OPTIMIZATION OF LNG REGASIFICATION TERMINAL PROCESSES M. TECH. SEMINAR REPORT
By Shashwat Omar (Roll Number – 133020005) 133020005)
Under the guidance of Prof. Ravindra D. Gudi
DEPARTMENT OF CHEMICAL ENGINEERING INDIAN INSTITUTE OF TECHNOLOGY BOMBAY – 400 400 076 APRIL, 2014
1
Acceptance Certificate
Department of Chemical Engineering Indian Institute of Technology Bombay
The seminar report entitled “ Optimization of LNG Regasification Terminal Processes” submitted by Mr. Shashwat Omar (Roll No. 133020005) has been
corrected to my satisfaction and can be accepted for being evaluated.
Date: 25-04-2014
Signature
(Prof. Ravindra D. Gudi)
2
Acceptance Certificate
Department of Chemical Engineering Indian Institute of Technology Bombay
The seminar report entitled “ Optimization of LNG Regasification Terminal Processes” submitted by Mr. Shashwat Omar (Roll No. 133020005) has been
corrected to my satisfaction and can be accepted for being evaluated.
Date: 25-04-2014
Signature
(Prof. Ravindra D. Gudi)
2
ACKNOWLEDGEMENT First of all, I would like to thank the almighty who gave me the strength to complete this report successfully. I would like to convey my heartfelt thanks to the Indian Institute of Technology Bombay for providing me the opportunity to work on this project and for allowing me the access to copious amount of invaluable literature. I would like to convey my sincere gratitude to my guide Prof. Ravindra D. Gudi, for his invaluable guidance, supervision and encouragement throughout the course of my study. I thank all my friends and colleagues for their help which contributed to the completion of this report. At last but not the least, I want to thank my parents for being with me and giving me inspiration when I needed it the most.
Place: Place: Mumbai
Shashwat Shashwat Omar
Date: 25-04-2014
3
TABLE OF CONTENTS
1
Introduction...............................................................................................5
2
Literature Survey......................................................................................9
3
Problem Formulation and Work Done so Far......................................23
4
List of Figures...........................................................................................28
5
References.................................................................................................29
4
CHAPTER 1 INTRODUCTION From the tsunami in Japan and Thailand to polar vortex in USA and Canada, the effects of global warming are worsening day by day. The main cause of global warming is greenhouse gases such as water vapour, Carbon dioxide (CO 2) and methane (CH4 ). The concentration of Carbon dioxide and methane in surrounding air has increased 148 % and 36 % respectively in the past 250 years due to industrial revolution. The industrial revolution is caused by big power looms, railways, factories and new technologies. Fossil Fuels like petroleum and Coal fulfilled are the major part of the energy requirement of this industrial revolution. But in doing so, they have increased the concentration of greenhouse gases in the surroundings. With the increasing population, the energy requirements are also increasing and knowing the harmful effects of fossil fuels, we have to go for clean fuels. Natural Gas is one of the clean fuels. It can be found in deep underground rock formation or with and proximity to petroleum. It consists of methane (80-90 %) some higher alkanes and impurities like carbon dioxide, hydrogen sulphide and nitrogen.
Figure:1 Emission of combustion by-products from fossil fuels (http://www.cleanandaffordable.ca)
5
Figure 1 shows that, after burning, it generates only small amount of carbon dioxide, nitrogen oxide and almost zero emission of sulphur dioxide. So, the use of natural gas is getting popular all over the world as a clean environment friendly energy source. In figure 2, we can see that the countries with most natural gas reserves are Russia, Qatar, Iran, USA, Venezuela and Australia. The most populated countries in the world like India, Brazil, China, Indonesia and Pakistan do not have enough natural gas production to fulfil their energy needs. So, it is necessary for these countries to import natural gas. But, gas can be transported only via pipeline and it is very costly to lay pipeline from one country to another and various legal and natural constraints are also involved in it.
