OTC 17700 API RP2FB, 1st Edition - Design of Offshore Facilities against Fire & Blast Loadin g P.E. O’Connor, BP; P.E. Versowsky, ChevronTexaco; and J.R. Bucknell and M. Chakravorty, MSL Copyright 2005, Offshore Technology Conference This paper was prepared for presentation at the 2005 Offshore Technology Conference held in Houston, TX, U.S.A., 2–5 May 2005. This paper was selected for presentation by an OTC Program Committee following review of information contained in a proposal submitted by the author(s). Contents of the paper, as presented, have not been reviewed by the Offshore Technology Conference and are subject t o correction by the author(s). The material, as presented, does not necessarily reflect any position of the Offshore Technology Conference, its officers, or members. Papers presented at OTC are subject to publication review by Sponsor Society Committees of the Offshore Technology Conference. Electronic reproduction, distribution, or storage of any part of this paper for commercial purposes without the written consent of the Offshore Technology Conference is prohibited. Permission to reproduce in print is restricted to a proposal of not more than 300 words; illustrations may not be copied. The proposal must contain conspicuous acknowledgment of where and by whom the paper was presented. Write Librarian, OTC, P.O. Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435. 01-972-952-9435.
Ab st rac t As the oil and gas production in the Gulf of Mexico began to extend into deeper water and larger and more congested topsides facilities became necessary, the US offshore Industry recognized the need for a Recommended Practice (RP) for design of these new facilities against fire and blast loading. This paper describes the background to the development of the RP, which has been put together with extensive contributions from industry experts with different areas of specialist knowledge. The paper discusses the consistency of the RP with recent work to update related standards, in particular the UKOOA/HSE initiatives in the UK. The paper addresses the issue of blast load determination and discusses the nominal loads provided in the new RP and their application to design. It also considers alternative methods for the calculation of blast loads, in lieu of applicable nominal load cases, including a promising new methodology analogous to the derivation of earthquake earthquake loading. The paper stresses the importance of good practice for fire and blast design and discuss the guidelines for facilities layout and structural connection detailing. The paper highlights the interface to other API documents that provide guidance on the implementation of safety and environmental management practices and hazard identification, event definition and risk assessment e.g. API RP 75 (1) and the API RP14 (2,3) series. It briefly outlines how the new RP incorporates hazard analysis output into the structural response assessment to determine whether the structure or its components meet the specified performance criteria. Background Deepwater developments to harness larger and deeper hydrocarbon deposits bring with them the need for larger and more complex topsides facilities. A key learning from explosion research over the past 10 years is that congestion as well as confinement can lead to high overpressures in an
explosion. The increased size of facilities increases the porabaility of releases and the increased congestion of facilities increase the consequences in the event that the release ignites, particularly if the ignition is delayed and an explosion occurs. Much can be done to manage this risk. Prevent the releases occurring is clearly preferable but the cosnsequences in terms of possible life-safety, environmental and economic consequences can be reduced by establishing performance objectives objectives and incorporating incorporating good practices. practices. Good practices have been developed in the industry and incorporation of thiose practices early in the design can significantly reduce the risks associated with fire and explosions. Hence one on the key objectives of this proposed RP was to capture those practices and share them with the rst of industry. In December 2000, API issued the 21 st Edition of RP2A (4). The document includes as Section 18, ‘Fire, Blast and Accidental Loading’ that provides, for the first time in API, a recommended practice for design of fixed steel offshore structures against fire and blast events. In October 2001, API SC2 formed a Fire and Blast Task Group under the Chairmanship of BP’s Patrick O’Connor. The objective of the Task Group was to create an RP for the Design of Offshore Structures for Fire and Blast. The focus of the document was to be structural loading and response to fire and blast events with RP2A Section 18 as the primary source document. API provided funding for a Consultant to draft the RP under the direction of the Task Group. MSL Services Corporation (MSL) was the selected Consultant. MSL was also coordinating a parallel MMS initiative to host an International Workshop on the subject of fire and blast of offshore facilities. The Task Group planned to use the Workshop as a forum for review of the draft RP and receive industry feedback. At the same time in the UK a joint operator (UKOOA) and regulator (HSE) team had begun a parallel initiative to update the UK Interim Guidance Notes for Fire and Blast. This work was also coordinated by MSL through their UK office thus ensuring effective communication with the API initiative. The scope of the RP encompassed the design of all new offshore platforms. In recognition of the specific issues associated with floating facilities a sub-Task Group was established to address fire and blast issues specific to floating structures. The MMS International Workshop on Fire and Blast was held in Houston in June 2002 (5). Over 150 delegates from
