid strength for each acid stage. This new staging solution can reduce acid consumption by up to 40%, which will represent significant savings for refiners. In addition to acid consumption, several other benefits, such as higher alkylate octane, lower alkylate distillation final boiling point and increased throughput, are possible by applying this improved technique for olefin and acid staging. It is important to note that there is no “one-size-fits-all” solution, and optimization within each stage is unique for each refiner depending on feed composition, alkylation unit configuration and the condition of the alkylation equipment. Case Study 1. As mentioned in Part 1, Case Study 1 involved

the conversion of a gravity-flow HF alkylation unit with a single TABLE 1. List

vertical acid settler and four acid coolers (FIG. 2). The fractionation equipment for this unit included an isostripper column and a depropanizer column. Effluent from the acid settler was split into a reflux stream to the isostripper and a feed stream to the depropanizer. The depropanizer overhead product was propane, and the bottoms product provided feed to the isostripper. The depropanizer also included a side stream, which provided approximately 20% of the recycled isobutane to the reaction section. The isobutane-rich isostripper overhead stream was condensed and sub-cooled, then pumped to the reaction section to provide the remaining 80% of the recycled isobutane. A side stream from the isostripper was the n-butane product stream, and the bottoms stream was the alkylate product. The design capacity for this unit was 13.5 Mbpsd of alkylate product, with a feed stream consisting of mixed C3 and methyl tertiary butyl ether (MTBE) raffinate. The proprietary conversion solution for this case involved adding four reactors,4 a refrigeration section and feed/effluent

of major equipment in Case Study 1

Existing equipment reused or repurposed

New equipment required for conversion

Existing equipment NOT used for conversion

Acid blowdown drum (remote)

Alkylation reactors

Acid coolers

Acid neutralization pit

Dry alumina adsorption

Acid rerun

Acid recontactor

Feed/effluent exchangers

ASO heater

Acid relief neutralizer

Isostripper accumulator

ASO pumps

Acid relief neutralizer circulation pump

Refrigerant accumulator

Propane KOH treater

Acid settler

Refrigerant condensers

Propane stripper

Acid storage drum (remote)

Refrigerant recycle pumps

Stripping IC 4 superheater

Alkylate cooler

Refrigeration compressor

ASO surge drum

Suction trap/flash drum

ASO surge drum circulation pump ASO washer ASO washer circulation pumps Depropanizer Depropanizer accumulator Depropanizer charge pumps Depropanizer condensers Depropanizer feed/bottoms exchangers Depropanizer feed/recycle IC4 exchangers Depropanizer reboiler Depropanizer reflux pumps Depropanizer side stream cooler Feed dryers Isostripper Isostripper charge pumps Isostripper condensers Isostripper overhead pumps Isostripper reboiler Normal butane condensers Propane condensers Propane defluorinators Hydrocarbon Processing | OCTOBER 2017

55

Process Optimization Refrigeration Compressor Suction trap/ flash drum

Fresh acid

Original equipment New equipment Acid Hydrocarbon


Acid settler

PCV

PCV

Propane Depropanizer Propane treating (defluorinators)

Olefin feed Feed Acid coalescer M/U isobutane dryers (acid recontactor)


n-Butane

Spent acid aftersettler (ASO washer)

HC relief header Acid relief header

Alkylate Isostripper


Acid blowdown drum (ASO surge)

Recovered HC

heat exchangers, while repurposing existing equipment to be used for the acid coalescer, dry alumina adsorption, propane product treatment and acid blowdown. TABLE 1 provides a list of existing and new major equipment in the alkylation unit. Most modern HF alkylation units also have remote HF acid storage vessels and remote HF acid blowdown drums. Depending on the location of these vessels and the requirements of the conversion solution offered, these may be utilized as spent acid aftersettlers, acid blowdown drums, suction trap/flash drums or supplemental acid settlers. Potential opportunities to reuse these vessels will be determined on a case-by-case basis. To maximize product quality and to minimize acid consumption, the technology provider used patent-pending designs to segregate olefin feed and to stage acid flows between the reaction zones. FIG. 3 illustrates a simplified process flow diagram of the conversion solution for this HF alkylation unit; however, olefin feed segregation and sulfuric acid staging are not depicted in this drawing. As the isostripper was designed to provide 80% of the required recycled isobutane, a significant capacity increase was achieved by using the existing equipment. By optimizing the isobutane-to-olefin ratio and maximizing existing equipment capacity, the technology provider increased capacity by nearly 70%, while maintaining a similar product quality. TABLE 2 pro vides a summary of feed and product streams for the conversion solution offered for Case Study 1. Since a conventional sulfuric acid alkylation technology was employed in this conversion solution, the technology provider can accurately predict alkylate

Spent acid TABLE 2. Feed

and product streams in Case Study 1

Olefin Isobutane Propane n-Butane Alkylate feed feed product product product

FIG. 3. HF alkylation unit converted to a sulfuric acid alkylation unit in Case Study 1.

Volume flow, bpsd 17,727

3,634

473

22,601

Composition, vol%

Acid recontactor

Acid settler Propane stripper Defluorinators Acid cooler

14,343

Ethane

0.09

0

0.43

0

0

Propane

13.19

8.53

98

0

0

i-Butane

9.04

87.63

1.57

2.5

0.03

n-Butane

5.04

3.85

0

92.59

4.46

0

0

0

0

0

Propylene

31.35

0

0

0

0

Butylenes

39.85

0

0

0

0

Amylenes

0.97

0

0

0

0

C5+

0.47

0

0

4.91

95.52

1,3-butadiene

Acid cooler Propane Main fractionator

KOH treater

Olefin feed

M/U isobutane

Defluorinators

Feed dryers

Rerun

n-Butane

ASO ASO/KOH separator KOH treater Alkylate

HC relief header Acid relief header Acid relief neutralizer

Reuse Debutanizer Repurpose Abandon/demo Acid Hydrocarbon

FIG. 4. The original HF al kylation unit in Case Study 2.

56


TABLE 3. Alkylate

properties in Case Study 1

% Propylene/total olefins (feed)

43.4

% Isobutylene/total butylenes (feed)

0.5

RON

95

(R+M)/2

93.5

D-86 T90, °F (°C)

< 255 (124)

D-86 EP, °F (°C)

< 380 (193)

Acid consumption, lb/gal

0.45–0.5

Alkylate Reid vapor pressure (RVP), psi

6

Alkylate sulfur, parts per million (ppm)

<2

Process Optimization properties utilizing proprietary correlations developed from decades of pilot plant testing and real-world empirical data. TABLE 3 provides a summary of predicted alkylate properties. Case Study 2. This study also involved the conversion of a

gravity-flow HF alkylation unit with a single vertical acid settler and four acid coolers ( FIG. 4). The fractionation equipment for this unit included a main fractionator column and a debutanizer column. Effluent from the acid settler provided feed to the main fractionator. The main fractionator overhead product was the propane product, and the bottoms product stream pro vided feed to the debutanizer. The main fractionator included a side stream, which provided 100% of the recycled isobutane to the reaction section. The debutanizer overhead stream was the n-butane product stream, and the debutanizer bottoms stream was the alkylate product. The design capacity for this unit was 10 Mbpsd of alkylate product, with a feed stream consisting of mixed C3s and fluid catalytic cracking-mixed C4s. The proprietary conversion solution for this case involved adding six reactors,4 a refrigeration section, feed/effluent heat exchangers and dry alumina adsorption, while repurposing existing equipment to be used for the acid coalescer, propane product treatment and acid blowdown. TABLE 4 provides a list of TABLE 4. List

existing and new major equipment in the alkylation unit. As discussed in Case Study 1, most modern HF alkylation units have remote HF acid storage vessels and remote HF acid blowdown drums. Depending on the location of these vessels and the requirements of the conversion solution offered, these may be utilized as spent acid aftersettlers, acid blowdown drums, suction trap/flash drums or supplemental acid settlers. The potential opportunities to reuse these vessels will be determined on a case-by-case basis. To maximize product quality and to minimize acid consumption, the technology provider employed patent-pending designs to segregate the olefin feed and to stage acid flows between reaction zones. FIG. 5 illustrates a simplified process flow diagram of the conversion solution for this HF alkylation unit; however, olefin feed segregation and sulfuric acid staging are not depicted on this drawing. As the main fractionator was designed to provide 100% of the recycled isobutane requirement, a significant capacity increase was achieved by using the existing equipment. By optimizing the isobutane-to-olefin ratio and maximizing existing equipment capacity, the technology provider increased capacity by more than 175%, while maintaining similar product quality. TABLE 5 provides a summary of feed and product streams for the conversion solution offered for Case Study 2. Because a conventional

of major equipment for Case Study 2

Existing equipment reused or repurposed

New equipment required for conversion

Existing equipment NOT used for conversion

Acid blowdown drum (remote)

Alkylation reactors

Acid coolers

Acid neutralization pit


Acid rerun

Acid recontactor

Feed/effluent exchangers

ASO heater


Net effluent pumps

ASO pumps

Acid relief neutralizer circulation pump

Recovered hydrocarbon pump

n-Butane defluorinators

Acid settler


n-Butane KOH treater

Acid storage drum (remote)

Refrigerant condensers

Propane KOH treater

Alkylate cooler

Refrigerant recycle pumps

Propane stripper

ASO/KOH separator

Refrigeration compressor

Stripping IC4 superheater

Debutanizer

Spent acid pump

Debutanizer accumulator

Suction trap/flash drum

Debutanizer reboiler Debutanizer reflux pumps Feed dryers Main fractionator Main fractionator accumulator Main fractionator condensers Main fractionator feed pumps Main fractionator feed/bottoms exchangers Main fractionator feed/recycle IC 4 exchangers Main fractionator reboiler Main fractionator reflux pumps Main fractionator side stream cooler Normal butane condensers Propane condensers Propane defluorinators Hydrocarbon Processing | OCTOBER 2017

57

Process Optimization TABLE 5. Feed

Refrigeration Compressor

and product streams in Case Study 2

Olefin Isobutane Propane n-Butane Alkylate feed feed product product product Volume flow, bpsd 26,416

13,500

3,509

1,399

27,810

Suction trap/ flash drum

Fresh acid


Acid settler

Composition, vol% Ethane

0.03

0

0.24

0

0

Propane

10.66

1.99

97.5

0

0

i-Butane

20.04

95

2.25

13.22

0.11

n-Butane

6.42

3.01

0.02

85.78

3.24

1,3-Butadiene

0.06

0

0

0

0

Propylene

28.67

0

0

0

0

Butylenes

30.42

0

0

0

0

Amylenes

0.65

0

0

0

0

C5+

3.05

0

0

1

96.65

PCV

PCV

Main fractionator Feed dryers Olefin feed Dry alumina adsorption

M/U isobutane TABLE 6. Alkylate

properties in Case Study 2

% Propylene/total olefins (feed) % Isobutylene/total butylenes (feed)

HC relief header

48 32.5

RON

93

(R+M)/2

92 < 275 (135)

D-86 T90, °F (°C)

< 400 (204)

D-86 EP, °F (°C)

0.5–0.55

Acid consumption, lb/gal

6

Alkylate RVP, psia

Propane Propane treating (defluorinators)

Acid coalescer (acid recontactor)

Acid relief header

Acid relief neutralizer Spent acid

n-Butane

Spent acid aftersettler (remote acid storage) Acid blowdown drum (ASO/KOH separator) Recovered HC

Alkylate Debutanizer Original equipment New equipment Acid Hydrocarbon

FIG. 5. HF alkylation unit converted to a sulfuric acid alkylation unit in Case Study 2.

<2

Alkylate sulfur, ppm

2

sulfuric acid alkylation technology was employed in this conversion solution, the technology provider can accurately predict alkylate properties utilizing proprietary correlations developed from decades of pilot plant testing and real-world empirical data. TABLE 6 provides a summary of predicted alkylate properties. Takeaway. The technology developments described in this ar-

ticle offer refiners a viable, safe alternative to the operation of existing HF alkylation units, which are attracting unprecedented negative pressure from communities and regulatory agencies due to safety and environmental concerns. At 40%–60% the cost of a new unit, the cost to convert from HF to sulfuric acid alkylation using these new conversion solutions is significantly lower than any other option available on the market. Additionally, by taking advantage of a key difference in how isobutane is circulated in HF and sulfuric acid alkylation units, a sizable capacity increase at the same time as the conversion is feasible with little additional capital investment. Finally, the solutions discussed here are designed for maximum reliability and operability, utilizing equipment that is very familiar to refinery process operators and maintenance personnel. Whether refiners are interested in conversion or expansion, these solutions are worth exploring. End of series. Part 1 of this article appeared in September. NOTES 1

Conventional reactor design refers to DuPont’s STRATCO® Contactor™ reactor technology.

58


Refers to Monel 400 nickel-copper alloy. Refers to a DuPont STRATCO alkylation unit. 4 Refers to DuPont’s STRATCO Model 74 Contactor reactors. 3

CORRECTION

Due to a drafting error, Fig. 3 in Part 1 shows an incorrect stream location. The isobutane recycle stream is shown to connect to the ASO stream. However, it should connect to the feed stream immediately downstream of the feed dryers. JASON NUNEZ is a Senior Technical Service Engineer at

DuPont Clean Technologies for the alkylation and hydroprocessing businesses. He has more than 14 yr of experience in the refining and petrochemical industries. Prior to joining DuPont in 2014, he served in various roles at Citgo Petroleum Corp. and the Saudi Aramco Mobil Refinery. Mr. Nunez holds BS degrees in chemical engineering and environmental science from McNeese State University. SHANE PRESLEY is the Technical Service and Development

Manager at DuPont Clean Technologies for the alkylation and hydroprocessing businesses. Mr. Presley has more than 18 yr of refining and technology licensing experience, including roles at Chevron, ExxonMobil and DuPont. He holds a BS degree in chemical engineering from Mississippi State University. DIWAKAR RANA is the Research Investigator at DuPont Clean

Technologies for the STRATCO alkylation business. Dr. Rana has more than 11 yr of experience in the oil, gas and biofuels industries, including roles with Alfa Laval, Fluor, Lloyd’s Register, Black and Veatch and DuPont. He holds a BCh (Honors) in chemical engineering from Deenbandhu Chhotu Ram University of Science and Technology, India, an MS degree in chemical engineering from the Illinois Institute of Technology and a PhD in chemical engineering from Washington State University. Since 200 9, Dr. Rana has been a licensed Professional Engineer in the state of Texas.

Regional report

L. NICHOLS, Editor/Associate Publisher

Diversification: The catalyst for the Middle East’s downstream transformation Over the past several years, the Middle East has made substantial investments to increase its downstream processing capacity. These investments will continue through the early 2020s. The region is making a deliberate move to diversify its products portfolio in the refining, petrochemical and gas processing/LNG sectors. Hydrocarbon Processing’s Construction Boxscore Database is tracking nearly 400 active downstream construction projects in the Middle East. These projects equate to nearly $280 B in capital expenditures (CAPEX) through the early 2020s. More than 85% of the region’s active projects are in the refining and petrochemical sectors, and more than half of these active projects are in the preconstruction stages. A full breakdown of active projects in the Middle East by activity level is: • Engineering�16% • FEED�12% • Proposed/planning�21% • Study�6% • Under construction�45%. Saudi Arabia continues to hold the greatest share of total active projects in the region (FIG. 1). According to BP’s Statistical Review of World Energy 2017, the Middle East’s oil consump-

tion is greater than 9.4 MMbpd. Over the last decade, the region’s appetite for oil has increased by more than 2.7 MMbpd. In turn, the Middle East plans to expand its refining capacity by approximately 1.5 MMbpd by 2022. Traditionally, Middle Eastern refineries have simple configurations and high fuel oil yields, partly due to high power generation requirements. However, this trend has seen a complete reversal in recent years, with the region building a new generation of highly complex plants. These facilities, combined with upgrades and expansions at existing plants, are radically changing the product mix. New unit configurations include hydrocracking, catalytic cracking and hydrotreating capacities designed to minimize fuel oil output and maximize production of middle distillates, diesel and gasoline. The region’s major crude oil refining initiatives focus on refining and petrochemical integration and clean-fuels programs. The Middle East is building some of the largest and most complex refineries in the world. These high-capacity mega-facilities will enable the region to produce higher-quality refined products to diversify the region’s products portfolio away from crude oil exports. According to the International Energy Agency (IEA), Mid-

Opening photo: Saudi Arabia will be the Middle East’s leader in downstream capital investments. The country has already completed major

capital-intensive projects. The 400-Mbpd SATROP refinery was developed by a JV of Saudi Aramco and Total SA. Saudi Arabia also commissioned the $8.5-B Rabigh 2 mega-project (pictured) in late 2016. The complex produces a variety of high-value-added petrochemical products. Hydrocarbon Processing | OCTOBER 2017

59

Regional Report dle East oil demand is forecast to increase from 8.5 MMbpd in 2016 to nearly 9.7 MMbpd by 2023. With the increase in regional oil consumption, some refined products destined for export may be diverted to domestic markets. According to the Gulf Petrochemicals and Chemicals Association (GPCA), the region’s total petrochemical production capacity is 150 MMtpy. The region is slated to see a substantial increase in petrochemical production. Several nations are building, or plan to build, mixed-feed crackers, along with capacity for ethylene derivatives, ammonia/urea and other petrochemicals. The Boxscore Database is tracking nearly $90 B in petrochemical CAPEX in the region. The Middle East has enjoyed a petrochemical feedstock ad vantage against the rest of the world, but is facing competition from cheap US shale gas. The Middle East is also starting to restrict the flow of ethane. Regional countries, such as Saudi Arabia, have even raised ethane prices to cut back on subsidies. The ethane price increase has forced some producers to utilize propane and naphtha feedstocks in their crackers. The Middle East still holds a price advantage, but the region’s competitiveness against other producers, such as the US, has dwindled. Regardless, the Middle East will add a substantial amount of new petrochemical capacity through the early 2020s. The petrochemical investments are part of the region’s plans to di versify its products portfolio to include more refined fuels and petrochemicals. The following sections contain detailed overviews of the ma jor refining and petrochemical trends happening in the region. Bahrain. The country’s state-owned oil and gas company,

Bahrain Petroleum Co., is progressing with the Sitra refinery expansion project. The $5-B expansion will increase the refinery’s distillation capacity from 267 Mbpd to 360 Mbpd. The project, referred to as Bahrain’s “Refinery Master Plan Project,” will produce high-quality refined products, as well as increase the plant’s energy efficiency. Iran. Over the past year, Iran has made great strides to develop

its downstream processing infrastructure. However, the future of the country’s downstream buildout hinges on foreign investment, technology and construction expertise. These elements are also critical for expanding exploration and production op-

Other—1% UAE—9% Turkey—8% Saudi Arabia—25% Qatar—6% Oman—11% Kuwait—11% Iraq—17% Iran—10% Bahrain—2%

FIG. 1. Market

share analysis of active projects in the Middle East. Source: Hydrocarbon Processing’s Construction Boxscore Database.

60


erations in the country’s South Pars gas fields, as well as for expansion, upgrading and grassroots capacity buildouts in its refining, petrochemical and gas processing/LNG sectors. With the easing of Western sanctions, Iran is making a great effort to attract foreign investment and technology, and that effort is starting to see positive results. Iran is investing heavily to raise its domestic refining capacity from 1.9 MMbpd in 2016 to 3.2 MMbpd by 2020. The increase in domestic refining capacity is part of the country’s 20-yr Outlook Plan between 2005 and 2025. Known as the Iran Project, the country’s goal is to increase its refining capacity to 3.4 MMbpd by 2025. The country hopes to accomplish this ambitious goal by constructing new refineries, as well as by expanding and upgrading those plants already in operation. The bulk of the country’s planned grassroots facilities focus on processing condensate from the country’s South Pars fields. The country’s grassroots refinery buildout will cost more than $20 B. This program includes the construction of the Persian Gulf Star, Siraf, Siraz Gas Condensates, Bahman Geno Heavy Oil and Anahita refineries. Iran is also seeking approximately $14 B in investments to upgrade its refining network to produce high-quality, low-sulfur fuels. This program focuses on upgrading the Esfahan, Tabriz, Tehran, Bandar Abbas and Abadan refineries. The country is also seeking more than $30 B in foreign in vestment to nearly triple its domestic petrochemical capacity. The country seeks to monetize vast reserves of natural gas from its South Pars fields to produce petrochemicals. According to Farnaz Alavi, Director of Planning and Development for Iran’s National Petrochemical Co., Iran will invest up to $60 B in more than 25 petrochemical projects to increase domestic petrochemical capacity from 60 MMtpy in 2017 to 160 MMtpy by 2025. This investment would include the construction of new ethylene and ethylene derivatives capacity, ammonia/urea projects, methanol-to-olefins (MTO) plants, methanol capacity, and others. The new production capacity will not be possible without foreign investment and technology. With the easing of Western sanctions, the country is increasing its cooperation with more international firms. One of the country’s most ambitious projects is the Mokran Petrochemical Complex in Chabahar. According to the Iran Project’s website, the facility will include two olefins plants, an aromatics plant, an MTO plant, a crystal melamine plant, four urea/ammonia plants, four methanol/ammonia plants and five methanol plants, along with utilities and terminal infrastructure. The three-phase construction plan could house up to 30 units, with a total production capacity of 25 MMtpy. Phase 1 is expected to be completed in 2020, with the remaining two phases to begin operations in 2022 and 2024, respectively. The total cost of the three phases is estimated at nearly $12 B. Iraq. Within the last decade, the nation’s crude oil consumption

has more than doubled to 4.5 MMbpd. Iraq’s crude oil production is more than adequate to meet demand; however, the nation’s refining network suffers from low utilization rates. This predicament hinders Iraq’s ability to satisfy domestic demand. To mitigate refined product imports, the Iraqi government announced an ambitious plan to boost domestic refining ca-

Regional Report pacity. The plan was dubbed the second National Development Plan (NDP) for 2013–2017, and detailed the construction of four grassroots refineries. These four facilities were envisioned with a collective price tag of approximately $20 B, and would have boosted the country’s domestic refining capacity by 740 Mbpd. However, due to in-country fighting with ISIS and continued economic and political instability, only a fraction of the planned refining capacity is expected to begin operations by the early 2020s. The four refineries would have been located in Karbala, Kirkuk, Missan and Nassiriya. At the time of publication, only the $5-B, 140-Mbpd Karbala refinery has seen tangible progress. Although the country is experiencing difficulty in building the projects, Iraq’s oil ministry has proposed the construction of five additional refineries. These facilities, which could add up to 610 Mbpd of new refining capacity, would be constructed in Anbar, Diwaniya, Fao, Kut and Samawa. These refineries would operate under a build-own-operate or build-owntransfer model. Each of the plants would be able to produce Euro 5-standard fuels. These projects are in the early planning stages. If they are greenlighted, operations are unlikely to begin until the early- to mid-2020s. Kuwait. Domestic

consumption has risen steadily due to increased petroleum-fired electricity generation. Rising demand and a failing infrastructure have prompted Kuwait to develop the Clean Fuels Project (CFP) and the New Refinery Project (NRP)�also known as the Al-Zour refinery project. These projects represent a collective CAPEX of approximately $33 B. The $17-B CFP is designed to upgrade and integrate the Mina Abdullah and Mina Al-Ahmadi refineries. The project includes the installation of 39 units, the revamp of seven units, and the closure of seven units. With the completion of the CFP, Kuwait will be able to produce high-quality, low-sulfur transportation fuels. The sulfur content in gasoline will decrease from 500 ppm to less than 10 ppm. Benzene and aromatics concentrations will also decrease. Bunker fuel oil sulfur content will decrease from 4.5 ppm to 1 ppm, and the maximum sulfur content of full-range naphtha will drop from 700 ppm to 500 ppm. Kuwait is also constructing the largest refinery in the Middle East. The $16-B grassroots Al-Zour project will process domestic heavy crude oil, supply power generation plants in the country with environmentally friendly fuel and provide alternatives to gas imports for heavy fuel use. The mega-project is under construction, with operations scheduled to begin in late 2019. The Al-Zour refinery will also provide propane and naphtha feedstocks to the Olefins III-Aromatics II complex. The $8-B project will consist of a mixed-feed cracker to produce 940 Mtpy of polypropylene (PP), 1.4 MMtpy of paraxylene, 420 Mtpy of gasoline and 209 Mtpy of additional fuels. The complex will be integrated into the Al-Zour refinery, with operations scheduled to begin in 2Q 2022. Oman. The country is

investing more than $14 B in new downstream infrastructure. These investments will help diversify the nation’s products portfolio, which is a major focus of Oman’s Vision 2020 plan. The country plans to take advantage of its strategic location on the Arabian Peninsula to build additional refining, storage terminal and petrochemical capacity.

The nation is investing nearly $13 B in the Duqm Refinery and Petrochemicals and Liwa Plastics Industries complexes. The $6-B Duqm project consists of a 230-Mbpd refinery and petrochemicals complex. The refinery will produce diesel, jet fuel, naphtha and LPG. Operations are scheduled to begin in 2020. Once completed, the complex will accomplish the major goal of developing the Duqm industrial area, which is a pillar of the nation’s Vision 2020 plan. The industrial zone’s strategic location will allow cargoes to travel in and out of Oman without having to traverse the Strait of Hormuz. Oman Oil Refineries and Petroleum Industries Co. completed the $2-B, 60-Mbpd Sohar Refinery Improvement Project in 2017. The expansion of the Sohar refinery allows Oman to increase gasoline, diesel, jet fuel, naphtha and LPG production to meet increasing domestic demand, as well as to enhance the country’s refined fuel export offerings. The facility will also provide most of the feedstock for the $6.5-B Liwa Plastics project. Located in Sohar, the new petrochemical complex will consist of an 800-Mtpy naphtha cracker, along with derivatives capacity. Also in Sohar, Oman International Petrochemical Industries Co. is developing a worldscale polyethylene terephthalate/purified terephthalic acid (PET/PTA) plant. The $680-MM facility will produce up to 1.1 MMtpy of PTA and 250 Mtpy of PET. Saudi Arabia. Leading the way in downstream capacity invest-

ments in the Middle East, Saudi Arabia’s Transformation Program�also known as Vision 2030�calls for the Kingdom to diversify its economy away from a reliance on oil exports. This plan includes slashing subsidies, boosting renewable energy usage, developing a financial hub in the country, investing heavily in building out the country’s downstream refining, petrochemical and gas processing capacities, and other initiatives. By 2020, Saudi Aramco plans to spend as much as $150 B to become the global leader in refined fuels and petrochemicals production. This plan includes increasing domestic refining and petrochemicals capacity, as well as boosting ownership stakes in downstream processing facilities internationally. The company’s goal is to increase its global refining capacity from 5.4 MMbpd to 8 MMbpd–10 MMbpd by 2025. Saudi Arabia’s domestic refining capacity is set to grow from 2.9 MMbpd to 3.3 MMbpd by 2019. The additional 400 Mbpd of refining capacity will come from the startup of the $7-B Jazan refinery. The plant will provide refined products to the country’s western and southern regions, with excess refined products to be exported. The Jazan refinery will be the third mega-scale refinery to be completed in the Kingdom since 2014. Over the past 4 yr, Saudi Arabia has added 800 Mbpd of new refining capacity with the startup of the SATORP and YASREF refineries. Saudi Aramco also plans to become a world leader in the production of clean fuels, and has established a program to achieve mandatory sulfur specifications for gasoline and diesel. The Kingdom is seeking to reduce sulfur content in diesel and gasoline to 10 ppm, and to lower benzene content in gasoline to 1%. Saudi Arabia has instituted multiple projects for the increased production of high-quality, ultra-low-sulfur transportation fuels at the Jazan, Ras Tanura, Riyadh, PetroRabigh and Yanbu facilities. Saudi Arabia will be a leader in the region’s petrochemical capacity buildout. According to the GPCA, Saudi Arabia has a Hydrocarbon Processing | OCTOBER 2017

61

Regional Report total petrochemical production capacity of nearly 100 MMtpy, and the country is investing billions of dollars to further expand domestic chemical production. Although the Kingdom enacted price reforms on ethane and methane in 2016, significant petrochemical capacity additions are underway. Due to the severe drop in crude oil prices in 2014 and 2015, the Saudi Arabian government enacted new reforms to reduce costs. One of the most significant reforms was the increase in ethane and methane prices. The price of ethane increased from $0.75/MMBtu to $1.75/MMBtu, and the price of methane increased from $0.75/MMBtu to $1.25/MMBtu. The ethane and methane price reforms represent an increase of 133% and 67%, respectively. The country is still the most cost-effective location for ethylene production, but Saudi Arabia has lost some of its competitiveness to the US, which utilizes cheap shale gas to produce ethylene. Regardless, Saudi Arabia is continuing its petrochemical expansion program. The $20-B Sadara mega-complex started up its mixed-feed cracker in 2016, and the last unit was put into operation in August. The complex can produce more than 3 MMtpy of petrochemical products. Saudi Arabia also commissioned the $8.5-B Rabigh 2 megaproject in late 2016. The project included an expansion of the existing ethane cracker, the construction of an aromatics complex and the expansion of a facility to process 30 MMcfd of ethane and approximately 3 MMtpy of naphtha as feedstock for a variety of high-value-added petrochemical products. Additional petrochemical infrastructure is under development in the Kingdom. In mid-2017, Saudi Aramco and Total announced their plan to conduct a feasibility study for a new, world-scale, mixed-feed cracker in Jubail. The 1.5-MMtpy cracker would utilize feedstock from the two companies’ 400Mbpd Satorp refinery. If built, the plant is estimated to cost $3 B–$5 B. However, the project would need to secure a vast amount of new natural gas feedstock. Total does not plan to begin detailed studies on the project until the partners are able to obtain additional gas supplies. Saudi Aramco has also teamed up with Saudi Basic Industries Corp. (SABIC) to conduct feasibility studies on oil-tochemicals technology. Phase 1 of the study has been completed. If greenlighted, the project could cost up to $20 B. The preliminary timeline would see the plant beginning commercial operations in 2024. The products would be exported to markets in Asia and Europe. The Kingdom is expanding its phosphate fertilizer production. Saudi Arabian Mining Co.’s (Ma’aden’s) subsidiary, Ma’aden Wa’ad Al Shamal Phosphate Co., has completed construction on Phase 2 of its mega-scale phosphate fertilizers complex. The $8-B project can produce up to 3 MMtpy of phosphate fertilizer products. The company plans to add a third phase that will increase production by an additional 3 MMtpy. The $6.4-B Phase-3 project is scheduled to be completed in 2024. To meet the growing demand for methyl methacrylate (MMA), Saudi Methacrylate Co. (Samac) invested more than $1.1 B in the construction of the world’s largest MMA plant. Dubbed the Alpha 2 project, commercial operations are expected to begin in late 2017. The facility can produce 250 Mtpy of MMA and 40 Mtpy of poly methyl methacrylate (PMMA). 62


To satisfy increasing regional and domestic demand for petrochemicals and detergents, Farabi Petrochemicals is building a $1-B plant in Yanbu to produce linear alkyl benzene (LAB) and n-paraffins. The plant will utilize diesel and kerosine from the Saudi Aramco and ExxonMobil refineries in Yanbu. Once completed in 2020, the facility will be able to produce 120 Mtpy of LAB and 246 Mtpy of n-paraffins, as well as de-aromatized specialty oils, asphalt, sulfonates, mining chemicals, process oils and lubes. Rufayah Chemical Co.’s $500-MM chemical complex in Jubail Industrial City is also scheduled to go online in 2020. The plant will produce a wide range of specialty chemical products, with a total production capacity of 350 Mtpy. United Arab Emirates (UAE). Under Abu Dhabi National

Oil Co.’s (ADNOC’s) 2030 Strategy, the UAE plans to: • Expand crude oil and sour gas production • Double domestic refining capacity • Triple domestic petrochemical production capacity. This program is needed to meet increasing demand for refined products in the country. Energy demand in the UAE is increasing by approximately 9%/yr, and with the country’s population expected to double by 2030, additional downstream infrastructure will be needed to satisfy demand. The UAE has announced multiple downstream processing capacity investments. Emirates National Oil Co. (ENOC) is investing $1 B to expand its Jebel Ali refinery. The project will add 70 Mbpd of new domestic refining capacity by 2020. Once completed, the Jebel Ali refinery’s total refining capacity will reach 210 Mbpd. Renewed interest has been seen in the Fujairah refinery project. The $3.5-B, 200-Mbpd project was put on hold while the project’s developers�International Petroleum Investment Co. (IPIC) and Mubadala Investment Co.�completed their merger. The company is assessing its long-term goals, which include a possible restart of the Fujairah project. The revived interest in the Fujairah project was sparked by a fire that knocked out production at the Ruwais refinery, which increased domestic transportation fuels demand; as well as a spat with Qatar that has closed the country to Qatar’s fuel imports. If greenlighted, the refinery would not begin operations until the early- to mid-2020s. As part of the UAE’s Vision 2030 plan, the country is making significant investments to boost domestic petrochemical production capacity from 4.5 MMtpy in 2016 to 11.4 MMtpy by 2025. To help accomplish this ambitious goal, ADNOC and Borealis are developing the Borouge 4 complex. The facility will utilize a mixed-feed cracker to produce polyolefin and nonpolyolefin products. The project’s FID will be taken in 2018. If greenlighted, Borouge 4 is scheduled to begin commercial operations in 2023. At the JV’s Borouge 3 complex, ADNOC and Borealis are developing a fifth PP plant. The PP5 project will use surplus feedstock from the Ruwais refinery’s propane dehydrogenation unit to produce 500 Mtpy of PP. Both the PP5 and Borouge 4 projects will help the UAE increase its domestic petrochemical production capacity to 10 MMtpy. Additional optimization and debottlenecking projects at Borouge 1, 2 and 3 will help the country move closer to reaching its petrochemical capacity goal of 11.4 MMtpy by mid-2025.

Maintenance and Reliability W. K. ALLAH, A. AL-ABOOD, F. BALLARD and K. BRASHLER, Saudi Aramco, Dhahran, Saudi Arabia

Troubleshooting the repetitive failure and shaft seizure of a hot oil circulation pump The Shaybah NGL recovery plant has a single-stage, doublesuction between bearings, radial-split design for hot oil circulation. The original design had four pumps in operation and two on standby. The pumps are rated for 3.16 Mgpm, with a total dynamic head (TDH) of 483 ft. The pumps are driven by 500hp induction motors at an operating speed of 3,570 rpm. The pumps utilize API-610 Plan M for the cooling water piping inside the bearing housing water. These pumps were experiencing several thrust bearing failures, including pump seizure events, since commissioning in October 2015. Typically, pumps start up with high-thrust bearing temperatures and operate for 2 wk–3 wk, followed by a significant increase in bearing metal temperature and, in most cases, a thrust bearing failure. Observations and findings. The primary factors that con-

tributed to the reliability issues mentioned with these pumps are outlined here. These pumps have a single rolling-element radial bearing on the drive-end (DE), and a duplex set of rollingelement bearings mounted in a back-to-back configuration on the non-drive end (NDE).

noticeable difference with visible splashing. Based on this observation and discussions with the pump vendor, it was decided to raise the oil level from the existing 50% sight glass level (recommended level) to 75% and recheck the oil ring function. Upon raising the level to 75%, no significant improvement in splashing occurred. The oil level was then slowly raised until visible splashing was observed, which occurred at a sight glass level of approximately 90%. The observed splashing and oil ring function was similar to the DE oil ring. FIG. 3 shows the NDE sight glass level at approximately 90%. The oil level was left at 90%, and the NDE thrust bearing metal temperature was trended to identify any decrease in bearing metal temperature. The Unit E pump NDE thrust bearing experienced a significant reduction in metal temperature from approximately 186°F to 156°F over a two-hour period. This observation was another indication that the bearings were not being properly lubricated.

Inadequate NDE bearing lubrication. All NDE thrust bear-

ings exhibited relatively high bearing metal temperatures� some as high as 200°F�upon reaching stabilization temperature after startup. Three of the six pumps have experienced complete thrust bearing failure and, in some cases, resulted in pump seizure due to the severity of the failure. In all cases, the inner bearing of the NDE duplex bearing set, which is furthest from the oil ring, failed first. A visible inspection indicated that the cage experienced a severe failure, which was followed by skidding of the rolling elements. FIG. 1 shows the damaged NDE inner thrust bearing on Unit E. A visible inspection of the rolling elements showed discoloration, which is consistent with a lubrication failure. All of the rolling elements showed two flat spots that were 180° apart (FIG. 2), which is consistent with the suspected skidding condition following the cage failure. Based on the suspected lack of lubrication, Unit E’s breather caps were pulled to check for proper oil ring function. Upon inspection, the oil rings appeared to be relatively dry, without significant splashing. The DE’s oil ring was also inspected to compare the DE and NDE oil ring function, which showed a

FIG. 1. Unit

E thrust bearing failure. Hydrocarbon Processing | OCTOBER 2017

63

Maintenance and Reliability Another indicator of improper lubrication was the significant difference�approximately 40°F�between the top and

flow was provided back through the pump discharge line, for an undetermined length of time, at the discretion of the operator. No restrictive orifice (RO) was provided in this line, so the bypass line globe valve was fully opened A consequence of exceeding the initial to provide an undetermined amount of warmup flow back through the pump. The warmup rate from the warmup rate is differential thermal expansion pump casing’s initial ambient temperature of apbetween the pump casing and the rotor, proximately 90°F was monitored. Upon opening which can result in damage to the the bypass valve, the pump casing temperature went from approximately 90°F to 280°F within a 10 min– pump’s internal components. 15 min period, which significantly exceeded the recommended warmup rate and resulted in the pump thermal shock. The recommended warmup flow and rate for bottom bearing housing skin temperatures. After the oil level was raised, the temperature differential between the top and this size pump and process temperature is about 6 gpm, with bottom of the NDE bearing housing was reduced to approxino more than a 50°F/hr increase, respectively. mately 15°F–20°F. This drop in temperature indicated that Based on these measurements, it was determined that the the bearings were now receiving lubrication, resulting in heat pump was experiencing a thermal shock condition during the initial warmup. A consequence of exceeding this warmup rate transfer to the oil. is differential thermal expansion between the pump casing and Inadequate warmup system and procedure. A review of the rotor, which can result in damage to the pump’s internal components, including the thrust bearings. The existing conthe warmup and startup commissioning was performed prior to the startup of pump C. After reviewing the piping and infiguration utilizes a bypass line around the discharge check strumentation diagram (P&ID) and conducting field verifica valve, without an RO providing uncontrolled warmup back flow through the discharge line and back out of the suction noztion, it was determined that these pumps lacked provisions for a controlled warmup. zle. This configuration does not provide a controlled warmup The warmup of the pumps was accomplished by opening the flow (i.e., no RO), which resulted in more warmup to the top of the casing. It will not uniformly warm up the pump casing from bypass valve around the discharge line check valve (FIG. 4). This the bottom to the top. Based on the recommended warmup flow of 6 gpm and the maximum warmup rate of 50°F/hr, the initial warmup should take a minimum of 4 hr. The difference between the top and bottom casings should not exceed 70°F. After implementing the warmup orifice at the site, as a temporary solution, an acceptable increase in rate casing temperature was achieved. FIG. 5 is an illustration of increasing casing temperature over a 2-hr period. TABLE 1 shows the temperature ranges over the testing period.