Figure: 2 Country wise Natural Gas Reserves in trillion cubic meters (http://thomaspmbarnett.com 2012) LNG (Liquified Natural Gas) is the solution for this problem. LNG is natural gas liquefied at atmospheric pressure when cooled to -162 oC. One cubic metre of Natural Gas is equal to the 600 cubic metre of LNG. So, for ease of transportation, large volume of Natural Gas can be converted into small volume of LNG. LNG is colourless, odourless, non-toxic and non-corrosive. LNG is relatively costly to 6
produce and it should be stored in a cryogenic tank because of its very low temperature. The process of natural gas exploration to LNG regasification is called LNG supply chain. It is divided into four parts – natural gas production, Liquefaction, LNG transportation and LNG storage and regasification. First of all natural gas is produced from gas field wells, and then it is transported to liquefaction plants. These plants are built at marine terminals so the LNG can be loaded onto specially made tankers for transporting it overseas. After delivering of LNG to importing terminals, the LNG is stored into storage tanks, regasified and then sent into pipeline systems for delivery to downstream customers. Next, we look at all the four processes one by one -
Figure: 3 LNG Supply Chain ( Deybach 2012)
Production: - Natural gas is extracted from oil wells or deep rock formations. This gas has impurities like carbon dioxide, nitrogen and hydrogen sulphide which should be removed before liquefaction otherwise they will freeze and damage the equipment in which phase change is taking place besides it is necessary to meet the demand of importer.
Liquefaction: - In this process, natural gas is converted to LNG via cooling it down to -162 oC at atmospheric pressure. The volume of natural gas shrinks 600 times in the phase change process that means the energy content of 1 7
cubic metre of LNG at -162 oC is same as 1 cubic metre of natural gas atmospheric pressure and ambient temperature. So, it becomes easy to transport natural gas from one place to another because of less volume.
LNG Transportation: - LNG is transported in specially designed ships with double hulls so as to minimize the risk of potential leakage. There are three types of LNG carriers as per the storage tank design – membrane tank, IHI prismatic and spherical tanks. Some Cargos use Boil -off Gas (BOG) generated from the tanks as a fuel to run ships. A cargo can take 2 to 8 days to reach its destination. So, it is necessary to calculate the BOG volume consumed by cargo to determine actual LNG volume that can be unloaded.
Storage and Regasification:- At the importing terminal, LNG is unloaded from carrier and stored in big cryogenic storage tanks at atmospheric pressure and 162 oC. The quantity (100,000 m 3 to 160,000 m 3) and number of storage tanks in a terminal depend on the frequency of cargos visiting to terminal and the demand of downstream customers. The regasification process consists of gradually converting LNG to natural gas at 0 oC and 80-100 bar pressure. The mechanisms to convert LNG into natural gas depend on the geographical nature of terminal and composition of the LNG imported. Sometimes, treatment of natural gas can also be done to meet the caloric value demand by changing the composition of propane, butane and nitrogen components before sending it to customers.
LNG production, transportation and converting it to natural gas are a costly and energy-intensive process. This thesis will focus on optimization of LNG storage and regasification operation. However, optimization of the complete LNG supply chain is beyond the scope of this thesis. For upcoming and existing LNG regasification plants, integration of new technologies with blend of optimization will help to minimize costs and maximize profit margins and this thesis will work in this direction.
8
CHAPTER 2 Literature Survey This chapter focuses on various optimization works that has been done previously in the field of LNG storage and regasification. A LNG carrier ship takes 2-9 days to reach its destination based on the distance of importing terminal to exporting terminal. LNG is stored inside specially fabricated and insulated cryogenic tanks. Because, LNG is at its boiling point inside the tanks and the insulation is not 100 % perfect, some heat ingresses into tank and LNG is converted into Boil-off Gas (BOG) which is nothing but natural gas. This phenomenon is called weathering. M iana et al. (2010) devised various calculation models for prediction of LNG weathering during ship transportation. Accurate prediction of LNG composition before unloading helps to determine density of LNG precisely and consequently we can prohibit rollover phenomenon and stratification. The other benefit is as per the quality specifications for transport, distribution and utilization of natural gas, we can carry out blending or mixing operations.