2
around the world attended the three-day event. Deliberations and output from the seven working group was compiled into a Proceedings (5) that was used by the Task Group as input to the API RP. The proposals contained in this paper are still being developed and have not yet been approved within API. As such this should be viewed as work in progress Alternative Codes and Standards The UKOOA/HSE study (6) has concluded the first phase of their initiative, which provides updated explosion design guidance. The methodology and general guidance is largely consistent with the API RP. One major difference is that nominal blast loads are not provided in the UKOOA/HSE code. The second phase, for fire loading and response is presently underway. The relevant ISO Standards, ISO 13702: Control and mitigation of fires and explosions and ISO: 19901-3 Topside structures, are currently on hold pending completion of the UKOOA/HSE and API work. Norwegian, Norsok Standards (Z-013 Risk and preparedness analysis and N-003 actions and action effects) adopts a more rigorous and probabilistic approach including establishing leak scenarios, cloud size distribution, explosion loads, from Computational Fluid Dynamics (CFD) simulations and detailed risk assessment. The procedure requires a large number of complex CFD and QRA analyses. Canadian Petroleum Board Regulations, in particular CSA standard S471 has a section dealing with accidental loads and refers to the Norsok N-003 and Z-013 / ISO for loadings and other NORSOK documents. This is currently being updated in reference to the UKOOA/HSE proposed methodology. Hence the proposed API approach whilst capturing a lot of the knowledge gained in Europe attempts to simplify its application within the API code.
Introduction The Recommended Practice for the Design of Offshore Facilities against Fire and Blast Loading is based on sound engineering principles and was developed with the help and extensive contributions of the owners, operators, designers, fabricators, suppliers, and certifiers of offshore facilities, with many years of experience. In no case is any specific recommendation included that could not be accomplished by presently available techniques and equipment. Foremost consideration is given to the safety of personnel, compliance with existing regulations, and prevention of pollution. The RP includes a simple qualitative risk assessment process to assist in the determination of the need for consideration of fire and blast in the design of the facility. Guidance is also provided for establishing performance criteria. The user is referred to alternative sources for more rigorous risk assessment outside the scope of the RP. The recommended practice includes detailed Commentary. The commentary provides design guidelines for the evaluation of structural response to fire and blast loads. Nominal blast load cases are provided for certain classes of facilities. Guidance is also provided for the calculation of fire loads. Discussion of alternative methods for the calculation of blast loads, in lieu of applicable nominal load cases, is included with reference to sources of detailed guidance. The
OTC 17700
commentary also includes examples of good practice for fire and blast design including guidelines for facilities layout and structural connection detailing.
Screening The proposed RP recognizes the excellent industry safety record for fire and especially blast events in the Gulf of Mexico. The platforms are typically open design with low complexity and low congestion of facilities. The proposed RP establishes a class of un-manned fixed structures; typical of many Gulf of Mexico fixed steel structures, as low risk facilities for which explicit modeling of fire and blast loading is not required beyond the adoption of good practice. Guidelines are provided in the proposed RP Commentary for good practice in layout of facilities and equipment to minimize likelihood and consequences of fire and/or blast events. Performance Criteria The operator is responsible for the overall safety of the platform and performance criteria should be established consistent with the operators overall safety management philosophy. In the structural evaluation of a defined fire and/or blast event, the structure should be designed to the adopted performance criteria. The propsed RP provides guidance to assist the user in selection of performance criteria, as follows: 1. For structural evaluation of loads associated with low probability, fire and/or blast events (infrequent occurrences) performance criteria should ensure defined ‘survival’ of the platform. 2. Any blast wall and/or firewall should remain in-place without rupture or disconnection from their supports. Deformations of the wall should be limited to avoid escalation. 3. Safety critical elements (SCEs) that are designed to mitigate the effects of a major accident, such as, those necessary for (a) the safe shut down of the installation, (b) personnel protection and escape, (c) fire protection, suppression and control, (d) communications, and for (e) hydrocarbon containment including transport and storage; should remain intact. 4. For platforms with the potential to be manned during the defined event, performance criteria should ensure defined ‘survival’ of the platform. For structural evaluation of loads associated with medium or especially high probability fire and/or blast events (more frequently occurring but smaller in magnitude) the RP recommends that performance criteria be modified to limit the extent of damage to the facility. For example, the platform may be designed to permit restarting of operations in a reasonable timeframe and following appropriate integrity checks.