FIG. 2. Damaged

FIG. 3. Unit

64

balls with discoloration and flat spots 180° apart.

E pump NDE sight glass oil level.


FIG. 4. View

of the bypass line globe valve for warmup flow.

Maintenance and Reliability Improper mounting of NDE thrust bearings. A visible in-

spection of the outer race of one of the failed bearings indicated evidence of improper mounting or installation of the bearings. FIG. 6 shows the outer race, which shows the visible evidence of these bearings being forced (hammered) during the mounting. The marks on the outer race at the rolling elements are an indication of this improper mounting issue.

do not provide adequate warning of the increased vibration amplitude associated with the pending bearing failure. Another contributing factor to the bearing failures without a machine trip was the bearing temperature shutdown logic. Existing logic requires both the DE and NDE RTD faults for machine shutdown. Changing the oil ring. The vendor has recommended

Inadequate thrust setting or end float. After observa-

tion during bearing replacement activities, it was determined that the established thrust setting (axial float) was set at approximately 1 mm. The manufacturer recommended a thrust setting of 1 mm–5 mm. The tighter setting can also contribute to higher bearing operating temperatures. It was observed that pumps set at 3 mm have relatively lower operating temperatures compared to pumps that have a tighter clearance of 1 mm. The thrust setting is changed by adjusting the thickness of the bearing end cover shim.

changing the oil ring size, which was implemented on all six pumps. The bearing temperature indicated improvement, but the oil rings were inspected during the oil change and some minor rubbing marks were found. Takeaways. After an anaylsis of the equipment, the following

aspects were noted: • Inadequate bearing lubrication was identified as a

Insufficient vibration and bearing temperature shutdown systems. The DE and NDE vibration accelerometers

are not directly mounted to the bearing housing and do not meet API 670 requirements ( FIG. 7). This was another contributing factor to the pumps experiencing complete failure of the NDE bearing without a shutdown of the machine. Without direct contact to the bearing housing, the accelerometers FIG. 6. View of the outer race. 300 F ° , e 200 r u t a r e p m e T 100

Top casing temp Bottom casing temp

0

2:30 2:35 2:40 3:00 3:10 3:20 3:30 3:40 3:50 4:00 4:10 4:20 4:40

FIG. 5. Illustration of the temperature increase in the casing during the warmup period.

TABLE 1. Top and bottom casings temperature difference

Time

Bottom casing temperature, °F

Top casing temperature, °F

Temperature difference, °F

2:30

80

83

3

2:35

81.6

84.5

2.9

2:40

81.7

88

6.3

3:00

83

93

10

3:10

96

109

13

3:20

104

130

26

3:30

111

162

51

3:40

123

180

57

3:50

140

202

62

4:00

141

203

62

4:10

145

210

65

4:20

150

218

68

4:40

175

237

62 Select 157 at www.HydrocarbonProcessing.com/RS

65

Maintenance and Reliability significant contributing factor to the frequent thrust bearing failures with these pumps. • The inadequate lubrication appears to be related to an improper oil level in the bearing housing, as evident from the oil level adjustment and observed splashing,

FIG. 7. NDE

vibration accelerometer and RTDs.

Others simply sell you a product – we offer a solution.

h t u m r e t u G L H O . n o y l e r n a c u o y e u q i n h c e t f f o t u h s d n a l o r t n o c A

Customized Valve Design

„MADE IN GERMANY“

ESD SOLAR POWER

s Valve Best 150 since s r Yea

OHL Gutermuth Industrial Valves GmbH Helmershäuser Str. 9+12 · 63674 Altenstadt / Germany Phone +49.60 47. 80 06-0 · Fax +49.60 47.80 06-29 www.ohl-gutermuth.de · [email protected]

and confirmed with the observed significant reduction in bearing metal temperatures. • The inadequate warmup system and the observed thermal shocking of the pumps during warmup were also identified as significant contributing factors to the bearing failures. • The thrust setting or axial float was identified as an improvement area to reduce the operating temperatures of the thrust bearings. • An improper bearing-mounting technique� the basis of the preliminary bearing failure analysis� was identified, and was a contributing factor to the poor reliability of the thrust bearings. • Inadequate bearing temperature shutdown logic required both DE and NDE bearing temperature faults to shut down the machine. • Inadequate vibration shutdown protection, due to improper mounting of the seismic accelerometers at both the DE and NDE locations. This configuration also contributed to the bearing failure events without a machine trip. The following are some recommendations: • Revise the warmup system to incorporate an RO to maintain a proper pump warmup rate. It is recommended to revise the warmup system per the manufacturer’s recommendation. The startup procedure for these pumps should include warmup requirements to ensure that the pumps do not experience thermal shocking and maintain a continuous warmup in the standby condition. • Ensure that the warmup is conducted in a controlled manner, as per the recommended warmup rate of less than 50°F/hr. The difference between the top and bottom casings should not exceed 70°F. • The thrust setting, or axial float, should be set at 3 mm–4 mm on future NDE thrust bearing replacements to ensure clearance for these bearings and to minimize the bearing operating temperature. • Reconfigure the bearing RTD shutdown logic, which requires both DE and NDE faults to trigger a pump shutdown. This configuration resulted in complete fai lure of the NDE bearing without a shutdown of the machine, due to requiring faults on both DE and NDE bearings. • Revise the DE and NDE bearing housing accelerometers, which were incorrectly mounted. • Future thrust bearing replacements should be performed by following the procedure outlined in the maintenance and repair manual to ensure proper installation and mounting. WESAM KHALAF ALLAH is a Rotating Equipment Engineer with a BCh degree

from the Purdue School of Engineering and Technology. He has 5 yr of experience with Saudi Aramco. KEITH BRASHLER has 26 yr of rotating equipment experience in refining,

pipeline, power generation, pulp and paper, and nuclear facilities. His expertise focuses on field pump and system troubleshooting. FERNANDO BALLARD has 11 yr of experience, with 9 yr as a Fixed Equipment

Specialist with ExxonMobil. ABDULLAH AL-ABOOD is a Vibration Engineer. He has 3 yr of experience

with Saudi Aramco.

66


Maintenance and Reliability J. THARAKAN, Suncor Energy Products,

Calgary, Alberta, Canada

Minimize engineering err error ors s with compe competenc tency y and proper engineering review Industry has seen numerou numerouss engineering errors during the commissioning phase of projects. Errors that slip through re views could lead to costly equipment failures. How do these errors creep in? Are organizations learning from incidents and capturing those learnings in their standards? How many times has corporate memory failed us? Some interesting examples of errors that will provide a taste of the broad and diverse range of potential errors are discussed here. Standards and specifications can address repeatable errors, but competency and the right level of engineering review are required to minimize random errors. Reliability vs. low initial cost. A set of spherical pressure vessels for butane storage were fabricated using carbon steel. However, corrosion allowance was not specified because the service was deemed non-corrosive. After 30 yr of operation, the sphere experienced a low rate of internal pitting and corrosion, which made it thinner than the design thickness. It is difficult to categorize it as a design error if the original specified design life was 25 yr. The truth is that pressure vessels vessels are meant to operate for much longer than 25 yr, so it is prudent to consider a corrosion allowance for all carbon steel equipment. In this instance, the operator was forced to resort to fitness-for-service and the application of protective internal coatings to extend the life of the vessel. The standards of this organization require corrosion allowance for all carbon steel equipment,t, regardless of ser vice. equipmen Steam methane reformers use high-nickel cast high-performance (HP) alloys for the catalyst tubes, which have a flanged opening at the top for filling catalyst. Some designs have a flange at the bottom, as well. The manufacturers used low-alloy steel flanges at the top end of the catalyst tubes and carbon steel flanges at the bottom end. This cost-reduction measure resulted in dissimilar welds on the catalyst tube, which are prone to failures. The dissimilar welds at the top cracked due to thermal fatigue (FIG. 1). The bottom flange is a dead-ended section outside of the furnace. The dissimilar weld suffered a combination of hydrogen (H2)-assisted cracking and carbonic acid (H2CO3 ) corrosion.1 The designer of a carbonate regenerator adopted a unique design premise to reduce the initial cost. This tower was 140 ft. tall, had a diameter of 10 ft. and was filled w ith random packpack-

ings. The top 24 ft. of the tower had a thickness of only 0.25 in. Tall, thin, cylindrical, stack-like structures are susceptible to wind-induced oscillation due to vortex shedding. This vi bration occurs occur s in a direction direct ion perpendicular per pendicular to the direction directi on of wind flow. f low. When the th e frequency frequenc y of vortex vor tex shedding approaches the natural frequency of the tower tower,, the oscillations gain amplitude. The wind velocity at which the frequency of vortex shedding matches matches the natural frequency of the tower is called critical wind velocity. Vortex shedding occurs at relatively lower wind velocities. Wind speeds in excess of 50 mph always contain gusts that will disrupt uniform vortex shedding.2 The presence of attachments like platforms and ladders lower the chance of vortex shedding, as well. The original designer of the tower had calculated that the critical wind velocity of the tower as 32 mph in an empty erected condition without packings. With packings, the critical wind velocity improved to 45 mph. The vessel designer included a note in the drawings that the tower should be supported with guy-wires each time the packing is removed. The guy-wires turned out to be an expensive undertaking during turnarounds when the packings were dumped. This situation could have been eliminated by using a slightly thicker plate for the fabrication of the tower. Organizations should place emphasis on reducing lifecycle costs, not upfront costs. Wrong metallurgy. Companies rely on a positive material

identification (PMI) program to capture deviations prior to commissioning. Most mix-ups are in small-bore fittings in low-alloy steel piping/equipment. It may be hard to put this in the engineering error category. An engineering error is when the wrong metallurgy is selected without a full understanding of the potential damage mechanisms. Such errors have

FIG. 1. Cracking

of dissimilar welds at the top of the catalyst tube. Hydrocarbon Processing | OCTOBER 2017

67

Maintenance and Reliability led to many serious incidents in industry. Operating within integrity operating windows (IOWs) is equally important to avoid unforeseen material damages. The operating manuals willl not indicate the rational wil rationalee behind material selecti selection on and the corrosion control documents (CCDs) being created to fill this gap. The American Petrole Petroleum um Institute (API) published RP 584 IOWs in 2014, and RP 970-CCD is expected to be released in 2017. Undersized Undersiz ed equipment. Gas-liquid separators are sized for

a recommended residence time to enable the separation of the vapor and liquid phases. If the sizing is not appropriate, liquid can carry over to the downstream systems. These are not easy fixes. FIG. 2 shows the failure of a compressor’s diaphragm (stationary guide vanes) due to liquid carryover from an undersized separator from which the compressor took suction. It may be worth noting that no knockout drum was present between the separator and the compressor suction. suct ion. Misinterpreting flange assembly loads. Flange joints

are generally the weakest component in a pressurized system.

The gasket seating stresses relax after assembly, and it is recommended to impart sufficient assembly load to compensate for future relaxation. For example, the A193 B7 bolt should be stressed from 40 kilopounds per square inch (Ksi) to 70 Ksi for joint integrity; although, the flange design calculation uses an allowable bolt stress of 25 Ksi. Many leaks on heat exchanger body flanges could be traced back to low-assembly loads selected by vendors thinking that they should stay within the design allowable stress of the bolt. With the release of ASME PCC-1 in 2010 and training and qualification requirements in the 2013 revision, this knowledg knowledgee gap has narrowed. Lack of understanding of the design intent. At times,

we fail f ail to communicate co mmunicate the design intent to the th e user. us er. During Dur ing an engineering review, the author came across a case where a pressure relief valve (PRV) was twinned, but the operations lined up both sets of valves, which doubled the relief capacity. If a relief scenario were to occur, this setup could potentially lead to valve chattering and seat/disk damage. Only one set is supposed to be in line, and the other should be a spare. Per API 520 Part 2, when spare relief devices are provided, a mechanical interlock or administrative controls�which manage the proper opening and closing sequences of the i solation valves to ensure that only one set is lined up at a ti me� should be provided. Inadequate margin. Spring-loaded PRV PRVss require a 10% mar-

FIG. 2. View of

a broken compressor’s diaphragm due to liquid carryov carryover. er.

gin between the operating pressure and the set pressure to pre vent valve simmering and leakage. For tank relief reli ef valves, valves , there is no spring loading, so the required margin is approximately approximately 20%. Typically, tank pressure vacuum relief valves (PVRVs) open and vent to the atmosphere during normal tank operation, as the level changes. Therefore, in a conventional tank design, the PVRVs have the dual function of regulating the tank vapor space pressure and overpressure protection. Whenn tank vapors contain toxic materials such as hydro Whe gen sulfide (H2S), the function of regulating the tank pressure is detached from the PRV. These tanks have a blanketing gas supply to the vapor space through a pressure controller. The normal venting from these tanks is either to a vapor recovery system or to a flare. A PRV and emergency vent will be open to the atmosphere on these tanks. The purpose is overpressure protection only, and the valve is expected to remain shut during normal tank operation. The author has encountered multiple cases on these tanks where the PRV has leaked due to an inadequate margin between the normal operating pressure and the set pressure of the PRV. The root cause of this issue is non-conservative design pressure for the tank, squeezing the available margin to set the PRV. High-pressure drop. In a Canadian facility, a hydrocracking

FIG. 3. The

68

original piston (left) and the modified piston (right).


unit was started up after debottlenecking was completed. In the piping circuit, more than a dozen piston check valves were not opening enough at normal flow. The check valve’s piston would have the upstream up stream pressure at the bottom side and the downstream pressure on the top side. The differential pressure acts against the weight of the piston. Since the original piston was heavy, the valve could not fully open at the available pressure differential. This setup restricted the capacity.

Maintenance and Reliability The pistons were replaced with lighter ones (FIGS. 3 and 4). Takeaway: The check valve’s ability to fully open at available differential pressure must be checked at the design stage. Design/fabrication errors. Reciprocating water-injection

pumps in a hydrocracking unit suffered a catastrophic fai lure. The failure began at the welding on the plunger attaching its cover to the barrel. It is not recommended to use a welded plunger in the first place, as the plunger sees severe cyclic stresses that could fail a weld. To make matters worse, the weld was defect defective. ive. A failure investigation investi gation revealed that the vent hole was miss ing in the failed plunger ( FIG. 5). This vent hole is required to relieve the welding fumes, without which the welding is prone to be defective. The plunger was replaced with a solid plunger to eliminate eliminate the welded joint in this severe cyclic ser vice. Another notable example was lamellar cracking of pump boxes�tanks that hold slurr slurry�in y�in a mining project (FIG. 6). The pump box had a rectangular shape with a conical bottom, and was fabricated from a 1-in.-thick carbon steel plate conforming to A 572 Gr. 50. The major contributors leading to the lamellar cracking were the use of high-strength steel (yield strength of 50 Ksi), with potentially low-transverse ductility and a poor weld joint design. Weld shrinkage stresses perpendicular to the rolling direction promote lamellar cracking. FIG. 7 shows how a joint design can offset the direction of weld shrinkage stresses. Preheating helps reduce weld residual stresses, as well. The joint design principle would be applicable for fabrication of thick header boxes for fin-fan exchangers.

when product is pumped out of the tank after the roof lands. A client’s standard required that the bleeder vent be located close to the center of the roof, with wi th an additional requirement on the projection of its sleeve above the tank roof. The seating face of the valve was at the top of the sleeve. The tank designer’s practice was to place the bleeder vent closer to the outer periphery of the roof. They attached the bleeder vent near the tank center, but failed to compensate the length of the sleeve for the roof sagging. When in service, the accumulation of snow/water on the roof caused the roof sagging to increase ( FIG. 8). The additional weight on the roof increased product displacement as per the floatation law. The resulting level difference (h) lifted the bleeder valve leaking product above the roof. Ligament failure. As per API 590, the convection tube sheet

of heaters requires a minimum 0.5-in thickness. The spacing between the tubes is determined by the center-to-center spacing of a 180° return bend, which is twice the nominal diameter. The holes on the tube sheet should be large enough to suit the outside diameter of the fin (or stud) of the convection tubes. This leaves a thin ligament, which is often inadequate to with-

FIG. 5. Cross-section

of the plunger.

Misinterpretation of the standard. Floating roof tanks

have bleeder vent(s), which are meant to prevent vacuum

FIG. 6. Cracking

FIG. 4. Cross-section

of the modified check valve’s piston.

FIG. 7. The

in a pump box.

improved improv ed joint design eliminates lamellar cracking. Hydrocarbon Processing | OCTOBER 2017

69

Maintenance and Reliability stand the operating load. The author has encountered multiple cases of cracking of 0.5-in.-thick tube sheets at the narrow ligaments. FIG. 9 shows a cracked tube sheet that was repaired with stiffeners. Increasing the tube sheet’s thickness is required where the ligament width is narrow. A company’s standard must supplement industry standards, based on experience. Technology selection. When selecting technology, avoid be-

ing “guinea pigs.” Two examples regarding proven technology

in service, the accumulation of snow/water on the roof caused the roof sagging to increase, which increased product displacement.

are discussed here; however, the designer failed to distinguish the specific limitations for their application. The first example is the selection of a vortex flowmeter to measure the flow on individual water-injection lines in a hydrocracker during a unit revamp. Orifice meters previously used in this serv ice were successful. The piping where the flowmeters were mounted ran through the platform of a bank of fin-fan coolers. Vortex meters are sensitive to vibration, and the fin-fan platform, where the meters were mounted, had sufficient vibration to affect the accuracy of its readings. A large upstream oil company selected oil-flooded screw compressors for wet gas compression in its gathering centers. In oil-flooded screw compressors, the lube oil provides the sealing between the casing and the rotor lobes. The lube oil gets mixed with the process gas, is separated in an oil/gas separator, and is then pumped back to the compressor through the oil cooler and the lube oil filter. The compressors experienced frequent plugging of the lube oil filter. The takeaway was that oil-flooded compressors were not an appropriate selection for wet gas compression due to the potential liquid entrainment in the gas, despite having a gas scrubber upstream of the compressor.3 The entrained liquid degraded the oil, which created gooey material/solid particles that choked the oil filter.

FIG. 8. When

Dry gas seals for pumps. Multiple pumps in light hydrocar-

bon/high-vapor-pressure fluids with a dry gas seal developed an interesting integrity issue. The seal plan was 74 for this dualpressurized seal (FIG. 10). The barrier gas, nitrogen (N), was at a pressure higher than the pump stuffing box pressure. The N leaked past the inner seal face and into the pump casing of the pumps that were on standby. This leakage caused vapor lock when the standby pumps were started. This was resolved by connecting the seal vent to the vapor space of the pump suction vessel. This solution channeled the leaking N to the suction drum, which prevented it from getting into the pump. Asymmetrical piping. Two pumps (A/B) at the bottom of a

fractionating tower (FIG. 11) had asymmetrical piping. If kept on standby, Pump A would accumulate corrosion products in the suction strainer. Pump A’s strainer required cleaning each time the pump was started. The operations team adopted a solution that entailed treating Pump B as an emergency spare and

FIG. 10. The FIG. 9. View

70

of a cracker tube sheet repaired with stiffeners.


connection of a seal vent to the vapor space of the pump suction vessel channeled the leaking nitrogen to the suction drum.

Maintenance and Reliability running Pump A almost continuously. This plan is an example where limitations can be managed, but a flip side to this example does exist. The plant follows a pump swinging program, which enables the detection of hidden failures and maintains even run hours between the machines. These pumps needed to be singled out from the pump swinging program. Takeaway. The author has not attempted to cover all of the

design errors experienced in this article. A few other examples are discussed in other technical articles.4,5 Aside from initial design, errors can creep in when changes are implemented. People easily visualize changes associated with any alteration of hard ware, but can fail to identify other changes that affect integrity. A change in the feed specification to a unit, in the heat duty of a furnace, or an increase in flow through a heat exchanger can impact integrity. A rigorous management of change (MOC) system and training are required to recognize and manage changes. Major engineering errors can lead to process safety events (PSEs), which result in an unplanned or uncontrolled loss of primary containment (LOPC). Examples to illustrate how design norms have changed over the years include: 1. Hydrocarbon liquid relief of a process vessel was routed to the oily water sewer. Once, an over-pressure scenario resulted in a massive release to the sewer, which led to a fire. 2. Due to the failure of a level controller, gas blowby occurred from a two-phase separator to the atmospheric storage tank through the rundown line. Both examples occurred on 50-yr-old plants. Present design standards require the relief valve outlet to be connected to the hydrocarbon flare. Gas blowby is a situation evaluated seriously through process hazard analysis (PHA), and

that entails additional levels of protection, such as a chop valve that closes at a low low level in the separator. Process safety requires multiple layers of protection. Engineering control is the top tier of the hierarchy of these layers of protection. All incident investigations require adequate probing to identify hidden engineering errors, if any. Organizations can keep both engineering and administrative errors in check through periodic PHA, damage mechanism review (DMR), pre-startup safety review (PSSR) and a robust MOC system. Networking and competency development of engineers are equally important. REFERENCES 1

El Ganainy, O., “Failure of dissimilar metals weld in reformer tubes,” Process Safety Progress, July 1985. 2 Moss, D. and M. Basic, Pressure Vessel Design Manual, 4th Ed., Elsevier, 2012. 3 Mirza, B. and F. Al-Kharqawi, “Key engineering highlights of facilities and equipment for oil and gas production in Kuwait Oil Co.,” Journal of Energy and Power Engineering, January 2016. 4 Tharakan, J., “Vibration studies offer valuable lessons for correcting problematic operations,” Hydrocarbon Processing, March 2017. 5 Tharakan, J., “A deeper examination of the thermal expansion issues in process equipment and piping,” Hydrocarbon Processing, May 2017. JOHN THARAKAN is a

corporate static equipment specialist at Suncor Energy Inc. in Canada. He specializes in mechanical integrity, with a focuses on troubleshooting, failure analysis and fitness for service. He holds an MS degree in mechanical engineering design and has more than 30 yr of experience in the oil industry.

FIG. 11. The

setup of Pumps A and B, located at the bottom of a fractionating tower. Select 159 at www.HydrocarbonProcessing.com/RS

71

Heat Transfer M. BADSHAHWALA, R. C. PATIL and A. GUPTA,

Reliance Industries Ltd., Jamnagar, Gujarat, India

Ensure uniformity of air preheater flow to obtain maximum heat transfer efficiency Uniform flow distribution inside process equipment is desirable for maximizing performance. However, flow maldistribution occurs even in seemingly symmetrical design because of asymmetric inlet and outlet piping. Quantification of the equipment efficiency loss caused by this non-uniform distribution is an important field of study in chemical engineering. A case study is presented of a shell-and-tube air preheater that experienced a maldistribution of the gas heating medium on the shell side. A simple methodology has been used to compute heater efficiency in terms of the gas maldistribution. Quantification of efficiency with regard to maldistribution. In an air preheater, heat transfer between hot gas and cold

air generally occurs by convective mode of heat transfer. The cold air flows through the tubes, and the hot gas flows over the tubes, as shown in FIG. 1. When designing an air heater, the designer assumes that the hot gas flow over the tubes and the cold air flow through the tubes will be uniform and will lead to the desired heater duty for which it is designed. However, in practice, the inlet duct for the flow of the hot gas may lead to non-uniform flow over the tubes.

As an example, the flow of gas for the air heater explained in FIG. 1 would be as shown in FIG. 2. The sharp 90° turn at the inlet would cause greater flow of gas toward the right side of an air heater, leading to greater gas flow over the tubes on the right half as compared to the design value. The left half would then have less gas flow as compared to the design value. To illustrate the effect of this non-uniform flow bifurcation on the heat transfer efficiency, an air heater is divided into two equal halves with an equal area of heat transfer, and simulated as two different heat exchangers, using commercial software. In the design case for an air heater, the air and gas are equally divided into two parts and sent to two separate heat exchangers. In the case with the non-uniform distribution of gas, 70% of the mass flow of the gas goes to one heat exchanger, and the remaining 30% of the mass flow of the gas goes to the other heat exchanger, while the air flow distribution is even to both heat exchangers. The non-uniformity can be defined in terms of standard de viation in gas flow in two heat exchangers, as shown in Eq. 1: 2

Standard deviation

72

of an air preheater.


i=

2

i

=

− x)

/2

(1)

x

where x is the percentage of gas flow to the ith heat exchanger and x is the average percentage flow of gas to two heat exchangers. For the design case where the gas flow is equally divided (50% to both heat exchangers), the standard deviation is zero. For the aforementioned case where the gas flow is non-uniform, the standard deviation value is 0.4. With an increase in the flow of gas to one heat exchanger, the rise in the heat duty is due to the increased heat transfer coefficient. The increase in the coefficient is the result of both the increase in the Reynolds number and the increase in the temperature difference caused by the high heating capacity of the gas. Similarly, in the other heat exchanger where the gas flow is reduced, the decrease in heat duty is expected because of the reduction in the heat transfer coefficient. The decrease in the coefficient is due to the reduced Reynolds number and the decrease in the temperature difference, which results from the low heating capacity of the gas. Overall, when the combined performance of both heat exchangers is compared with the design value, the reduction in heat duty (and, therefore, the lower heat transfer efficiency as compared to the design case) is anticipated. The heat transfer efficiency is defined in Eq. 2: i

FIG. 1. Diagram

∑ 2 1( x

Heat Transfer

100

95

% , y c n e i c90 ffi e r e f s n85 a r t t a e H

80

75 0.0

0.1

FIG. 3. Impact

0.2

0.3

0.4 0.5 0.6 Standard deviation in flow

0.7

0.8

0.9

1.0

of flow distribution on heat transfer efficiency.

100

2 heat exchangers 3 heat exchangers 4 heat exchangers

95

FIG. 2. Flow

% , y c n e i c90 ffi e r e f s n85 a r t t a e H

of gas at a sharp 90° bend in the inlet duct.

Combined heat duty of heat exchangers with non-uniform gas flow Efficiency i = Combined heat duty of heat exchangers with uniform gas flow

80

(2)

The heat transfer efficiency for the design case will be 100%, owing to the equal heat transfer area and the equal hot and cold fluid flow to each heat exchanger. In the case of the non-uniform distribution of gas, the heat transfer efficiency is calculated as 96.3%. The heat transfer efficiency can be simulated for multiple cases where the standard deviation in the hot gas flowrate can vary from zero to the maximum possible value of one, in this case. These calculations are mapped on a plot of standard deviation vs. heat transfer efficiency in FIG. 3. As explained previously, it can be seen that the heat transfer efficiency decreases with the increase in flow maldistribution. The heat transfer efficiency falls as low as 79.9% in an unlikely worst-case scenario of maldistribution where the entire flow of hot gas goes to one part of the air heater. It should be noted that, when the heat exchanger is divided into two parts, the maximum value of the standard deviation is that for which the lowest heat transfer efficiency is achieved. Heat transfer efficiency may fall to even lower values when the air heater is divided into a greater number of parts, and the entire gas flow goes to one part of the heater (i.e., when the standard deviation value is greater than one). However, in this work, the standard deviation values are limited to a maximum of one through the selection of practical cases with flows in all parts of the heat exchanger. To verify the aforementioned observations, the exercise is repeated by dividing an air heater into three and four heat exchangers to represent the maldistribution. The simulation points from each are charted on a single plot in FIG. 4. The important observation confirmed by FIG. 4 is that a variation of the heat transfer efficiency with the standard deviation in gas flow is almost the same when an air heater is divided into two, three or four heat exchangers to illustrate the effect of mal-

75 0.0

0.1

0.2

FIG. 4. Standard

0.3

0.4 0.5 0.6 Standard deviation in flow

0.7

0.8

0.9

1.0

deviation in gas flow vs. heat transfer efficiency.

distribution. This data suggests that a single correlation can be deduced between heat transfer efficiency and maldistribution in terms of standard deviation of the gas flow. A single equation (Eq. 3) between standard deviation (SDV) and heat transfer efficiency (HTE) has been established: HTE(%) = –17.0614 (SDV 2) – 2.8256 (SDV) + 100.00 [R 2 = 0.9972]

(3)

If the total heat duty of the designed air preheater and the actual heat duty offered by it are known, then the process engineer can use this correlation to calculate the heat efficiency and the amount of maldistribution occurring in the gas flow of an air preheater. Takeaway. A simple methodology has been developed to

quantify the effect of maldistribution of hot gas over the tubes on the heat transfer efficiency of an air preheater. A correlation has been established that equates the standard deviation in the hot gas flow to the heat transfer efficiency of an air preheater. Process engineers can use this correlation to determine the extent of maldistribution in air preheaters. MUFFAZAL BADSHAHWALA is a Research Scientist in the refining research and

development (R&D) division of Reliance Industries Ltd. in Jamnagar, Gujarat, India. RAHUL C. PATIL is a Lead Research Scientist in the refining R&D division

of Reliance Industries Ltd. in Navi Mumbai, Maharashtra, India. AJAY GUPTA is the Assistant Vice President in refining R&D at Reliance Industries

Ltd. in Jamnagar, Gujarat, India. Hydrocarbon Processing | OCTOBER 2017

73

Heat Transfer J. COLANNINO, ClearSign Combustion,

Seattle, Washington

New burner structure targets significant NOx reduction Delek US comprises both refining and logistic segments. The refining segment operates four moderate complexity refineriesEl Dorado, AK; Krotz Springs, LA; Big Spring, TX ; and Tyler, TXrepresenting a combined production capacity of more than 300 Mbpd. Delek’s logistics segment gathers, transports and stores crude oil. It also distributes, stores and transports refined products to West Texas and the Southeastern US. The Tyler facility processes local light sweet crude (West Texas Intermediate and similar) and produces a complete range of refined products. These products include LPG, NGL, gasoline, jet and diesel fuels, with the largest volume being light, high-value products, such as gasoline and diesel. Delek has historically been forward-leaning with respect to technology and the environment. For example, the El Dorado facility produces low-sulfur gasoline and ultra-low sulfur diesel fuel that meet or exceed current clean-fuel standards. To date, Delek has invested nearly $1 B in technology, operational and environmental spending. Delek has been forward-leaning with respect to technology and the environment. For example, the El Dorado facility produces low-sulfur gasoline and ultra-low-sulfur diesel fuel that meet or exceed clean fuel standards. To date, Delek has

Furnace

invested nearly $1 B in technology and operational and environmental spending in pursuit of its goals. In its continued search for new technology to improve operational and environmental excellence, Delek began a program to improve the flame profile in key heaters. One heater of particular interest was a fluid catalytic cracking (FCC) preheater at its Tyler facility. The heater was originally sized for 15 MMBtu/hr in 1978, but it was operating at a reduced firing rate due to flame impingement issues. Flame impingement occurs when wayward flames impact process tubes. The high heat causes the process fluid to lay down, insulating sediment or forming oligomers known as coke. Over time, these layers thicken, further insulating the tube wall from the cooling effect of the process fluid. If coking continues unabated, the tube wall may become so hot that it loses critical strength and ruptures. In a refinery, this issue is of particular concern because the process fluid is flammable. In some instances, emissions reduction techniques can impede heater operation. For example, low-nitrogen oxide (NO x ) burners generally reduce NO x by elongating flames to allow more time for the flame to exchange radiant heat with the cooler process tubes in the furnace. Such radiative cooling lowers the flame temperature and reduces NO x since NO x is strongly affected by three factors: temperature, time at temperature and available oxygen (O2 ) concentration. To understand why this happens, an overview of NO x formation mechanisms is needed. NOx formation. NO x can be formed in three ways: 1) by the

Flame Burner tile

Fuel orifice

Stabilizing cone

Burner Air

oxidation of nitrogen (N2 ) in a parent fuel molecule, termed fuelbound NOx ; 2) by the fusion of N 2 and O2 in the combustion air to create NO x , termed thermal NOx1; and 3) by the fusion of N 2 from the combustion air with partially decomposed fuel early in the combustion process, termed prompt NOx.2 Nitric oxide (NO) is the majority forma of NO x in heaters and boilers; the discussion focuses on that species of NO x . For gaseous refinery fuels, N2 is generally not present as part of the fuel molecule. Moreover, prompt NO x makes a small contribution to the total NO x budget. That leaves thermal NO x , which is represented in Eq. 1. b

[ NO ] Fuel

Air register

FIG. 1. Elements

of a typical burner, including fuel orifices, an air register, a pilot (not shown) and a means of flame stabilization.


=

[

A N 2

] ∫ e

T

[O ]dt 2

(1)

and b are constants, the brackets represent concentrations of the enclosed species, T is the temperature, and t is the time. The exact time-temperature-O2 history in industrial boilNote: A

Heat Transfer

ers is too complex to allow for solution of Eq. 1. Notwithstanding, the equation is instructive in showing that NO x formation increases with temperature increases (exponentially weighted), time (linearly weighted) and O 2 concentration (square-root weighted). A technique known as distal surface combustion can reduce the contribution of all three factors. Distal surface architecture. The elements of a typical burn-

er are shown in FIG. 1. All combustion equipment requires metering of the fuel, metering of the air, a means of ignition and a means of stabilizing the flame. In a conventional burner, these functions are fulfilled by the fuel orifice, air register, burner pilot and one or more burner tiles, stabilizing cones or other bluff bodies or swirlers. In FIG. 2, a combustion surface has been added downstream (distal) from the burner, and the flame has been transferred from the burner to the distal surface, which now radiates heat to the process. The distal surface dramatically reduces NO x . For example, a conventional burner generates about 50 parts per million (ppm) of NO x , while NO x from the distal surface is usually 5 ppm (i.e., an order of magnitude less). Why does this happen? First, a porous ceramic body has an effective emissivity close to that of a perfect blackbody radiator; therefore, it transfers heat to the surroundings with much higher efficiency than the flame. This effect is made apparent by comparing gaseous species radiation from a natural gas flame with blackbody radiation from a solid (FIG. 3). The area under the curve of the blackbody radiator is much greater than the area under the curve for gaseous species radiation, and is proportional to the radiative heat transferred. Comparing blackbody radiation to gaseous species radiation shows that the blackbody surface has a greater ability to reduce flame temperature via radiative cooling. The result is dramatically lower NO x .

Second, burner flame lengths in process heaters and boilers are typically measured in tens of feet. A long flame length, at high temperatures, provides ample combustion to generate NO x . Conversely, with a distal surface, combustion is completed in inches, representing an order of magnitude less time for NO x formation. Ultra-short flame lengths are possible because com bustion is not kinetically limited, but rather mixing limited. In other words, the time it takes to mix the fuel and air is the principal determinant of flame length. Once fuel and air are mixed, the combustion reaction proceeds very quickly. By supporting com bustion only after the fuel and air mixing are complete, the flame is confined to within and near the porous distal surface. Short flames eliminate flame impingement, as well. Another possible benefit of delayed ignition is the ability of the fuel and air to entrain flue gas. This process dilutes the O2 and fuel concentrations before ignition�the third important factor in the NO x equation.

FIG. 3. Comparison of blackbody vs. spectral radiation for a gaseous

fuel. A blackbody radiates with greater intensity than gaseous species radiation. The total radiant energy is given by the areas under the respective curves. By inspection, the area under the blackbody radiation curve is much greater than the area under the gaseous species radiation curve, indicating a much higher amount of radiant energy transferred.