Figure: 4 – Dahej LNG regasification terminal of Petronet LNG Ltd. ( Ahuja 2005) Figure 4 above shows a typical LNG regasification terminal looks like. A LNG regasification terminal consists of various heat, mass and fluid transfer equipment and 9
operations. We can divide the whole plant into individual equipment and operations and then will optimize them separately. First of the lot comes unloading and recirculation operation. For most of the LNG regasification terminals, storage tanks are located 4-5 km. far from the LNG unloading platform. The LNG transporting pipelines must be insulated to keep LNG at liqui d state. A running LNG regasification plant is operated in two modes: first, when unloading is happening from a LNG carrier ship and second, when there is no ship at the LNG unloading platform. In the second situation, some LNG is pumped from storage tanks and circulated to unloading transportation lines and then comes back to tanks. This is necessary to keep transportation lines cool before LNG unloading. In the LNG regasification terminals, a constant rate of recirculation is always maintained. It determines the operating cost of the LNG pump which is needed to maintain the flow rate into transportation lines. However, depending on the LNG send-out flow rate, the handling method and costs of BOG to flow into the storage tanks are varied. So, the recirculation flow rate should be decided after taking BOG handling cost into account. Park et al. (2010) devised an optimal recirculation method to reduce total operating costs by adjusting the recirculation flow rate dependent on demand of downstream customers. During unloading process, flashed BOG from the pipeline flows into the storage tank. Due to increasing amount of BOG, pressure in the storage tank rises. BOG compressor is used to maintain pressure in storage tank. BOG from compressor meets the LNG transported to the recondenser to meet demands. This BOG inflow rate determines the operating cost of the BOG compressor to handle the BOG in the storage tank. Therefore, recirculation flow rate creates a trade-off between the cost of BOG compressor and operating cost of the recirculation flow pump. If the necessary recirculation flow rate is high, the pump operating cost increases, but the compressor operating cost decreases due to the reduction of BOG inflow and vice-versa. After unloading, LNG is stored in cryogenic storage tanks. These are of different kinds based on the natural climate and necessity. The most popular one is full containment tank. LNG is characterized by its density. Less dense LNG tends to have lower chain hydrocarbon content and less calorific value and denser LNG tends to have higher chain hydrocarbon content and high calorific value. During the unloading operation, terminal operators have to face three different cases. In the first case, incoming LNG is lighter than the LNG stored in the tank to be filled. A complete mix 10
of two LNG quantities is ensured by tank bottom filling operation with a limited BOG generation and stratification risk is eliminated which can potentially lead to a rollover event. For the second case, incoming LNG is heavier than the stored in storage tank. By tank top filling operation, stratification and risk of rollover is avoided but it usually results in excessive BOG production and tank pressure increases as shown in figure 5.
Figure: 5 – Total Boil-off gas flowrate and pressure evolutions during tank top filling in a 120,000 m 3 LNG tank ( Zellouf and Portannier 2011 ) The last situation is tank bottom filling with a heavy LNG under a light stored LNG heel producing less BOG but leading to stratification which will need to be managed in order to avoid rollover. Figure 6 shows how a stratification profile looks like inside a tank. Rollover phenomenon is the overturning of two LNG stratified layers.
Figure: 6 – Characterization of LNG stratification ( Zellouf and Portannier 2011 11
Rollover occurs when the two LNG layers densities equalizes mainly further to the LNG expansion in the lower layer due to the heat ingress. It is accompanied by a sudden rise in BOG which is due to the sudden release of the energy accumulated in the lower density layer through time. Zellouf and Portanni er (2011) showed that by optimizing the operating pressure in
the storage tank, the total BOG rate generated during filling can be reduced by about half. This can be done in three easy steps. First, pre-cooling of the tank heel before unloading occurs by lowering the operating pressure. Secondly, before unloading, the operating pressure is increased above the atmospheric pressure to sub-cool the LNG and this pressure is maintained throughout the unloading process. Third, once tank is filled, the pressure is then lowered progressively to the atmospheric value.
Figure: 7 – Pressure optimization of BOG rate during tank filling of heavy LNG under/over light heel LNG at a filling rate of 10,000 m 3 /h, obtained using the GDF
SUEZ “LNG MASTER ®” software ( Zellouf and Portannier 2011) Figure 7 shows that by operating pressure optimization, the total BOG rate generated during filling can be decreased by about 50 %. This highlights the advantages gained from this procedure, not only in terms of the costs saving by reducing compressor input, but also in terms of avoiding the use of safety equipments like flares. Deshpan de et al. (2011) presented a lumped parameter model in order to predict time
for the rollover phenomenon and to investigate its sensitivity to heat and mass transfer coefficients. The originality about this work is its ability to estimate heat and mass
12
transfer coefficients from the real time data using an inverse methodology. The real time data of Level, Temperature and density from storage tanks is assimilated in order to estimate heat and mass transfer coefficients from the densities of the stratified layers. Then, the optimized heat and mass transfer coefficients are used in prediction of time of rollover as shown in figure 8.