Risk Assessment The new RP includes guidance for the assessment of the risk level associated with a fire or blast event. The recommended approach is a simple qualitative assessment and the user is referred to alternative API documents (6) for more rigorous risk assessment methodologies outside the scope of the RP.
OTC 17700
3
The three-by-three risk matrix provided in the RP is shown in Figure 1. The matrix defines low, medium and higher risk categories for fire or blast events. The risk is defined by consideration of the probability of the event occurring and the consequences of the event if it were to occur.
e c n e r r u c c O f o y t i l i b a b o r P
h g i H
Medium Risk
Higher Risk
Higher Risk
m u i d e M
Low Risk
Medium Risk
Higher Risk
w o L
Low Risk
Low Risk
Medium Risk
Low
Medium
High
Consequence of Occurrence Figure 1: Risk Matrix
Low risk events do not require specific consideration of the event in the structural design of the facility. However, the design of all offshore structures should adopt good practice in layout of facilities and equipment to minimize likelihood and consequence of possible events. Higher risk events shall be evaluated in the structural design or reduced by implementation of prevention and/or mitigation measures or through change(s) in layout etc. Alternatively, more rigorous assessment of probability and/or consequence of the event may be undertaken. Medium risk events should be evaluated in the structural design unless they are determined to be as low as reasonably practicable i.e. the effort and/or expenses of mitigation become disproportionate to the benefit. If this is determined to be the case it should be suitably documented as such. The significant events requiring consideration and their probability of occurrence levels (that is low, medium or high) are normally defined from a fire or blast process hazard analysis. Such analyses are outside the scope of the RP. General guidance is provided to assist the user in assigning the probability and consequence of the event, as follows: Higher Probability: The event is likely to occur during the life of the platform OR has occurred more than once on a similar platform in the past. Medium Probability: The event is not expected to occur during the life of the platform, and the event does not meet the criteria for Higher Probability or Low Probability. Low Probability: The event is extremely unlikely to occur during the life of the platform OR no such occurrence of the event is reported on similar platforms. (Note for novel concepts the lack of historical evidence needs to recognized) The consequence level for a defined fire or blast event should be assigned as low, medium or high as applicable for either life-safety, environmental, or other consequences. The life safety of personnel in the direct vicinity of the event is an issue for operational safety procedures and management and not of structural design and is outside the scope of the RP. In the selection of life safety consequence for structural design, therefore, the designer should be considering life safety
consequence of personnel away from the immediate vicinity of the event.