Porous ceramic surface Furnace Support structure Burner tile

Fuel orifice

Stabilizing cone

Burner Air

Fuel

Air register FIG. 4. Integral burner/distal surface assembly. Distal surfaces are

FIG. 2. Addition of a distal combustion surface. A distal porous

ceramic surface has been added downstream of the burner. The flame is omitted. All combustion takes place within and nearby the distal surface. Under such conditions, NO x is generally less than 5 ppm.

added downstream of the burner proper as part of an integral burner assembly. This allows for a burner-for-burner replacement, whereby high NOx of flame-impinging burners are swapped one-for-one with the integral burner/distal surface assembly. Hydrocarbon Processing | OCTOBER 2017

75

Heat Transfer Delek burner upgrades. The typical way that the distal sur-

face is installed in a heater is shown in FIG. 2. It is also in commercial use in once-through steam generators (OTSGs). In these configurations, the technology operates in units with fired capacities as high as 62.5 MMBtu/hr. However, Delek preferred to install an integral burner-distal-surface assem bly, because such an assembly would give Delek the ability to perform burner-for-burner replacements without the need to modify the heater. FIG. 4 shows the integral assembly that was developed. In this case, no additional surfaces are installed in the heater. Rather, the burner assembly contains both the burner and the distal surface. One of the challenges to installation was refiner y fuel that contained varying amounts of hydrogen (H2 ). H2 has markedly different combustion properties compared to hydrocarbons. For example, H2 has triple the flame speed and burns over an air-to-fuel range that is more than six times wider than natural gas. These characteristics conspire to form flames prematurely, lengthen flames and inflate NO x . Moreover, bringing the distal surface closer to the burner requires faster mix ing than afforded by conventional burner designs. To ensure that the final design would operate successfully under these conditions, the vendor4 performed extensive testing in its facility, which varied excess O2 , H2 content, turndown and flame detection locations. These variables were tested with various burner and structural support configurations, with Delek approving the final test matrix. The final design required the

A D V A N C E D

S U L F U R

development of a novel technology to enhance mixing, stabilize the flame and reconfigure specially designed distal surfaces. State of the art. The new product has successfully proven it-

self in full-scale laboratory testing over H2 concentrations ranging from 0% to 70%, and under anticipated turndown conditions and a wide range of excess air scenarios. The next step is installation in Delek’s Tyler, Texas refinery before the end of the year. Delek’s FCC preheater is a six-burner heater. Two of the six burners are to be replaced in the operating heater. The goals of this demonstration are twofold: to demonstrate that the burners can be safely swapped while the unit is in operation, and to demonstrate that flame lengths are reduced and flame impingement is eliminated. Pending a successful result, the remaining four burners will be replaced and NO x emissions will be measured. With this and other efforts, Delek seeks to further establish its commitment to environmental stewardship and enhanced value for shareholders and society. NOTE a

The other major species containing NO x is nitrogen dioxide (NO 2). However, for industrial combustion systems operating at 2%–5% excess O2 , NO2 is a relatively minor component and not considered further. REFERENCES

1

Zeldovich, Y. B., Acta Physicochim, USSR, Vol. 21, Iss. 577, 1946. 2 Fenimore, C. P., “Formation of nitric oxide in premixed hydrocarbon flames,” 13th International Combustion Symposium, The Combustion Institute, Pittsburgh, Pennsylvania, 1970. 3 ClearSign Combustion Corp., Seattle, Washington, www.clearsign.com.

A N A L Y S I S

The best in sulfur analysis on the bench or online.

From ultra-low sulfur measurement in gasoline and diesel fuels to high-pressure process and pipeline sulfur analysis, AMETEK has the solution for you. Our ASOMA Phoenix II XRF analyzer ofers a detection range down to 1.5 ppm sulfur, which covers both current and proposed federal EPA limits for sulfur in gasoline. And our ASOMA 682T-HP online sulfur analyzer eliminates the need for sample handling while providing low cost of ownership. It’s ideal for crude and other highly viscous hydrocarbons in pipelines, terminals, and blending operations. For the best in benchtop and online sulfur analyzers, visit ametekpi.com today. © 2017 AMETEK Inc. All rights reserved.



682T-HP XRT Online Sulfur Analyzer

Phoenix II XRF Analyzer for Elemental Analysis

Environment and Safety S. ZARDYNEZHAD, Contributing Writer,

Calgary, Alberta, Canada

Design atmospheric aboveground storage tanks for accident prevention The design and construction of oil, gas and petrochemical facilities is continuously improving, as increasing compliance with ever-changing codes and standards, such as the American Petroleum Institute (API) and the National Fire Protection Association (NFPA), is resulting in fewer atmospheric aboveground storage tank (AAST) accidents. The frequency of tank accidents has decreased with better design considerations and improved inspection, operation and maintenance activities, but fires and other incidents still occur. Due to the size of most AASTs, fire accidents involving large AASTs carry increased consequences in terms of cost, property damage, health, safety and environmental impacts, the interruption of operations and damaged public opinion. The following will review some main causes of fires or other types of accidents, and discuss AAST design requirements that impact safe and reliable tank operation. In addition, accident prevention methods will be detailed. AASTs are used to store or mix/blend flammable and com bustible liquids in various ways, and may also be used for crystallization, phase separation and heat exchange. Each year, numerous AAST fires or other types of accidents occur in the US and around the world, despite different and updated studies and techniques�such as risk studies, hazard and operability (HAZOP) studies, failure mode effects analysis (FMEA), etc.� that have been conducted during design or operation stages. Many factors must be considered during AAST sizing, selection and design to construct AASTs that are reliable for the entire service life of the plant (FIG. 1). It is important for designers and for operations and maintenance personnel to be familiar with the causes of fires and the respective prevention methods. At first glance, it seems that the major causes of an AAST fire are operational and maintenance errors, corrosion, lightning or sparks, or static electricity. All signs may indicate that these causes point to plant operators and the maintenance team. However, by using a detailed root cause analysis (RCA), the main causes of a fire can be easily captured. For example, the generally held belief is that the sinking of the Titanic was caused by the impact of an iceberg. Actually, a post-tragedy RCA determined that more than 100 factors caused the great ship to sink. Poor AAST design can trigger operation and maintenance errors, corrosion or static electricity that may cause a fire or other incident.

To be considered “competent” to perform design activities and tasks, designers must be qualified, trained and experienced. They should understand codes, standards and regulations, and interpret them correctly. Designers should participate in the identification of the project’s technical risks related to AAST design and plant layout. TABLE 1 shows modified input, tools and techniques, and output for identifying risks related to AAST design activities,1 with some modifications. Incorrect roof selection for an AAST can be a potential cause of fire. According to a study of 242 AAST accidents that occurred in industrial facilities over a 40-yr period (1960–2003), the external floating roof tank (EFRT) was the most frequent type of roof that experienced fire, followed by the atmospheric cone fixed-roof tank.2 Roof selection. Fixed-cone/dome-roof storage tanks include

vent nozzles to atmosphere that normally allow the tank to breathe during loading, unloading and temperature changes. Vent fires are typically related to fixed-roof tanks, such as cone/ dome and internal floating roof tanks (IFRTs). The main cause of fire is a lightning strike that can ignite fugitive vapors that might be present at the vent. Depending on the location of the tanks in the plant and their proximity to the local community, Fire protection and fighting study

Sizing and range of liquids Causes of accidents (lesson learned)

Hazard identification Material selection Corrosion prevention Risk study/FMEA

AAST selection and design

Layout study

Engineering and technical requirements Codes, standards, regulations Environmental requirements Local fire department review and feedback

Community firefighting facilities FIG. 1. Key

factors that must be considered during AAST design and selection. Hydrocarbon Processing | OCTOBER 2017

77

Environment and Safety the vents are equipped with environmental controls and flame arresters/diverters to capture emissions. An EFRT comprises vertical steel cylinders with a roof that floats on the surface of the liquid in the tank. It is open to the atmosphere above; i.e., there is no fixed roof above the floating roof. The risk of a fire in this type of tank is very high and probable because the wetted shells are open to atmosphere when the floating roof travels downward. In addition, any leak could result in stock on the deck, which can cause a fire. Pontoon failure of the EFRT is another source of accident and can be caused by several factors, including an incorrect selection of weld design. FIG. 2 shows welding inside the pontoon, which is normally incomplete in length. Welding outside the pontoon ( FIG. 3) cannot be done due to limited access and space between the tank shell and outer ring of the pontoon, which is 300 mm–400 mm. This clearance will not provide suitable access to weld the outer ring plate to the deck plate, as shown in FIG. 4.

This design is most common in EFRTs, based on the author’s experience. To weld the outside angle between the deck plate and the outer ring plate ( FIG. 5), the design of the angle should be modified, as shown in FIG. 6. Basically, the outer ring plate of the pontoon is extended vertically to make a 90° angle with the deck plate. This will allow the welder to weld the angle from below the deck when the deck sits on supports. In FIG. 3, the deck plate is extended horizontally rather than the outer ring plate extending vertically, as shown in FIG. 6. The details shown in FIGS. 3 and 5 may cause leakage of the liquid to the pontoon and sink the roof accordingly, causing many problems (even during hydrostatic test operations). Rim-seal fires are the main cause of a large majority of EFRT fires, and can occur in IFRTs, as well. Lightning is the primary source of ignition for rim fires for EFRTs, and an induced electrical charge may also spark ignition. In addition to manual semi-fixed or fully-fixed, rim-seal fire

FIG. 2. Welding inside the pontoon is normally incomplete

FIG. 3. Limited access between the tank shell and outer ring of the

in length.

pontoon prohibits welding.

TABLE 1. Inputs,

tools and techniques, and outputs of risk-identification processes for AAST design activities

Inputs

Tools and techniques

Outputs

Equipment criticality table

Expert judgment

Risk register

Technical document and drawing

Design review

Decision criteria

Information gathering techniques

FMEA for each tank

Group creativity techniques

Hidden failure study

Analytic hierarchy process (AHP)

Mean-time between failures (MTBF)

Assumption validity analysis

Causes of accidents/lessons learned

Diagnostic techniques

Technical deviations Prototype design identification Plant layout Plant hazard identification/HAZOP Plant safety philosophy Stockholder register and analysis Community/neighborhood study Codes, standards and regulations

78


Environment and Safety protection systems, an automatic foam injection system mounted on the floating roof deck in different locations should be considered to cover 360° and quickly inject foam to the rim seal. Rim-seal fires for IFRTs are more difficult to combat compared to EFRTs�particularly when semi-fixed or fully-fixed systems are not considered by the designer�because the only access for fire extinguishing is through the vents or access covers on the fixed roof (FIG. 7). Layout. One key aspect of AAST design is determining the best

location (for cost minimization, required safety, etc.) for the AAST during the preparation of the plant plot plan. If a plant is on a slope and must be designed over several terraces, designers sometimes install the feed and intermediate products of the AAST upstream of the process plant to minimize costs and pumping power requirements. However, if the large-feed AAST collapses due to fire, then the fire and hydrocarbon liquid may reach the plant and create a dangerous and out-of-control situation. It is better to install the tanks downstream of the plant on the slope�the overall plant layout should be carefully studied on a case-by-case basis. The surrounding areas and a suitable number of access routes to the tank’s location have significant impacts on firefighting activities and success. Safe distances from other process equipment must be calculated. The layout and distance of the hydrant and monitors around/to each tank should be carefully studied, and decisions should be based on the number of tanks in each area, wind direction, migration of fire risk to adjacent tanks, etc. Water supply. A large volume of water is critical during AAST

firefighting operations for cooling and extinguishing purposes. If the plant is located near the sea, connecting the main fire-water piping system to the seawater in case of emergency is highly recommended. Also, during the design phase, consider suitable portable or fixed monitors around the tank to cool it evenly. Uneven cooling of the AAST will allow the tank shell to soften and ultimately fail where the uneven cooling was applied ( FIG. 8). The fixed cooling piping rings and associated supports installed around the tank perimeter are suitable for working under high-temperature conditions as a result of fire, and are carefully

limited space between the tank shell and the outer ring of the pontoon.

welded to the tank surfaces (FIG. 9). At the very least, the num ber of fixed cooling piping rings should cover from the top to near the middle of the tank surface. Tank height. Careful selection of tank height is important, as

it affects firefighting operations and productivity. Designers prefer to reduce AAST diameter and increase the height for the following reasons: • To provide positive head for unloading pumps [The AAST must always remain filled above the “low pump-out” level (LPO) during operation, which is a fixed distance above the bottom of the tank] • The cost of a floating roof is proportionally higher than that of a shell for a given tank capacity • To reduce the bottom surface area, shorten the maximum Outer ring plate

Welding is not possible

Inner ring plate Shell plate

Pontoon FIG. 5. A

300 mm–400 mm

method of pontoon welding design.

Outer ring plate Inner ring plate Shell plate

Pontoon 300 mm–400 mm Welding is now possible from under the deck FIG. 6. Recommended

method of pontoon welding design.

FIG. 4. The

FIG. 7. Automatic

foam injection-deck mounted. Hydrocarbon Processing | OCTOBER 2017

79

Environment and Safety distance to sump and minimize the formation of “bird baths” • To reduce vapor space in fixed-roof tanks • To reduce the cost of foundation, cathodic protection, earth work, etc. The recommended height of an AAST is a maximum of 15 m, depending on land acquisition and cost limitations. It is important that the fire-water stream’s trajectory can reach to the top of the tank. During a fire, obstructed full liquid surface fires can occur in a fixed roof, EFRT and IFRT. The roof or pan blocks access to the burning surface, and this becomes more difficult when the tank’s height is increased. If the tank height cannot be limited at 15 m, then an elevated monitor tower should be considered to allow fire water or foam to reach the top of the tank for firefighting and cooling. Open flames. Hot particles can ignite flammable vapors around

an AAST. The risk of explosion of process equipment, such as large process towers, should be considered during plant layout, and a safe distance should be determined following risk and other required studies. In a fire case at a mega petrochemical plant, a large process tower, with a height that exceeded 120 m, exploded due to an increase of internal pressure caused by overheating. Hot flying particles reached an EFRT that was used to store naphtha, igniting a large fire. If the tank had been an IFRT rather than an EFRT for this service, then the accident may not have happened. Static electricity. Friction between two sliding or rolling parts

made of different materials or substances generates static electricity. The electrical charges on the surface of the objects balance each other by the free flow of electrons. However, when they are separated, each part is left with an extra or shortage of electrons, causing both parts to become electrically charged. When the produced charges lack a rout to the ground (e.g., by grounding probe or wire), the charges create static electricity, which, if not eliminated from the system, will build up and change to a spark to a grounded part or another less highly charged part. If a flammable or combustible vapor or dust mixture exists, as in a hydrocarbon AAST, the result can be a fire or an explosion. Static electricity can build up during liquid transfer and through contact with other materials, such as when liquid is pumped to a tank, agitated inside, stirred or transferred in piping. Designers should consider suitable designs and materials to eliminate the risk of fire due to static electricity.

ing AAST operations: slop-over, froth-over, boil-over and pyrophoric ignitions. Dike designs. Dikes around the AAST serve two main purposes:

• Provide a physical barrier to prevent the spread of tank contents if the tank overflows, or if the tank structure fails due to corrosion or other causes • Segregate and group tanks according to their content classifications. Dike types and materials (i.e., compacted earth, concrete or other material) are carefully selected and designed. The dike’s dimensions (height, perimeter, etc.) are functions of the volume of the tanks enclosed within a particular dike area. Dike design must consider containment of the tank’s total contents, and it should be designed according to the volume of the largest tank, plus an additional percentage as a safety margin (FIG. 10). The dike’s waste and rainwater drain piping system should be furnished with a suitable gate valve located safely outside the dike area: that valve should be normally closed with a proper locking device. Equipment, lighting towers, transformers, etc., should also be installed outside the dike area, and all openings on the dike walls should be carefully and completely sealed by proper materials. Emergency pumpouts. During firefighting, it is sometimes

necessary to decrease the volume or level of flammable and com bustible liquid inside the tank as quickly as possible, particularly for floating-roof tanks where decks or pans block access to the burning surface. Eliminating combustible material greatly assists firefighting efforts. Designers should check the capacity of tank unloading pumps�particularly large tanks�and consider proper margins that allow higher flowrates when needed, such as using variable frequency drives (VFD) and parallel operations. The temperature of the tank’s contents during a fire will be increased, and designers should consider this when considering materials and pump seal selections, as well as the transfer of hot liquid to other tanks during the fire. However, when the product is removed from the tank that is on fire, air is drawn into the vapor space as liquid is removed, creating a hazard and bringing more oxygen into the tank. This must also be considered. Roof plate joint designs. The details of fixed-roof AAST

joints (including the fixed roof for an IFRT) are shown in 3 FIG. 11. In the author’s opinion, this joint design is unsuitable because the vapor inside the tank can be diffused and trapped between the gap of two roof plates (e.g., A and B) and condensed by changing the ambient temperature or roof plate temperature. Special design considerations. Designers should remain aware of, and plan to prevent, the following phenomena durThis configuration helps sever the corrosion condition that can happen between the two plates when the vapor contains a corrosive element (e.g., the high sulfur content in gasoil). As the author was inspecting a gasoil tank in an oil refinery with a high sulfur content run for 30 yr, part of the roof beneath his feet broke. A detailed root study showed high corrosion in the gap between the two plates. An alternative to prevent trapped vapor in the gap is to install plate FIG. 8. Steel tank softens and fails when FIG. 9. Cooling piping supports soften. A on plate B, and plate C below plate B, heated. (Photo courtesy of Shana.) (Photo courtesy of Shana.) and so on. This configuration causes the 80


Environment and Safety condensed vapor to be drained from the gap between the roof plates. This is not in compliance with API 650 (11th Ed.) and should not be used without API and designer approval. However, the author believes that this alternative supplements the standard design where highly corrosive vapor exists.

level, different shell courses may have different corrosion allowances. The corrosion allowances for removable parts, such as pipe spools, and permanently welded parts, such as columns, should be considered. The roof structure may also require a corrosion allowance, particularly if the service and vapor are corrosive.

Frangible roof joint designs. Clause 5.10.2.6 of API 650

Column design. The following example illustrates the design

(11th Ed.) allows a frangible roof joint design for fixed-roof AASTs, after securing purchaser approval. This is not a mandatory requirement as per API 650, but rather a recommended practice for emergency venting. (This causes the roof-to-shell joint of the tank to fail before the shell-to-bottom plate or annular plates in the event of excess internal pressure, preventing tank collapse and catastrophic failure.) In this case, the design pressure is limited to the equivalent pressure of the dead weight of the roof plates, including structural rafters. The designer should explain this API 650 requirement and its respective advantages to the purchaser. If the frangible joint is designed and fabricated carefully, the roof will only be separated, causing a “fish mouth” opening and not rocketing the roof into the air.

of columns. A project has two aboveground atmospheric AASTs with IFRTs, each with a diameter of 70 m and a height of 15 m. Each AAST is being designed by a different engineer. One engineer considers a fixed dome roof with 130 columns of 6 in. to 12 in., and the other is designing a dome roof without columns and with a self-supported roof. Both meet code requirements in terms of roof stability and strength. The fixed-roof columns pose the following disadvantages: • They must penetrate into the floating roof/pan, causing vapor or liquid leakage from the opening and requiring extra costs for sealing around the opening. • They occupy the volume of the tank. • They require inspection during operation and extra cost for scaffolding. • Corrosion may occur on the surfaces of the columns. • They need adjustment and added shims after a field hydrostatic test to prevent roof damage due to foundation settlement. • They increase the cost of the project. • They may obstruct future inspections of the bottom, roof plates and respective welds. It is recommended to minimize the number of columns for the AAST and, preferably, design the tank without the column.

Brittle fractures. Knowing the type and range of fluid, operat-

ing temperature, design life, geographic location and minimum design metal temperature (MDMT) for the plates is important. MDMT defines which grade of steel is necessary to prevent brittle fracture, and is the basis for establishing the required toughness for the selected steel. The susceptibility of the material to brittle fracture�the tensile failure of a material showing little deformation or yielding�is one of the most important material selection considerations for an AAST. Literature provides evidence regarding tank failure due to brittle fracture, even during hydrostatic tests.4 The designer should be aware of two ways to ensure that selected steel for an AAST has adequate toughness for the tank’s MDMT. The first is to proof-test each plate by impact toughness testing samples at or below the MDMT (e.g., using the Charpy V-notch method). The second method is purchasing a material with adequate toughness (preferably one with a “transition temperature” that is below the MDMT), eliminating the need for expensive impact testing, which can add 5%–10% to the cost of the plate. Older codes, along with some newer codes, do not have toughness requirements, such as the API 12 series standards and the Underwriters Laboratories (UL) tank standards. API 650 has clearly defined a basis for impact testing and toughness requirements of the plates in clauses 4.2.8 and 4.2.9, as briefly explained in FIG. 12 for plate material of A 283-C (Group 1) and an MDMT of 14°F (–10°C). The best practice for selecting steel for tank construction is to use steel above its ductile-to-brittle transition temperature. This helps ensure that the steel has adequate toughness for the temperature conditions at the specified geographical location.

Nozzle stress relief. API 650 (clauses 5.7.4.2 and 5.7.4.3) re-

quires post-weld heat treatment (PWHT) for all nozzle openings of 12 in. and larger that are fabricated in a certain shell material group. API requires that nozzle openings are prefabricated into the shell plate or thickened insert plate, and that the prefabricated assembly is thermally stress relieved together prior to installation. In the author’s experience, most contractors and vendors prefer to install nozzles in the field, do local PWHT and request a deviation to API 650, as they normally erect and weld the shell plates onsite without the nozzles to accelerate construction activity.

FIG. 10. Common

dike for several AASTs.

Plate C Plate B

Corrosion allowance. The designer should clearly define the

corrosion allowance for AAST components to meet both economic and safety requirements. Normally, 1.5 mm is the minimum thickness for the shell, and 0.75 mm for the roof. However, based on the water/corrosive elements content and the water

Plate A Roof-plate joint FIG. 11. Roof-plate

No seal weld; vapor/gas can enter between plates and condense by temperature change

joint as per Figure 5-3A of API 650. Hydrocarbon Processing | OCTOBER 2017

81

Environment and Safety

Thk. > 9-mm Figure 4-1

Impact test is required

A 283-C MDMT = –10°C Thk. < = 9-mm Figure 4-1

Impact test is not required

Average value of 3 nos test < 41 J Table 4-4 Average value of 3 nos test > = 41 J Table 4-4

One value of 3 nos test < 20 J and > = 2/3 (20J) = 13.3 J (4.2.8.3 and Table 4-4) Two values of 3 nos test < 20 J and > = 2/3 (20J) = 13.3 J (4.2.8.3 and Table 4-4) One value of 3 nos test > = 2/3 (20J) = 13.3 J (4.2.8.3 and Table 4-4) FIG. 12. Example

Reject

Accepted

Reject, re-test new 3 nos specimens (4.2.8.3)

All of 3 nos test > = 20 J (4.2.8.3 and Table 4-4)

Accepted

One value of 3 nos > = 20 J (4.2.8.3 and Table 4-4)

Reject

of impact test chart according to API 650.4

to remove a larger portion of the shell, weld the nozzle into a new plate, conduct a PWHT, butt-weld the inserted assembly into the shell and then weld a re-pad onto it. This requirement is unclear and must be addressed by API in detail. Takeaways. Many studies have been conducted regarding the

causes of accidents (mainly fires) on AASTs in oil, gas and petrochemical facilities. Fishbone diagrams are used to categorize and list causes and sub-causes in groups such as operational error, lighting, static electricity, tank crack/rupture, maintenance errors and others. However, design is rarely reviewed and addressed�it is assumed that, if the design is based on API and other standards, then it meets all technical requirements and the designer’s job is completed. It should be noted that standards, codes and regulations only cover the minimum (and not all) technical requirements. Most tank accidents could have been avoided if good engineering practice in design had been implemented and executed. LITERATURE CITED 1

FIG. 13. Nozzle

to inserted plate assembly.

However, API does not allow local stress relief of a nozzle after it has been welded into the shell plate. A designer cannot accept this deviation because, as FIG. 13 shows, the local stress relief in the field cannot be conducted for a nozzle-to-shell weld (Point 1). However, it can be done for a flange-to-pipe weld (Point 2). Schedule and cost are not valid reasons for deviating from codes, standards and regulations. The requirement is clear for a new AAST, but, when API 653 is used regarding nozzle replacement of an existing tank, the PWHT requirement refers to API 650. For an existing tank that requires a shell nozzle replacement for any reason, it is required 82


“A Guide to the Project Management Body of Knowledge: PMBOK Guide,” Project Management Institute Inc., 5th Ed., January 2013. 2 Chang J. and C.-C. Lin, “A study of storage tank accidents,” Journal of Loss Prevention in the Process Industries, December 2004. 3 “Welded steel tanks for oil storage,” American Petroleum Institute, API 650, 11th Ed., Figure 5-3A, June 2007. 4 “Design and construction of large, welded, low-pressure storage tanks,” American Petroleum Institute, API 620, 11th Ed., January 2008. SHAHAB ZARDYNEZHAD is an

Alberta-based Registered Senior Mechanical Equipment Engineer with 25 yr of experience working on many of the world’s largest oil, gas and petrochemical projects. His areas of specialization and industrial experience include the engineering, design, procurement, inspection, installation, startup, maintenance and troubleshooting of AASTs. He holds a BS degree in mechanical engineering from the University of Petroleum, an MS degree in industrial engineering from the Iran University of Science & Technology (IUST), and an MEng degree in project management from the University of Calgary.

Environment and Safety K. DEAN, MSA Senscient, Poole, UK

Turn to open-path laser detectors for HF gas safety Process engineers at a refinery in one of the Rocky Mountain states were concerned about detecting the potential accidental release of hydrogen fluoride (HF) gas at their plant as part of a recent series of upgrades to the facility. HF is a catalyst commonly used in petroleum refining alkylation processes that create octane-boosting additives for gasoline produced at the plant. The process engineers take great pride in living and working in their community, and the refinery is located near one of the region’s major cities, which is recognized as a regional hub for its strong economy, including oil and gas production and refining. Maintaining a safe working environment at the refinery is the top priority for the company’s process engineers. The potential presence of hazardous toxic, flammable and combustible gases requires the plant to employ a highly reliable safety monitoring system with advanced detection sensor technologies. The refinery (FIG. 1) has a crude oil capacity of more than 50 Mbpd. Crude oil is purchased from local producers and imported from Canada via pipeline. The plant’s refined products are distributed primarily to the eastern slope of the Rocky Mountain region, including the Denver metro area, eastern Wyoming and western Nebraska.

The solution. To help the refinery solve its HF gas detection

concerns, the process engineers at the plant contacted a manufacturer and told it that they were seeking a highly reliable HF gas monitoring solution that would provide the maximum level of protection with the highest service uptime availability and no maintenance. The manufacturer’s team recommended a gas monitoring system composed of four individual enhanced laser diode spectroscopy (ELDS) open-path HF gas detection sensors, creating a full detection perimeter around the alkylation unit’s processing area. The open-path gas detector (OPGD) shown in FIG. 2 relies on ELDS to detect specific toxic and flammable gases. In the event of a gas leak, the sensor’s laser technology recognizes and analyzes a gas’ specific harmonic fingerprint�which, in this case, is HF gas�and issues an alarm when gas is present. During normal operation, some of the laser light is reflected continuously through a sample of the target gas contained by a hermetically sealed reference cell. This design ensures that the laser remains locked on the selected gas wavelength for the specific target gas, such as HF. The detector’s harmonic fingerprint technology (FIG. 3) helps ensure precise gas recognition, eliminating the potential for false alarms, even during adverse environmental conditions.

The problem. Alkylation is a conversion process used in many

petroleum refineries, creating additives that boost the octane of gasoline by converting isobutanes into alkylates. This process requires the presence of a strong acid acting as a catalyst, and in many refineries (including this facility), HF is the acid used. Refinery alkylation processes with HF pose two process safety operating hazards. First, large amounts of hydrocarbons are processed, which are both flammable and explosive. The accidental release of toxic HF gas is also a serious potential threat to plant employees and the surrounding area, should a gas cloud form and be carried away on a windy day. The US Occupational Safety and Health Administration (OSHA) and the Environmental Protection Agency (EPA) regulate HF as a highly toxic substance. If HF is released into the atmosphere, it often rapidly forms dense vapor clouds that hover near land and can travel over great distances. Like other powerful acids, HF can cause deep, severe burns and damage the eyes, skin, nose, throat and respiratory system.

FIG. 1. The

potential presence of hazardous toxic, flammable and combustible gases requires refineries to employ a highly reliable safety monitoring system. Hydrocarbon Processing | OCTOBER 2017

83

Environment and Safety

FIG. 3. Multipoint harmonic

FIG. 2. A

gas monitoring system composed of four individual enhanced laser diode spectroscopy (ELDS) open-path HF gas detection sensors was recommended to surround the alkylation unit’s processing area.

False alarms pose a serious problem with many gas and flame detection technologies. They can result in excessive plant downtime, which often requires complex investigations and regulatory reporting. From a safety perspective, frequent false alarms lead to a lack of confidence by employees in the detection technology, and a culture of apathy that can cause employees to fail to act promptly during an actual emergency event. Class 1 eye-safe lasers are used to penetrate thick fog, heavy rain and snow beyond the capability of traditional open-path infrared (OPIR) detectors. With its automated safety integrity self-check, the typical OPIR sensor gas checks and recalibrations that usually require field technician time to address are not needed. Unlike electrochemical cells, ELDS sensors are also immune to sensor poisoning and interferent gases, thanks to their gas-specific harmonic fingerprint detection. The refinery plant team reported that the new HF gas detectors were relatively easy to install, commission and operate. Bluetooth wireless technology with advanced smart diagnostics are an additional feature. No technician or physical inter vention is required for local interrogation, event log downloading and troubleshooting. The process engineers at the refinery considered this gas detector’s performance to be excellent for their application, 84


fingerprint illustration.

with a fast response time (< 3 sec) compared to other toxic gas detection technologies. The ELDS gas detector features separate transmitter and receiver assemblies, which are certified for use in potentially explosive atmospheres and can detect HF over distances of 5 m–120 m. The ELDS gas detectors are constructed of high-grade, corrosion-resistant 316 stainless steel. Unlike traditional point detectors and other field instruments, HF corrosion is of less concern as this transmitter/receiver arrangement allows them to be located away from the high HF concentrations found close to a potential leak source. ELDS detectors are ideally suited for onshore, open and enclosed environments, including the cold winter temperatures in the Rocky Mountain states, or the high temperatures in the Middle East or Africa. Heated optics provide service over a wide temperature range, from –67°F–140°F (–55°C–60°C), and the detector is hazardous area-approved to CSA, UL Class 1, ATEX, IECEx, EAC and INMETRO standards. Results. The

refinery process engineering team continues to be satisfied with the performance of the open-path gas detector for HF monitoring, and no missed events or false alarms have occurred since the detection system was put into service. The engineers also appreciate having virtually zero maintenance detectors, which are expected to provide a significant operational maintenance cost savings over other types of HF gas detection technologies. KEVIN DEAN is a Product Manager with MSA Safety, and

has more than 35 yr of experience in industrial gas detection. He has held numerous sales, marketing and product management roles within the industry before joining Senscient, which is now part of MSA Safety. Mr. Dean has an HNC in engineering and a postgraduate diploma in marketing with the UK Chartered Institute of Marketing.

SULFUR

MANAGEMENT

Special Supplement to

COSO case study: 22 years of reliable sulfur removal S–86

CORPORATE PROFILES

Veolia S–91 COVER PHOTO

Moving sulfur granules into piles. Image courtesy of WorleyParsons.

SULFUR MANAGEMENT

COSO CASE STUDY: 22 YEARS OF RELIABLE SULFUR REMOVAL M. KOLAR, COSO Operating Co., California; W. ECHT and S. OSGOOD, Merichem Co., Houston, Texas

Coso Operating Co. LLC operates a injected into the geothermal field with 300-MW electricity generation facility at water (H2O). Over time, this H2S abatethe China Lake Naval Weapons Station, ment method became costlier, mostly due approximately 170 mi northeast of Los An- to compressor maintenance. In 1993, the geles, California, at Coso Junction. After first of three LO-CAT units was installed. steam wells are tapped and gathered, they After startup, the non-condensable carproduce electricity from the renewable bon dioxide (CO2 ) and H2S were flashed, geothermal energy source. The produced compressed and routed to the LO-CAT steam is passed through a set of turbines/ unit for sulfur removal before being emitgenerators. Non-condensable vapors are ted into the atmosphere. separated from the condensed steam (waThe LO-CAT process has been removter) at low pressure. Finally, the brine is re- ing H2S at the site for the past 22 yr. This injected into the geothermal field. technology greatly reduced sulfur emisThe non-condensable vapors cannot sion exceedances and operating costs rela be vented to the atmosphere until small tive to technologies used prior to installing amounts of hydrogen sulfide (H2S) are the LO-CAT unit.1 removed. When the plant was started up The site has a total of four power initially, the H2S-laden vapors were regeneration facilities, with two of them TABLE 1. LO-CAT

Plant name Navy 1 Navy 2 Navy 210

Power generated, MW 75 90 90

TABLE 2. LO-CAT

Plant name Navy 1 Navy 2 Navy 210

units in operation at the facility LO-CAT installation date 1993 1993 1996

unit design parameters

Raw gas flowrate, sft 3m 3,688 6,774 6,228

Inlet H2S, vol% 1.18 1.55 3

*Long tons per day

FIG. 1. Simplified flow diagram of the LO-CAT unit.

S–86

Operating status 24/7 On standby 24/7

SULFUR MANAGEMENT | OCTOBER 2017 | HydrocarbonProcessing.com

Recovered sulfur, ltpd* 2.4 5.8 10

containing LO-CAT units: the Navy 1 power plant and Navy 2 power plant. The site has a total of three LO-CAT units (TABLES 1 and 2). The Navy 2 site has two LO-CAT units, the Navy 2 unit and the Navy 210 unit. Only the Navy 210 is discussed here, since the Navy 2 unit is only periodically operated. Overall, the units have operated well throughout their history.

LO-CAT process description and process flow. The LO-CAT process converts H2S contained in the raw feed gas into elemental sulfur (FIG. 1). Before entering the LO-CAT unit, raw feed gas passes through an activated carbon bed to absorb mercury and other heavy metals. The raw gas then enters the autocirculation vessel where the H2S is absorbed into a proprietary LO-CAT catalyst solution. The catalyst is deactivated in the absorber section where H2S is con verted into elemental sulfur. Subsequently, the catalyst is regenerated in the oxidizer section of the same autocirculation vessel. Regeneration is achieved by contacting the LO-CAT solution with oxygen contained in air. The air and sweetened gas exit to the atmosphere as vent gas. The LO-CAT solution is circulated between the absorber and oxidizer sections via a system of baffles and weirs, with density difference as the driving force. Elemental sulfur formed via the reaction becomes suspended in the catalyst solution. To remove the elemental sulfur from the process, a circulation pump sends a slipstream of solution to a settler vessel, which allows the sulfur to concentrate and form a slurry. The slurry is routed to a filter, which separates the sulfur from the LO-CAT solution and washes the filter cake. The sulfur is discharged into a sulfur bin while the clarified solution (i.e., filtrate) is returned to the autocirculation vessel. Even with H2O washing of the sulfur filter cake, some LO-CAT solution exits

SULFUR MANAGEMENT

with the solid sulfur. Makeup catalyst is 8,000 ppmv and 13,000 ppmv), with averadded to maintain the solution at opti- age concentrations close to 1 vol%. mum concentrations. A surfactant is addNavy 1 initially produced sulfur at deed to help prevent foam and floating sul- sign rates of 2.4 long tons per day (ltpd), fur. Potassium hydroxide (KOH) is added with spikes of up to 2.7 ltpd. Over time, for pH control. the average sulfur production actually increased before declining to the current Operations review. Two key parameters sulfur production of 1.25 ltpd–1.75 ltpd. ensure consistent LO-CAT operations: The reason for sulfur production decreasing over time is that H2O condensed from 1. Prevent sulfur from settling in incorrect places the steam wells (which are now free of 2. Maintain proper solution sulfur) is injected into the geothermal reschemistry. ervoir. This process dilutes the sulfur conOperating practices keep sulfur tent of the produced steam. The LO-CAT from settling in the wrong places within unit was adapted to turndown conditions the unit. The main method is to use via the sparger shuffling procedure men“air blasts” that are placed strategically tioned previously. throughout the unit in regions of low During the early years of operation, flow. Nozzles send bursts of air into stag- the Navy 1 unit periodically experienced nant areas within the autocirculation high H2S in the vent gas. However, these and settler vessels, which prevents sulfur few instances of exceedance were lower buildup. When feed gas flows through than other technologies used to meet the unit at the process design rate, sulfur environmental standards before the LOin the solution is less likely to settle in the wrong places within the unit. Coso and Merichem have developed special flushing and “sparger shuffling” methods to prevent sulfur settling when the unit is operating at low flowrates. The gas flow to each sparger head (internal vapor distributor) is blocked, allowing gas pressure buildup. Water is then periodically flushed through the spargers to keep them clean. This “shuffling” is done approximately every 4 hr–8 hr to each sparger in rotation. As a result of this attention to detail, Coso is able to run both active LO-CAT units consistently for a full year, until the entire plant takes the mandated geothermal field shutdown. The need for a shut- FIG. 2. Sulfur data for the Navy 1 unit. down is determined by the backpressure on the raw gas compressors. An increase in raw gas backpressure indicates sulfur buildup on the floor or at the spargers of the autocirculation vessels. Unplanned outages due to high backpressure are very rare. Outages are typically due to low- or no-flow from the upstream power plants, which causes sparger plugging. The H2S removal performance of the Navy 1 and Navy 210 units is summarized in FIGS. 2 and 3. The Navy 1 unit was designed for 1.2 vol% H2S in the feed gas, but experienced highs of 1.4 vol%–1.5 vol% during its first 5 yr of operation. Those peaks came less often over the last 17 yr. The inlet H 2S has been consistent at between 0.8 vol% and 1.3 vol% (between FIG. 3. Sulfur data for the Navy 210 unit.

CAT unit was installed. One incident occurred in December 2000, when the vent gas H2S was reported at 90 ppmv. As shown in FIG. 4, this incident happened because the solution chemistry was out of balance. All readings above 30 ppmv correlate to rapid changes in the feed gas conditions that require operator response. Since early 2001, the vent gas H 2S has rarely exceeded 30 ppmv. Typical performance measures at approximately 15 ppmv, which maintains the unit within environmental permit requirements. The Navy 210 unit was designed for 3 vol% H 2S in the feed gas, but H 2S levels have not exceeded 2.8 vol%. At present, inlet H2S has declined from an average of 2 vol% to approximately 1 vol%. At startup, the Navy 210 LO-CAT unit produced sulfur at rates of 4 ltpd–6 ltpd, with spikes of up to 7.2 ltpd. Over time, the average sulfur production declined to 3 ltpd–4 ltpd.