Figure: 8 – procedure used to infer HTC and MTC from the real time LTD data profile ( Deshpande et al. 2011 ) BOG generation is uncalled for but inevitable in a LNG regasification terminal. BOG generation happens in LNG storage tanks and all LNG transporting lines due to some heat ingress. Querol et al. (2010) proposed number of methods to handle BOG so as to
minimize
the
operating
cost.
Figure: 9 – Diagram of the unloading procedure of a LNG regasification terminal (Querol et al. 2010) 13
As seen in figure 9, during the unloading operation, heat is absorbed by the carrier (QC), tank (Q T), unloading arms (Q A) and LNG transporting lines (Q L). For a normal LNG regasification terminal with BOG generation at its worst case and normal case, these methods are discussed to handle BOG –
Flaring: This is used when there is no other option to handle BOG and safety of equipment and plant is necessary. No additional investment is needed for it because it is installed compulsorily in each terminal.
Recondenser: All the regasification terminals have this equipment to handle the total amount of generated BOG in the normal case.
BOG compressor: This equipment is installed to compress BOG as per the requirement of recondenser and LNG carrier pressure.
Most of the LNG terminals buy the electricity supplied by the state or national grid. In the plant, LNG is converted into natural gas which is a very good fuel and it can be used to generate electricity inside the terminal. For generation of electricity, a cogeneration power plant should be set up. In this gas fired power plant, gas will enter inside the equipment at very high pressure and will be burned. The burned gases will generate heat and pressure that will be used to rotate a Gas turbine. By rotation of this Gas turbine, electricity will be generated. The hot gases after burning can be used for producing hot water from cold water in a heat exchanger. This cogeneration plant are very effective in terms of energy saving and cost minimization. In addition to that, Variable costs occurring due to the submerged combustion vaporiser could be reduced if submerged combustion and heat recovered from cogeneration plant are integrated. This equipment succeeds to obtain nearly 100 % efficiency in reducing electricity bill of LNG terminal and its energy dependence on outer sources. In the LNG storage tank the liquid temperature is approximately -162 oC and the pressure is nearly above the atmospheric pressure. Part of LNG stored in tanks is continuously evaporated into BOG (Boil-Off Gas) because of heat transfer from surroundings. This BOG should be removed by compressors because it is necessary to maintain tank pressure within a safe range. However, excessive operation of compressors spends excess energy, and even sometimes generates excess BOG. So, proper handling of BOG is essential for the safe and efficient LNG terminal operation. Shi n et al. (2008) proposed an empirical BOR (boil off rate) model a nd a simplified 14
dynamic tank model. Proposed algorithm generates an optimal operation schedule for BOG compressors which minimizes the power consumption given potential failure of one of the operating compressors. They used following empirical equation in order to estimate the rate of BOG generation in a LNG storage tank
(2-1)
Where, CR = rollover coefficient (≥ 1) BS = boil off rate on specification (h -1)
ρL = LNG density (kg/m 3) VL = LNG volume (m3) K 1 = correction factor for the offset of the tank pressure from the LNG vapour pressure K 2 = correction factor for the LNG temperature K 3 = correction factor for the ambient temperature A range of correction of correction factors was definedbased on real situations and also defined an equation for calculating LNG vapour pressure from a Antoine model equation. Using all these equations, the BOG generation rate can be calculated from the tank pressure. Under the assumption that, LNG volume is very large so the temperature and volume rate of change will be very slow and thus can be taken as constant. So, for a target steady state tank pressure P s, the target compressor load F 0 can be calculated using the MILP optimization problem solving. A simple dynamic model based on the ideal gas law is used for the prediction of the tank pressure change which can be effectively applied to safety analysis for preparing against the potential failure of one of the running compressors. Based on a case study, the performance evaluation indicated that the energy consumption could be reduced by half as compared with the conventional method by operating the tanks at a higher pressure within the safe range. 15