Risk Assessment Process The risk assessment process provided in the RP comprises a series of tasks to be performed to identify facilities at significant risk from specific fire and/or blast events. The process is illustrated in Figure 2. The steps in the process are defined as, follows: Task A1: If facility meets the definition of a low-risk facility as defined by the screening process describes above and characterized by low equipment counts, limited to wellheads and manifold with few vessels and little associated pipe work, which would lead to low congestion and inventory, consideration of specific fire and blast loading in the design of the structure is not required. Task A2: Establish the performance criteria for the facility to comply with the overall safety and environmental management philosophy, as well as relevant regulations and company standards. Task A3: Implement measures to reduce fire and blast risk in accordance with good design practice. Task A4: Establish whether nominal load cases for blast loading are available for the facility. If nominal load cases are not available for the facility, proceed to Task B1. Task A5: Evaluate the response of the critical structure and other key components to the nominal load cases. Critical structure and key components refer to elements of the facility that must survive for a specified duration of time following the occurrence of the event in order that the performance criteria for that event are met. Task A6: If the performance criteria set in Task A2 can be met for the nominal load cases the structural design for the event is complete for the facility. If it has been established from Tasks A1 to A5 that the facility does not meet the low-risk platform definition and that either nominal load cases are not available or structural evaluation indicates that the facility does not meet the performance criteria for the nominal load cases applied, it is necessary to consider fire and blast risk on an event-by-event basis. Task B1: Consideration of event-by-event fire and blast risk requires a formal hazard identification study for the definition of credible events (scenarios) and determination of their associated risk. Some guidance for the determination of the probability of events and their consequences is provided in the RP. More detailed Guidance is available within the API RP 14 series (2,3) and other sources (7,8). Task B2: Determine whether the level of risk associated with the event is low-risk, medium risk or higher-risk. If the risk is low, the assessment is complete for the defined event. Task B3: For events, where a higher-risk is identified, consideration may be given to modifying the design concept or adopting an alternative concept. This may be especially applicable during the early stages of a project. In this case, the assessment process is repeated from Task B1 for the modified or alternative concept. Task B4: This task provides the choice for the engineering team to explore prevention and mitigation options to reduce the risk associated with the event.
4
Task B5: If the probability of the event or its consequence or both were reduced such that the risk becomes low-risk, the assessment is complete for that event. Task B6: This task involves the calculation of the fire or blast loads associated with the event and the evaluation of the survival of the critical structure and other key components required to meet the performance criteria set in Task A2. Task B7: If the performance criteria set in Task A2 can be met for the load cases for the specific event, the assessment is complete for the event. Task B8: In the case that the structural evaluation indicates that the performance criteria cannot be met, the engineering team must consider whether further risk reduction/mitigation options exist. If so, these should be implemented and the process reverts to Task B4. If no further risk reduction options are available, the RP stresses the need to modify the design concept or, adopt an alternative concept. In this case, the assessment process is repeated from Task A1 for the modified or alternative concept.
Blast as a Load Condition The loading generated by a blast depends on many factors, such as the type and volume of hydrocarbon released; ignition source, type and location; the degree of congestion in a module; the amount of confinement; and the ventilation conditions. The RP discusses alternative means for developing the design blast event(s). Blast Simulation Modeling A blast scenario developed from a process hazard analysis establishes the make up and size of the vapor cloud, and the ignition source for the area being investigated. The blast overpressure from explosions in congested volumes can be predicted using various models, as follows: Empirical models based on the correlation of experimental data and their accuracy and applicability relating to the experimental database. Phenomenological models based on modeling the underlying physical processes interpolating more accurately between data and extrapolating with more certainty to situations not addressed by experimental work. Numerical models that solve the underlying equations describing gas flow, turbulence and combustion processes. Numerical models following the principles of Computational Fluid Dynamics (CFD) have the potential for providing a higher accuracy and of addressing any blast scenario (11). Figure 4 illustrates the distribution of blast overpressures in a closed compartment. Nominal Blast Loads There is a reasonable experience base across industry from which ‘nominal loads’ have been established for certain classes of structures. The Commentary section of the RP provides available data sets for ‘nominal loads’. In lieu of the availability of applicable nominal overpressures, some level of blast simulation modeling is recommended for computation of blast loading on offshore structures The RP introduces the ‘nominal load’ concept for blast events on selected known facility types. Nominal loads for •
•
•
OTC 17700