HYDROCARBON PROCESSING | OCTOBER 2017 |

SULFUR MANAGEMENT

S–87

SULFUR MANAGEMENT

The startup and operations experi- working catalyst solution within a defined ence from Navy 1 helped the startup and range of alkalinity, catalyst concentration operation of the Navy 210 unit. Initially, and sufficient regeneration. Trending daily solution analysis results the H2S in the vent gas averaged 15 ppm, with occasional spikes ranging from 30 help operations keep the solution within ppmv–40 ppmv. Since 2007, the H 2S an optimum range. However, rapid chang vent gas has averaged less than 10 ppm, es in the inlet gas rate and composition with occasional spikes of up to 20 ppm. can cause excursions that could result in Despite these spikes, the treated gas has off-specification treating. Additional soluconsistently complied with environmen- tion testing immediately after measurable tal permit requirements. inlet gas changes helps operations maintain solution chemistry during the “upset” LO-CAT solution maintenance. In gen- and mitigate any adverse treating effects. eral, LO-CAT units offer robust, consisMerichem recommends monthly detent performance that meet unit design tailed analysis, but Coso decided to send criteria over a wide range of operating a sample to Merichem every two weeks. conditions, including varying inlet gas This process generated a large amount of flowrates and composition. This per- operating data for analysis. Merichem’s deformance is possible by maintaining the tailed analyses show that the proprietary

FIG. 4. Solution analyses for the Navy 1 unit.

chemicals solution concentrations have stayed relatively stable throughout 22 yr of operating both LO-CAT units. Coso measures the alkalinity and oxidation-reduction potential (ORP) of the regenerated LO-CAT solution daily. These measurements ensure that the unit operates reliably because these measurements determine the chemical addition rates needed to keep the solution chemistry within the required guidelines. Coso performs a shake test that measures how quickly the sulfur particles sink to adjust the rate of surfactant addition. ORP measurements indicate whether the catalyst solution is properly regenerated in the oxidizer�a key parameter for determining catalyst activity. As shown in FIGS. 4 and 5, ORP typically stayed within the optimum operating range. If the catalyst is over-regenerated, then chemical usage is higher. Under-regenerating the catalyst reduces the active catalyst available for sulfur reaction, and may cause off-specification treating in the absorber section. This issue appears to have been the case early in the operation of the Navy 1 unit, which may have caused some of the high H2S in the vent gas that was noted in FIG. 2. One discovery during the early years of the Navy 1 unit operations was the importance of alkalinity as an operating variable over pH. Most LO-CAT units monitor pH only; however, that is not the case when treating streams with high concentrations of CO2. Large amounts of acidic CO 2 in solution reach equilibrium with the basic LO-CAT solution and serve as a buffer, stabilizing the pH. This scenario makes pH measurement a less responsive indicator of the solution’s H2S solubility. In addition, increased alkalinity is known to reduce the solubility of CO2 in the LO-CAT solution. In December 2000, data from the Navy 1 unit (FIG. 4) indicated that the solution alkalinity was at an all-time low. This reading corresponded to the highest treated gas H2S content of 90 ppmv, as shown in FIG. 2. The low alkalinity caused the solubility of the H2S in the LO-CAT solution to fall so low that significant amounts of H2S were measured in the vent gas. The alkalinity and ORP in the Navy 210 unit has been more tightly controlled vs. the Navy 1 unit, due to the lessons learned during the early operation of the Navy 1 unit.

LO-CAT sulfur product. The two outFIG. 5. Solution analyses for the Navy 210 unit. S–88

SULFUR MANAGEMENT | OCTOBER 2017 | HydrocarbonProcessing.com

puts from the LO-CAT unit are the vent

SULFUR MANAGEMENT

gas and the sulfur cake. The sulfur cake is approximately 65% sulfur and 35% moisture. Washing the slurry during the drying process minimizes the loss of chemicals (and operating cost). The elemental sulfur produced by the LO-CAT unit is sold to Hondo Inc. in lieu of being sent to a landfill. Hondo blends the sulfur with gypsum, and sells it as a soil “amendment.” The Coso LO-CAT sulfur has recently been certified as meeting the guidelines for use in fertilizing organically grown crops.

Cost of operation. The LO-CAT units’ two largest operating cost components are chemicals consumption and electrical usage. TABLES 3 and 4 show the operating costs of the two units at the Coso facility. The electrical demand is constant, even when changing the sulfur load. The Navy 210 unit has a larger inventory of solution; therefore, it requires more air to regenerate the catalyst. The basis for the values is a total of 5.6 ltpd–1.6 ltpd produced by Navy 1 and 3.9 from Navy 210. Minor costs include 2

hr/d–3 hr/d of operator time to conduct lb of sulfur removed. Additionally, the solution testing and other operator tasks. Navy 210 LO-CAT unit achieved signifiOperator responsibilities include activities cant throughput turndowns of 25%–35%, for other process units within the power without adverse process effects and with plant, in addition to the LO-CAT unit. only minor operational adjustments. LITERATURE CITED

Takeaway. The data shows that the units

1

consistently meet H2S specifications, with operating costs in the range of just $0.29/ TABLE 3. Electrical

Mason, T., “CalEnergy’s Coso operation: A substantial commitment to the environment pays off,” Geothermal Resources Council Bulletin, June 1996.

demand for both LO-CAT units

Major electricity user

Navy 1

Navy 210

Air blowers, kW

50

300

Slurry pump, kW

20

30

Belt filter, kW

15

20

Total electricity, kW

85

350

TABLE 4. Operating

cost per amount of sulfur removed

Operating cost

$US/long ton

US/lb

420

0.19

93

0.04

Electrical ($0.07/kW-hr)

130

0.06

Major operating cost

643

0.29

Merichem catalyst/chemicals KOH

LIVE WEBCAST: Tuesday, November 7, 2017 11 a.m. EST / 10 a.m. CST / 4 p.m. UTC

Designing Corrosion Out of Your Fireproofing System SPEAKER

Russell Norris Intumescent Coating Specialist Sherwin-Williams

Hydrocarbon processing plants are challenged with protecting assets against both corrosion and fire damage. Corrosion poses a great threat to structural integrity – especially in a fire scenario. Corrosion underneath fireproofing can be hard to detect, and when it’s found, the costs to mitigate the corrosion and replace fireproofing materials can overwhelm plant maintenance budgets. This presentation highlights differences in fireproofing types and how each type correlates with corrosion potential. It will help owners and specifiers select the proper fireproofing materials to last the desired life of their assets, while eliminating the threat of corrosion.

Register For Free: HydrocarbonProcessing.com/Webcasts MODERATOR

Lee Nichols Editor/Associate Publisher Hydrocarbon Processing

HYDROCARBON PROCESSING | OCTOBER 2017 | SULFUR MANAGEMENT

S–89

I am resourcing the world through sulfuric acid regeneration.

witer: @veolia_na www.veolianorthamerica.com


VEOLIA

DRIVING A MORE CIRCULAR INDUSTRIAL ECONOMY THROUGH SULFURIC ACID REGENERATION According to research published by Accenture, “The circular economy could generate $4.5 trillion of additional economic output by 2030.” The research also finds that “today’s business practices will contribute to a global gap of 8 billion tons between the supply and demand of natural resources by 2030.” As the global leader in water, waste and energy management, Veolia’s a major player in driving a more circular economy for industry. This translates into services and technologies for diverting waste from landfills, moving materials up the waste hierarchy, beneficial reuse strategies, and producing quality materials and products that are competitively priced and have a smaller environmental footprint than those made with virgin materials. INVESTING IN SULFURIC ACID MANAGEMENT

Sulfuric acid, for example, is the most widely used—and reused— chemical compound in the world, with applications in nearly every industry, including refining, fertilizers, chemical manufacturing, pharmaceuticals, personal care and electronics, among others. Since acquiring the Sulfur Products business from Chemours in 2016, Veolia now owns seven operating sites across the United States that are dedicated to sulfuric acid regeneration and production. At these plants, elemental sulfur, spent sulfuric acid and recovered acid (sulfur) gases are processed into clean fuming and non-fuming sulfuric acids and other high-value sulfur derivative (HVSD) products. Four of these plants are in geographies that make strategic sense to customers, product transportation and other supply chain logistics. The other three plants in Veolia’s network are located on-site or adjacent-to a major refiner y customer. For example, Veolia’s Morses Mill sulfuric acid regeneration plant is located inside the gate at the Phillips 66 Bayway refinery in New Jersey. All of the refinery’s spent acid, plus its acid gases, are directed to this plant. The spent acid is regenerated into fresh sulfuric acid and returned to the refinery. The acid gases are processed into sulfur products and supplied to other customers in the merchant market. The plant also produces and supplies high-pressure steam to the refinery. Additionally, this plant takes spent acid from other refineries to process and return as fresh acid. RAMPING ALKYLATION MARKET

One of the primary drivers for sulfuric acid in the refining segment is the demand for alkylate. As consumer and regulatory demand for gasoline with higher octane and lower vapor pressure ratings increases, alkylate has emerged as the preferred additive. For the approximately 100 refiners with alky units in North America, alkylate is high-value and in constant demand. Sulfuric acid is one of two catalyst options for refiners running alkylation units. The other is hydrofluoric acid, in which Veolia also plays a regenerative role through its potassium hydroxide (KOH) manufacturing process. Since adding sulfuric acid regeneration capabilities, Veolia is positioned to offer a circular economy solution to every refinery with an alkylation unit. SPONSORED CONTENT

What makes Veolia’s sulfuric acid business different from other players in this market is its Acid Technology Center (ATC), which is a group of over 30 engineers, scientists and technicians who exclusively support the company’s sulfur products plants and customers. The nature of the ATC is to maintain, improve and provide customer support and to bring experts together to solve problems. This nimble group travels to customer sites as needed, supported by “assistance to operations” engineers located at every Veolia plant. OPTIONS FOR REFINERS

For refiners, sulfuric acid management is a required, but non-core, component of their operation. Available options include sending materials off-site for regeneration, or managing it through on-site regeneration plants. Both present opportunities for Veolia to leverage its capabilities, operations and maintenance experience and expertise. We can partner with a refinery to manage their spent acid at one of our existing regeneration plants, or build a new regeneration plant to meet their specific needs. Alternatively, if refiners want to outsource their existing regeneration systems, Veolia is open to operating them on their behalf. This allows them to focus on their core business of processing oil, not dealing with the challenges of running a sulfuric acid plant safely and reliably.

CONTACT INFORMATION

Veolia North America, Regeneration Services Main Office 131 Continental Dr. Suite 300 Newark, DE 19713 P: 302-444-9172 [email protected] veolianorthamerica.com HYDROCARBON PROCESSING | OCTOBER 2017 |

SULFUR MANAGEMENT

S–91

ACTIONABLE INSIGHTS INTO US NATURAL GAS FACILITIES The Energy Web Atlas is a comprehensive source of data for facilities and installations in the US natural gas industry. The web-based GIS application includes detailed data on more than 500 US gas processing plants, 400 natural gas underground storage locations, and 58 natural gas market hubs.

Data Points The platform’s mapping, search and query tools allow users to interrogate up-to-date facility and project data and find the information needed for major decisions. Data points include: • Key contacts (address, email, phone) • Plant type/flow, BTU content, infrastructure • Plant status • Owner/Operator • Gas storage capacity • Market hub/Underground storage data.

Comprehensive Functionality Built for ease-of-use, the platform facilitates access to the most relevant data, with functionality including: • Custom reporting, charting and graphing • Layer/attribute access and modification • Data exports • Production quality maps, PDF exports • Interactive charts/statistics.

Additional Data/Maps Available. Global LNG/pipeline data

To schedule a personal demonstration,

contact J’Nette Davis-Nichols at [email protected], or at +1 (713) 520-4426.

DISCOVER MORE AT ENERGYWEBATLAS.COM

Show Preview

ADIPEC 2017 A. BLUME, Executive Editor

ADIPEC 2017 to forge ties and drive growth with new downstream program 13–16 November 2017 Abu Dhabi National Exhibition Centre, Abu Dhabi, UAE • 20th edition of Abu Dhabi International Petroleum Exhibition & Conference • More than 10,000 delegates, 900 speakers and 2,000 exhibiting companies from 135 countries • Enhanced opportunities for dialogue with ministers and global business leaders, and for networking with business partners and colleagues • New in 2017: Downstream global business leaders speak at plenary and technical sessions geared toward the downstream • Returning programs: Women in Energy, Security in Energy, ADIPEC Awards, Young ADIPEC, VIP conference for Middle East Petroleum Club members www.adipec.com

The 20th edition of Abu Dhabi In- ebrate excellence in energy; the Women ternational Petroleum Exhibition & in Energy conference, which addresses Conference (ADIPEC) will take place gender balance in the energy sector; the 13–16 November at the Abu Dhabi Security in Energy conference, which exNational Exhibition Centre. Hosted by amines the real-life security threats and Abu Dhabi National Oil Company (AD- challenges facing the industry; Young NOC), and supported by the UAE Min- ADIPEC, a dedicated “edutainment” istry of Energy, the Abu Dhabi Chamber program designed to encourage students and the Abu Dhabi Tourism & Culture to choose a career in energy; and a VIP Authority, ADIPEC 2017 will promote conference program for members of the dialogue between ministers and global Middle East Petroleum Club. business leaders. ADIPEC 2017 conference sessions More than 10,000 delegates, 2,000 include Offshore and Marine, and Secuexhibiting companies and 900 speak- rity. The sessions will feature presentaers from 135 countries are projected to tions on a number of prominent topics: gather in the UAE capital for ADIPEC • Exploration and production 2017 (FIG. 1). Also, 25 Country Pavilgeoscience ions will encourage business generation • Production facilities technologies • Field development and extensions of existing bilateral trade • Operational excellence opportunities. In line with the event’s theme for 2017, • Drilling and completion technology “Forging ties, driving growth,” ADIPEC • Health, safety and environment 2017 will feature two ministerial sessions, four global business leader sessions, four (HSE) • Project engineering and downstream global business leader sessions, eight panel sessions, three offshore management plenary panels, nine C-suite dialogue ses• Gas technology • Unconventional resources sions (FIG. 2), three industry breakfasts and three topical luncheons. Conference program. The ADIPEC

conference program is organized in collaboration with the Society of Petroleum Engineers (SPE). It will feature more than 900 speakers at 200 technical and strategic conference sessions. New this year is the introduction of global downstream technical sessions to drive the conversation on downstream expansion, diversification and technology innovation, research and development. Returning conference features include the ADIPEC Awards, which cel-

FIG. 1. The

ADIPEC Exhibition is anticipated to draw more than 10,000 delegates from 135 countries to speak with oil and gas companies operating worldwide. Hydrocarbon Processing | OCTOBER 2017

93

ADIPEC 2017

• Improved and enhanced oil recovery • People and talent • Petroleum advanced analytics. The full-day Women in Energy program will focus on the opportunities for, and achievements of, women working in the oil and gas industry. The co-located Security in Energy conference will return for a second year, recognizing the growing and critical importance of cy-

FIG. 2. C-level

ber and infrastructure security within oil and gas operations. New downstream program. ADI-

PEC’s global downstream plenary panels are new in 2017. At a time when integration and collaboration across the industry’s value chain continue to become increasingly significant, ADIPEC’s global downstream technical sessions are set to drive the conversation on down-

executive speakers at ADIPEC 2016.

stream investment, expansion, diversification and development. The panels will include top executives from ADNOC, Borealis, BP, CEPSA, EQUATE Petrochemical Co., Mubadala Investment Co., Nova Chemicals, OMV, Petronas, PARCO, TAKREER and Total. Day 2 of ADIPEC will feature a panel on driving growth and investment into downstream refining and petrochemicals. Day 3 will include panels on refining and petrochemicals integration and downstream expansion for sustainable growth, redefining feedstock allocation and diversification, and rethinking technology to enhance industry performance and counteract market disruptors. The downstream technical tracks of the conference will feature sessions on refining and petrochemicals technology and operations; marketing and distribution of refined products; connected performance technology and services; gas processing and hydrogen technology and operations; project management, engineering technology and implementation; and operational excellence, maintenance and HSE.

BE INFORMED. BE INSPIRED. BE THERE.

11 – 15 June Frankfurt am Main, Germany ›

World Forum and Leading Show for the Process Industries

›

3,800 Exhibitors from 50 Countries

›

170,000 Attendees from 100 Countries

www.achema.de

94


BOB ANDREW, TECHNICAL EDITOR [email protected]

Innovations

New testing facility for process pumps

has been increased significantly to yield The Series 2000 digital converter has more accurate results. When limit values an onboard microcontroller. It is ready are exceeded, the integrated monitoring to use with no programming. The Petfunction triggers an alarm and, if neces- rochemical software provided with the sary, utilizes direct communication with converter calculates specific gravity, API the drive unit to safely switch off the test gravity and can also display corrected motor. In addition, there is additional gravities using information generated data that was not available previously. All from the density cell. Select 2 at www.HydrocarbonProcessing.com/RS measured values are stored synchronously so they can be used for follow-up analyses and serve as the basis for condition Real-time answers to monitoring and predictive maintenance systems. Furthermore, the low-voltage engineering challenges system, with a completely new design and ANSYS has released a technology preits own medium voltage transformer, en- view of a new product called ANSYS Dissures malfunction-free power supply to covery Live (FIG. 3). This new simulation the test objects and devices. environment provides real-time, instantaThanks to increased test capacity, re- neous answers to engineering challenges. placement of safety devices and the pro- Prior to its 1Q 2018 release, interested curement of specially developed, state-of- users can download (or can use a limited the-art measurement technology, Lewa experience cloud option) at no charge, at has created the best possible conditions ansys.com/discovery. ANSYS would like for innovative product developments. In feedback so they can learn how to best use the future, prototypes of products, par- this revolutionary new technology and ticularly newly developed process pumps make improvements. Select 3 at www.HydrocarbonProcessing.com/RS and individual assemblies such as drive units or pump heads, will be completed and validated in the testing room. Endurance testing with various diaphragm materials or clamping methods can be used to examine long-term behavior of the products at a customer’s facility. This guarantees reliability and durability.

At the beginning of 2017, Lewa GmbH commissioned its newly designed process pump testing room (FIG. 1). The previous testing facility was outdated, so it was completely refurbished and the technology was modernized. The new testing room can accommodate up to four large process pumps, with a total power consumption of 800 kW (and each separately driven by a variable frequency de vice). These pumps can be operated and monitored in parallel. In this process, all relevant process and pump data are continuously recorded. This data validates test operation of the machines through endurance testing, and also provides the necessary initial data required for analyses during the product development process. In addition to its high-function density, the testing facility features comprehensive system integration of process/drive/automation and measurement technology. The process water system in the testing room supplies the conditioned con veying fluid for test operation of test objects in a fully automated process. An adiabatic heat exchanger system, with a redundant design, has been installed to discharge heat in the conveying fluid that is generated through pumping. The main Select 1 at www.HydrocarbonProcessing.com/RS aim was to achieve hygienic, environmentally-friendly system operation, ensuring that no legionella or deposits can Petroleum software form. Motors with an electric connected displays corrected load exceeding 500 kW can be used as a pump drive. Regardless of the load curve gravities and type of electrical machines used, the Measure the density of crude oil, use of state-of-the-art frequency inverter butane, diesel fuel, distillates, gasoline, technology guarantees zero circuit feed- ethane gas, isobutene, jet fuels, LPG, back during power train operation. Tem- kerosine or propane using a Dynatrol CLperature increases in the area around the 10-HY density cell and Series 2000 dentest field are minimal, thanks to liquid- sity digital converter (FIG. 2). cooled power electronics. Dynatrol cells come in a full range of A data detection and recording system corrosion-resistant materials that are both developed in-house is used to record the weather-tight and explosion proof. They measured values required for short- and also serve broad temperature and preslong-term testing to determine the condi- sure ratings, with no moving parts for a tion of the test object. The scanning rate long operating life.

FIG. 1. Process

FIG. 2. Liquid

pump testing facility at Lewa.

and gas density measurement.


94–A

Innovations

Convert HF alkylation to H2SO4 alkylation DuPont Clean Technologies has launched DuPont’s ConvEx HF alkylation conversion technology, the first costeffective solution that enables refiners to convert volatile and toxic hydrofluoric acid (HF) alkylation units to the safer sulfuric acid alkylation technology ( FIG. 4). The technology offers refiners the opportunity for significant capacity increases at minimal additional cost. Historically, the expense of converting from HF to sulfuric acid alkylation was estimated by the industry at 80% of the cost of a grassroots sulfuric acid alkylation unit of a similar size. This perceived highconversion cost and the lack of any other economic benefits deterred refiners from committing to this change. By reusing much of the existing equipment, conversion with ConvEx technology is estimated to be significantly lower. At approximately 40%–60% of the cost of a grassroots sulfuric acid alkylation unit, this new technology represents a step-change reduction in cost. In situations where plot space is available near the existing facility, downtime can be reduced by installing new

FIG. 3. 3D

model showing fitting flow regime.

equipment during normal operations. With the installation of new equipment, the remainder of the conversion work can be finalized within a typical 30 d–45 d turnaround window. The suite of HF conversion technologies offered by DuPont was developed with flexibility to adapt to the strategic objectives of refiners operating HF alkylation units. One of the options that was de veloped utilizes conventional STRATCO Contactor reactors to achieve optimum product quality. A second option incorporates a novel reactor design, allowing for significant cost savings with only a minor performance debit. By taking into consideration the current HF alkylation unit technology and configuration, DuPont’s customized solutions can be tailored to meet the specific business needs of the refinery. Fundamental to the expansion aspect of the ConvEx technology is the difference in the way in which isobutane is recycled between the HF and sulfuric acid alkylation technologies. HF alkylation units provide all isobutane to the reaction zone by recycling it from the fractionation section, while sulfuric acid alkylation units provide half of the required isobutane from fractionation and half from the refrigeration section. This difference is significant, as conversion from HF to sulfuric acid alkylation means isobutane required from fractionation is cut in half, freeing up fractionation space; therefore, effectively doubling the capacity of the alkylation unit without requiring any significant changes to the fractionation equipment. HF alkylation conversion using DuPont’s ConvEx technology is unique and innovative, but the design elements and know-how that have made STRATCO alkylation the leading global alkylation technology are incorporated in these con version solutions. The resulting product includes proven technology applications, a robust design, and equipment that is familiar to refinery operators. Select 4 at www.HydrocarbonProcessing.com/RS

Programmable Logic Controller (PLC) awarded critical certification Honeywell Process Solutions’ ControlEdge PLC ( FIG. 5) has received ISASecure Embedded Device Security Assurance (EDSA) Level 2 certification by exida, the ISO 17065 accredited certification body who conducted the assessment. The certification recognizes the integrity of both the PLC and its development lifecycle, and includes rigorous communication robustness testing, as well as functional and software development security assessments. ControlEdge PLC is the first PLC to be awarded the ISASecure Level 2 certification, which builds on the Level 1 capabilities by providing additional security features such as the confidentiality and integrity of data at rest and in motion, detection of unauthorized changes and additional protection against Denial of Service attacks. The ISASecure-compliant controller has advanced features including a secure boot capability to pre vent uploading of unauthorized software and a built-in firewall. PLC communication can also be secured using IPSec to prevent man-in-the-middle attacks and unauthorized access. The ISASecure program has been de veloped by the ISA Security Compliance Institute (ISCI) based upon the ISA/IEC 62443 series of standards, with a goal to improve cyber security for industrial automation and control systems. ControlEdge PLC is used in a variety of balance-of-plant control applications, such as water treatment, pump systems and material handling. It is part of a next generation family of controllers offering unprecedented, secure connectivity through all levels of process and business operations. Honeywell’s integrated Distributed Control System (DCS)/PLC approach provides a significant reduction in integration efforts and project costs. These benefits are achieved through seamless integration of a DCS with PLC, remote terminal unit (RTU), HART field device asset management, and HMI panel PC. Select 5 at www.HydrocarbonProcessing.com/RS

FIG. 4. HF

alkylation vs. H2SO4 alkylation

schemes.

94–B

FIG. 5. certified


programmable logic controller.

An expanded version of Innovations can be found online at HydrocarbonProcessing.com.

MARKETPLACE / [email protected] / +1 (713) 412-2389

Why should you filter your water? SALE

RENT

▪

▪

LEASE

Superheat & Saturated Boilers to 250,000pph 750psig Custom Design & Manufacture Too! In Stock Assorted Sizes Ultra Low Nox Available SCRs Available

▪ ▪

▪

Scale formation reduces the heat transfer rate and increases the water pressure drop through the heat exchanger and pipes. In fact, one study has shown that .002" fouling will increase pumping needs by 20%.


2672 S. La Cienega Blvd, Los Angeles, CA 90034 USA (800) 336-1942 (310) 839-2828 www.tekleen.com [email protected]

Epoxy Resists Harsh Chemicals


Epoxy Compound EP41S-1HT

Designed to Meet Specific Application Requirements • Fuel, alcohol & solvent resistant • Serviceable from -60 F to +400 F °

°

Hackensack, NJ 07601 USA +1.201.343.8983 • [email protected]

www.masterbond.com



Call +1 (713) 412-2389 for details about Hydrocarbon Processing’s Marketplace

Detailed and up-to-date information for active construction projects in the refining, gas processing, and petrochemical industries across the globe. ConstructionBoxscore.com


95

ADVERTISER INDEX / HydrocarbonProcessing.com The first number after the company name is the page on which an advertisement appears. The second number is the Reader Service Number. There are two ways readers can obtain product and service information: go to www.HydrocarbonProcessing.com/RS, follow the instructions on the screen, and your request will be forwarded for immediate action, or go online to the advertiser's website listed below.

Company

Page

RS#

Website

www.info.hotims.com/66267-152

(65)

(74)

Ametek Process Instruments....................... 76 (160) www.info.hotims.com/66267-160


China Petrochemical Technology Co. Ltd .......18

(69)

Neste Jacobs Oy .......................................... 13

Construction Boxscore Database............... 46


Cudd Energy Services ................................. 39 (156) www.info.hotims.com/66267-156

(151)

OHL ........................................................... 66 (158) www.info.hotims.com/66267-158

Prosernat .................................................. 32 (153) www.info.hotims.com/66267-153

HP Subscribers ........................................ 52

Shell Global Solutions .................................16

(92)


(56)


Idrojet........................................................ 71 (159) www.info.hotims.com/66267-159

(84)


Gulf Data—EWA ................................. 34, 92

Harsco Industrial Air-X-Changers/Hammco .. 22



HP Webcast .......................................49, 89 (59)

(62)


Merichem Company......................................5


HP Marketplace....................................... 95 (51)

RS#

Maire Tecnimont SpA ................................... 31 (154)


Events—WGLC......................................... 99


Page

Kobelco ...................................................... 2

Gulf Publishing Company


Company Website

Dechema ...................................................94

ExxonMobil Chemical Company ..................... 6

Ametek Process Instruments....................... 27 (155)

Axens.......................................................100

RS#

Evonik Oil Additivies USA Inc ....................... 65 (157)


Ametek Process Instruments.......................40

Page

Website

American Petroleum Institute ..................... 24 (152) Ametek Process Instruments........................ 11

Company

Veolia.........................................................91

(72)


ZymeFlow Decon Technology ...................... 20

(93)


This Index and procedure for securing additional information is provided as a service to Hydrocarbon Processing advertisers and a convenience to our readers. Gulf Publishing Company is not responsible for omissions or errors.

Catherine Watkins, Publisher Phone: +1 (713) 520-4421 E-mail: [email protected] www.HydrocarbonProcessing.com

SALES OFFICES—NORTH AMERICA

RUSSIA/FSU Lilia Fedotova Anik International & Co. Ltd. Phone: +7 (495) 628-10-333 E-mail: [email protected] TURKEY, WESTERN EUROPE Hamilton Pearman Phone: +33 608 310 575 [email protected]

EASTERN UNITED STATES, EASTERN CANADA Merrie Lynch Phone: +1 (617) 357-8190 Mobile: +1 (617) 594-4943 E-mail: [email protected]

UNITED KINGDOM/SCANDINAVIA Patrick Djuma Phone: +44 20 3409 2243 E-mail: [email protected]

GULF COAST, SOUTH TEXAS Keith Smith Phone: +1 (713) 520-4449 E-mail: [email protected]

SALES OFFICES—OTHER AREAS

MIDWEST, NORTH TEXAS Josh Mayer Phone: +1 (972) 816-6745 E-mail: [email protected] WESTERN UNITED STATES, WESTERN CANADA Rick Ayer Phone: +1 (949) 366-9089 E-mail: [email protected] SALES OFFICES—EUROPE ITALY, EASTERN EUROPE Fabio Potestá Mediapoint & Communications SRL Phone: +39 (010) 570-4948 E-mail: [email protected]

96


AFRICA, MIDDLE EAST Tanya Mbaluli Twiga Media Partner Phone: +254 722 376 972 Email: [email protected] CHINA—HONG KONG Iris Yuen Phone: +86 13802701367 (China) Phone: +852 69185500 (Hong Kong) E-mail: [email protected] INDIA Manav Kanwar Phone: +91-22-2837 7070/71/72 Mobile: +91-98673 67374 E-mail: [email protected]

INDONESIA, MALAYSIA, SINGAPORE, THAILAND Peggy Thay Publicitas Singapore Pte Ltd Phone: +65 6836-2272 E-mail: [email protected] JAPAN Yoshinori Ikeda Pacific Business Inc. Phone: +81 (3) 3661-6138 E-mail: [email protected] KOREA YB Jeon Storm Associates Inc. Phone: +82 (2) 755-3774 E-mail: [email protected] MEXICO, CENTRAL AMERICA, SOUTH AMERICA Marco Antonio Monteiro Mobile: +55 21 99616-4347 E-mail: [email protected] CLASSIFIED SALES Laura Kane Phone: +1 (713) 412-2389 E-mail: [email protected]

DATA PRODUCTS J’Nette Davis-Nichols Phone: +1 (713) 520-4426 E-mail: [email protected]

REPRINTS Jill Kaletha, Foster Printing at Mossberg & Co. Phone: +1 (800) 428-3340 x 149 E-mail: [email protected]

ALISSA LEETON, CONTRIBUTING EDITOR [email protected]

Events

OCTOBER AFPM Operations and Process Technology Summit, Oct. 2–4, JW Marriott Austin, Austin, Texas P: +1 202-457-0480 [email protected] www.afpm.org/Conferences Polyurethanes Technical Conference, Oct. 2–4, New Orleans Marriott, New Orleans, Louisiana [email protected] www.americanchemistry.com Emerson Global Users Exchange, Oct. 2–6, Minneapolis Convention Center, Minneapolis, Minnesota EmersonExchange@ Emerson.com www.emersonexchange.org Coalescence, Filtration & Separations Technical Training Seminar, Oct. 3–4, Research Forest Convention Center, The Woodlands, Texas P: +1 832-510-8191 [email protected] www.nexosolutions.com Lubricants and Base Oils Symposium, Oct. 18–20, Hotel Esplanade, Zagreb, Croatia P: +385 1487-3549 [email protected] www.lubricants.goma.hr Gas Asia Summit, Oct. 25–27, Marine Bay Sands Expo and Convention Centre, Singapore P: +65 6422-1159 [email protected] gasasiasummit.com Chem Show, Oct. 31–Nov. 2, Javits Convention Center, New York, New York P: +1 203-221-9232 [email protected] chemshow.com

NOVEMBER Women’s Global Leadership Conference in Energy, Nov. 1–2, Gulf Publishing Company Events, Hyatt Regency Houston, Houston, Texas WGLConference.com/2017 (See box for contact information)

International Gas Union (IGU) Safety Summit, Nov. 6–7, Washington, D.C. www.igu.org CRU, Sulphur Conference & Exhibition, Nov. 6–9, Hilton Atlanta, Atlanta, Georgia P: +44 0-207-903-2444 [email protected] www.crugroup.com API Cybersecurity Conference for the Oil & Natural Gas Industry, Nov. 7–8, Marriott Woodlands Waterway, The Woodlands, Texas (See box for contact information) Latin American Petrochemical Annual Meeting, Nov. 11–14, Royal Tulip Hotel, Rio de Janeiro, Brazil [email protected] www.apla.com Abu Dhabi International Petroleum Exhibition & Conference (ADIPEC), Nov. 13–16, The Abu Dhabi National Exhibition Centre, Abu Dhabi, UAE P: +971 2-444-4909 adipec.delegate@ dmgeventsme.com www.adipec.com API Fall Refining and Equipment Standards Meeting, Nov. 13–16, Hyatt Regency Dallas at Reunion, Dallas, Texas (See box for contact information)

JUNE 2018

Turbomachinery & Pump Symposia, Dec. 12–14, George R. Brown Convention Center, Houston, Texas P: +1 979-845-7417 [email protected] tps.tamu.edu

IRPC Europe, June 5–6, Gulf Publishing Company Events, Milan, Italy HPIRPC.com/Europe (See box for contact details)

FEBRUARY 2018 Egypt Petroleum Show (EGYPS), Feb. 12–14, New Cairo Exhibition Centre, Cairo, Egypt P: +971 0-2697-0504 [email protected] www.egyps.com

MARCH 2018

ACHEMA, June 11–15, Messe Frankfurt, Frankfurt, Germany P: +49 69-756-4100 www.achema.de World Gas Conference, June 25–29, Walter E. Washington Convention Center, Washington, D.C. P: +44 20 -7978-0775 www.wgc2018.com

AUGUST 2018

Asia Turbomachinery & Pump Symposium, March 12–15, Suntec Singapore Convention & Exhibition Centre, Singapore P: 979-845-7417 [email protected] atps.tamu.edu

ONS, Aug. 27–30, Stavanger, Norway P: +47 932-56-125 [email protected] www.ons.no/2018 HP Awards, Aug. 30,

Eastern Mediterranean Gas Conference (EMGC), March 21–22, Gulf Publishing Company Events, Nicosia, Cyprus EMGasConference.com (See box for contact information)

APRIL 2018

Downstream Caspian and Central Asia, Nov. 28–30, Baku, Azerbaijan www.globuc.com/ downstream-centralasia

NACE Corrosion Conference & Expo, April 15–19, Phoenix Convention Center, Phoenix, Arizona P: +1 800-797-6223 nacecorrosion.org

Easyfairs, Tank Storage Germany, Nov. 29–30, Hamburg Messe und Congress, Hamburg, Germany P: +44 0-208-843-8800 [email protected] www.easyfairs.com

STRATCO Alkylation Best Practices Workshop, April 23–26, Omni Amelia Island Plantation Resort, Amelia Island, Florida www.dupontbpw.com

DECEMBER

MAY 2018

ASME Gas Turbine India, Dec. 7–8, Sheraton Grand Bangalore at Brigade Gateway, Bangalore, India [email protected] www.asme.org

Petrochemical & Refining Congress (PRC) Europe, May 7–8, Berlin, Germany P: +312 0808-7321 [email protected] prceurope.com

Gulf Publishing Company Events, Houston, Texas HydrocarbonProcessing.com/ Awards (See box for contact information)

SEPTEMBER 2018 IRPC Americas, Sept. 25–26, Gulf Publishing Company Events, Houston, Texas HPIRPC.com/Americas (See box for contact information) International Pipeline Expo, Sept. 25–27, Telus Convention Centre, Calgary, Alberta, Canada P: +1 888-799-2545 [email protected] internationalpipeline exposition.com Hydrocarbon Processing /

Gulf Publishing Company Events P: +1 713-520-4475 [email protected] [email protected] American Petroleum Institute (API) P: +1 202-682-8195 [email protected] www.api.org Hydrocarbon Processing | OCTOBER 2017

97

MIKE RHODES, TECHNICAL EDITOR [email protected]

People

EnLink Midstream has named Rhonda L. Price to succeed Jennifer K. Johnson as Senior VP of human resources. Ms. Johnson is retiring after more than 13 yr of service to EnLink and its predecessor, Crosstex Energy. Ms. Price has more than 30 yr of global and national-level HR experience, including 7 yr as VP of HR at EnLink, where she has been involved in every mergers and acquisitions transaction and set EnLink’s employee benefits strategy. Prior to her career at EnLink, she served as VP of HR at Centex Corp.

Braun Intertec has appointed Keith Linton as Senior Environmental Consultant. Mr. Linton brings decades of environmental site assessment and remediation experience working with refinery and oil and gas clients, oilfield service companies and chemical manufacturers. For 20 yr, he has performed risk-based assessments on the Resource Conservation and Recovery Act (RCRA), Superfund, Voluntary Cleanup Program (VCP) and other corrective action sites, and developed and implemented strategies to successfully achieve site closure. He has managed large, complex remediation projects for energy clients, including one that recently successfully received approval for an extensive cleanup at a former refinery site.

98

Mehdi Adib has joined

SNC-Lavalin as VP of downstream for oil and gas. He oversees all downstream activities for SNC-Lavalin, including chemicals, petrochemicals and syngas, refining, fertilizers, utilities and related infrastructure. Mr. Adib joins SNC-Lavalin from his role as CEO of Best Consulting and Enterprise Co., where he helped the refining, petrochemical and related infrastructure industries to develop and expand their international markets. He has also supported the development and implementation of major projects in the US, Canada, Brazil, France, Saudi Arabia, Kuwait, Abu Dhabi, Russia, Singapore and Malaysia. Prior to Best Consulting, Mehdi worked for major international companies such as C.E. Lummus (now CB&I Lummus), Bechtel and Sipchem.

BCCK Holding Company (BCCK) has appointed Matt Amilian as Director of business development to support company growth in the Oklahoma, Appalachian and Rocky Mountain regions. He joins BCCK from Condit Co., where he served as gas processing market manager. Mr. Amilian is VP of the Mid-Continent Gas Processors Association (GPA) Midstream Chapter, and a member of the Rocky Mountain and Permian Basin GPA Midstream Chapters.