As indicated in the paper, the recondenser is the heart of a LNG regasification terminal. Like, for a body when heart stops beating, it becomes dead. Same is true for a LNG plant. Recondenser is the most vital equipment of LNG terminal. It converts BOG to LNG and gives positive head to High pressure pumps. In storage tanks, LNG is at 1 bar pressure and -162 0C. It is send to recondenser at 8 bar pressure, so it becomes subcooled. This energy difference between subcooled and boling state is used to condense BOG into LNG inside recondenser. BOG recovery is an essential thing and also from that perspective, this equipment is pretty valuable. Yajun et al. (2012) proposed an optimized control theory to control pressure and level inside the
recondenser and optimum flow rate of LNG to condense BOG. In the figure 10, it is showed that in some existing BOG terminals, pressure is controlled via recondenser pressure controller PIC-1 and to prevent High Pressure pumps from cavitations by regulating the pressure control valves PCV-1 and PCV-2 also used for LNG bypassing. Condensing LNG flow rate is maintained by the control valve FCV. The Condensing LNG flow rate is governed by the following equation:
(2-2)
Flow rate is fixed by the ratio calculation module in FX1. Where, Q LNG (m3/h) is
(Nm /h) is standard volume flow of BOG; (MPa) is pressure of HP suction header: is 0.58, a parameter referring volumetric flow rate of condensing LNG,
3
to LNG composition. Value of Q LNG will be transformed into the setting point of FIC-1. The normal high set point of recondenser liquid level is 60 %. Once liquid level exceeds 60 %, LIC-2 will sendout signal to LCV to send high pressure make up gas inside recondenser to lower the liquid level. This move will increase the pressure of recondenser and can also prevent sufficient amount of LNG not to enter recondenser. If the level still keeps on rising, there is a interlocked level controller LT-2 which will automatically give HH-SD a signal to shut down control valve XV-1. When the recondenser level goes below the set point, LIC-1 could start override control to reduce BOG flow rate entering into recondenser.
16
Figure: 10 – Existing control system to BOG recondenser (Yajun et al. 2012) The demand of natural gas from downstream customers fluctuates with different time and different seasons. So, frequent liquid level instability will be there. The disadvantage with this control strategy is there is lag in response of PIC-1 to PCV-1 and PCV-2. System takes long time in returning to the normal position with fluctuation of demand in downstream. This control is a very high cost operation. Make-up gas is inserted in recondenser frequently. Pressure in LNG storage tanks can be increased. If LNG output is very low, whole BOG cannot be condensed. After analysing the existing system, the main reason causing frequent changes in recondenser level and pressure control signals from the High pressure pumps suction header pressure. So, it seems more logical that signals should be sourced from the measured values of liquid level and pressure directly. So, instead of proposing a new structure of recondenser, control theory is changed instead. In the figure 11, For flow rate control, LNG/BOG ratio control mechanism and FX-1 are removed and after optimization, flow rate and discharge pressure of recondenser are not controlled anymore. 17
Figure: 11 – Optimized control system to BOG recondenser (Yajun et al. 2012) Pressure will be controlled by the new valve PIC-3. If pressure inside the recondenser keeps on increasing PC-1 sends signal to PCV-1 to open and decrease the pressure. When PCV-1 is fully open and pressure inside recondenser still keeps on increasing, then pressure controller PIC-2 gives command to PCV-3 to vent BOG. For the condition of decreasing pressure inside recondenser, PC-1 will send signal to PCV-1 to close and if this move is not sufficient then, PCV-4 will operate PCV-2 and make up gas will enter recondenser and maintain the pressure. For liquid level control, LCV-1 and LCV-2 will come into use now. When the required LNG output is more than summation of BOG condensate and condensing LNG stream, the liquid level will drop down. The liquid controller LIC-1 will increase the opening of valve LCV-1 and LCV-2 to maintain a desired liquid level. Conversely, in the case of liquid level rise, the opening of LCV-1 and LCV-2 will be decreased. The advantage of this control is LNG storage tank pressure is held constant. Recondenser comes into normal position sooner. It is more energy efficient because makeup is not entered into system frequently. New system can operate at a flexible operating pressure of recondenser. BOG can be totally liquefied even for the low flow rate of LNG.