fires have been in use since the publication of the Interim Guidance Notes (8) in 1993, and have been updated and extended in more recent references (6). For fires, these take the form of recommended radiation levels and flame temperatures for pool and jet fires in confined and open conditions. For blast, the nominal loads provided in the RP are peak spatially averaged blast overpressures determined for specific platform types from a set of data. The methodology for the derivation of ‘nominal overpressures’ was first developed based on limited data extracted from a review of about 30 recent safety cases (about 50% post 1997) prepared in the UK and carried out in April 2002 (12). These were extended with additional data sets provided by Task Group members. If available and considered suitable for use for the particular facility, these nominal loads may be used for the structural design of the facility. Nominal loads may be particularly useful at an early project phase where detailed geometry of the layout, and congestion is not known. Response Spectrum Method A further attempt to simplify the assessment method is presently under development (10) and is discussed briefly in the proposed RP. The response spectrum approach takes into account the variations in response of structural elements resulting from their natural periods and differing dynamic properties for a given pressure-time history. The proposed method also enables the reserves of strength released when elements are allowed to deform plastically to be taken into account. The purpose of the method is to enable equivalent static blanket loads to be derived for use in a conventional design check for blast loads of a given severity and duration. The method has been in use for decades in the earthquake response context. Figure 3 shows the application of blast response spectra in determining a static design pressure. The severity of the blast loading is determined from local conditions by the use of nominal overpressures, previous experience, risk classification, simulations or experiment. The structural element is represented by its natural period and resistance at effective yield. A further important parameter is the allowable ductility of the element, which is a measure of the amount of plasticity that the element can sustain before rupture and is related to the peak displacement. Allowance for local plastic deformation is an essential part of efficient blast resistant design. The element natural period determines the position on the horizontal axis in Figure 3 and the design pressure or required static resistance may be read off the relevant curve representing the allowable ductility. Continuing research work in this area may lead to better data sets and simpler analytical procedures in the near future.
Fire As Load Conditio n The treatment of fire as a load condition as provided in the RP requires the definition of the following: The fire event or scenario. Heat flow characteristics from the fire to the unprotected and protected steel members. • •
OTC 17700
Properties of steel at elevated temperatures where applicable. Where applicable, properties of fire protection systems (active and passive). The fire scenario may be identified from a process hazard analysis. The fire scenario establishes the fire type, location, geometry, and intensity. The fire type will distinguish between a hydrocarbon pool fire and a hydrocarbon jet fire. The fire’s location and geometry defines the relative position of the heat source to the structural steel work, while the intensity (heat flux) defines the amount of heat emanating from the heat source. The Commentary in the RP provides guidance on the calculation of temperature/time histories of the member and/or the steady state temperature. For load bearing members, the temperature determines the appropriate values for the material properties to be used in the structural analysis. For panels and firewalls, which are usually, not load bearing; the important parameters are the temperature of the cold face and the time to reach certain limiting temperatures, which determine the walls’ rating. •
•
Structural Assessment Against Blast For higher consequence facilities, the RP recommends that the operators may consider two levels of explosion loading by analogy with earthquake assessment, i.e., the ductility level blast and the strength level blast. The design level blast load (derived for the blast event using one of the options described above) is referred to as the Ductility Level Blast (DLB); defined as a low-probability high-consequence event, which is to be evaluated to confirm whether performance criteria are met. The ductility level blast is the design level overpressure used to represent the extreme design event. A reduced blast load, sometimes referred to as a Strength Level Blast (SLB) by analogy with earthquake design, is defined as a higher probability, lower consequence event. Performance criteria associated with the SLB may include elastic response of the primary structure, with the safety critical elements remaining functional, and with an expected platform restart within a reasonable period. In lieu of a more detailed hazard process study or other means to define the magnitude of the SLB, the RP recommends the load may be taken as 1/3 the DLB overpressure. The SLB load case may be desirable for the following reasons: The SLB may detect weaknesses in the structure at an early stage of the design improving the likelihood of meeting performance criteria for the DLB. The prediction of equipment and piping response in the elastic regime is better understood than the conditions that give rise to rupture. The SLB enables these checks to be made at a lower load level often resulting in good performance at the higher load level. It is quicker to perform SLB load case. If performed correctly, the assessment will provide good assurance of adequacy of structure under DLB loads. The SLB load case provides a degree of additional asset protection. •
•
•
•
5
Ductility Level Assessment The performance of the primary framing of an installation during the ductility level load can be assessed with either linear elastic structural response analysis or a more accurate non-linear assessment. In the case of the linear elastic analysis, code checks may be accepted with higher than normal utilization factors to allow for member plasticity, strain rate and strain hardening effects. The RP provides guidance on acceptable code check utilization ratios e.g. 2.5 for a tension member and 2.0 for members under bending and/or compression so long as the member does not buckle, under the blast loading. The preferred assessment is to use a non-linear ductility level analysis that will account for large displacements, load re-distribution, material property changes, non-linear effects and the dynamics of the structure. This method of analysis is recommended in the RP where it is required to account for the reserve strengths of the structure. The Commentary in the RP provides guidance on deformation checks, buckling checks, and rupture checks. The Commentary in the RP describes details of the analysis procedures and provides guidance on properties of material behavior under blast loads including strain rate effects, strain hardening and ductility. Figure 10 illustrates the progressive collapse of a deck structure in a blast scenario.