The board of directors of Kingsbury Inc. has appointed Michael Brawley to the position of President and CEO. He will succeed William Strecker, who has held the position since May 2007. Mr. Brawley began his Kingsbury career in 1985 as a Sales Applications Engineer, and held multiple positions within the sales department before being promoted to Division Manager, and then Director, of Kingsbury’s newly formed magnetic bearings division. Following roles as Division Manager of the repair and service division and then VP of the repair and service division, he assumed the title of Chief Operating Officer (COO) of North America Operations, where he was responsible for Kingsbury’s stateside manufacturing operations.

ValvTechnologies Inc. has appointed Michael Teele as Director of global quality management. His 25 yr of experience includes applying core quality, lean, Six Sigma and continuous improvement disciplines and leveraging industrial engineering, manufacturing, R&D, testing and reliability competencies. He previously served as Regional VP of operational excellence at Proserv, Global Quality Assurance Manager at Array Holdings Inc., and Quality Assurance Manager at Tyco Valves and Controls.

Schneider Electric has appointed Carlos Villa to VP of its industry business (US). In his new role, he will oversee the continued development of Schneider Electric’s digitally-enabled solutions and network of distribution partners nationwide to support US industrial customers. Mr. Villa joined the company as Industry VP, Mexico, in 2014.

Jacobs Engineering Group Inc., as part of the acquisition of CH2M HILL Co. Ltd., has formed an integration management office (IMO) to oversee the integration of Jacobs and CH2M. The IMO will be jointly led by senior executives from both companies. Gary Mandel, who most recently served as Jacobs President of petroleum & chemicals (P&C), has been appointed to lead the integration for Jacobs. Jacobs also announced that Vinayak Pai has been appointed interim head of global P&C. Mr. Mandel joined Jacobs as Executive VP of operations overseeing various global portfolios, including the upstream and downstream oil and gas businesses. Prior to joining Jacobs, he served as Executive VP for Aker Solutions. Mr. Pai served as Senior VP of global operations for Jacobs’ P&C business, and has more than 28 yr of experience in the oil and gas and petrochemical industries.

Gabriela Koren has joined

industrial cyber security company Indegy as VP, Americas. She brings more than 20 yr of international experience in enterprise technology sales, business development, alliances management and field marketing. Prior to joining Indegy, Ms. Koren was Executive VP at Panaya, an Infosys company, where she was responsible for sales and marketing in the Americas. She also served as General Manager for North America at Radvision, an Avaya company.

Saeid Rahimian has joined

Aereon as Chief Executive Officer. He joins Aereon after serving as President and CEO of Gardner Denver Energy Group, a division of KKR-sponsored Gardner Denver Inc.

Javier Jimena has been

appointed President and General Manager of Spirax Sarco Inc., and is responsible for the company’s US business. Mr. Jimena has more than 32 yr of experience in leadership roles with demonstrated success in strategic development. He comes from Spirax Sarco Argentina, where he was the Regional General Manager leading Argentina, Chile, Peru and Latin America. Before joining Spirax Sarco, he held a number of general and strategic management positions in Argentina and Brazil.

November 1–2, 2017 | Hyatt Regency Houston | Houston, Texas

Join the Discussion at Women’s Global Leadership Conference (WGLC) Engage with key industry leaders at WGLC for presentations and discussions on:

Keynote Speakers

— The Essentials of Engineering Leadership and Innovation, Dr. Pamela McCauley, Professor, Department of Industrial Engineering and Management Systems, University of Central Florida

•

DAY 1 KEYN OTE

•

DAY 2 KEYN OTE

— Discipline Career Management and Knowledge Management,

Susan Rosenbaum, Director , Schlumberger •

DAY 2 PANEL DISCUSSION

— STEM: Are we really making progress?

Moderator: Ray Dempsey, Vice Presi dent and Chief Dive rsity Offi cer , BP America Panelists: Pam Darwin , Vice President Africa, ExxonMobil Exploration Company Dr. Reagan Flowers , Founder and CEO, C-STEM Teacher and Student Support Services, Inc Johanna Haggstrom , Director of Technology , Baroid Product Line, Halliburton Natalia Shehadeh , Vice Presid ent & Chief Compl iance Offi cer , Weatherford •

DAY 2 PANEL DISCUSSION —

Keynote Day One

Pamela McCauley

Digital transformation: The impact on the industry

Moderator: Soumya Seetharam, Vice President , Enterprise Solutions, Schlumberger Panelists: Morag Watson , Vice President, Digital Innovation, BP Maria Claudia Borras , President & CEO, Oilfield Services, Baker Hughes, a GE company Dr. Satyam Priyadarshy, Technology Fellow & Chief Data Scientist , Halliburton Mehrzad Mahdavi , Vice President Digital Solutions, Weatherford WGLC is the oil and gas industry’s largest and most important women’s event, which makes it the ideal venue for networking with like-minded individuals and making new industry connections. Take advantage of these opportunities by attending our speed networking and topical luncheons. For more information on sponsorship opportunities, please contact Hortensia “Tish” Barroso, Business Development Manager , at [email protected]

PROUDLY SPONSORED BY:

Keynote Day Two

Susan Rosenbaum


Technology and Business Information for the Global Gas Processing Industry

GasProcessingNews.com | SEPTEMBER/OCTOBER 2017

LNG TECHNOLOGY Innovative terminal design to boost process efficiency Precooling strategies for increasing LNG production capacity ENGINEERING AND CONSTRUCTION Debottleneck and troubleshoot molsieve unit for gas dehydration COMPRESSION Meet compression train design requirements for FPSOs Special Supplement to

Our equipment is efficient, proven, robust, and now, seaworthy. When Petronas built the world’s first FLNG vessel, they brought Air Products aboard. We put 45 years of LNG expertise and 20 years of FLNG development to work in our process technology and equipment for offshore use. So whether you have a small peak-shaving plant or a large base load facility, our proven capabilities will make any LNG project — especially offshore — a success. Call +1-610-481-4861 or visit us online.

Coil Wound Heat Exchanger

tell me more

airproducts.com/LNG

Photo credit: Petronas | ©2017 Air Products

SEPTEMBER/OCTOBER 2017 GasProcessingNews.com

SPECIAL FOCUS: LNG TECHNOLOGY

13

Embracing innovation and diversity in liquefaction technology—Part 1 J. G. Baguley and L. A. Clark

19

Precooling strategies for ef�icient natural

gas liquefaction G. Krishnamurthy, M. J. Roberts and C. M. Ott

13

31

Floating cryogenic hoses unlock new opportunities in LNG transfer V. Lagarrigue

6

ENGINEERING AND

37

CONSTRUCTION

33

31 37

Optimal design and operation of molecular sieves for gas dehydration—Part 2 R. H. M. Herold and S. Mokhatab

9

COMPRESSION

37

41

Meet compression train base package design requirements for FPSOs—Part 1 E. Abraham and H. Miller

DEPARTMENTS

Gas Processing News ........................................................................... .........6

PIPELINES

US Industry Metrics ....................................................................................... 8

45

COLUMNS

Manage risks for trans-country pipeline projects: India case study—Part 2

Editorial Comment ...................................................................................................4

A. Moitra

New in Gas Processing Technology ................................................... ...... 49

US to dominate LNG market through next decade

Regional Focus .............................................................................................. .9 Russia and Iran join forces to expand Southeast Asia’s LNG market

Cover Image: INPEX operates the Ichthys LNG project onshore processing facility at Bladin Point, near Darwin, Northern Territory, Australia.

EDITORIAL COMMENT

www.GasProcessingNews.com

US to dominate LNG market through next decade

P. O. Box 2608 Houston, Texas 77252-2608, USA Phone: +1 (713) 529-4301 Fax: +1 (713) 520-4433 [email protected] PUBLISHER

Catherine Watkins [email protected] EDITORIAL

The ongoing development of shale gas resources in the US has spurred infrastructure construction for both natural gas processing capacity and LNG export terminals. New gas processing projects include the construction of multiple cryogenic and gas processing plants, NGL fractionators and multibillion-dollar pipeline infrastructure. This abundance of natural gas has fueled an unprecedented surge in the construction of new LNG export capacity. In total, the US has announced more than 36 LNG export terminal projects. These projects represent more than 330 MMtpy of LNG export capacity at an investment A. BLUME, Editor of more than $200 B by 2030. Nearly 80% of the nation’s LNG export capacity projects are located along the US Gulf Coast. Due to the global oversupply of LNG, it is highly unlikely that the majority of these projects will be completed. Regardless, the US is expected to become one of the largest LNG-exporting nations by 2020. By the early 2020s, the country will challenge Australia and Qatar to become the largest LNG exporter in the world. The US is scheduled to start operations on approximately 71 MMtpy of LNG export capacity by 2020 ( TABLE 1). These projects constitute total CAPEX of approximately $50 B. A second wave of US LNG export capacity could contribute an additional capacity of 30 MMtpy. These projects include LNG trains at Sabine Pass, Corpus Christi, Cameron and Freeport. Multiple LNG export projects are also expected to be built. It is imperative that these projects secure LNG offtake agreements to reach final investment decisions. Nearly all of the US LNG projects under construction have secured customers for their LNG capacity. US LNG export volumes will vary in response to global demand and the difference between domestic and global natural gas prices. At present, LNG is traded under oil price-linked contracts. However, as the LNG market expands, contracts are expected to change, weakening their ties to oil prices. As global LNG trade is forecast to increase substantially over the next 15 yr, so is the amount of LNG export capacity. The projects that secure offtake contracts quickly will be in better shape to obtain financing and build their facilities. Regardless, the US is quickly becoming one of the leading nations in the LNG industry. GP TABLE 1. US

LNG export capacity expected to come online by 2020

Project

Location

No. of liquefaction trains

Capacity, MMtpy

Sabine Pass LNG

Sabine Pass, LA

5

27.5

Cameron LNG

Hackberry, LA

3

13.5

Freeport LNG

Freeport, TX

3

13.2

Cove Point LNG

Lusby, MD

1

5.75

Corpus Christi LNG

Corpus Christi, TX

2

9

Southern LNG (Elba Island)

Elba Island, GA

2

2.5

16

71.45

Total

4SEPTEMBER/OCTOBER 2017 | GasProcessingNews.com

Editor Adrienne Blume Managing Editor Mike Rhodes Technical Editor Bob Andrew Editor/Associate Publisher, Hydrocarbon Processing Lee Nichols MAGAZINE P RODUCTION

Vice President, Production Sheryl Stone Manager, Advertising Production Cheryl Willis Manager, Editorial Production Angela Bathe Dietrich Assistant Manager, Editorial Production Lindsey Craun Artist/Illustrator David Weeks Graphic Designer Andreina Keller ADVERTISING SALES

See Sales Offices, page 50. Copyright © 2017 by Gulf Publishing Company. All rights reserved.

President/CEO John Royall CFO Alan Millis Vice President Andy McDowell Vice President Ron Higgins Vice President, Finance and Operations Pamela Harvey Vice President, Production Sheryl Stone Other Gulf Publishing Company titles include: Hydrocarbon Processing ® , World Oil ® , Petroleum Economist © , Pipeline & Gas Journal and Underground Construction.

SHELL TURBO TRAYS Novel tower internal to enhance the capacity and efficiency of your gas absorber

Find out more at www.shell.com/gasprocessing or email us at [email protected]

GAS PROCESSING NEWS B. ANDREW, Technical Editor

LNG carrier sets Northern Sea Route record On August 17, the Sovcomflotowned icebreaking LNG carrier Christophe de Margerie completed its first commercial voyage, transporting LNG through the Northern Sea Route (NSR) from Norway to South Korea. During this voyage, the vessel set a new time record for an NSR transit of just 6 d, 12 hr and 15 min. Moreover, Christophe de Margerie has also become the world’s first merchant vessel to travel the full length of the NSR without any icebreaker escort. During the record-setting voyage, the LNG carrier covered 4,060 km from Cape Zhelaniya of the Novaya Zemlya archipelago to Cape Dezhnev at Chukotka, Russia’s easternmost continental point. The average speed during the passage exceeded 14 knots, even though the Christophe de Margerie had to sail through ice fields 1.2 m thick in some areas. The total time of the voyage from Hammerfest in Norway to the port of Boryeong in South Korea was 19 d, approximately 30% faster than the regular southern route through the Suez Canal. The speed achieved demonstrates the economic potential of using the NSR for large-capacity vessel transits. Christophe de Margerie is the world’s first (and, at present, the only) icebreaking LNG carrier. The vessel was built to order for Sovcomflot to serve the Yamal LNG project and transport LNG year-round through the difficult ice conditions of the Kara Sea and the Gulf of Ob. The vessel was delivered to Sovcomflot on March 27, following successful ice trials in the Kara and Laptev Seas. The LNG carrier is capable of sailing independently through ice of up to 2.1 m thick. Christophe de Margerie has been assigned an ice class of Arc7, the highest ice class among existing merchant vessels. The ship’s propulsion system has a power of 45 MW, which is comparable to the capacity of a modern nuclear-powered icebreaker. Christophe de Margerie also became the world’s first vessel with a high ice class to have three Azipod units installed, providing the vessel with a high ice-breaking capability and pronounced maneuverability.

6

Gas to be primary energy source by 2035 DNV GL’s Energy Transition Outlook (ETO), a forecast that spans the global energy mix to 2050, predicts that global demand for energy will flatten in 2030, then steadily decline over the next two decades, thanks to step changes in energy efficiency. The fossil fuel share of the world’s primary energy mix will decrease from 81% at present, to 52% in 2050. Demand for oil i s expected to peak in 2022, driven by expectations for a surge in the use of light-duty electric vehicles, which are expected to account for 50% of new car sales globall y by 2035. However, the stage is set for gas to become the largest single source of energy to 2050, and the last of the fossil fuels to experience peak demand, which DNV GL expects will occur in 2035. Gas will continue to play a key role alongside renewables in helping meet future, lower-carbon energy requirements. Major oil companies intend to increase the share of gas in their reserves, and DNV GL expects an accelerated shift by 2022 as companies decarbonize their business portfolios. While demand for hydrocarbons will peak over the next two decades, significant investment will be needed to add new oil and gas production capacity and to continue operating existing assets safely and sustainably. However, the results of DNV GL’s model reinforce the need to maintain strict cost efficiency to achieve the margins necessary for future capital and operational expenditures.

Shell to monetize gas via Egypt LNG Royal Dutch Shell is considering options to bring gas from Noble Energy and Delek Drilling’s giant Leviathan gas field offshore Israel and Cyprus to market. Under discussion is a proposal wherein Shell would buy natural gas from Leviathan, combine it with gas production from the Aphrodite field (in which Shell owns a 35% stake) offshore Cyprus, and then send it to Shell’s Idku LNG export terminal in Egypt. Combining the production from the two fields could serve to improve the economics of the projects. Noble Energy and Delek Drilling have cited an estimated development cost for Leviathan of $3.75 B. If the partners strike a deal with Shell, then they will need to secure additional funds for development to increase Leviathan’s output. Noble Energy remains in negotiations to supply natural gas to the Egyptian market. If settled, the multi-partner, multi-field deal could go a long way toward fulfilling the Eastern Mediterranean region’s aspirations to become a major gas-exporting hub. Egypt was a net exporter of LNG until 2014, when dwindling gas production and power shortages resulting from political uprising forced the country to reserve the fuel for domestic use. Cyprus is also eager to develop its gas sector. The concept of building floating LNG (FLNG) vessels to process the Aphrodite gas may be a viable option if regional politics prevent the construction of crossborder gas pipelines.

Gazprom: Sanctions will not stop pipeline The $11-B Nord Stream 2 expansion project to add a second Russian trans-Baltic gas export pipeline will be implemented, regardless of how new US sanctions related to the Ukraine crisis and the 2016 US presidential elections are interpreted, according to Gazprom. Earlier this year, Uniper, Wintershall, Shell, OMV and Engie agreed to each loan 10% of the cost of the pipeline. Gazprom will remain the sole shareholder in Nord Stream 2, shouldering 50% of the cost. The conduit will have the capacity to carry 55 Bcmy of gas to a terminal in Germany, and is due to start operating in 2019.

SEPTEMBER/OCTOBER 2017 | GasProcessingNews.com

Petronas eyes pipelines for Canada gas

Malaysia’s state-owned oil and gas company, Petroliam Nasional Berhad (Petronas), is considering investing in a pipeline to market its Canadian gas assets, after scrapping plans for an LNG project in Canada. Petronas would use the pipeline to connect and market natural gas from an area of western Canada that holds 22.3 Tcf of proven unconventional gas to the rest of Canada and North America. Petronas scrapped its proposed $29-B LNG export terminal in western Canada last month due to weak global prices, in a blow to i ts ambitions to expand its LNG portfolio beyond Malaysia. However, industry analysts said the decision was expected and would bring the company long-term benefits, as it frees up funds for other domestic projects. Petronas is working diligently on its Refinery and Petrochemical Integrated Development (RAPID) project in Malaysia’s southern state of Johor.

LNG peak shaving...are you ready? The Linde Group, a world leader in LNG technologies, has applied product standardization and modularization to the LNG industry. Our experience addresses your complete plant needs, including pre-treatment, core liquefaction, LNG product storage and loading, and LNG vaporization.

Learn more! www.leamericas.com/peakshave

Linde Engineering North America Inc. [email protected]

We deliver:

Pre-engineered process designs → Modularized plant layout → Short delivery time → Minimal on-site construction → Safe, simple and robust technology → Optimized energy eciency → Capacity expansions → Upgrades, revamps, repairs and replacements of existing equipment - liqueers, plate n heat exchangers, vaporizers →

US INDUSTRY METRICS A. BLUME, Editor

In the US, Henry Hub natural gas spot prices hovered around $3/ MMBtu in August, according to US EIA data. In the week ending September 6, dry natural gas production grew by 2% on the week, in spite of Hurricane Harvey reducing production in the Gulf of Mexico by an average of 409 MMcfd between August 30 and September 4, according to the Bureau of Safety and Environmental Enforcement. Meanwhile, US production of NGL, LPG, ethane/ethylene and propane/propylene rose in July and August, after across-the-board decreases in May and June. GP

US natural gas spot prices at Henry Hub and NGL spot prices at Mont Belvieu, $/MMBtu 25

20

15

u t B M M / $

Natural gasoline Isobutane Butane NGPL composite Propane Ethane Natural gas spot prices (Henry Hub)

10

5

US gas production (Bcfd) and prices ($/Mcf) 100

7 6

80 d f c B , n o i t c u d o r P

5 60

4

40

3

20 0

Monthly price (Henry Hub) 12-month price avg. Production

2

f c M / $ , s e c i r p s a G

1 0

J A S O N D J F M A M J J A S O N D J F M A M J J A 2015

2016

2017

Production equals US marketed production, wet gas. Source: EIA.

US natural gas plant field production of NGL, LPG, ethane/ethylene and propane/propylene, Mbpd 40 d p b M , n o i t c u d o r p d l e fi t n a l p s a g S U

35 30

NGL LPG Ethane/ethylene Propane/propylene

25 20 15 10

0

Aug. Sept. Oct. Nov. Dec. Jan. Feb. Mar. April May June July Aug. Sept. 2016 2016 2016 2016 2016 2017 2017 2017 2017 2017 2017 2017 2017 2017

8


June- July- Aug.- Sept.- Oct.- Nov.- Dec.- Jan.- Feb.- Mar.- April- May- June2016 2016 2016 2016 2016 2016 2016 2017 2017 2017 2017 2017 2017 Source: US EIA

REGIONAL FOCUS

Russia and Iran join forces to expand Southeast Asia’s LNG market E. GERDEN, Contributing Writer

Russian natural gas monopoly Gazprom is strengthening its presence in the gas market of the Middle East through the planned construction of an 11-metricMMtpy–12-metric-MMtpy LNG plant in Iran. The plant will source gas from the South Pars gas field ( FIG. 1), Iran’s largest. The majority of the LNG product will be exported to Southeast Asia and the EU. The expansion into Southeast Asia will allow both Russia and Iran to form the so-called Middle East-Asian Energy Ring. It will also make the two countries the largest suppliers of LNG to one of the world’s most densely populated regions, where gas demand has significantly increased in recent years. Gazprom has reached a preliminary agreement with state-run National Ira- FIG. 1. The South Pars field, from which the new LNG plant will source its gas. nian Oil Co. (NIOC), the operator of the South Pars gas field, for the establishment of a JV that will be responsible for LNG project strategy. A foreseen chal- sian Ministry of Energy, the majority of the implementation of the project. The lenge for the project is low global gas the funds will be provided by Gazprom. company has been operating in South prices. According to assessments by the South Pars’ total gas reserves are estiPars since 1997, when it participated in Russian Ministry of Energy, global LNG mated at 13.8 Tm 3. Iran’s reserves are just the development of the second and third prices have declined by 45% since the over 12 Tm3 , with the remaining reserves phases of the field. beginning of 2017, compared to aver- located in the territorial waters of Qatar. However, due to the imposition of age LNG prices in 2016. The decrease Iran holds approximately 17% of global economic sanctions on Iran by the UN is due to an abundance of LNG on the gas reserves and is the third-largest gas Security Council in 2006, further opera- global market, as well as to the link be- producer in the world. In 2016, Iran’s gas tions of Gazprom and other oil and gas tween natural gas and oil prices, which output totaled 192.5 Bm 3 , while domesmajors within Iranian territory were sus- are also weak. If prices for natural gas tic consumption was measured at 191.2 pended. On January 16, 2016, sanctions and LNG remain low, then payback pe- Bm3. To make use of its massive gas re were lifted, which allowed foreign energy riods for the project may be longer than serves, Iran plans to commission up to 60 companies to resume operations in Iran. originally envisioned. metric MMt of LNG production capacity Gazprom signed a memorandum Another issue is that approximately over the next 10 yr–12 yr. of understanding with the Iranian gov- 90% of Iran’s gas transportation fleet is Gazprom considers the LNG project ernment in March 2017 outlining the operated by Chinese firms. China is not in Iran to be of strategic importance for planned cooperation of Russia and Iran expected to encourage deliveries of LNG diversifying its operations. The imporin the global gas industry, and in particu- produced in Iran to Southeast Asia, as tance of the project, and the impacts of lar the LNG sector. The new venture will the region falls under China’s sphere of Russia’s ongoing financial crisis, are un be registered in the form of a service en- energy influence. derlined by Gazprom’s recent decisions terprise. This strategy should help avoid NIOC is expected to aim for a con- to suspend or cancel several similar dolegal problems in Iran, as production of trolling stake in the LNG venture with mestic LNG projects. The company has oil and gas in the country can be con- Gazprom; however, this will depend on suspended implementation of the Vladiducted only by state-owned companies, adequate funding for the project and the vostok LNG project in the Russian Far under state law. technologies used. According to the Rus- East. Gazprom already operates an LNG Gas Processing | SEPTEMBER/OCTOBER 2017

9

REGIONAL FOCUS

plant in Sakhalin with the capacity to produce 10 metric MMtpy of LNG. From a strategic point of view, investments in Iran might justify themselves, since Gazprom will be able to redirect a portion of the volume of Iranian LNG destined for Europe to Southeast Asia instead. This would also minimize Iran’s expansion into the EU gas market, which is largely controlled by Gazprom. Work on the new LNG plant could be started by the end of the current year, as the partners may speed up the process to secure contracts for future LNG supplies in a competitive marketplace ( FIG. 2). According to some forecasts, LNG ca-

FIG. 2. Preparations

pacity to be commissioned in 2018 will exceed 133 metric MMt. Of this volume, 60 metric MMt will come from US projects, while another 53 metric MMt will come from Australia. Southeast Asia to need more LNG.

Demand for gas is increasing in the region. Delivery via LNG carrier is thought to be a more viable option than pipelines for many countries, due to their geographical location and the high costs associated with pipelines. According to predictions by the Russian Ministry of Energy, the volume of LNG imported to Southeast Asia may

for building an LNG plant at South Pars.

double during the next several years, as local production cannot satisfy growing regional demand. This expanding consumption provides an opportunity for both Russia and Iran to significantly increase their shares in the global LNG market. For example, Thailand has recently announced its plan to cut domestic gas production and to increase imports of low-cost spot LNG in an effort to conserve its declining gas reserves. The Thai government aims to reduce gas production from the Gulf of Thailand by 10%, from approximately 79 MMm3d at present to 71 MMm 3d by 2018 ( FIG. 3). At the same time, Indonesia is considering LNG imports by 2019 if no new oil fields are developed, according to the Indonesian Ministry of Energy and Mineral Resources. The country could become a net importer of LNG by 2030, as its domestic gas reserves are steadily declining. Vietnam is expected to become another major LNG consumer in the coming years. To date, the country has relied mostly on coal-fired power plants, which provide about 40% of its electricity needs. However, amid growing environmental concerns and declining coal reserves, the country’s government has shifted its attention to gas as an alternative energy reserve. On top of this, Bangladesh has announced its plan to begin LNG imports after 2020. Before the imposition of sanctions, Iran had planned to build at least two LNG plants on the basis of South Pars. One, referred to as Persian LNG, would have included two LNG trains, each with a capacity of 8.1 metric MMtpy. The liquefaction process initially selected for the project was Shell’s double mixed-refrigerant process. Another project, Pars LNG, would have seen the construction of an LNG plant with a capacity of 10 metric MMtpy. France’s Total and Malaysia’s Petronas had planned to partner on the project. The implementation of both projects was suspended when sanctions were imposed. GP is an international contributing writer specializing in the global oil refining and gas industry. He has been published in a number of prominent industry publications. EUGENE GERDEN

FIG. 3. Gas

10

production platform in the Gulf of Thailand.


More ways to measure moisture in natural gas than anyone else. It’s in your pipeline and we’ll find it. AMETEK offers the industry’s widest array of natural gas analyzer technologies to meet your exact application needs—including quartz crystal microbalance, tunable diode laser absorption, electrolytic sensors and chilled mirror dew point moisture analyzers. We also offer analysis solutions for hydrogen sulfide, carbon dioxide, BTU content, specific gravity and gas density. Eighteen technology platforms in all—backed by AMETEK’s global service and support. With so many proven scientific instruments at hand, why settle for anything less than the optimum solutions for your application needs.

ametekpi.com

© 2017 by AMETEK, Inc. All rights reserved.


Embracing innovation and diversity in liquefaction technology—Part 1 J. G. BAGULEY and L. A. CLARK, LNG Ltd., Houston, Texas

The LNG liquefaction business has witnessed many transitions as it has moved from the remotest corners of the world to settle in more developed industrial environments, including the US Gulf Coast (USGC). The innovative combination of elements from existing technologies, together with creative engineering designs, introduces an optimized facility process and execution profile to enable the successful realization of projects in these new market conditions. Departure from the “tried and true” is viewed with a healthy and appropriate dose of technical and commercial skepticism. However, the application of proven technologies�including one with over 100 yr of industrial experience�makes it possible to introduce enhancements to the LNG production process and generate surprisingly beneficial results. The increasing importance of thermal efficiency in LNG liquefaction facilities is addressed, including the appropriate selection of refrigeration compressor drivers as a means to lower GHG emissions, enhance production and facilitate startup. Using an ongoing USGC project as a case study, the effects of recovering waste heat to drive the precooling process are discussed and quantified. A discussion on the available alternatives for precooling refrigerant is also presented. In Part 2, the application of a proprietary single mixed-refrigerant (SMR) liquefaction process in a mid-scale LNG project will be explored, in which existing technologies are combined in an innovative manner to improve key project attributes, enabling the delivery of a creative, reliable and efficient design. The foundations of the global LNG industry were established on the firm bedrock of safety and reliability, consistent with the Evolution of the LNG market.

FIG. 1. A

typical remote LNG project site.

demands of a closed market with carefully balanced supply and demand. For the producers, the capital investments necessary in the LNG value chain have always been substantial, often with extended payout periods requiring a high degree of certainty in the revenue streams to ensure the financial viability of the projects. For consumers, access to a reliable supply of LNG has been vital to electricity generation, industrial consumption and domestic use underpinning national economies. The solution was a closed market with point-to-point deliveries from specific producers to specific consumers using dedicated shipping. These operations were locked under long-term, highly restrictive contracts intended to protect the interests of all parties. Energy and capital efficiencies were important, but some what secondary, considerations within the greater equation of project viability. Today, the LNG market is changing. While long-term “take-or-pay” contracts still dominate the industry, an emerging market of spot cargoes, short-term con-

tracts and arbitrage opportunities is freeing up opportunities for nimble producers. Nominally, 70% of the world’s LNG still trades (at least initially) under the historic long-term contract model; however, this percentage is slowly and steadily dropping as more producing nations and facilities enter the market.1 Throughout this market evolution and increasing flexibility in the value chain, LNG baseload production facilities have remained substantially unchanged. This extended design inertia in the face of changing industry dynamics has created the opportunity to take a fresh look at production facilities to generate innovation and diversity in the production market. LNG construction trends. Since

the inception of the commercial LNG industry in the 1960s, LNG projects have maintained certain defining characteristics: • Historically, LNG projects have been limited to the domain of large, integrated national and international oil companies. Gas Processing | SEPTEMBER/OCTOBER 2017

13


Only major organizations of this nature have had access to the technical, financial, logistical and managerial resources to successfully deliver projects of this cost and complexity. • Facility owners have traditionally been involved in the development of the entire gas supply value chain, from gas field discovery and assessment through production well installation, feed gas pipelines, LNG facilities, shipping and marketing of the product. • LNG proponents have been (or have partnered with) highly mature organizations with extensive global engineering standards, processes, procedures and existing operational petrochemical facilities. • LNG facilities have been located in some of the world’s harshest and most remote environments� challenging to reach, let alone to operate successfully within (FIG. 1). • Extensive logistics programs have been vital to operations and to the success of the projects. The absence of local or regional facilities are a hallmark of LNG sites, and excellence in logistics planning and execution is essential to mere existence. • Massive infrastructure development programs are necessary, sometimes requiring years to complete. The project sites can be hundreds of km from existing regional infrastructure. They may lack roads, docks, marine facilities, food, water, shelter, electricity, communications, emergency response and security capacity. They may also be subject

to extreme weather conditions. The costs of these infrastructure developments can run into the billions of dollars while providing zero direct return on their investment. • Regulatory programs at many of the traditional sites are loosely defined (if existing at all), with no previous industrial development of a similar nature. Overlaps, conflicts and gaps between local, regional and national requirements can exist. • LNG facilities in these locations potentially represent life-changing and culture-changing transitions in local communities. Great care and planning must be taken to ensure that the new facilities and the associated construction and operational workforces can harmonize with the region, providing a positive catalyst for necessary and positive changes without altering the nature of the communities being served. With the emerging migration of LNG projects to the US Gulf Coast and other economically and structurally developed regions, the project and proponent profile has changed radically and rapidly. Key differential characteristics of projects under development include: • Brownfield or brownfield-type environments. Many of the new US export facilities are converted import terminals with existing storage and marine facilities. • Small, lightly funded, lightly staffed emerging owner organizations with total company headcounts in the 100s rather than the tens of thousands.

TABLE 1. Development of LNG plants

Typical 1980 baseload facility (remote)

Typical 2015 baseload facility (remote)

Typical 2015 baseload facility (USGC)

Capacity

2.3 metric MMtpy– 2.5 metric MMtpy

4.5 MMtpy–5 MMtpy

Same

Acid gas removal

MEA or sulfinol

aMDEA

Same

Cooling medium

Seawater

Air

Same

Heating medium

Steam

Hot oil

Same

Compressor drive

Steam turbine

Gas turbine

Gas turbine or motor

Precooling refrigerant

Propane

Propane

Same

Liquefaction refrigerant

Mixed refrigerant (MR)

MR or C2–C1 cascade

Same

14


• Limited infrastructure development requirements. Sites may have access to nearby regional power grids, water, roads, ports, telecoms, security and hospitals. • Ready access to well-established regional supporting industries, often located amidst the world’s largest petrochemical complexes. • Ready access to supporting expertise�suppliers, contractors, subcontractors, engineers, craft skills, vendor service centers and representatives. • Broad choice of readily available specialty chemicals and refrigerants. • Third-party gas supply from a national grid accessing multiple providers. • Well-established regional and national regulatory processes with extended schedules. • Increased sensitivity to greenhouse gas emissions. (Note: Whether one agrees or disagrees with the impact of greenhouse gases on the global environment, there is certainly an enhanced interest in the topic). • Background air emissions from existing regional industries that must be incorporated into plant emission modeling and limits. • Greater attention paid to plant energy efficiency. Energy efficiency has always been a consideration, but at remote sites the gas is essentially “free,” as it has no regional value unless converted into a transportable product such as LNG. On the USGC, LNG facilities must pay for every Btu of gas they receive from a pipeline, so efficiency and its impact on project finances become more important and visible. Fundamentally, this is not LNG “business as usual.” Yet, despite these substantial structural changes, the vast majority of LNG plant development in North America is proceeding on the same technical and execution bases as the historical remote site facilities (TABLE 1). From TABLE 1, the most significant changes over a 35-yr period from 1980 to the present include: • Nominal increase in capacity by 100% • Migration from water-cooled to air-cooled • Migration from indirect-fired

Don’t Compromise…

Kobelco Oil Injected Screw Compressors are flexible custom-made solutions to the small scale LNG market. Small scale LNG operators take note, Kobelco’s oil injected screw compressor meets all your requirements – no more standard packages.

...Customize.

Kobelco’s screw compressor has high turndown capability, allowing for a wide operating range and power savings, leading to lower total cost of ownership over the long term. Don’t Compromise, call Kobelco. • Custom Manufactured • Competitive Price • High Quality/Reliability • Low Cost of Ownership • Full Project Lifecycle Support

SCR EW

RECI PRO CAT ING

CENTRIFUGAL

Kobelco Compressors America, Inc. [email protected] 713.655.0015 www.kobelcocompressors.com

To k y o , J a p a n • H o u s t o n , Te x a s • M u n i c h , G e r m a n y • J u r o n g , S i n g a p o r e • D u b a i , U . A . E . • S a o P a u l o , B r a z i l


steam turbine drive to direct-fired gas turbine drive • Emergence of the cascade liquefaction process as a baseload technology. TABLE 1 also shows that the migration of LNG facilities from remote environs to the substantially more industrially developed USGC has resulted in virtually no changes. The US/USGC facilities being constructed or newly in operation at Sabine Pass, Corpus Christi, Cameron, Freeport, Cove Point and Lake Charles Trunkline are virtually indistinguishable from their remotesite cousins, despite the substantial structural differences highlighted previously. Humankind inherently resists change. A report in the No vember 2010 Journal of Experimental Social Psychology 2 concluded that the longer something is believed to have existed, the more highly it is regarded. The nature of the LNG industry�characterized by high development costs, long development schedules, delivery contracts of up to 20 yr, criticality of supply reliability to importers and limited alternate sources�does not lend itself to risk taking. Several key groups must be satisfied with the project design concepts and arrangements to enable a project to move forward to market: • Investors: For small companies requiring access to external funding sources, investors must be confident in the technology. This confidence enables the initial development funds to be secured, thereby allowing the project to be advanced through the design, estimation and regulatory processes. For an LNG project, this level of investment can be significant (in excess of $100 MM from inception to start of EPC when bank funding can be secured), and the entire development cycle, from initial activities through to LNG production, can span 6 yr–7 yr. High costs and long durations require a high degree of confidence by investors that the project and its technology will eventually generate a return on their investments. • LNG purchasers: When an LNG buyer signs up with a producer to provide LNG (through a tolling arrangement or a conventional sales purchase agreement), the buyer takes itself out of the market for that Barriers to innovation.

16

volume of LNG near the start of the project’s EPC phase, nominally 3 yr–4 yr prior to the actual delivery. The buyer makes onward commitments for the sale or use of that LNG and, consequently, must have confidence that it will be available as contracted, as alternate sources may be unavailable or characterized by unacceptable economics. As with investors, the purchasers of LNG cast a wary eye on anything considered novel or new, as this represents a risk to their business models and onward commitments. • Lenders: Lending institutions are vital to developing projects, providing access to the significant capital requirements of LNG projects amounting to billions of dollars. Obviously, these organizations are not charities; they expect timely repayment and take a dim view of any perceived risks to reimbursement. The lenders must be confident that the facility will reliably and safely deliver LNG over its design life. • Insurers: LNG facilities must obtain insurance coverage during the construction, startup and operational phases. This coverage should include construction risk, delayed startup, business interruption and operations coverage. Insurance organizations closely review the risk profile of a facility to understand how it compares to the historical industry experience. Step-out technologies can make it challenging for insurers to perform their risk assessments to develop this profile, potentially impacting availability and/or premiums for insurances. • Regulatory agencies: Stepping outside of a well-traveled technical or project execution pathway can result in the need for additional time for regulatory agencies to complete their activities and satisfy their obligations under governing laws. Conversely, innovations that inherently improve safety or reduce plant emissions can be viewed favorably by regulators once they have gained familiarity with the arrangements.


• EPC organizations: LNG plant developments rely heavily on the skills and knowledge of highly qualified EPC contractors to deliver these complex facilities to market. Lending institutions require project developers to obtain guarantees from the EPC contractors for plant capacity, fuel consumption and schedule. These guarantees often include “makegood” obligations on production rates and contract terms, with substantial performance damages for failure to achieve. Within this commercial environment, the EPC contractors must be comfortable with the expected performance of the technology to avoid taking on excessive risk. Drivers for innovation. Albert

Einstein is quoted as saying, “We cannot solve our problems with the same thinking we used when we created them.” With this advice in mind, the question “why innovate?” seems almost silly. Drive an automobile from the 1970s or 1980s, replete with an AM radio and perhaps an optional eighttrack player, hand-crank windows, biasply tires, solid rear-axle suspension and hard-starting carburetors, and the benefits of innovation become quickly obvious. Companies innovate to stay in business and respond to the demands of an evolving market. In the case of LNG, the market has indeed changed, although in more subtle, fundamental ways: • Efficiency matters: Purchasing feed gas from third parties makes plant energy efficiency an economic necessity. For a 10-metric MMtpy LNG facility, a nominal change in the retainage (quantity of feed gas consumed as fuel) from 8% to 6% of feed gas represents an annual savings to the facility of $35 MM, at a gas purchase price of $3/MMBtu, to produce the same quantity of LNG. • Emission matters: According to the US Energy Information Administration (EIA),3 burning 1 MMBtu of natural gas generates 117 lb (53.1 kg) of CO 2. Reducing the retainage from 8% to 6% of feed gas in a 10-metric MMtpy LNG facility reduces CO2 emissions by 1.38 B lb/yr (626 metric Mtpy of CO2 emissions reduction).

Protecting You During LNG Transmission From well head or tanker to terminal or pipeline, the transportation of natural gas or Liquefied Natural Gas (LNG) is an extensive process that brings risk to all personnel involved. During the processing of well head gas and/or the unloading, storage and vaporization process of LNG, there is a potential for fire and/or explosion. HUNTER understands the need for protecting the facility personnel engaged in these transitions. HUNTER engineers and manufactures modular, blast-resistant buildings designed to safeguard personnel and critical equipment during all phases of the processing and transporting this explosive material. All HUNTER buildings have undergone physical blast tests and meet ASCE guidelines for “Design of Blast-Resistant Buildings in Petrochemical Facilities.” Buildings feature climate control, fully-furnished interiors, flexibility, and meet all applicable codes and building standards.

As a one-stop solution for blast-resistant buildings, HUNTER buildings can be configured for the following applications:  n Compressor Stations  n Operator Stations  n Metering and Instrumentation Stations  n Shelter in Place  n And More

Design / Manufacture / Customization / Installation / Site Services / Leasing

HunterBuildings.com

14935 Jacinto Port Boulevard / Houston, Texas 77015 / +1 281.452.9800


• Reliability matters: Engineering to achieve high levels of simplicity and provide facilities with inherent redundancy increases reliability. • Flexibility matters: With the nature of the US gas supply market (not accessing a designated gas field), there is a need for production flexibility. These emerging projects benefit from having LNG trains with designs that enable significant turndown and feed flexibility. • Safety matters: Simplification also promotes inherent safety. The safest piece of equipment in a process facility is the one that was eliminated. Reducing quantities of flammable and explosive hydrocarbons through improved configurations and use of alternate refrigerants promotes inherent safety. Overall, the status quo is comfortable but inherently incapable of advancement. Innovation is a necessity. Key elements impacting LNG plant efficiencies. LNG plant efficiencies can be

addressed in two fundamental ways: process selection and plant configuration. Process selection garners the most attention, representing the core of the LNG technology licenses and technologies commonly applied and including the choice of refrigerants. Numerous articles have been published on the topic. Not all have performed comparisons of process efficiencies on the same basis, nor have they demonstrated a consensus agreement on the results.4,5,6 Comparing inconsistent elements of plant configuration can contribute to differences, and others may not have had access to the latest optimizations applied to the designs. An additional complication is that, in some cases, the differences in process efficiencies are quite small, such that variations in the assumptions used in the modeling (ambient temperature, refrigerant compositions, etc.) can impact the results. Plant configuration can have an even larger impact on the effective efficiency of a given LNG installation. Plant configuration in this context refers to several elements: • Driver selection (steam turbine, industrial frame gas turbine, aeroderivative gas turbine, motor) • Waste heat integration (process 18

heating duty oil/water coils, combined-cycle steam systems) • Application of liquid expanders • Application of LNG product flash gas as a heat sink • Number of refrigerant stages selected • Cooling medium (air vs. water) • Where system boundaries are drawn (i.e., including or excluding infrastructure utility loads). Setting aside the hyperbole, some generalizations can be made: • The difference in efficiency between facilities with the greatest and the least efficiency is not numerically large, representing only a few percent of the feed gas supply to an LNG plant. The best achievable LNG plant efficiencies run in the range of 92%–94%, while the lowest run in the range of 88%–90%. • While these differences are small numerically, considering the large gas flowrates into an LNG facility, they are significant on an absolute basis. • Mixed refrigerants are generally more efficient for liquefaction than pure component refrigerants, as the natural gas cooling curve can be more closely approximated. Depending on the composition of the natural gas and the number of refrigeration stages selected, this difference can be minor or it can be more significant. • Precooled liquefaction arrangements are generally more efficient than arrangements using a single refrigerant. It is challenging to maintain a close approach to the natural gas cooling curve over the entire range of ambient down to –160°C (–260°F) with a single refrigerant selection. • Nitrogen expansion processes are some of the least efficient. • Gas turbine plants deliver their greatest efficiency when the turbines are run at full capacity. • Aeroderivative gas turbine drivers are more efficient than industrial gas turbine drivers, although some of the newer-generation industrial machines deliver efficiencies as high as, or higher than, those of competing aeroderivative machines. • Newer-generation gas turbines are more efficient than earlier-


generation machines. • Addition of waste heat recovery can greatly increase overall gas turbine cycle efficiency. A common industrial frame gas turbine efficiency is on the order of 32%–34%. Adding “light duty” process heating recovery can raise efficiency to 36%–38%. Aeroderivative and high-efficiency industrial machines provide efficiencies in the range of 40%–45%, while incorporation of combined cycle can increase delivered efficiencies to the range of 50% and higher. • Motor drives introduce additional inefficiencies. Instead of directly coupling a gas turbine to a compressor, the gas turbine drives a generator, which then drives a motor, which drives the compressor. Some of the additional, inherent inefficiencies induced by the generator and motor can be recovered if the power plant efficiency is high enough. • Steam turbine-driven LNG plants, while highly reliable, are among the least efficient. Efficiency elements.

Part 2 will address two of the key efficiency elements in greater detail, followed by a case study of an innovative USGC LNG design and delivery project opportunity. GP LITERATURE CITED

Complete literature cited available at GasProcessingNews.com. JOHN G. BAGULEY serves as

Chief Operating Officer for the Magnolia LNG and Bear Head LNG projects, and Chief Technical Officer for LNG Ltd. His involvement in international LNG project development and delivery spans nearly 37 yr and includes project management, engineering, construction and commissioning roles. He holds a BS degree in chemical engineering from Michigan State University and is a registered Professional Engineer in Texas. LINCOLN CLARK is the Group

Engineering and Operations Manager for LNG Ltd. He has 25 yr of experience in oil and gas projects covering the design, construction, commissioning and operation of LNG plants, gas processing facilities, oil production facilities and power stations. Mr. Clark joined the LNG Ltd. team in 2005 and has been closely involved with the development of the Gladstone LNG, Magnolia LNG and Bear Head LNG projects.


Precooling strategies for efficient natural gas liquefaction G. KRISHNAMURTHY, M. J. ROBERTS and C. M. OTT,

Air Products and Chemicals Inc., Allentown, Pennsylvania

The invention of the propane precooled mixed refrigerant (C3MR) natural gas liquefaction process in 1970 brought about better efficiency and greater economies of scale than had previously been available to the young LNG industry. Splitting the refrigeration duty between two refrigerants allows each to be individually optimized; the first is used to precool the natural gas feed to a moderate temperature, and the second is used to liquefy and subcool the precooled gas to produce LNG. Due to these advantages, precooling is used in a majority of baseload LNG trains today. Various precooling refrigerants have been proposed in the LNG industry, such as propane, mixed refrigerant (MR), ammonia, carbon dioxide (CO 2), ethane, propylene and fluorocarbons. Guidelines are provided here for selecting an optimal precooling refrigerant, considering multiple factors, such as efficiency, precooling temperature, operability, refrigerant availability, flammability, ambient temperature and safety. Additional thermodynamic factors that are relevant in the selection, such as specific heat ratio and critical temperature, are also discussed. A direct comparison is also provided of several precooling refrigerants, highlighting the benefits and drawbacks of each, and demonstrating that optimal selection of both the precooling refrigerant and the precooling configuration plays a critical role in driving project success. Selection of a liquefaction process cycle is a key component in the design of a natural gas liquefaction facility. It is driven by various factors including design capacity, ambient temperature, plot space availability, floating or land-based applications and others. A crucial decision that impacts this selection is whether the cycle should involve a dedicated precooling circuit and, if so, which precooling refrigerant should be used. FIG. 1 shows a simplified schematic for an LNG facility. Natural gas is first pretreated to remove contaminants including acid gases, water and mercury, after which it is optionally sent to an NGL removal unit for removal of C 2–C5 hydrocarbon components. NGL removal may be performed prior to precooling, as shown in FIG. 1, or it may be integrated with the precooling and liquefaction processes. The NGL stream may be sent to a fractionation unit, where individual components are separated for sale or used as refrigerant makeup in the facility. Heavy hydrocarbons (C6+ components) may be removed with the NGL stream. Some facilities include a separate heavy hydrocarbons removal unit to prevent freeze-out of these components during liquefaction. The pretreated gas is then precooled, liquefied and subcooled against at least one cold refrig-

erant stream to produce LNG. The refrigerant is compressed and cooled using air or cooling water prior to providing the refrigeration duty. The LNG is reduced in pressure and may be sent to an endflash drum before it is sent to the storage tank. Any endflash vapor generated during pressure letdown is typically used as fuel gas. The refrigeration duty for the precooling, liquefaction and subcooling processes may be provided by a single refrigerant, as shown in FIG. 1, or it may be split among multiple cooling sections, each with individually optimized refrigerants. FIG. 2 shows a precooled gas liquefaction process with two refrigerant circuits, a precooling circuit with a precooling refrigerant for cooling the pretreated gas to –20°C to –50°C, and a liquefaction circuit with a main refrigerant for liquefy ing and subcooling the precooled natural gas to an LNG temperature of approximately –130°C to –150°C. The main refrigerant is cooled to ambient temperature with air or cooling water, and then to about –20°C to –50°C against the precooling refrigerant prior to providing liquefaction duty to the feed. To fuel Pretreatment unit (AGRU, dryer, Hg removal)

NGL removal

Endflash drum

Precooling, liquefaction and subcooling

LNG to tank

Fractionation unit

FIG. 1. Simplified schematic of a natural gas li quefaction facility.

To fuel Pretreatment unit (AGRU, dryer, Hg removal)

NGL removal

Precooling

Fractionation unit

Liquefaction and subcooling

Endflash drum LNG to tank

FIG. 2. Simplified schematic of a precooled natural gas liquefaction

facility. Gas Processing | SEPTEMBER/OCTOBER 2017

19


The primary benefits of a dedicated precooling circuit are as follows: 1. It debottlenecks the liquefaction circuit, since both the natural gas feed and the main refrigerant are precooled prior to being introduced into the main exchanger. This reduces the main refrigerant circulation rate, liquefaction heat duty requirement, and main exchanger size. 2. It allows for independent optimization of each refrigerant. These benefits lead to an improvement in overall efficiency of up to 20%, depending on feed conditions and project requirements, and make dedicated precooling circuits desirable for baseload LNG facilities. Heat transfer mechanism and precooling configuration.

Two heat transfer mechanisms by which precooling and liquefaction take place are boiling and sensible. In boiling heat transfer, liquid refrigerant is vaporized, taking advantage of the latent heat of vaporization to provide cooling duty. In sensible heat transfer, refrigerant is warmed without undergoing phase change to provide the same cooling duty. For most fluids, the latent heat capacity of vaporization is 4–6 times greater than the sensible heat capacity; therefore, more refrigeration per mass of refrigerant can be obtained by latent heat transfer than by sensible heat transfer. Boiling heat transfer reduces the refrigerant circulation rate, which results in LNG to end-flash drum

Precooling refrigerant

Pretreated natural gas

Precooling

MR FIG. 3. Four-pressure,

pure-component precooling and MR liquefaction

configuration.

Feed HP

Feed MP

Feed LP

Feed LLP

Feed MR HP

Feed MR MP

Feed MR LP

Feed MR LLP

FIG. 4. Four-pressure

20

precooling process configuration.


higher efficiency and lower equipment size. Only boiling precooling refrigerants are considered here. Another proposed configuration is an absorption-based cycle, such as using aqueous LiBr. The precooling temperature achievable by aqueous LiBr is limited to 0°C, the freezing point of water. Deeper precooling is generally preferred, as it maximizes the overall process efficiency of the liquefaction process. Other absorption-based processes, such as ammonia absorption, have been proposed but are not evaluated here due to operational considerations that are outside the scope of this study. Single- and multiple-pressure evaporative precooling configurations are widely used due to their high efficiency, simple operability, versatility and low cost. Many refrigerants have been employed in these processes, such as pure components, multi-components and azeotropic mixtures. With a pure-component, single-pressure configuration, all of the refrigerant evaporates at a fixed temperature. The efficiency of this process can be significantly improved by adding additional pressure levels. Each pressure level provides refrigeration at a progressively lower temperature, so only a fraction of the refrigeration is provided at the coldest temperature. Three- and four-pressure precooling configurations are ad vantageous for LNG train capacities of 3 MMtpy–5 MMtpy. An example of a pure-component, multi-pressure precooling configuration is shown in FIG. 3. The pretreated natural gas is precooled against a precooling refrigerant prior to being liquefied and subcooled against mixed refrigerant (MR) in the main cryogenic heat exchanger (MCHE), which is shown as a coil-wound heat exchanger. Coil-wound heat exchangers are state-of-the-art exchangers for gas liquefaction and are known for their high thermal stability, safety, small footprint and high heat transfer area. They contain helically wound tube bundles housed within an aluminum or stainless steel pressurized shell. The MCHE in FIG. 3 contains two bundles: warm (bottom) and cold (top). Precooled natural gas feed enters the bottom end of the MCHE, where it is cooled, liquefied and subcooled as it flows through tubes and exits as LNG. Refrigeration is provided by boiling MR on the shell side of the MCHE. MR is a mixture of nitrogen, methane, ethane or ethylene, propane, butanes and other hydrocarbons. A warm, low-pressure MR stream withdrawn from the warm bundle is compressed in a series of compressors, cooled against ambient air or cooling water, and further cooled against the precooling refrigerant. It is then sent to a phase separator to produce an MR vapor (MRV) stream and an MR liquid (MRL) stream, both of which are cooled in the tubes of the MCHE. At the top of the warm bundle, the cooled MRL is flashed across a Joule-Thomson (J-T) valve and sent to the shell side of the warm bundle to provide refrigeration duty. At the top of the cold bundle, the liquefied MRV is flashed across a J-T valve and sent to the shell side of the cold bundle to provide refrigeration duty. The shell-side fluids vaporize and provide cooling duty to the fluids in the tubes. FIG. 4 shows the details of the precooling circuit. The precooling refrigerant is a pure component, such as propane in the case of the proprietary C3MR cycle, and is vaporized at four pressure levels. Natural gas and MR are cooled against the boiling refrigerant in parallel heat exchangers, as shown in FIG. 4. The refrigerant is condensed and subcooled prior to being di-

Every compressor we ship endures a vigorous inspection process to ensure maximum build and quality. Once approved, one of our highly trained mechanical inspectors proudly affixes his nameplate to a compressor – personalizing his approval, a commitment built on more than 50 years of quality engineering, manufacturing, service, and support. Quality: embedded in everything we do.


vided into two streams, one for each parallel set of heat exchangers. The subcooled refrigerant is then let down in pressure and partially vaporized in the high-pressure (HP), medium-pressure (MP) and low-pressure (LP) exchangers. The vapor streams produced from each stage are mixed and introduced into the precooling compressor as HP, MP and LP side streams, while the liquid streams produced in each stage are let down in pressure and sent to the subsequent exchangers. The precooling fluid is fully vaporized in the final low-low pressure (LLP) exchanger, sent to a suction drum to remove any liquid, and subsequently sent to the suction of the precooling compressor. Several considerations must be taken into account when selecting a precooling refrigerant. This section highlights key factors that need to be evaluated and their implications for the overall process. Desired precooling temperature. A key optimization variable for a liquefaction process is the temperature of the natural gas and MR leaving the precooling section, which is called the “precooling temperature” or “cut temperature.” The lowest prePrecooling refrigerant selection.

0

NH3 Propane Propylene R-410A

Ethane

CO 2

MR

-20 -40 -60

C ° , e r -80 u t a r e p -100 m e T

-120

-140 -160 -180 FIG. 5. Lowest

precooling temperature achievable without a vacuum

design. 101

Optimum 100 % , y c n e i c 99 ffi e l l a r e v o e98 v i t a l e R

Transfer load into precooling Transfer load into liquefaction

97

96 -10

-20

-30

-40

-50

-60

Precooling temperature, °C FIG. 6. Optimal

precooling temperature for a DMR process at specific process conditions.

22


cooling temperature for a specific configuration depends primarily on the refrigerant composition and the lowest pressure at which it boils. Typically, the lowest pressure is kept at a positive pressure (i.e., non-vacuum), to prevent air ingress in the event of minor leaks. Air ingress will affect the precooling performance. For flammable refrigerants, it is a potential safety issue. Assuming typical process parameters (a pressure of 1.1 bara at the compressor suction, a 0.2-bar pressure drop in the suction drum, and a 3°C approach temperature in the LLP exchanger), FIG. 5 presents the lowest precooling temperature achievable for seven different precooling refrigerants: propane, ammonia, propylene, ethane, R-410A, CO 2 and MR. R-410A is a 50–50 wt%, near-azeotropic blend of hydrofluorocarbons (HFCs) R32 and R125. In this group, ammonia has the warmest precooling temperature of approximately –25°C. Propane, propylene and R-410A have lower precooling temperatures, while ethane and CO2 can be precooled to as low as –80°C. MR may be used as a precooling and/or main refrigerant, and has the ability to cool the natural gas, liquefy it to LNG, and subcool the LNG to colder temperatures. As a result, it does not have a precooling temperature limit; however, beyond the liquefaction temperatures, it is no longer considered a precooling refrigerant. Optimal precooling temperature. Although a refrigerant can cool natural gas to a certain temperature, it may not be optimal to do so. An optimal precooling temperature exists that results in the highest overall process efficiency. The optimal precooling temperature is driven by the ambient temperature and relative efficiencies of the precooling and main refrigerants. For cooler ambient conditions, it may be preferable to precool the natural gas to lower temperatures. FIG. 6 shows the overall process efficiency vs. precooling temperature for a dual MR (DMR) process that utilizes separately optimized MR streams as the precooling and main refrigerants. The optimal precooling temperature is –37°C, based on specific process conditions for the scenario considered. The optimal precooling temperature varies as the ambient temperature and process conditions of the gas feed change. As the precooling temperature warms, refrigeration load is transferred into the liquefaction system, and the main refrigerant must be optimized for a larger temperature range. Conversely, as the precooling temperature cools, refrigeration load is transferred into the precooling system. The shape of the optimum curve and sensitivity to warmer vs. colder precooling temperature varies on a case-by-case basis and is a strong function of the choice of precooling and liquefaction refrigerants, configurations, ambient temperature and other factors. FIG. 7 shows the power split for the precooling and liquefaction circuits as the precooling temperature differs from the optimum. At a warmer precooling temperature, the precooling power requirement is reduced and the liquefaction power requirement is increased. The temperature difference between hot and cold sides in the MCHE increases, lowering the overall efficiency. On the other hand, at a lower precooling temperature, the precooling power requirement is increased and the liquefaction power requirement is reduced. In this case, the temperature difference between the hot and cold sides of the precooling exchanger increases, reducing the overall efficiency of the process.

Modular purpose-built equipment and turnkey solutions under one roof We execute EPC projects and offer industry-leading off-the-shelf equipment by bringing together our engineering, procurement, fabrication and construction capabilities. Through our modular equipment design and in-house fabrication capabilities we mitigate risk with guaranteed pricing and delivery. snclavalin.com/en/market-services/oil-gas/processing-treating


Refrigerants with deeper precooling ability may be advantageous for applications in colder ambient conditions 1 because lower precooling temperature provides a better balance between the precooling and liquefaction heat duties. The precooling temperature impacts the driver power split desirable for a given project. Selection of the driver 2 and machinery configurations,3 while not within the scope of this article, are also important factors that require consideration during the design phase. Process efficiency. One of the most important factors to be evaluated is the overall process efficiency, since the need for higher efficiency is often a justification for precooling. The efficiency of a precooling process measures how much power is required to provide a given amount of refrigeration. For a precooled process using a boiling main refrigerant, such as MR, about 75% of the total precooling heat duty goes toward partially condensing the main refrigerant. Approximate100

ly 25% of the heat duty provides precooling to the feed. In other words, precooling reduces the load on the liquefaction circuit by precooling both the feed and the MR. Several factors influence the process efficiency of a refrigerant. The ratio of the constant-pressure specific heat and the constant-volume specific heat, also known as the specific heat ratio, is an important parameter (Eq. 1). Most of the commercially practical refrigeration processes compress the vapor refrigerant, which can be modeled as a polytropic process. For a polytropic compression process, Eq. 2 gives the relationship between the pressure and temperature of a fluid. The combination of Eqs. 1, 2 and 3 illustrates that, for a given pressure ratio, the temperature rise of a refrigerant during compression is determined by the specific heat ratio of the refrigerant. As the ratio increases, the temperature at the compression outlet warms, which indicates that more work is expended during compression. Conversely, as the specific heat ratio decreases, the temperature at the compression outlet cools, thereby making the process more efficient. Generally, refrigerants with a low specific heat ratio are preferred. k

80

=

Liq P

% , r e60 w o p e v i t a l e R

Liq

Liq

1−n

where: k = c p = cv = ηp = P = T =

Pre Pre Pre 0 -57

-37

FIG. 7. Optimal precooling temperature: Precooling (Pre) and

liquefaction (Liq) power requirements.

140

Hot ambient Average ambient Cold ambient

100 C ° , e 80 r u t a r e p m60 e T

40

20 0

CO2

Ethane

R-410A

=

Constant

k =

k −1

η P

(2) (3)

Specific heat ratio Specific heat at constant pressure Specific heat at constant volume Polytropic efficiency Pressure Temperature

-77

Precooling temperature, °C

120

n

T

n −1

20

(1)

cv

n

40

Propylene

Propane

FIG. 8. Precooling refrigerant critical temperatures.

24

c p


Ammonia

Critical temperature. The critical temperatures of various pure-component refrigerants are shown in FIG. 8. The critical point is the point on the phase diagram (see FIG. 9) where the dew and bubble lines meet. Above the critical temperature, both gas and liquid phases have the same density and cannot be separated. FIG. 8 also includes lines that represent an average ambient temperature of 25°C, a hot ambient temperature of 40°C and a cold ambient temperature of 10°C. CO 2 and ethane have the lowest critical temperature of approximately 31°C. For CO2 and ethane applications at cold ambient temperature, it may be possible to operate the process below the crit ical temperature so that the process functions similarly to any other subcritical process. However, for hot and likely average ambient cases, the refrigerant temperature will be higher than the critical point, causing the process to be transcritical in nature. Another scenario where CO2 and ethane would function as subcritical refrigerants is when the natural gas arrives cold� for instance, if it is already precooled to –30°C, and if CO 2 or ethane is used to liquefy or further cool the natural gas. FIG. 9 shows a pressure-enthalpy (P-H) diagram for subcritical and transcritical processes. In a subcritical process (FIG. 9, left), the refrigerant vapor (A) is at a pressure of P1 and a temperature of T1 , and is compressed to pressure P2 and tem-


perature T2 (B). The compressed vapor is then desuperheated to the dewpoint (C), condensed to the bubble point (D), and subcooled to produce subcooled liquid (E). The subcooled liquid is then flashed to the original pressure, P1 (F), and vaporized to complete the cycle and return to vapor phase (A). During steps B through E, the process rejects heat to ambient air or cooling water. During steps F through A, the process provides cooling duty to a process stream. In a transcritical process (FIG. 9, right), the cycle diagram looks similar; however, the heat rejection steps B through E occur above the critical point. The process starts with refrigerant vapor (A) at pressure P1 and temperature T1 below the critical temperature. It is then compressed to pressure P2 and temperature T2 (B), which is above the critical temperature. Above the critical point, a fluid does not possess distinct vapor and liquid phases. Therefore, when it is cooled from point B to point E, it does not condense. The fluid exhibits vapor-like properties at point B and liquid-like properties at point E. However, unlike the subcritical condensing process, where temperature stays constant during the condensation process (steps C through D), the temperature decreases continually during the transcritical heat-rejection step.

T3

E

P2

Bubble line Dew line Constant temperature lines P2

T2

D

Liquid

e r u s s e r P

T1

The temperature at point E after heat rejection is set by the ambient temperature plus a heat exchanger approach temperature. Due to the vertical nature of the constant temperature lines above the critical point, point E is in the central porti on of the graph. Therefore, when refrigerant is let down in pressure from point E to point F, a two-phase stream with large amounts of vapor is produced. Point F typically has a higher vapor fraction in a transcritical process than in a subcritical process, which leads to a lower process efficiency. FIG. 10 shows a graph of hot- and cold-side temperatures vs. heat duty, also known as “cooling curves,” for the heatrejection steps B through E in a subcritical process using propane, as well as in a transcritical process using CO2. For a subcritical process, desuperheating, condensing and subcooling zones are distinct, and a majority of the cooling is provided during condensation. In a transcritical process, the zones are not distinct and the temperature difference between the hot and cold sides increases at the warm end, leading to inefficiency. This process further reduces the efficiency of transcritical processes. Generally, refrigerants with critical temperatures above the ambient temperature plus approach temperature are preferred

C

T1

E

T2

T3

B

B

Liquid + vapor

Vapor

e r u s s e r P

Liquid

P1

Vapor F

A

Liquid + vapor

P1

F

A

h0

h1

h2

h0 Enthalpy

Enthalpy

h1

h2

FIG. 9. P-H diagram for subcritical (left) and transcritical (right) processes.

80

80

Hot (precooling refrigerant) Cold (ambient)

70 60 C ° 50 , e r u t a r 40 e p m e T 30

B

C D E

60 C ° 50 , e r u t a r 40 e p m e T 30

20

20

10

10

0

B

Hot (precooling refrigerant) Cold (ambient)

70

E

0

Duty

Duty

FIG. 10. Cooling curves for subcritical (left) and transcritical (right) cooling. Gas Processing | SEPTEMBER/OCTOBER 2017

25


to ensure stable subcritical operation. Some challenges associated with transcritical operation include: 1. Lower process efficiency 2. Challenges with supercritical refrigerant storage in the accumulator 3. Lack of references with transcritical refrigerants in baseload LNG facilities. Refrigerant availability. Several natural gas liquefaction facilities are in remote locations where importing refrigerants is difficult and/or expensive. For such projects, it is preferable to use refrigerants available in the natural gas feed. Some refrigerants are present in the natural gas itself, such as propane, ethane and components of MR, and may be extracted using a fractionation system, depending on the feed gas composition. Flammability. Refrigerant flammability is also an important consideration, especially for FLNG projects. It is desirable to minimize the onboard flammable inventory, specifically the propane inventory. Therefore, certain precooling refrigerants are better suited for FLNG applications than others. Additional considerations. Other factors that should be taken into account include corrosiveness, toxicity, global warming potential, equipment count and size. Corrosive refrigerants are detrimental to heat exchangers and other equipment, reduce equipment life, and increase the frequency of shutdowns for repairs. They are also unfavorable for overall plant availability and economics. Safe, nontoxic and environmentally friendly refrigerants with low global warming potential are preferable. Equipment size and count impact capital cost, and are driven by the precooling configuration, refrigerant, operating pressure and other process conditions. Equipment size is a function of A1 P1

P4

P3

P2

Feed HP

Feed MP

Feed LP

Feed LLP

Feed MR HP

Feed MR MP

Feed MR LP

Feed MR LLP

FIG. 11. Optimized parameters in four-pressure precooling circuit.

both the specific heat capacity and the density of the refrigerant. As the specific heat capacity increases, the mass of refrigerant required decreases. Lower refrigerant mass flowrate and higher density lead to smaller pipe sizes. Site- and feed-specific precooling refrigerant evaluation.

Various precooling refrigerant options are presented for a land based natural gas liquefaction project of 5 MMtpy nominal capacity. The ambient temperature is 25°C, and air coolers with an air temperature rise of 10°C are used. Natural gas feed is at a pressure of 70 bara and contains 90% methane and a significant amount of NGL components. As shown in FIG. 2, the process in volves an acid gas removal unit (AGRU), followed by removal of water, mercury, heavy hydrocarbons and NGL. The pretreated natural gas is precooled in the precooling system, followed by liquefaction and subcooling in the MCHE. LNG is let down to storage pressure, and the vapor produced by flashing is sent to fuel. Precooling circuit. The precooling configuration is illustrated in FIG. 4 and involves four pressure levels and parallel precooling of the natural gas and main refrigerant. Six cases with different precooling refrigerants were performed. A seventh case was done using MR as the precooling fluid at a single pressure level. The following precooling parameters were held constant in all cases: 1. A fixed precooling temperature of –25°C; this value is based on ammonia that has the warmest achievable precooling temperature among the refrigerants considered, as shown in FIG. 5 2. Heat transfer coefficient times heat exchanger area (UA) for the desuperheater, condenser and subcooler heat exchangers, which ensures that the analysis is based on the same number of air cooler bays 3. Pressure drop across each unit operation 4. Approach to ambient temperature for the desuperheater, condenser and subcooler heat exchangers 5. Air temperature rise for the desuperheater, condenser and subcooler heat exchangers 6. Minimum allowable temperature difference between the refrigerant and the process stream in all eight heat exchangers. The following precooling parameters were optimized in each case, as highlighted in FIG. 11: 1. Four pressure levels: P1, P2, P3 and P4 2. Air flowrate A1 , which was varied to yield a fixed air temperature rise.

TABLE 1. Precooling refrigerants score card

Propane

Ammonia

CO 2

Propylene

Ethane

R-410A

MR

Relative precooling efficiency at –25°C precooling temperature

100%

97%

82%

100%

81%

100%

100%

Coldest precooling temperature without vacuum design, °C

–33

–25

–81

–39

–81

–43

No limit

Critical temperature, °C

97

132

31

92

32

72

No limit

Flammability

Yes

Low

No

Yes

Yes

No

Yes

Available in feed

Yes

No

No

No

Yes

No

Yes

Global warming potential

Low

None

Low

Low

Low

High

Low

Toxicity and corrosiveness

No

High

No

No

No

No

No

26



Liquefaction circuit. For all cases considered, the liquefaction and subcooling heat duty are provided by MR in the configuration shown in FIG. 3. Several parameters were held constant in the liquefaction circuit: 1. Natural gas and LNG pressure, temperature and flowrate throughout the liquefaction circuit 2. MR pressure, temperature and flowrate throughout the liquefaction circuit 3. Pressure drops across all equipment 4. Fuel flowrate 5. Air temperature rise 6. Air cooler UAs 7. J-T valve outlet pressure. In essence, no parameters in the liquefaction circuit were allowed to vary, and the resulting liquefaction power is the same for all cases. Machinery. The precooling and liquefaction compressors were modeled as polytropic, with a fixed efficiency of 83% in all cases. No machinery constraints were imposed in any of the cases. Therefore, the analysis is not biased toward any specific machinery selections, such as driver choice and driver-compressor arrangements.

INNOVATIVE VALVE TECHNOLOGY ENGINEERED + PATENTED QUARTER TURN MECHANICAL SEAL ZERO LEAKAGE ON/OFF & FLOW CONTROL PROTECTED SEAT DESIGN PIGGABLE DESIGNS CARTRIDGE FEATURE REDUCES DOWNTIME HIGH TEMPERATURE/METAL SEAT 2”-24”

Results and discussion. A

score card comparing the seven precooling fluids against the key selection parameters is shown in TABLE 1. Precooling efficiency. The relative precooling process efficiency of the refrigerants at a fixed precooling temperature of –25°C is shown in FIG. 12. Among this group of refrigerants, propane, propylene, MR and R-410A have the highest efficiency. One reason for this is the low specific heat ratio,C p /Cv , for these refrigerants. Ammonia has 3% lower precooling efficiency than propane. One of the factors causing this is its higher C p /Cv ratio compared to propane. Ethane and CO2 have the lowest efficiencies because of their higher specific heat ratios and the transcritical nature of these processes, as demonstrated in FIG. 9 and FIG. 10. Although ethane and CO 2 have similar critical temperatures, CO2 has a slightly better efficiency than ethane, partly due to its slightly lower C p /Cv ratio compared to ethane. Of the seven refrigerants compared in TABLE 1, propane, propylene, ammonia, ethane and CO2 are pure-component refriger-

DOWNSTREAM SEAT UPSTREAM SEAT

DOUBLE BLOCK & BLEED/DIB “VERIFIABLE SHUTOFF” VALVE DOUBLE BLOCK AVAILABLE IN HIGH TEMP

100

95 % , y c 90 n e i c ffi e e v i t a 85 l e R

80

75

Ethane

CO2

NH3

Propylene R-410A

Propane

FIG. 12. Relative precooling efficiencies at a fixed precooling

temperature of –25°C.

MR

15B S TRADE CENTER PKWY, CONROE, TX 77385 PH (936) 539-5770 FAX (936) 539-2990 EMAIL [email protected] WWW.CHROMATICINDUSTRIES.COM API & ISO CERTIFIED Gas Processing | SEPTEMBER/OCTOBER 2017

27


ants. R-410A, being a near-azeotropic blend of R32 and R125, functions similarly to a pure-component refrigerant. These refrigerants were evaluated in a four-pressure precooling configuration. MR, on the contrary, is a multi-component refrigerant comprising (in this instance) nitrogen, methane, ethane/ethylene, propane and butane, and was evaluated in a single-pressure precooling configuration. Even for single-pressure MR precooling, the overall process efficiency is comparable to multiplepressure propane precooling. This is because the hydrocarbon mixture boils over a wide temperature range, leading to a smooth cooling curve, and provides an opportunity to optimize cooling curves with small temperature differences and high efficiency. The efficiency for a precooled process can be further improved by adding features such as inlet air chilling, gas turbine heat recovery, endflash gas recycle, hydraulic turbines, boiloff gas (BOG) reliquefaction and others. However, these features are not limited to specific precooling refrigerants and may be applied across the board. When comparing processes with different precooling refrigerants, it is crucial to compare them based on the same features. Precooling temperature. As shown in FIG. 5, the precooling temperature achievable varies with the refrigerant. FIG. 13 shows the optimum precooling temperature that yields the highest overall process efficiency for this application. Note that the optimal temperature is a strong function of ambient temperature, process conditions and liquefaction process selection, and must be determined on a case-by-case basis.

LIQUEFIED PETROLEUM GAS Firefighting Training for Bulk Storage Facilities

With the exception of ammonia, which is limited to a precooling temperature of approximately –25°C, all of the other precooling refrigerants can precool to the optimal precooling temperature and offer an advantage over ammonia. As mentioned previously, the relative efficiencies in FIG. 12 are shown for a fixed precooling temperature of –25°C for all of the fluids to ensure a common basis. However, all of the fluids, other than ammonia, have the advantage of precooling to the optimal temperature. For instance, propane precooling at –33°C provides an additional improvement in overall efficiency as compared to propane precooling at –25°C. Although fluids like CO2 , ethane and MR can precool to very cold temperatures, it was not optimal to precool to these temperatures. Propane precooling temperature was found to be close to optimal for typical ambient conditions. As the am bient temperature cools, the optimal precooling temperature decreases, and refrigerants with colder precooling ability are advantageous. Considerations for the liquefaction cycle selection in light of project conditions are discussed in the following sections. Critical temperature. The critical temperatures of the refrigerants are shown in FIG. 8. A higher critical temperature implies that the process can operate at high ambient temperatures within the subcritical regime. Propane, propylene, ammonia and R-410A have critical temperatures higher than 70°C, which is well above typical hot ambient temperatures. Being a mixture of multiple hydrocarbons, critical temperature is not an issue for MR. CO 2 and ethane have a critical temperature of approximately 31°C, causing these processes to involve transcritical cooling at cold ambient temperatures. Flammability. An advantage of non-hydrocarbon refrigerants, such as CO2 , R-410A and ammonia, is that they reduce flammable inventory onsite. Propane, being an LPG component, is typically not preferred for FLNG service. MR also includes flammable hydrocarbons; however, they can be designed to operate efficiently without propane, and have gained prominence for FLNG applications. Refrigerant availability. Propane, ethane and components found in MR have the advantage of being present in typical natProject conditions and liquefaction cycle selection.

0

This is a excellent firefighting training video for facilities that handle LP-Gas. The program was developed and videotaped with the assistance of the Texas A&M University Firefighting Training School in College Station, Texas. Series parts include: Part 1: Small vessel fire Part 2: LPG rail car fire Part 3: Tank truck fire Part 4: Flange leak fire Part 5: Pump seal fire Part 6: LPG leak and gas cloud Part 7: LPG fire in stacked bottles ORDER TODAY

28

l

+1 (713) 520-4426

-20

NH3 Propane Propylene R-410A

Ehtane

CO2

Optimal precooling temperature

-40 -60

C ° , e r u-80 t a r e p -100 m e T

-120

-140 -160

Price: $495 DVD

l

[email protected]


-180

FIG. 13. Optimal precooling temperature based on ambient

temperature, site-specific.