Structural Assessment Ag ainst Fire Three successively more complex methods of analysis requiring different analysis tools with increasing complexity are recommended in the RP for structural response assessment against fire by using one, or a combination of, the following methods. Zone (or screening) method Strength level method Ductility level method For the zone method, the temperature of the structural steel member due to a fire scenario is checked against the maximum allowable temperature that a steel member can sustain without reducing its yield strength below 60% of the yield strength (F y) at ambient temperature. If this is confirmed for the design event then further checks are not required. A strength level analysis consists of a conventional linear elastic structural analysis. Depending on the maximum temperature attained by individual structural members for the duration of the fire, the reduced stiffness and yield stress of the member are used in the structural analysis. A ductility level analysis, which is a progressive collapse analysis, allows redistribution of structural load from failed members and can indicate collapse of the structure after no further load distribution is possible. Facilities, which do not pass the strength level analysis, may be evaluated using ductility level analysis if prevention and mitigation measures are not available or desirable. This level of analysis will also allow sequences of failure and time histories of events to be established. This may be important for emergency evacuation and escape assessments. Figure 9 illustrates progressive collapse of a support structure in a fire scenario. • • •
6
The Commentary in the RP describes details of the analysis methodologies and guidance on thermal properties of steel subject to fire loads.
Mitigation Implementing preventative measures has historically been, and will continue to be, the most effective approach in minimizing the possibility of occurrence of a fire or blast event and the resulting consequences of the event. The RP draws attention to the design of the structure and in the layout and arrangement of the facilities and equipment to minimize the effects of these events. The commentary in the RP details the various mitigation measures including the use of passive fire protection systems, and use of properly designed blast walls. Figures 5, 6 and 7 illustrate details of commonly used blast wall profiles, penetrations and connections. The RP stresses the need for using good engineering practice particularly in detailing connections to ensure adequate design ductility. These are shown in Figure 8. Floating Structures, Storage and Offloading Systems An increasing number of large floating installations with high inventories, storage and/or throughput are operated and being planned around the world including the Gulf of Mexico. There are several special features associated with floating installations, which cannot be dealt with by simple extrapolation of current practices in use on fixed installations. These features relate to the differing geometry, methods of construction, compartmentation, operations, fire and blast scenarios, response characteristics of marine construction to fire and blast, and special features associated with the motion, station keeping and stability of the installation. The RP emphasizes the need to check on the stability of the installation due to the effect of any fire or blast event, which has the potential to bring on the instability of the floating system. Other considerations include structural integrity of the hull, maintenance of evacuation capabilities, and prevention of secondary events and escalation. Floating Offshore Installation require the use of Marine Systems and a Marine Operations Manual as well as personnel for safe operation. The marine operations depend on a number of sub-systems that are important in evaluation of safety against fire and explosion events. The RP provides details of all such sub-systems. The specific issues that should be addressed during risk analysis of a floating offshore installation including FPSOs have also been categorized in the RP. The features of a floating offshore installation that may contribute to the change of risk associated with fire and blast events have been identified in the RP. The specific design issues for a floating offshore installation against fire and blast loading including hull design, hull compartmentation, etc., are highlighted in the RP. Conclusion In response to the needs of US industry API established a Task Group, and funded a specialist consultant, to develop a Recommended Practice for the ‘Design of New Offshore Structures for Fire and Blast Loading’. The RP provides a consistent approach for the design of future offshore