MR


ural gas feed. Depending on the feed gas composition, it may be possible to generate sufficient quantities of these components onsite, using an NGL removal and fractionation system. While CO2 has a slightly higher efficiency than ethane, the availability of ethane in the feed gas is a significant advantage for ethane over CO2. Global warming potential. HFC refrigerants, including R-410A, typically have high global warming potential, which must be kept in mind when making a selection. This is not an issue for the other refrigerants considered. Toxicity and corrosiveness. Unlike the other precooling refrigerants evaluated, ammonia has the significant disadvantage of being both highly corrosive and toxic in nature, which may have serious implications for safety, equipment life, cost, plant availability and overall facility performance. Over the past few decades, precooling has been implemented in a majority of baseload gas liquefaction facilities and has improved the efficiency, increased the capacity and enhanced the overall performance of these processes. Multiple precooling refrigerants and configurations have been proposed in the LNG industry. Various considerations must be taken into account when making a selection for a project. Here, seven precooling fluids have been directly compared: propane, MR, ammonia, CO 2 , ethane, propylene and fluorocarbons. The refrigerants were evaluated on a fixed basis against several important parameters, such as efficiency, precooling temperature, refrigerant availability, flammability, operability, ambient temperature and safety. These factors, along with key thermodynamic parameters (such as specific heat ratio and critical temperature), have a significant impact on the precooling process and must be closely evaluated on a case-by-case basis. Of the precooling refrigerants proposed, propane and MR possess several advantages, such as high efficiency, deep precooling temperature, subcritical operation and simplicity. They are also well proven and easily available. They are both nontoxic and noncorrosive, and have low global warming potential. MR possesses the additional benefits of low propane inventory, a small footprint and deeper precooling temperature, which make it a preferred choice for FLNG and cold ambient applications. The actual selection of precooling refrigerant for a specific application should be carefully determined based on the parameters highlighted here, along w ith project-specific requirements and site conditions. The optimal selection of both the precooling refrigerant and the precooling configuration plays a crucial role in overall plant economics and performance. GP Recommendations.

LITERATURE CITED 1

Schmidt, W. P., C. M. Ott, Y. N. Liu and J. W. Wehrman, “Arctic LNG plant design: Taking advantage of the cold climate,” LNG17, April 2013. 2 Schmidt, W. P., C. M. Ott, Y. N. Liu and W. A. Kennington, “How the right technical choices lead to commercial success,” LNG16, 2010. 3 Krishnamurthy, G., M. J. Roberts and J. W. Wehrman, “Drive your LNG project to success with optimal machinery selections,” LNG Journal, May 2015. ACKNOWLEDGM ENT

The authors would like to thank William P. Schmidt for his valuable contributions to this work.

GOWRI KRISHNAMURTHY is a Senior Research Engineer in the

LNG Process Technology group at Air Products. She has participated in the development of liquefaction process cycles and heat exchanger designs for several LNG projects, including both land-based and FLNG, and has also participated in plant performance tests. Dr. Krishnamurthy is actively involved in various research and development (R&D) initiatives for the development of new and improved natural gas liquefaction processes and heat exchanger designs, and is listed as an inventor on 15 patent applications. In addition, she is responsible for intellectual property strategy and development for the LNG business at Air Products, and is a registered patent agent with the US patent office. Dr. Krishnamurthy also has experience in R&D for reaction engineering, adsorption and renewable energy. She holds a BS degree in chemical engineering from the Indian Institute of Technology (IIT) at Chennai and a PhD in chemical engineering from Purdue University in Indiana. MARK J. ROBERTS is an Engineering Associate in LNG at

Air Products. He joined Air Products in 1996, and has 26 years of experience in developing cryogenic cycles for gas separation and liquefaction. Several US and international patents have been issued in his name, including the patent for the AP-X process. His responsibilities include developing and evaluating new approaches to LNG liquefaction and technical support of sales estimates for Air Products’ LNG business. is the Lead Process Engineer for the LNG Process Technology Group at Air Products. He joined Air Products in 2001 after receiving his BS degree in chemical engineering from Villanova University. He began working for Air Products in process design work for LNG and hydrocarbon cryogenics. He then moved to assignments in air separation, chemicals technology and electronics before returning to the LNG process group in 2005, where he has worked on many recent LNG projects, including several mid-size LNG and FLNG facilities. He has also participated in the startup and performance testing of many baseload and mid-size LNG plants. CHRISTOPHER M. OTT

eltherm – Specialist for electrical heat tracing systems Trace heating for all industrial applications Turnkey heat tracing solutions from Engineering to Commissioning Secure, reliable systems for high process temperatures and hazardous areas

Customised energy saving and monitoring systems International operating company with 40 years technology experience High quality and flexibility Own production site

Products include: - heating cables and tapes - heating mats and jackets - heated hoses - measurement and control systems - accessories eltherm GmbH Ernst-Heinkel-Str. 6-10 57299 Burbach, Germany Phone: +49 27 36/ 44 13-0 E-Mail: [email protected]

www.eltherm.com

a t V is i t us

Ad ipec, a t iona l ha b i In tern D u b A n ference, e h T i t ion & Co b i h x E m Pe tro leu 4 tand 1 3 4 6 Ha l l 1 3 • S 1 7 0 2 vem ber 1 3 - 1 6 No

Gas Processing | SEPTEMBER/OCTOBER 2017

29

REAL-TIME LNG DATA Any project. Any time. Any place. The Energy Web Atlas is a comprehensive dataset covering the global LNG market. Updated in real-time, the web-based platform enables users to access worldwide facility and project data.

Real-time Data

Comprehensive Functionality

The platform’s mapping, search and query tools allow users to interrogate real-time facility and project data and find the information needed for major decisions. Data points include:

Built for ease of use, the Energy Web Atlas facilitates access to the most relevant data available, with various functionality including:

• Owner/Operator

• LNG shipping/vessel information

• Custom annotation, reporting and graphing

• Shareholder data

• Interactive data filtering

• Process method.

• Layer/attribute access and modification

• Project status • Project storage and capacity

• Production quality maps, PDF exports • Interactive charts/statistics.

Questions and More Info

Additional Data Layers US gas processing plant data is also available.

• Data exports

AMERICAS: David Parker, +1 (713) 525 4680 or

[email protected] REST OF WORLD: Alastair Noakes, +44 (0) 203

409 2244

or [email protected]

D I S C O V E R M O R E AT E N E R G Y W E B AT L A S . C O M


Floating cryogenic hoses unlock new opportunities in LNG transfer V. LAGARRIGUE, Trelleborg Oil & Marine, Clermont-Ferrand, France

The LNG market is both growing and vessels would moor at the dockside and coming from the woven fabric layers. changing. According to a recent study, use a jetty platform for ship-to-shore The number and arrangement of mulglobal demand for LNG is projected to transfers, or use bridging arms for ship- tiple polymeric film and woven fabric increase by a factor of 50% by 2020, com- to-ship transfers between an FSRU and layers is specific to the hose size and appared to 2014. Much of this demand is a carrier, bunkering vessel or LNG-pow- plication (FIG. 1). spurred by the fact that LNG is usable in ered ship. Gas in its liquefied form presThe polymeric film and fabric matemultiple markets, regions and configura- ents a different challenge since it must be rials are selected to be compatible with tions, and is widely available. transported at a temperature of –163°C. the conveyed product and the operating At the same time, big changes are beLNG transfer solutions require spe- temperatures likely to be encountered. ing seen in demand patterns for LNG, cialized cryogenic hosing to safely trans- Insulated hoses can reduce boiloff by particularly gas power generation in the fer LNG to regasification plants. In many as much as 60%, equating to a savings Asia-Pacific region. India is leading the cases, there will be a need for traditional of 10 BBtu of energy over the course of charge, while China, Japan and Korea are jetties; however, due to the requirement 500 transfers. The outer protective hose both eyeing natural gas in a push to move for specialist cryogenic hoses for LNG draws on flexible rubber-bonded hose away from high-polluting fossil fuels. In transfer, more focus is being placed than technology, which is well known for its addition, remote regions of Indonesia ever before on hose technology and its high resistance to fatigue and its ability and the Philippines are looking to LNG capabilities. to withstand harsh environmental condito fill the energy gap where access to Composite LNG hoses typically con- tions. The flexibility and high flowrates the main grid is limited or unreliable, or sist of multiple, unbonded, polymeric achievable with cryogenic technology where power generation capabilities are film and woven fabric layers encapsu- mean that it increases the economic fearestricted. Reliable access to energy is vi- lated between two stainless steel wire sibility of power generation, terminal, tal to the success of economic growth in helices�one internal and one external. and marine bunkering projects that these remote regions, and LNG can play Essentially, the film layers provide a flu- are located away from existing infraa key role in delivering sustained growth. id-tight barrier to the conveyed product, structure�particularly in areas where Clear acceleration in demand is seen with the mechanical strength of the hose jetty-based transfer would be infeasible for small- and medium-scale LNG. In delivering LNG to these remote regions, it is necessary to be able to split it into smaller load parcels. To accommodate this need, the global fleet of LNG carrier vessels is diversify ing. Today, the world’s LNG fleet of around 500 vessels includes 26 FSRUs and 33 small-scale ships with a storage capacity of 30,000 m� or less. These vessels have specialized capabilities in oil and gas transfer. To carry out these operations successfully, they require innovative and flexible transfer solutions to adapt to the requirements of both the cargo and the recipient. This requires moving away from solutions that had facilitated the transfer of coal or fuel oil. Rethinking transfer operations. Tra-

ditional thinking has been that LNG

FIG. 1. Cryogenic

floating hoses. Gas Processing | SEPTEMBER/OCTOBER 2017

31


length of the transfer operation, further Floating transfer terminals or barges can reducing risk (FIG. 3). be connected to the shore using cryogenic hoses, which can then be easily connected Unlocking the transfer zone. A major with a vessel using a transfer system on a advantage of hose-in-hose technology barge. These solutions can be built, outis that it can negate the need for fixed fitted and commissioned offsite in paralonshore infrastructures; a concrete plat- lel, with relatively light civil engineering form onshore combined with hose trans- activity, thereby reducing the overall time fer solutions offers an alternative that scale from contract award to first gas. can be up to 80% more cost-effective As a self-contained mobile unit, a for locations where fixed onshore infra- floating barge or transfer unit can be readstructure would be prohibitive. ily adapted for alternative deployment in Collaborations between hose manu- the event of local changes or a desire to facturers and transfer equipment man- move location entirely. A floating soluufacturers demonstrate how certain tion also allows for refuge to be sought in ship-to-shore operations can be further safe harbor during storms or hurricanes, enhanced, as well as offer increased flex- for deep maintenance to be undertaken ibility and a choice of transfer options. at a shipyard, and for integration with a variety of LNGC mooring configurations. Moreover, the barge is only used when transfer is underway, minimizing the impact on the environment. Crucially, it is also suited to smaller carriers, for which many existing docking facilities are not configured. In conclusion, cryogenic hoses in floating configurations unlock new opportunities for LNG transfer, both at sea and in nearshore environments, reducing the risk of safety incidents and downtime, and lowering handling procedure times by as much as 60% in some instances. It is perhaps unsurprising that the prognosis for this technology is in rude health; a study conducted by Mordor Intelligence estimated that the global cryogenic equipment market would reach $23.5 B by 2021. FIG. 2. Cryogenic hoses in ship-to-shore configuration. As LNG’s role in the global energy mix increases, it is imperative that transfer technology keeps pace. Cryogenic hose technology is demonstrating that transfer technology can match the ubiquity and flexibility of LNG fuel itself. GP

because of harsh conditions or environmental concerns ( FIG. 2). Similarly, in offshore environments, cryogenic hose technology allows transfer to occur in deeper seas and in more challenging conditions. In comparison to traditional side-by-side transfer using loading arms, cryogenic hoses can be used in a tandem configuration, significantly increasing the distance between the vessels involved: approximately 100 m–150 m for FLNG-to-carrier transfers, and 300 m–500 m for carrierto-FSRU offloading transfers. This distance plays an important role in mitigating the risk of collision, as does the fact that the high flowrates afforded by the technology significantly reduce the

VINCENT LAGARRIGUE, Director of

FIG. 3. Tandem

32

ship-to-ship offloading.


Trelleborg Oil & Marine, is an expert in crude oil offloading solutions and LNG offshore transfer, and has worked at Trelleborg for more than 10 yr. Mr. Lagarrigue handles the Oil & Marine group’s complete product range of oil and gas transfer systems, including solutions for FLNG and small-scale LNG applications. At Trelleborg, he has been responsible for the design and qualification of flexible bonded hoses for offshore crude oil transfer (e.g., oil offloading lines, reel systems) and LNG transfer. Mr. Lagarrigue holds an MSc degree in mechanical engineering from the French Institute for Advanced Mechanics (IFMA) in Clermont-Ferrand, and an Executive MBA degree in finance and entrepreneurship from the EMLyon Business School in Écully, France.

ENGINEERING AND CONSTRUCTION

Optimal design and operation of molecular sieves for gas dehydration—Part 2 R. H. M. HEROLD, Contributing Writer; and S. MOKHATAB, Gas Processing Consultant, Halifax, Nova Scotia, Canada

Part 1 of this article examined the main design elements of a molecular sieve dehydration unit, although it is virtually impossible to design such a process unit without taking operational aspects into account. However, once a molecular sieve unit is constructed and the gas processing plant has started up, some elements during its operation require special attention. Part 2 focuses on the operation of a molecular sieve unit used for natural gas dehydration. Debottlenecking and troubleshooting options are also discussed. Addressing common issues. A contributing factor to molecu-

lar sieve deactivation is the deposition of liquids, such as amines, that originate from the upstream amine unit being deposited on the molecular sieve.1–6 Such liquids can enter the macropores of the molecular sieve and, during regeneration, cause severe degradation of the molecular sieve by destroying the binder material. This process can ultimately result in caking and coking, mechanisms discussed in Part 1. The deposition of liquid droplets can arise from poor upstream liquid/vapor separation, retrograde condensation or a regeneration profile that results in the formation of liquid water. To minimize this effect, the knockout (KO) vessel upstream of the dryer beds usually has advanced deentrainment internals, such as vane type, mist mat, or swirl deck and mist mat (SDMM) installed. While these internals should provide adequate separation, it is conceivable that they have not been installed properly, which can lead to liquid carryover in the form of entrainment. It is also possible that a slug from the upstream amine unit will enter the molecular sieve bed. This type of liquid carryover is difficult to prevent and can cause serious damage to the molecular sieve unit, which usually requires immediate changeout unless it is identified early. If operational indicators show that a large volume of liquid (e.g. through foaming of the upstream amine unit) has been deposited on the bed, then it can be mitigated only by a slow and careful ramp-up in the following regeneration step. Additional protection of the molecular sieve against small droplets can be provided by the installation of a guard layer on top of the bed. This layer is composed of a small coating of silica or alumina specifically installed to catch liquid droplets and protect the molecular sieve bed. If capillary condensation of hydrocarbons takes place in the bed, this can also lead to a rapid increase in the rate of deacti vation. Liquid hydrocarbons wet the sieve, covering it in a film, which adds extra resistance to the mass transfer of water mol-

ecules from the bulk gas phase to the molecular sieve active surface, reducing the water removal capacity of the sieve. Wetting of the molecular sieve also increases the sieve’s susceptibility to attrition, which will lead to dust formation, increased pressure drop and channeling. The only way to prevent these developments is to operate the molecular sieve beds a few degrees above the hydrocarbon dewpoint, which can be achieved by installing a heater downstream of the feed KO drum. Capacity test runs, which provide information on the amount of water adsorbed by the molecular sieve at the time of testing, are essential for evaluating the performance of the molecular sieve over its run length and for estimating the remaining lifetime.3,7 Once the capacity has been determined from a test run, it can be plotted against the number of regeneration cycles that the bed has experienced. If the results of the test runs are plotted against the expected deactivation curve, then an evaluation can be made as to whether it is possible to reach the planned changeout time. When sufficient data points are collected over time, the true deactivation profile can be determined and extrapolated to estimate the remaining lifetime. The results of several of these test runs for three trains are summarized in FIG. 1. As shown in the graph in FIG. 1, the results suggest that the molecular sieve beds are deactivating much faster than expected. Once some certainty is established that the beds’ performance is indeed lower than expected, an attempt can be made to identify the cause. The aim is to avoid an unplanned shutdown for changing out the molecular sieves and to keep the units run-

200

% t 150 w , y t i c a p a c e k a100 t p u r e t a W

Train A Train B Design Train C

50

0 0

200

400 600 No. of regeneration cycles

800

1,000

FIG. 1. Molecular sieve dehydration unit test run results. Gas Processing | SEPTEMBER/OCTOBER 2017

33


ning until the planned shutdown. At that moment, only a limited amount of data can be collected and analyzed in an attempt to identify the cause of underperformance of the beds.7 The first type of dataset to be collected is plant data�more specifically flow, feed temperature, pressure, temperature profiles during regeneration, regeneration gas flow, pressure drop over the beds and analyzer data. The analyzer data is the data collected by the moisture analyzers, generally located in the bottom of the bed and in the common outlet. Sometimes a mid-bed probe is also present. The dataset collected should be analyzed for anomalies by a comparison with historical data (i.e., data from a period of time where the plant was functioning well). If not already part of the standard sampling and analysis scheme implemented to monitor plant performance, condensate samples can be taken from the feed gas and regeneration gas KO drums for analysis. As before, the dataset collected should be analyzed for anomalies by a comparison with historical data, as well as with new plant data. The sampling scheme is presented in FIG. 2. Feed gas samples are collected and analyzed on a regular basis. However, this type of analysis will generate only a bulk composition�data that is needed for analysis but that does not generate information on trace components due to the limited amount of Gas samples Feed gas

4A sieve samples 280°C–320°C

Condensate samples

Bypass

Condensate samples Dry gas

FIG. 2. Sampling

options in molecular sieve dehydration unit.

External insulation 100-mm turn-up along wall of 20-mesh screen

Vessel wall

Hold-down ring 10-mesh screen Ceramic rope

5-mesh screen

Support grid

Support ring

FIG. 3. Detailed

34

sketch of a “classic” bottom support.


samples available. In general, when discussing the quality of gas samples, the subject of sampling reliability will inevitably arise. As the molecular sieve dehydration unit (DHU) is still running, it will not be possible to extract molecular sieve samples from the vessels. However, if the cause for the beds’ underperformance is not easily identified, then molecular sieve samples should be taken and analyzed at the first opportunity. Even when no problems are detected with the DHU, it is recommended to take and store samples during the planned changeout. In this manner, samples are available for comparison if the need arises. When vessels are opened and unloaded, close observation should be made as to whether signs are present of caking (clumps of molecular sieve particles), excessive dust formation, excessive discoloration and darkening of the sieve, and channeling (discoloration on the wall). Fresh spent molecular sieve samples can be analyzed by a variety of commonly used techniques:3,7 • Bulk-crushing strength and plate-crushing strength analyses to determine whether particles maintained their strength • Thermometric gravimetric analysis (TGA) adsorption measurements of capacity and mass transfer properties to determine if an excessive decrease has occurred in the water adsorption capacity of the adsorbent (i.e., to measure the degree of deactivation of the adsorbent) • Pyrolysis combustion mass spectrometric elemental (PCME) and/or flash combustion analysis to determine the amount of coking • Mercury porosity determinations provide information about the pore structure of a material and its degradation. One of the more common failures seen in DHUs is the failure of the bottom support structure of the vessel.1,3,4 The bottom support is an essential structure in the vessel, as it must ensure proper flow distribution during regeneration. The bed rests on a support, with the bottom dome of the vessel having a void space. A bottom dome filled with ceramic balls will achieve the same goal. The root cause of the problem is the continuous expansion and contraction of the bottom support during thermal cycles. If a weakness exists in the bottom support structure that can create a gap, then the molecular sieve particles will almost certainly find that hole. Once the molecular sieve starts flowing, it behaves like water and part of it will likely flow into the bottom dome. When opening a vessel, a clear indicator of an issue is the observation that a depression exists in the bed. Such a depression would create a channel�i.e., a path of least resistance manifesting itself as an early breakthrough of the bed. The molecular sieve in the bottom dome might also start to swirl, thereby grinding itself to dust and potentially creating problems for downstream equipment. A sketch of a “classic” bottom support structure is shown in FIG. 3. The most common causes of bottom support failure for the classic bottom support are mesh screen installation failures, incorrect sizes of ceramic balls used, and incorrect installation of ceramic rope packing in the gap. An example of the latter is that the ceramic rope is too tightly packed, to the point where it loses its flexibility. Accumulation of dust between the support grid and the wall can result in a deformed support grid. An alternative to the classic bottom support structure is the use of V-wire mesh screens, although these also have problems.


Care should be taken not to copy the design used in hydrotreater vessels. Although these reactors operate at high temperatures, they are not exposed to continuous temperature cycles. The other option is filling the bottom dome with ceramic balls, as shown in FIG. 4. The main advantage of filling the bottom dome with ceramic balls is eliminating issues with the bottom support as it is removed from the vessel. However, there are a few disadvantages to this method. The weight of the vessel increases, which can be a significant disadvantage for floating or offshore structures, where weight and plot space are major cost items. Also, due to the larger inventory in the vessel, more heat is required for regeneration. Using the same regeneration flow means that more time is needed for regeneration (heating and cooling)i.e., the minimum required uptake capacity increases. (Note: As discussed in Part 1, the minimum required uptake capacity is essentially determined by the minimum time needed for regenerationi.e., the fastest time the system can be heated up and cooled down while ensuring that the beds are fully regenerated.) Such a system also has a higher CO 2 footprint, another characteristic that is evaluated during modern design. Proper loading of the molecular sieve beds is a critical operation. Thorough checks should be carried out during installation of the ceramic rope, the mesh screens, the ceramic balls and the first layer of molecular sieve. Unloading of the vessels should be carried out quickly, but safely, to minimize downtime. Before unloading takes place, the beds should be regenerated and Note:

) • 5, 8 and 11mm probe S P capability D ( • Ceramic probes M for extreme pH E environments T S Y S Y T I M I X O R P L A T G I D

purged to ensure that coadsorbed species, such as hydrocarbons and sulfur species, have been removed to acceptable levels.

Ceramic balls, ½ in.

Adsorbent layer, ⅛ in.

Adsorbent layer, ⁄ in.

Ceramic balls, ⅛ in. Ceramic balls, ¼ in. Ceramic balls, ½ in. Ceramic balls, 1 in.

FIG. 4. Sketch

of an adsorber vessel with a bottom dome filled with ceramic balls.

• Crankcase Velocity

S R using the ST5484E O Velocity Transmitter S • Crosshead Vibration S using the IT6810 E R Impact Transmitter P • Rod Drop using the M MX2034 Transmitter O C G N I T A C O R P I C E R

MProximity probe E system that can ft T S in tight spaces and Y provide excellent S Y linear results T I M Normal probe I X O R P ™ W TightView™ E I probe V T H G I T

Machinery Monitoring and Protection Made Easy. With over 50 years of product and in dustry knowledge, Metrix has created and culvated a selecon of products that are the best in the market including transmiers, switches, probes, cables, and monitors. We oer a line of products that oer great reliability, dependability, and value. Our leading products are the backbone of our company and we connue to evolve and expand these products to t the needs of our customers.

Contact us for more informaon:

info@metrixvibraon.com 281.940.1802


35


One issue that is often neglected when operating DHUs is the disposal of spent adsorbent. Several aspects must be considered. Health, safety and environmental regulations demand that full personal protection equipment be worn during loading and unloading and during transport and storage. When loading and unloading, dust will be an issue. De-dusting before startup via piston purging is recommended. Local regulations for the handling and transport of possibly contaminated materials, as well as regulations for the safe destruction and recovery of spent adsorbents, will apply. The latter is usually not an issue wi th spent molecular sieves. In general, it is important to realize that public concerns with regard to the handling and storage of possibly contaminated materials might endanger the license to operate. For spent molecular sieves, several disposal options are available. Usually, the sieve will be stored onsite, which is not a permanent solution. Spent molecular sieve is disposed of through a landfilling procedure that is completely controlled by local regulations. One of the minimum requirements for landfilling is that a stabilization and leaching test confirms the safety of the landfilling option. Safe destruction is also an option, but this tends to be the most expensive one. In this context, vendors will usually assist with spent adsorbents. Some vendors even offer “cradleto-grave” support services. Other equipment items that create problems on a regular basis are the valves surrounding the vessels, which enable switching between the beds and guide the flow for adsorption and regeneration. 8 The valves used in this environment must adhere to stringent specifications to provide a tight seal over a wide temperature range at high pressure. In a worst-case scenario, a valve would leak and deposit hot, wet regeneration gas on a bed in adsorption, thereby severely reducing the uptake capacity of that bed and possibly even inducing caking. Few vendors can deliver these types of valves. In some designs (or applications, such as oxygen plants), regeneration is carried out at a lower pressure, which provides an additional driving force for desorption and thereby increases the efficiency of the regeneration. Such an operation raises the pressure differential over the valves, making a leak more likely to occur. This is one of the reasons that regeneration is carried out at the same pressure for temperature-swing adsorption (TSA) units, although it is not the main reason. The primary reason is that regeneration at a similar pressure occurs as the feed reduces the size of the compressor in the regeneration loop.

denser molecular sieve9,10 could be combined with variable cycling as a debottlenecking option. In summary, recommendations to help ensure that the DHU will run in a way that guarantees production of on-spec product and to ensure that the planned shutdown data will be available include: • Conduct performance test runs on a regular basis • Ensure proper design and operation of the upstream separator • Use a guard layer • Pay attention to the design and integrity of vessel internals • Select good-quality molecular sieves • Ensure proper loading of the molecular sieve • Implement a regeneration profile that prevents the formation of liquid water • Analyze spent molecular sieve samples. Based on the data presented, it might seem as though molecular sieve units are the source of many problems in a gas processing plant. In reality, that is not the case. These units tend to be reliable and require relatively little operational attention. That in itself can be problematic when issues arise, however, as experience with troubleshooting these units is sometimes difficult to find. GP LITERATURE CITED

Mokhatab, S., W. A. Poe and J. Y. Mak, Handbook of Natural Gas Transmission and Processing, 3rd Ed., Gulf Professional Publishing, Burlington, Massachusetts, 2015. 2 De Bruijn, J. N. H., M. A. Huffmaster, J. M. van de Graaf, P. F. A. van Grinsven and H. Grootjans, “Maximizing molecular sieve performance in natural gas processing,” presented at the 81st GPA Annual Convention, Dallas, Texas, March 11–13, 2002. 3 Carlsson, A. F., J. B. Rajani and A. J. Kodde, “Finding the fountain of youth for a molecular sieve dehydration unit,” presented at the GPA Europe Annual Conference, Provence, France, September 22–24, 2004. 4 Hawes, P., “Molecular sieves in natural gas processing,” presented at the GPA Europe Young Professional Training Day, Manchester, UK, Februar y 11, 2016. 5 Terrigeol, A., “Molecular sieves in gas processing: Effects and consequences by contaminants,” presented at the GPA Europe Annual Conference, Berlin, Germany, May 24–25, 2012. 6 Meyer, P., “Molecular sieves troubleshooting,” presented at the 27th GPA Europe Annual Conference, Lisbon, Portugal, September 22–24, 2010. 7 Herold, R., “Use analytical tools to investigate LNG molecular sieve underperformance,” Gas Processing, April 2015. 8 Pack, B. and A. Shackleford, “Gas conditioning failures show need for design scrutiny,” Oil & Gas Journal , Vol. 111, Iss. 5, May 2013. 9 Meyer, P., “Easy and sophisticated debottlenecking of molecular sieve plants,” Hydrocarbon World, Vol. 5, Iss. 1, 2010. 10 Meyer, P., “Cost reduction using high-density molecular sieves,” presented at the 90th GPA Annual Convention, San Antonio, Texas, April 3–6, 2011. 1

RUUD H. M. HEROLD was formerly senior process engineer

Takeaway. Despite careful design and operation of the

molecular sieve dehydration unit, a variety of reasons (e.g., different feed composition or upstream equipment not working properly) could make it impossible to run the unit in the manner for which it was originally designed. If “quick fixes” are not possible, then a few options are available for consideration. If the main purpose is to reach the next planned shutdown, and if some capacity is still left in two or more beds (assuming a 2 + 1 configuration), then one option is to continue operating at a reduced flow throughput. If one bed is severely underperforming, then another option is the combination of reduced flow throughput with running in 1 + 1 mode. Unfortunately, these options will always carry a large price tag. In the case of the beds structurally underperforming, a 36


at Shell Global Solutions International BV in Amsterdam, The Netherlands. He joined the company in 1986 and began working in the gas processing group in 20 01, where he specialized in adsorption and catalytic processes used in gas and liquid treating. Mr. Herold holds an MSc degree in chemical engineering from the U niversity of Amsterdam. SAEID MOKHATAB is one of the most recognizable names

in the natural gas processing industry. He has b een actively involved in the design and operation of several gas processing plants around the world, and has contributed to gas processing technology improvements through 300 technical papers and two well-known handbooks (published by Elsevier in the US). He founded Elsevier’s Journal of Natural Gas Science & Engineering, and has given invited lectures on gas processing technologies worldwide. As a result of his work, Mr. Mokhatab has received a number of international awards and medals, an d has been listed in prestigious biographical directories.

COMPRESSION

Meet compression train base package design requirements for FPSOs—Part 1 E. ABRAHAM and H. MILLER, the Dresser-Rand business, part of Siemens Power & Gas, Olean, New York

Floating production, storage and offloading (FPSO) vessels are used throughout the world for the processing of oil and gas, for oil storage and for offloading to a tanker or through a pipeline. FPSOs can be subject to high winds and accelerations from the pitch, roll and heave of the sea. Continued safe operation of onboard equipment under both normal and adverse conditions is essential. Base packages typically consist of a compressor, a gear and a driver, and are mounted on three anti-vibration mounts (AVMs) to minimize the loads and displacements being transmitted into the base package. The three-point mount bases require the analyses of a significant number of operational and upset load conditions to ensure safety and sustained equipment operation. Transport and package lifting must also be evaluated. The normal operating loads include dead weight; acceleration due to FPSO; pitch, roll and heave; unbalance; torque; wind; and nozzle loads. The upset loads could include motor short-circuit torque, maximum acceleration and survival wind loading. A modal and harmonic response analysis may also be required to ensure that response at key locations on the package remain within acceptable vibration limits due to rotor unbalance. It is important to perform these calculations early in the design phase, as design changes may be required to satisfy criteria. The analytical procedures presented can apply to any driver, although motor drives are presented in most of the examples. These procedures also apply to either a standard gear or a variable hydraulic gear. The three-point mount examples show the use of AVMs, and the procedures could apply to gimbal mounts. Single-body compressor train examples are shown in the examples, and the presented procedures have also been applied to base packages with multi-body compressors. vessels first emerged in the mid-1970s. Since that time, 186 FPSOs have been commissioned into service; 147 of these remain in operation today. FPSOs are widely deployed offshore in Latin America, Asia, West Africa, the Middle East, the North Sea and most recently in the Gulf of Mexico. The use of FPSOs appears to be still growing. The larger FPSOs have storage capacities in excess of 2 MMbbl of oil, and living accommodations for crews of between 100 and 200 people. They are also capable of processing up to 20 MMsm3/d of natural gas and injecting up to 300 MMbpd of water. 1 A typical FPSO layout is shown in FIG. 1 , and an actual FPSO is shown in FIG. 2. Several types of turbomachinery may exist

FIG. 1. Typical

FIG. 2. An

FPSO layout.

FPSO at a fabrication yard in Korea.

FPSO worldwide distribution and operation. FPSO

onboard, including gas injection compressors, gas lift compressors, export gas compressors, gas boosting compressors and fuel gas compressors. A view of two motor-driven compressor trains is shown in FIG. 3. Water-injection pumps and gas turbinedriven power generation trains may also be present, as shown in FIG. 4. The compressors and pumps are usually driven by mechanical-drive gas turbines or electric motors. In most instances, a speed-increasing gearbox is also used between the driver and the driven equipment. It is common to mount the compressor, gear and driver on a common, single-lift baseplate. The baseplate is fabricated from structural steel and contains mounting pedestals for each piece of equipment. Gas Processing | SEPTEMBER/OCTOBER 2017

37

COMPRESSION

In some cases, all of the auxiliary equipment needed to support the compressor and its drivers (such as a lubricating oil system, a dry gas seal system, instrumentation, and a local control panel) are also mounted on or within the baseplate. Some FPSOs have utilized steam turbine-driven electric generator sets. FIG. 5 and FIG. 6 show two steam turbine-driven generators of different powers. Both are mounted on the top decks of their respective FPSOs.

FPSO technology has matured significantly over the years, with the vessels gradually growing larger and more complex. As many as 50 risers can be connected through its mooring system. These risers have more sophisticated processing capability, with the latest evolution being the introduction of natural gas liquefaction to an FPSO. When an FPSO is utilized for the production of LNG, it becomes known as an FLSO. A typical FLSO design is shown in FIG. 7. The world’s first FLSO is undergoing commissioning at this time, but several more FLSOs are in the planning stages. This innovative method for producing oil and natural gas can have several advantages compared to conventional offshore platforms, the primary examples of which are cost-effective production of smaller-sized reservoirs and the ability to operate in waters considered too deep for conventional platforms and portability. Many FPSOs can disconnect from their risers, allowing them to be moved away from hurricanes and severe storms.2 The technology must overcome some challenges, including mooring system development, turret system development, flexible riser systems, safe handling of flaring, government regulations, financing, and coping with wave motion.

FIG. 3. Two

motor-driven, gas-injection compressor trains showing the drive motor, speed-increasing gear and compressor mounted on a common baseplate, together with a base-mounted lube oil system, dry gas seal system and local control panel.

FIG. 6. Three

24-MW condensing steam turbine generator sets on board the Peregrino FPSO. FIG. 4. A

typical aeroderivative gas turbine-driven electric generator destined for operation on an FPSO.

FIG. 5. Two

27-MW steam turbine generator sets on board the

Knock Allan FPSO for power generation.

38


FIG. 7. FLSO

Energy.

design concept. Used with permission from Excelerate

COMPRESSION

This last challengewave motionwarrants additional discussion. FIG. 8 illustrates the peak tilt angle experience by a typical FPSO during a 6-hr time period. Note the random fluctuation of the tilt, which achieves a maximum value of more than 18°. ( Note: For the sake of comparison, the cruise industry considers a tilt of 15° to be extremely severe. In such events, cruise passengers are usually injured by falls and sliding objects; some have even been thrown overboard.) On an FPSO, the ability to properly mount and secure rotating machinery is of paramount importance. Of critical importance are the mechanical design of the baseplates upon which the turbomachinery is supported, and the mounting of the baseplate to the topsides deck. The baseplate must be properly secured to and support the rotating equipment and the loads mounted on it, and it must be able to handle the forces and moments imposed by the FPSO hull and deck motions. Typical base package designs. The

typical base package design uses torque boxes or torque tubes to provide torsional and bending stiffness. The flexural stiffness is required for dead weight, package lift and ship heave. The torsional stiffness limits the overall base package twist resulting from both vessel pitch and roll, and from wind loads. Adequate torsional stiffness is required to limit the relative displacements between the shaft ends. This relative displacement must be limited both on the high-speed end between the compressor and high-speed gear, and on the low-speed end between the driver and the low-

speed gear. Adequate bending stiffness is required for package lifting and ship heave. FIG. 9 shows a motor, gear and compressor package supported by a torque box design. The lube oil console is cantilevered off the end of the base. A top view showing the base (skid) structure is shown in FIG. 10. The primary flexural and torsional mem ber is the fabricated torque box. FIG. 11 shows the bottom view and the positioning of the three AVMs. Two AVMs are located under the compressor, and one is located under the motor. An analytical model of a torque tube concept is shown in FIG. 12 and FIG. 13. This package includes two compressors and a 20 18 16 14 e e r g e d , l l o r . x a M

Max. roll 7° 10° 12°

12

6-hr average

10 8

7

6 4

2 0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 FIG. 8. Wave

motion roll angle experienced by an FPSO.

THE FUTURE IS REFINED HERE. 2017 Fall Refining and Equipment Standards Meeting Join us at the 2017 Fall Refining and Equipment Standards Meeting and have access to technical committee meetings covering API standards. Learn about the latest research and development on refining process equipment and obtain the latest information from industry suppliers while making contacts with your colleagues and establishing new business connections

Register at www.api.org/meetings November 13-16, 2017 Hyatt Regency Dallas at Reunion Dallas, Texas

It all starts with API Events ™

© 2017 – API, all rights reserved. API, the API logo, and the “It All Starts” tagline are either trademarks or registered trademarks of API in the United States and/or other countries.


39

COMPRESSION

gear. The base is bolted to another base that includes the driver, so this package includes only the two AVMs under the compressor. The base with the driver includes one AVM under the driver. The torque tube provides flexural and torsional stif fness. A third design concept does not use either a torque tube or a torque base; instead, wide flange beams are used on the perimeter of the base and for the main transverse beams. This design typically results in higher torsional and bending stiffnesses,

but it also results in a heavier base. FIG. 14 shows a compressor, gear and motor package supported by wide flange beams with the lube oil console cantilevered off the end of the base. The flexural stiffness is provided through the two large, wide flange beams that run in the longitudinal direction. These two beams also provide the support for the cantilevered lube oil console. The longitudinal beams, together with the transverse beams, provide the torsional stiffness.

On an FPSO, the ability to properly mount and secure rotating machinery is of paramount importance. Of critical importance are the mechanical design of the baseplates upon which the turbomachinery is supported, and the mounting of the baseplate to the topsides deck.

Base design considerations. A number of considerations af-

fect the base design. The required flexural strengths must be met, and this requires the detailed analyses discussed in this article. Bases fabricated from wide flange beams have been shown to reduce shaft-end relative displacements, but they have also been shown to be 18%–22% heavier. It is important that the shipbuilder has a good estimate of the total package weight. Costs of material and fabrication, including welding, are important. Total base costs are typically obtained from a number of base fabricators. The cost to manufacture a beam base vs. a box or torque tube base varies by manufacturer. In many cases, it is advantageous to include the lube oil console under the gear as opposed to cantilevering it off the end of the base. FIG. 15 shows a base beam with the lube oil console under the gear. Top and bottom views of the base are shown in

FIG. 9. Typical

torque box base package design with motor, gear, compressor and lube oil console.

FIG. 11. Bottom

view of torque box base.

FIG. 10. Top

FIG. 12. Typical

FPSO base package with torque tubes.

40

view of torque box base.


COMPRESSION

and FIG. 17. The entire base package is shorter, and space is at a premium on the FPSO. The other advantage is that it is easier to meet the API 2.5° drain requirement from the gear to the lube oil console. Typical ship roll, pitch and heel are 12°, 3° and 1°, respectively. For a 3° ship pitch, the pipe slope must be 3° plus 2.5°, or 5.5°. If the lube oil console is positioned off the end of the base, it might be too long to achieve the required drain angle. If the ship roll is specified as 12°, then the drain would be 14.5° in the lateral direction if the pipe needs to be run laterally for a certain distance. The torque tube design can also accommodate a lube oil console under the major equipment. However, the disad vantage of including the lube oil console under the gear for a torque tube design is that there is no large center torque tube extending from one end of the base to the other. It must be replaced by two smaller torque tubes that run along the sides of the lube oil console. FIG. 16

Three-point mounts. Three mounts are used for each pack-

age, and these are key to the successful operation of the package onboard the FPSO. Typical AVM designs are shown in FIG. 18. The top pad of each AVM is bolted to the base, and the bottom pad is bolted to the ship deck. Wire mesh cushions (WMCs) are used to provide the stiffness. These cushions are stacked and positioned to provide the proper stiffness in each direction per the package requirements. The AVMs also pro-

FIG. 13. Top

view of base showing torque tubes.

vide rotational flexibility. Cushioned uplift restraints provide stiffness in the vertical uplift direction during ship heave or during ship pitch and roll. The vertical movement at the mount position is limited to 3 mm–6 mm. All mounts can gimbal by 0.35° in all three rotational directions. For a 4.6-m-wide base, one side could displace upward by 13 mm, and the other side could displace downward by 13 mm. WMCs have high damping (15% to 20% of critical damping). Three-point mounts are easier to install than a multi-point system, which is difficult to properly install. The bolts between the base and pads, and between the pad and deck, are slip-critical. Shear pins are used in some designs. The advantages of AVMs over gimbal mounts are: • AVMs are generally considered to offer better vibration isolation • AVMs are not as high as gimbals, which raises the height of the overall package • AVMs do act as gimbals in that they allow rotation, but the rotation is limited compared to a gimbal • No significant cost difference is observed • No sliding large forces can be transmitted into the skid from the deck, although gimbals can be designed with sliding.

FIG. 15. Wide

flange beam base with lube oil console under the gear.

FIG. 16. Top FIG. 14. Typical

FPSO base package with wide flange beams.

view of wide flange beam base with lube oil console under the gear. Gas Processing | SEPTEMBER/OCTOBER 2017

41

COMPRESSION

The advantages of gimbal mounts include: • More rotation is allowed if needed (15° for gimbals vs. 0.375° for AVMs), which is advantageous in situations where higher rotation is seen between the top and bottom plates of the mounting system • Gimbals do not add to displacements at pipe connections, and there is no relative (dynamic) movement between the deck and the package. AVMs are more often used because of their high damping and successful experience with their use. AVMs provide stiffness and damping in the axial, lateral and vertical directions. FIG. 19 shows typical AVM placement. Many early designs included two AVMs under the driver and one under the compressor; however, designs with two AVMs under the compressor and one under the driver have been shown to more easily meet displacement criteria. One reason for this is that incorporating two AVMs under the compressor limits the compressor rotation due to the nozzle loads, vessel pitch and roll, and

other operational loads. Additionally, displacement limits are more stringent for the high-speed coupling on the compressor side than for the low-speed coupling on the driver side. The AVMs isolate the base package from the vessel hull and deck in two ways. First, the AVMs are heavily damped, decreasing the amplitude of base package displacement. This large frictional damping is provided by the WMCs. As a result of this damping, amplification factors for an AVM are typically 2.5. As a comparison, amplification factors for coil springs and rubber are 20 and 10, respectively. Damping is lower for vibration loads like rotor unbalance, and very good vibration isolation between the base package and the deck is realized. The AVMs are particularly effective in preventing structure born noise from being transmitted to the package. Second, sliding is allowed in two directions, as shown in FIG. 19 where AVM No. 1 is allowed to slide in the axial (X) direction and AVM No. 3 is allowed to slide in the lateral (Y) direction. This sliding prevents deck twist from being transmitted to the base package. As the deck bends and twists, the package has the capability to slide in the axial and lateral directions, minimizing the twist and bending that are transmitted into the base. The AVM sliding is activated under normal operational loads and upset loads. Sliding does not occur as a result of vibrational loads because the smaller vibrational loads cannot overcome the friction. For this reason, the sliding is activated in the analytical model for the static analyses of the Mount No. M1 Fixed +Y -X B

Mount No. M2

+X Global coordinates -Y

AVM No. 1

A

Sliding FIG. 17. Bottom

AVM No. 2

Sliding

view of wide flange beam base with lube oil console Plan view

under the gear.

AVM No. 3 B Mount No. M3 FIG. 19. AVM

fixed and sliding directions to isolate base package deflection from FPSO deck twist and bending

194 2,458 A 1,450

P2A

1,675 1,675

P3A

209

7,261

AVM No. 3

3,781

AVM No. 1 P1A

AVM No. 2

4,275 P2B

1,675 1,675

P3B

1,975 A FIG. 18. Typical

42

AVM designs.


FIG. 20. FPSO

AVM No. 3

AVM No. 1 P1B

Joint P1A P2A P3A P1B P2B P3B

AVM No. 2

deck location where stiffness is required.

Kz (kN/cm) 1,118 2,207 1,054 1,013 907 1,748

COMPRESSION

operational and upset loads. For dynamic analyses (harmonic response), the rotor unbalance loads are not high enough to overcome the friction, even in the sliding direction. The threepoint mount also serves to keep the package level. A typical arrangement of base packages on an FPSO deck is shown in FIG. 20. The axial direction of the equipment is generally installed parallel to the ship’s longitudinal direction, and the package’s lateral direction is parallel to the athwartship direction. The vessel’s deck stiffness under each AVM is pro vided by the shipbuilder for inclusion in the analytical model. If not provided, deck stiffnesses from similarly sized jobs are used until the f inal deck stiffnesses are available. Inclusion of the deck stiffness in the operational and upset load analyses (pseudo-static analyses) have been shown to increase shaftend relative displacements by as much as 8%. Therefore, it is conservative to include them. AVM stiffness values are determined from load-deflection curves (FIG. 21). The individual WMCs are tested to determine their load-deflection curve. The load-deflection curve is then determined analytically, based on how the WMCs are stacked and arranged inside the AVM. A linear stiffness value is extracted from this curve and used in the analysis. This process is accomplished by using the tangent stiffness at the typical load. The AVM vendor requires load data on each AVM for all load cases to properly design the AVM. The AVM is designed and built concurrently with the base build and the analysis. Therefore, preliminary values of AVM stiffness are used early in the analysis phase. This can be accomplished in one of two ways. AVM load-deflection curves from similar packages can be utilized, or the AVM stiffness can be estimated. Since the AVMs are designed to give a response of 12 Hz–15 Hz in the vertical direction, the preliminar y vertical stiffness for each AVM can be calculated from Eq. 1: Fn

1 =

2π

Kv / M

(1)

where: Kv = AVM stiffness in vertical direction Fn = 12 Hz to 15 Hz M = R/g = total mass supported by AVM (R = AVM vertical reaction) The AVM load-deflection curves, such as the one in FIG. 21, are typically supplied late in the analysis phase. Then, the most critical cases are rerun using the final AVM stiffness values. If the preliminary AVM stiffness values are adequately estimated, then the final results typically do not vary from the preliminary results by more than 1%–2%.

3,000

2,000

1,000 N K , d a o L

0

-1,000

-2,000

-3,000 -5.0

-4.0

-3.0

-2.0

-1.0

0.0 1.0 Deflection, min.

2.0

3.0

4.0

5.0

6.0

FIG. 21. AVM load vs. deflection curve supplied by AVM vendor.

detail is included in the finite element analysis (FEA) models to accurately calculate the shaft-end relative displacement. These details include more accurate modeling of the rotors, bearing connections, compressor pedestal sliding, and keel blocks. The importance of initiating the analysis while using preliminary data is emphasized as the base manufacture and analysis phases are conducted concurrently. Base modifications that are identified early in the manufacturing cycle are much easier to implement than those identified later. Improvements in data gathering, FEA model preparation and the automation of worst load case combinations have resulted in a 40% reduction in analysis time. The analytical models provide a valuable tool in assessing the suitability of threepoint base package design for operation on FPSOs. Part 2, to be published in December, will examine data gathering for the FEA and construction of the FEA model. GP LITERATURE CITED 1

“Axial and centrifugal and expander compressors for petroleum, chemical and gas industry services,” American Petroleum Institute (API), Vol. 617, 7th Ed., July 2002. 2 Mastrangelo, C., K. Barwick, L. Fernandes and E. Theisinger, Petrobras America Inc., “FPSOs in the Gulf of Mexico,” January 9–11, 2007, Kenner, Louisiana, online: http://offshorelab.org/documents/FPSOs_in_the_GoM.pdf NOTES

This article was presented at the Asia Turbomachinery and Pump Symposium in Singapore, February 22–25, 2016, and is published with permission of the Turbomachinery Laboratory. EDWARD ABRAHAM is Principal Solid Mechanics Engineer

Takeaway. Worldwide distribution of

FPSOs and typical applications have been discussed. The three AVMs dampen the response and isolate the base package from the FPSO deck. Three base designs have been discussed. Torque box and torque tube designs provide torsional stiffness and result in lighter base packages. Larger I-beam designs are heavier, but they provide higher torsional stiffness and allow for a shorter package by including the lube oil reservoir under the base. The shaft-end relative displacement criteria have been shown to be more limiting than the stress criteria. Significant

for the Dresser-Rand business, which is part of Siemens Power & Gas. He has performed numerous analyses of turbomachinery components, including cases, impellers, lifters, rotors and base packages.

HARRY MILLER is Director of Emerging Technology for the

Dresser-Rand business. He ha s authored several technical papers and articles, has been awarded and has contributed to several patents, and has received the Dresser Industries Annual Technical Achievement Award. He is also a member of the American Society of Mechanical Engineers. Gas Processing | SEPTEMBER/OCTOBER 2017

43

86 GAS INDUSTRY TOGETHER YEARS OF BRINGING THE

27th WORLD GAS CONFERENCE J U N E 25 -2 9, 2 018

| W A S H I N G TO N D C , U S A

EARLY BIRD REGISTRATION NOW OPEN SAVE $500 WHEN YOU REGISTER BEFORE JANUARY 31, 2018

500

12,000

600

100

100

350

SPEAKERS

ATT ENDE ES

ORGANIZATIONS

COUNTRIES

SESSIONS

EXHIBITORS

REGISTRATION INQUIRIES Email registration @ wgc2018.com or call +44 20 7978 0006 SPONSORSHIP OR EXHIBITION INQUIRIES Email sponsorship @ wgc2018.com or call +61 2 9556 7996

HOST PARTNERS

PRINCIPAL SPONSORS

FUELING THE FUTURE

OFFICIAL SHOW DAILY

FOLLOW US

wgc2018.com

PIPELINES

Manage risks for trans-country pipeline projects: India case study—Part 2 A. MOITRA, Indian Oil Corp., New Delhi, India

Sourcing gas by investing in a trans-country pipeline is an option being considered for India. It has potential to improve the country’s energy mix, economy and per-capita energy consumption. The outcome of a situational analysis and a SWOT analysis (discussed in Part 1 of this article) supported the in vestment to meet project objectives. According to an Energy Politics in Eurasia (ENERPO) journal report from May 2016, the opportunity for India exists primarily in sourcing gas from large producers such as Russia, Qatar, Iran and others through trans-country gas pipeline projects. The Project Management Institute (PMI) describes project risk as: “An uncertain event or condition that, if it occurs, has a positive or a negative effect on at least one project objective, such as time, cost, scope or quality.” Multiple factors inherent in these projects create or increase project uncertainty, which mandates both risk analysis and financial analysis for decision-making on project appraisal. The following sections elaborate on both qualitative and quantitative risk analysis. The procedures and techniques used in this study may include other available project options; however, this case study is limited to two specific projects� the Turkmenistan–Afghanistan–Pakistan–India (TAPI) pipeline and the Iran–Pakistan–India (IPI) pipeline ( FIG. 1).

developed. The IPI pipeline project�also known as the “peace pipeline”�has drawn international attention as a rival to the TAPI project. It envisaged sourcing gas from Iran for delivery to Pakistan and India. The project had an estimated cost of approximately $7 B in 2005 and would bring a total of 5.2 Bft3d of gas to Pakistan and India. Under the plan, each country would build a pipeline in their respective territories. Iran constructed a 1,172-km pipeline from Asaluyeh to the Iran-Pakistan border and signed a formal agreement with Pakistan in 2010. However, due to international sanctions, progress on the project thawed. India has kept open the option of rejoining the negotiation. According to a 2013 ADB report titled, “An overview of energy cooperation in South Asia,” no significant further progress has been made on the project, and investors have expressed diminished interest. Both the TAPI and IPI projects have similar project success criteria and project rationale. Situational and SWOT analyses gave similar results. Mr. Klaus Gerhaeusser, Director General of the Central and West Asia Department at the ADB, said in 2012 of the TAPI project, “The pipeline represents a win-win scenario for each TAPI country, as it will give Turkmenistan�with the world’s fourthlargest [gas] reserves�more diverse markets and help fuel the energy-hungry economies to the South. Each country stands Project success criteria.

The TAPI pipeline is proposed to be approximately 1,800 km in length and would export up to 33 Bcm/yr of natural gas from the Dauletabad gas field in Turkmenistan. The Turkmenistan countries agreed, in principle, that India Dauletabad gas field Tehran and Pakistan would equally share approxProposed Turkmenistanimately 1 Bft3d, while Afghanistan would Afghanistan Herat Afghanistan-Pakistan3 Iran take 0.2 Bft d. India Pipeline (TAPI) TAPI Pipeline Co. Ltd., incorporated Proposed Iran-PakistanIraq Kandahar in November 2014, appointed Turk India Pipeline (IPI) Quetta mengaz as consortium leader. GAIL (InKuwait Dera Bugti District dia) Ltd. and Inter State Gas System Pvt. Bandar-E Persian Asaluyeh Abbas Iranshahr Gulf Ltd. signed a sales purchase agreement Khuzdar South as a first step. The Asian Development Pars gas Baluchistan field Bank (ADB) was appointed as transacQatar Karachi Gulf of Oman tion advisor. Saudi Arabia TAPI pipeline project.

pipeline project. Iran holds the world’s second-largest natural gas reserves. However, due to sanctions, proven reserves remain underIPI

United Arab Emirates

Islamabad

Pakistan Multan

Fazilka

Sui

New Delhi

India

Arabian Sea

Oman 0

FIG. 1. Proposed

China

Miles

500

routes for the TAPI and IPI pipeline projects. Gas Processing | SEPTEMBER/OCTOBER 2017

45

PIPELINES

to gain, making this not only the ‘peace pipeline,’ but a pipeline to prosperity, as well.” The Central Asia Regional Cooperation (CAREC) strategy for regional cooperation in the energy sector identified TAPI as one of the links in the energy interrelationship among CAREC countries. The project would accomplish this by ensuring energy security through balanced development of regional infrastructure and institutions, stronger integration of markets and economic growth through enhanced energy trading. The IPI project’s success criteria was also found to be strategically aligned with CAREC objectives. Project risk analysis. The project study analyzes and identi-

fies which of the projects is the best fit for the region, has minimal risk and satisfies multiple objectives. The Institute of Risk Management (IRM) has identified four key risks, which are externally driven and beyond control. • Financial risk • Strategic risk • Operational risk • Hazard risk. These four risks are examined in the following sections, using a qualitative risk analysis. TABLE 1. Qualitative

2

3

4

5

6

7

8

9

10

46

project are completing the project within estimated costs and envisaged revenues. Any significant departure from these estimates generates financial risk. The probability of cost overrun is high for complex projects due to technical uncertainties (e.g., challenging terrain, resource mobilization, etc.) and political uncertainties. Furthermore, multibillion-dollar projects rely on loans. The impacts of foreign exchange fluctuation and interest rates also contribute to financial risk. The mechanisms of public-private partnerships (PPP) and build-own-operate (BOO) implementation serve to manage the financial risk of projects. Strategic risk. Political uncertainties may change the priorities of investor countries, cause project delays or cancellation, or bring revenue loss. Adverse impacts to revenue inflow will lower the return on investment (ROI), thereby jeopardizing the project. The governments of Turkmenistan, Afghanistan, Pakistan and India have established a natural gas pipeline consortium under a PPP. This mechanism may mitigate risk for the proposed TAPI project if the owners implement the project slowly. However, the success of the project remains reliant on the strategic intents of the participating governments.

analysis of key risks for pipeline projects on a five-point scale

Sub-factors (from project point of view)

1

Financial risk. Prime factors for economic viability of a

Regulations/security

Geopolitical factors

Organizational factors

Financial factors

Selection criteria

Impact of insurgency and terrorist activity, current scenario, history Political perspective, country's priority, bilateral relations Board, vision and mission, structure of organization Financial investment, credits, international support

Resources (other than financial)

Labor, machinery, data handling, material, communication

Force majeure

Unstable political climate, earthquake-prone zone

Geographical factors

Impact of terrain, current scenario, historical data

Awareness of risk and environment

Political situation, education, awareness, demography

Technical factors

Technical knowledge, construction, inaccessible geographical terrain, maintenance

Market uncertainty

Market scenario of seller and buyer countries


Factor score

1

2

3

4

5

Very Good

Good

Satisfactory

Bad

Very Bad

IPI

X

TAPI

X

IPI

X

TAPI

X

IPI

X

TAPI

F i n a n c i a l r i s k

X X

IPI

X

TAPI

X

IPI

X

TAPI

X

IPI

X

TAPI

X

IPI

X

TAPI

X

IPI

X

TAPI

X

IPI

X

TAPI

S t r a t e g i c r i s k

X

IPI TAPI

Key Risks (as per IRM)

X

H a z a r d r i s k

O p e r a t i o n a l r i s k

PIPELINES

Another project organizing strategy is to adopt the BOO route. The BOO contractor of a project usually obtains revenue through a fee charged to the project owner. The project owner (investor) directly takes tariffs charged to consumers. For the proposed TAPI project, the BOO operating company, TAPI Ltd., would use its business expertise, technical skills ski lls and resources to mitigate risk. A similar strategy to circumvent risk is envisaged for the proposed IPI project. Operational risk. As the gap between be tween demand and supply of natural gas is high, the operational risk from consumers and competition is deemed low. However, technical risks run high for the two proposed projects. Almost 50% of the land area of Pakistan and Afghanistan is mountainous. This difficult terrain can be a challenge for workers and heavy machinery. In addition, the area’s snow-fed rivers are prone to flooding in the summer months, which creates a challenging climatic condition in which to operate the pipeline. Moreover, the portions of the planned pipeline route located in the seismic zone have an increased probability of disruption. The challenges to design, construction, regular maintenance and operation of the pipelines contribute to higher operating cost. The TAPI and IPI pipelines are characterized by similar operationall risks. These risks could be controlled or mitigated operationa by engaging experienced exper ienced expert expe rt consultants consulta nts and contractors. contractors . Hazard risk. India’s Health and Safety Authority states that hazard risk must be assessed for financial, legal, moral and ethical reasons. Accordingly, hazard risk must be eliminated, sub-

stituted, isolated or addressed with suitable control measures. To assess and eliminate el iminate such hazard risk, risk , the engaged agencies must have the requisite experience, competence and knowledge. It is imperative that efficient and effective projTABLE 2. Weighted score of key risk

sub-factors for pipeline

projects on a 10-point scale Sub-factors (fr fro om proje jec ct poin intt of vie iew w)

Wei eig ght

Weig ight htag age e fac acttor (a (a))

1

Regulations/security

9

0.14

2

Geopolitical factors

10

0.15

3

Organizational factors

5

0.08

4

Financial factors

8

0.12

5

Resources (other than financial)

4

0.06

6

Force majeure

4

0.06

7

Geographical factors

10

0.15

8

Awareness of risk and environment

7

0.11

9

Technical factors

5

0.08

10

Market uncertainty

4

0.06

66

1.00

Sum total

REGISTER TO ATTEND TODAY

Engage with leading industry stakeholders and discuss pathways to a liquid LNG market, smart financing for LNG projects and gas in the energy mix

Minister S Iswaran

Minister Bill Johnston

Sandeep Mahawar

Dulal Halder

Minister for Trade and Industry (Industry)

Minister of Mines and Petroleum

Director,, Commercial Director

Parliament of Western Australia

Singapore LNG Corporation Pte Ltd

Vice President - Strategy & Corporate Planning

HE Dr. Seyed Mohammad Hossein Adeli

Alan Lim

Denis Bonhomme

Deputy Director – Port Services

Senior Vice President Business Development Asia

Ministry for Trade and Industry (Industry)

Porrasak Ngamsompark

Acting Director for LNG Management Bureau Department of Mineral Fuels, Ministry of Energy, Thailand

Secretary General Gas Exporting Countries Forum (GECF)

Maritime Port Authority of Singapore

ONGC Videsh

ENGIE Global LNG

FOR MORE DETAILS, PLEASE CONT CONTACT ACT [email protected] [email protected] OR VISIT VISI T WWW.GASASIASUMMIT.COM/GP-1 TO VIEW UPDATED CONFERENCE PROGRAMME


47

PIPELINES

TABLE 3. Qualitative

Subfactors (from project point of view)

1

analysis of key key risk sub-factors for pipeline projects projects and aggregate scores

2

3

4

Regulations/ Geopolitical Organizational Financial security factors factors factors

5

6

7

8

9

10

Resources Awareness (other Geographical of risk Technical Market Force than factors and factors uncertainty majeure financial) environment

e r o c s e t a g e r g g A

k s i r %

Weightage factor (a)

0.14

0.15

0.08

0.12

0.06

0.06

0.15

0.11

0.08

0.06

–

–

Factor score

4

4

3

3

3

3

4

4

3

2

–

–

Calculated factor score

0.55

0.61

0.23

0.36

0.18

0.18

0.61

0.42

0.23

0.12

Factor score

4

5

2

1

2

4

4

4

2

1

Calculated factor score

0.55

0.76

0.15

0.12

0.12

0.24

0.61

0.42

0.15

0.06

ect management be established to ensure the mobilization of resources and expertise required for putting in place control measures and emergency preparedness to avoid safety incidents from flammable natural gas. Project risk assessment using a decision-making tool.

Project risk assessed for the two competing pipeline projects includes both qualitative risk and quantitative risk, using the weighted factor scoring scoring model model as a tool (TABLES 1, 2 and 3). With this methodology, the quantified score enables decision makers to assess risks, compare options and make informed decisions. Projects with higher scores involve higher risk. Risk assessment and outcome. Qualitative analyses are shown

in TABLE 1. Four key risks are identified and considered. Each key risk is broken down into sub-factors for each alternative, and each sub-factor is given a factor score under a five-point scale: very good (1), good (2), satisfactory (3), bad (4), very bad (5). Quantitative analyses are shown in TABLE 2 and TABLE 3. Each sub-factor is assigned weight in a 10-point scale, where the factor with the least risk and impact on the project is scored as “0” and the factor having the most risk and impact is scored as “10.” The weightage factor is calculated for each sub-factor, as summarized in TABLE 2. The factor score is then calculated, giving an aggregate score. The percentage of the aggregate score gives the percent risk factor for each project alternative, as summarized in TABLE 3. The outcome of the aggregate score�i.e., the percent risk factor�for the TAPI project is 3.2 (64%). For the IPI project, it is 3.5 (70%). The aggregate aggregate risk of the IPI project is comparatively higher than that of the TAPI project, although it is in a similar range. Recommendations. The diversification of India’s energy mix

and an increase in the volume of environmentally friendly fuel will pave the way for enhanced energy security. Risk analysis, 48


3.5 70 70

–

–

3.2 64 64

combined with financial analysis, is required to examine highCAPEX project proposals for different scenarios and countries. The likelihood of project success can be appraised with w ith the help of qualitative and quantitative risk analyses. Project risk must also be managed throughout the entire lifespan of the project, and this case study suggests that project risk management can be a strategic strategic initiative. Developing India aims to sustain its GDP growth and pro vide energy energy access access to a wider spectrum of the popula population. tion. A situational analysis shows that sourcing low-cost, environmentally friendly gas from neighboring nations will not only have a positive impact on per-capita energy consumption, industry and the economy, but it will also provide a platform for sustained trade relations, broaden employment avenues and encourage a win win solution for investing investing countries. countries. However, undertaking such mega-projects requires elaborate risk assessment for informed decision-making. Proposed projects like TAPI and IPI can offer avenues for pinpointing how developing countries with similar challenges may improve percapita energy consumption and sustain growth. As the future unfolds, it will become clear how stakeholding countries set priorities to successfully harness risk and capture opportunities. GP ANINDITA MOITRA, Deputy General Manager—Projects at Indian

Oil Corp. Ltd., has more than two decades of experience in project management. At present, present, she is pursuing a PhD in business management (corporate governance) and holds a firstclass MTech degree from Calcutta University in Calcutta, India, as well as a first-class MBA degree, with specialization in general management, from ICFAI Business School in Hyderabad, India. Ms. Moitra was a GATE scholar, and received honors with her BSc degree from Calcutta University. University. She was also the recipient of the Best Performer medal during the Hybrid Certificate Program in project management from U21 Global in Singapore, and holds an executive diploma in project management. She has also received a number of women executive awards for her work in the Indian oil and gas sector, particularly with regard to project management. Ms. Moitra has presented papers on project risk management at events organized by Hydrocarbon Processing in Houston, Texas; and on corporate governance at the Business School (Kolkata). She is a member of the All India Management Association (AIMA).

NEW IN GAS PROCESSING TECHNOLOGY B. ANDREW, Technical Editor

Compressor delivery for Russia’s Yamal LNG

China National Offshore Oil Corp. (CNOOC) delivered the final LNG core module for Novatek’s Yamal LNG project in Russia in mid-August. CNOOC, an offshore oil and gas specialist, was commissioned to build 36 key compressor modules for the Arctic LNG project. A compressor helps liquefy natural gas at an LNG plant. Novatek, with its Chinese partners and France’s Total, plan to launch Yamal LNG later this year. The terminal is designed to produce 16.5 MMt of LNG by 2019, as well as gas condensate. Novatek will export both the LNG and condensate, with the majority of the volumes going to Asia through Arctic waters via Russia’s Northern Sea Route.

ABS approves FLNG hull design ABS has granted approval in principle (AIP) to an FLNG hull design by Hyundai Heavy Industries (HHI). ABS and HHI collaborated with the objective of developing a safe, technically feasible and class-compliant FLNG hull design. The ABS AIP demonstrates the design substantially complies with class and regulatory requirements and minimizes risk to project schedules and safe operations, giving regulatory agencies and other key stakeholders confidence in the design. In addition, a newbuild conversion FPSO hull concept design is under development, following the same process.

Australia’s LNG boom enters final stretch The delivery of the last massive component of Australia’s $180-B LNG construction boom has stepped up a race between Anglo-Dutch giant Shell and Japan’s INPEX to start chilling gas for export in 2018. Company reputations are at stake, as well as first access to overlapping gas fields and Australia leapfrogging Qatar as the world’s largest exporter of LNG. The Ichthys Venturer, a floating production, storage and offloading facility, travelled 3,500 mi from a South Korean shipyard and will be moored 220 km off Western Australia to handle condensate from the Ichthys field. Japan’s top oil and gas explorer, INPEX Corp., is running Ichthys, which is both the country’s biggest overseas investment and its first LNG megaproject. First production, due by March 2018, will be more than 1 yr behind target. Costs have ballooned by more than 10% to $37 B since the project’s approval in 2012. Nearby, Royal Dutch Shell’s $12.6-B Prelude project—the world’s largest floating LNG (FLNG) facility— is also behind schedule. Shell lost out on becoming the world’s first producer of FLNG when Malaysia’s Petronas started up a smaller FLNG facility this year. Shell’s facility, six times the size of the biggest aircraft carriers and with a deck longer than four US soccer fields, arrived last month. Shell expects hookup and commissioning to take up to 12 mos, indicating possible startup between April and July 2018. Whichever project starts first will pump gas away from the other’s field, as the two straddle the same Prelude is smaller, according to Wood Mackenzie reservoirs. The race means more to Prelude than to Ichthys, asPrelude analyst Saul Kavonic. INPEX also has an each-way bet: It owns 17.5% of Prelude, as well as 62.2% of Ichthys.

Research for US gas storage hub in Appalachia The region around West Virginia, Pennsylvania and Ohio has abundant natural gas resources. The three states are working to turn those resources into economic growth. A West Virginia University (WVU)-led research team is helping with the effort. On August 29, WVU released to the public a study that shows how the region can support storage facilities that are critical for attracting petrochemical and related industries to the area. Researchers planned to present the data at a technical workshop in Canonsburg, Pennsylvania, hosted by the Eastern Petroleum Technology Transfer Council (PTTC), at WVU. Researchers from the geological surveys in West Virginia, Pennsylvania and Ohio studied geologic formations that could offer suitable locations for developers to build underground facilities to store NGL from Marcellus and Utica wells. The team identified and mapped all potential options for subsurface storage of NGL along the Ohio River from southwestern Pennsylvania to eastern Kentucky, and the Kanawha River in West Virginia. The researchers focused on three options for subsurface storage. One option includes areas where where the Salina F Salt is at least 100 ft thick and suitable for solution mining, a type of mining that uses a liquid, such as water injected through a borehole, to dissolve and extract salts and minerals. Another option includes areas where the Greenbrier Limestone is present 1,800 ft–2,000 ft below the surface and is at least 40 ft thick. Converting existing sandstone reservoirs in depleted gas fields and inactive gas storage fields to NGL storage is the third option. Previously, the consortium had conducted studies of the Marcellus and Utica shale gas plays. Results from those studies have been used by both small producers, such as Northeast Natural Energy, and large multinationals, such as ExxonMobil, and have helped fuel the region’s shale gas industry.

GE biogas engines for Turkey As Turkey looks to implement new energy targets under its economic development strategy, its focus is on increasing renewable energy sources and digitalization. Helping support these goals, GE will provide its Jenbacher biogas engine technology and myPlant Asset Performance Management (APM) solution for three new landfill power production sites in Turkey owned by ITC-Ka Enerji Uretim Sanayi Ve Ticaret AS (ITC), the largest landfill gas power producer in the country. The three landfill gas power facilities will convert organic waste into biogas that can generate renewable electricity and heat. The biogas produced will be used to fuel four of GE’s Jenbacher J420 biogas engines at Eskisehir, nine at Antalya and three at the Alanya landfill sites. The sites are configured to generate up to 50% of the common limit for nitrous oxide and will provide 22 MW of electrical power to the national grid. This represents enough power for 30,000 households in the three cities. The equipment will be built at GE’s Distributed Power facility in Jenbach, Austria, and it is expected to enter service during the second half of 2017.

Skangas to supply LNG for ESL carriers

Skangas, supplier of LNG in the Nordic markets, and ESL Shipping, a carrier of dry bulk cargoes in the Baltic region, have entered into a marine fuel agreement. Skangas will supply gas fuel to ESL’s LNGpowered bulk carrier newbuilds. The two new vessels will start operating during first half of 2018. The first of the two vessels was named Viikki in in June 2017. The new, ice-class 1A ships will be the first LNG-fueled large bulk carriers in the world, representing the latest in technology. They will carry out sea transports within the Baltic Sea and the North Sea. As part of the agreement, Skangas will also deliver ESL Shipping with the low-carbon liquefied biogas (LBG). LBG and LNG are interchangeable as fuel, and so the same LNG supply chain is used. The Viikki is is fitted with a dual-fuel engine. Shipowners are turning to LNG as their primary source of fuel. By 2020, nearly 200 vessels are expected to be running on LNG worldwide. This number is expected to rise significantly, as LNG is recognized by the marine industry as the most viable alternative fuel to reduce emissions.


49

SALES OFFICES—EUROPE

ADVERTISER INDEX

ITALY, EASTERN EUROPE Fabio Potestá Mediapoint & Communications SRL Phone: +39 (010) 570-4948 E-mail: [email protected]

TURKEY, WESTERN EUROPE Catherine Watkins, Publisher Phone/Fax: +1 (713) 520-4421 E-mail: [email protected] www.GasProcessingNews.com

Hamilton Pearman Phone: +33 608 310 575 [email protected]

UNITED KINGDOM, SCANDINAVIA Patrick Djuma Phone/Fax: +44 20 3409 2243 E-mail: [email protected]

SALES OFFICES—NORTH AMERICA EASTERN UNITED STATES, EASTERN CANADA Merrie Lynch Phone: +1 (617) 357-8190 Mobile: +1 (617) 594-4943 E-mail: [email protected]

Keith Smith Phone: +1 (713) 520-4449 E-mail: [email protected]

WESTERN UNITED STATES, WESTERN CANADA

Chromatic Industries...................................27 Eltherm GmbH ..............................................29 Gas Asia Summit ..........................................47 Gulf Publishing Company Events—WGLC ............................................ 51

JAPAN Yoshinori Ikeda Pacific Business Inc. Phone: +81 (3) 3661-6138 E-mail: [email protected]

Rick Ayer Phone: +1 (949) 366-9089 E-mail: [email protected]

Ariel Corporation ..........................................21

CHINA—HONG KONG

Manav Kanwar Phone: +91-22-2837 7070/71/72 Mobile: +91-98673 67374 E-mail: [email protected]

Josh Mayer Phone: +1 (972) 816-6745 E-mail: [email protected]

Ametek Process Instruments ....................12

Energy Web Atlas .............................30, 50

INDIA

MIDWEST, NORTH TEXAS

American Petroleum Institute..................39

SALES OFFICES—OTHER AREAS

Iris Yuen Phone: +86 13802701367 (China) Phone: +852 69185500 (Hong Kong) E-mail: [email protected]

GULF COAST, SOUTH TEXAS

Air Products & Chemicals Inc. .................... 2

Hunter Buildings ...........................................17 Kobelco Edti Compressors Inc ..................15 Linde Engineering North America ............ 7 Merichem Company......................................11 Metrix ..............................................................35 Pentair Filtration & Process ......................52 Redguard..........................................................8 Shell Global Solutions................................... 5 SNC-Lavalin ...................................................23 World Gas Conference .............................. 44

DATA PRODUC TS J’Nette Davis-Nichols Phone: +1 (713) 520-4426 E-mail: [email protected]

This index and procedure for securing additional information are provided as a service to advertisers and a convenience to our readers. Gulf Publishing Company is not responsible for omissions or errors.

ACTIONABLE INSIGHTS INTO US NATURAL GAS FACILITIES The Energy Web Atlas is a comprehensive source of data for facilities and installations in the US natural gas industry. The web-based GIS application includes detailed data on more than 500 US gas processing plants, 400 natural gas underground storage locations, and 58 natural gas market hubs.

Data Points The platform’s mapping, search and query tools allow users to interrogate up-to-date facility and project data and find the informat ion nee ded for major decisions. Data points include: • Key contacts (address, email, phone)

• Owner/Operator

• Plant type/flow, BTU content, infrastructure

• Market hub/Underground storage data.

• Gas storage capacity

• Plant status

Additional Maps Available. Global LNG/pipeline data

To schedule a personal demonstration, contact J’Nette Davis-Nichols at [email protected], +1 (713) 520-4426.

D I S C O V E R M O R E AT E N E R G Y W E B A T L A S . C O M

50


November 1–2, 2017 | Hyatt Regency Houston | Houston, Texas

Join the Discussion at Women’s Global Leadership Conference (WGLC) Engage with key industry leaders at WGLC for presentations and discussions on:

Keynote Speakers

— The Essentials of Engineering Leadership and Innovation, Dr. Pamela McCauley, Professor, Department of Industrial Engineering and Management Systems, University of Central Florida

•

DAY 1 KEYN OTE

•

DAY 2 KEYN OTE

— Discipline Career Management and Knowledge Management,

Susan Rosenbaum, Director , Schlumberger •

DAY 2 PANEL DISCUSSION

— STEM: Are we really making progress?

Moderator: Ray Dempsey, Vice Presi dent and Chief Dive rsity Offi cer , BP America Panelists: Pam Darwin , Vice President Africa, ExxonMobil Exploration Company Dr. Reagan Flowers , Founder and CEO, C-STEM Teacher and Student Support Services, Inc Johanna Haggstrom , Director of Technology , Baroid Product Line, Halliburton Natalia Shehadeh , Vice Presid ent & Chief Compl iance Offi cer , Weatherford •

DAY 2 PANEL DISCUSSION —

Keynote Day One

Pamela McCauley

Digital transformation: The impact on the industry

Moderator: Soumya Seetharam, Vice President , Enterprise Solutions, Schlumberger Panelists: Morag Watson , Vice President, Digital Innovation, BP Maria Claudia Borras , President & CEO, Oilfield Services, Baker Hughes, a GE company Dr. Satyam Priyadarshy, Technology Fellow & Chief Data Scientist , Halliburton Mehrzad Mahdavi , Vice President Digital Solutions, Weatherford WGLC is the oil and gas industry’s largest and most important women’s event, which makes it the ideal venue for networking with like-minded individuals and making new industry connections. Take advantage of these opportunities by attending our speed networking and topical luncheons. For more information on sponsorship opportunities, please contact Hortensia “Tish” Barroso, Business Development Manager , at [email protected]

PROUDLY SPONSORED BY:

Keynote Day Two

Susan Rosenbaum

Hydrocarbon Processing October 2017

Recommend Documents