OTC 17700
structures. The document has been balloted and approved by API Subcommittee 2: Offshore Structures. It will be re balloted with updates to address comments received during the first ballot in the second quarter of 2005. Some of the noteworthy features of the new RP may be summarized, as follows: Screening: The RP establishes a class of un-manned fixed structures as low risk facilities for which specific consideration of fire and blast loading is not required beyond the adoption of good practice. Risk Assessment : The RP provides a simplified qualitative risk assessment process for determining risk of fire and blast events. The RP also suggests alternative API and other documents for hazard studies and more detailed risk assessment outside the scope of the document. Applicability: The RP is applicable to the new design of future offshore production systems including fixed and floating structures. Specific recommendations are provided for floating systems as appropriate. Nominal Loads: The RP provides nominal blast loads for certain known facility types and configurations to reduce the requirement for complex blast simulation analyses. The Response Spectrum Approach: The RP introduces a promising new methodology for blast load derivation. Strength and Ductility Checks : The RP introduces the concept, analogous to earthquake engineering, of strength level and ductility level analyses. Strength level analyses are linear elastic checks that can be used at the early project stage and with basic design analysis tools. Standardization: The RP is consistent with the general approach to fire and blast design for offshore structures contained in the parallel UKOOA/HSE study (6) and intended for the future updates to the ISO Standards.
Ac kn ow led gem ent s The authors wish to thank their respective companies for the opportunity to publish this paper. Thanks are also extended to Steve Walker with MSL for his technical input to the drafting of the RP and to all members of the API Task Group on Fire and Blast for the contribution of their knowledge and experience. References 1. American Petroleum Institute, “Recommended Practice for Development of a Safety and Environmental Management Program for Outer Continental Shelf Operations and Facilities”, API RP75, 3rd Edition - May 2004. 2. American Petroleum Institute, “Recommended Practice for Design and Hazards Analysis for Offshore Production Facilities”, API RP 14J, 2 nd Edition - April 2001. 3. American Petroleum Institute, “Recommended Practice for Fire Prevention and Control on Open-type Offshore Production Platforms”, API RP 14G, 3rd Edition - December 1993. 4. American Petroleum Institute, “Recommended Practice for Planning, Designing and Constructing Fixed Offshore Platforms – Working Stress Design”, API RP 2A, 21st. Edition, December 2001. 5. Bucknell J., Chakravorty M. and Smith C., Editors, “Proceedings of International Workshop on Fire and Blast Considerations in the Design of Offshore Facilities”, June 12-14, 2002, Houston, Texas.
OTC 17700
6. UKOOA/HSE, “Preparation of Updated Guidance for Fire and Explosion Hazards – Part 1 – Guidance on Design and Operational Considerations for the Avoidance and Mitigation of Explosions, Commentary”, December 2002. 7. NORSK Standard, N-004, “Design of Steel Structures, Annexure ‘A’ – Design against Accidental Loads, 2000. 8. Selby C. A. and Burgan B.A., “ Blast and Fire Engineering for Topside Structures – Phase 2, Final Summary Report”, SCI Publication Number 253, Ascot, UK, 1998. 9. Steel Construction Institute, “Interim Guidance Notes for the Design and Protection of Topside Structures against Explosion and Fire”, SCI-P-112/005, Ascot, UK, 1993. 10. Private Communication from Brian Corr (bp), Vincent Tam (bp) and Steve Walker (MSL). 11. Steel Construction Institute, “Interim Guidance Notes for the Design and Protection of Topside Structures against Explosion and Fire”, SCI-P-112/005, Ascot, UK, 1993. 12. W.S. Atkins, Inc., ‘Preparation of Updated Guidance for Fire & Explosion Hazards’, 2002.
7
8
OTC 17700
Note 2
Task B.1 Task A.1
Does the facility meet screening criteria?
Yes
No
Task A.2
Assessment complete Ensure Good Practice
Consider fire and blast risk Event-by-Event
Note 1
Task B.2 Risk Matrix
Establish performance criteria
Task A.3
Low
Higher
Implement measures to reduce fire and blast risk
Task B.3
No Modify, or select new, concept or reassess risk
Yes
Reconsider or modify concept or reassess risk with more rigorous approach
Task A.4 Are nominal loadcases applicable?
No
No
Task B.4
Implement measures to reduce fire and blast risk Task A.5
Yes
Assess impact on safety critical elements
Task B.5 Risk Matrix
Low
Task A.6 Are performance criteria met?
Higher
Task B.6
Assess Load and Response for the Event Yes Yes Assessment Complete for the Facility
Task B.7 Are performance criteria met?
Note 1 : Interface to API 75 or relevant regulations or company standards. Performance criteria are established in line with the overall safety philosophy with due consideration of safe design practice and inherent safety - these can be re-evaluated at any stage provided the appropriate disciplines are involved in the decision process and the intent of the original criteria are maintained.
Yes
Assessment Complete for the Event
No
Modify, or select new, concept
No
Task B.8
Note 2: Interface to API 14 Series or alternative company practice Input required from other disciplines for hazard identification and selection of credible fire and blast events.
Figure 2: Risk Assessment Process
Are further risk reduction options available
OTC 17700
9
Generic respons e spectrum
Local conditions Risk level Ignition point Mass release Confinement Congestion
Impulse Energy Duration td
Required static resistance Design pressure Peak deflection Dynamic Reacti
Scale severity Impulse energy Local conditions
Idealise structure Mass M Stiffness Ke Resistance Rm Yield deflection Xe Allowable ductility mu
Natural period T= 2 pi sqrt (M/Ke)
Figure 3: Generic Response Spectra for a Hydrocarbon Blast (10)
Figure 4: Blast Pressure Distribut ion in a Closed Compartment
10
OTC 17700
Blast Pressure
Blast Pressure
Unstiffened
Stiffened (bent) Blast Pressure
Stiffened (welded) Figure 5: Blast Wall Profiles
Box section
Penetration plate
r o o D
Elevation Figure 6: Blast Wall Penetrations
OTC 17700
11
Girder Bent plate welded to bot. of girder
Bent plate welded to bot. of girder
Blast wall cap plate site welded to bent plate
Blast wall cap plate site welded to bent plate PFP if reqd.
(a)
(b) Blast Wall Top Support Details
Blast wall bot. plate welded to channel
PFP if reqd.
Blast wall bot. plate welded to deck plate
Channel site welded to deck plate
Channel
Deck plate
Deck plate
(b)
(a) Blast Wall Bottom Support Details Figure 7: Blast Wall Connections
12
OTC 17700
Not preferred
l l a w t s a l B
Preferred
Vent area restricted Equipment
l l a w t s a l B
Vent area maximized Equipment
Elevation
s l e s s e V
Vent area restricted by vessels
s l e s s e V
Vent area maximized
Plan
Vessels across module restricts venting and promotes turbulance
s l e s s e V
Vessels alongside module improves venting and promotes smooth flow
Vessels
Plan
Vessels
Venting through narrow side is restricted and promotes turbulance
Plan
Figure 8: Layout Options
Vessels
Venting through wide side allow smooth flow
OTC 17700
13
°C
Time = 240 sec. Location of thermally loaded region.
Time = 0 sec.
Time = 320 sec.
Time = 60 sec.
Time = 360 sec.
Time = 120 sec.
Time = 420 sec.
Time = 180 sec.
Time = 480 sec.
Figure 9: Progressive Collapse of a Support Structur e in a Fire Scenario
14
OTC 17700
T=1.000 ms
T=1.0592 ms
T=1.0382 ms
T=1.0722 ms
T=1.0502 ms
T=1.0812 ms
T=1.0562 ms
T=1.1612 ms
Figure 10: Progressive Collapse of a Deck Structure i n a Blast Scenario