18
High pressure Pumps are used to pressurize LNG that comes out of recondenser to 100-100 bar pressure. This high pressure is required because after LNG is converted into natural gas, it must be transported over a long distance. These pumps can be 4-5 in numbers based on the capacity of the plant. These pumps take considerable amount of energy to run. A single stage and multi-stage cryogenic pump is shown in figure 12. Presently, most of the LNG terminals have high pressure pumps running on fixed speed motors and produce a single flow versus head curve at that speed. These pumps are controlled by throttling the discharge pressure with a control valve or bypassing a portion of the flow to a secondary stream. Pumps running on a fixed speed have a single Best Efficiency Point (BEP) at which the pump is designed to run at the rated point. Pump flows that are throttled by reduction of the discharge pressure run off of the BEP and therefore the efficiency is reduced. The efficiency and range of operation for cryogenic pumps can be improved by used of variation of motor speed.
Figure: 12 – Multistage and Single Stage High Pressure Cryogenic Pumps (Lovelady et al. 2008) L ovelady et al. (2008) showed that by adjusting the speed user can accurately control
the operating characteristics of the pump over a greater range with better overall efficiency. Efficiency of pump operation at off design flow points can be improved by varying the pump speed to a point on the pump hydraulic curve which matches the 19
BEP to the desired flow rate. This paper discussed how the variable speed operation of cryogenic pumps resulted in improved operating costs with more effective control of the output flow and head. In addition, the paper also discussed other design, control and economic benefits of utilizing variable speed motors at LNG terminals. The vaporizers are used in LNG regasification terminal to convert LNG to natural gas. High pressure LNG at about -160 0C comes into the vaporizer and it is vaporized until a suitable temperature is achieved. This temperature is generally close to 0 oC and it can vary according to the demand of customers. There are several types of vaporizers that can be used according to the climate, cost and easily available vaporization fluid. Some of them are Open Rack vaporizer (ORV), Fired heater with Shell and Tube vaporizer, Submerged Combustion Vaporizer (SCV), Intermediate Fluid Vaporizer, Ambient Air Vaporizer etc. It is the most important process to select the right vaporizer system in designing a LNG terminal as the regasifying costs contribute major portion of an LNG terminal operation. Low operating cost with high reliability of the regasifying system is a key parameter for a successful operation of an LNG receiving terminal. I n-Soo Chun (2008) reviewed ORV performance and optimum heat recovery temperature under harsh seawater temperature. The practical optimization of vaporization with the unfavourable seawater conditions in winter is discussed. An economic comparison between conventional vaporization and optimized vaporization is also discussed.
As shown in figure 13, an ORV is a
vaporizer in which LNG flows inside a heat-transfer tube and exchanges heat with seawater that flows outside the heat-transfer tube to gasify the LNG. The LNG flows in form an inlet nozzle near the bottom and passes through an inlet manifold and a header pipe to be sent to a set of panels which each of them consisting of a curtainlike array of heat-transfer tubes. LNG exchanges heat with seawater that flows downward in a flim-like manner utside the heat-transfer tubes as it flows upward inside the heat-transfer tubes. This operation results in normal temperature gas to be sent out from an outlet nozzle via an outlet header and manifold pipe.
20
Figure: 13 – Schematic of an Open Rack Vaporizer (Egashira 2013) The paper discussed that for achieving low cost vaporization and to minimize impact on the environment, ORVs and SCVs should be combined as a backup. The seawater temperature drops to 0 oC after it is used for LNG vaporization, which is not recommended for operational purpose especially during the winter. The following parameters were investigated for the vaporization opti mization:
Performance of ORV with the lower seawater temperature
Optimum seawater operating temperature
Economic combination of ORV and SCV
Economic justification of seawater heating
Overall economic assessment of vaporization facility
In figure 14, it is shown that seawater requirement abruptly increases as the temperature of inlet seawater decreases, while seawater requirement is constant after a certain point even though its temperature increases. As the temperature of seawater decreases below design temperature, the SCV operation and seawater heating process need to be operated. This results an increase in operating cost. Fuel gas requirements
21
for the operation of SCVs increase linearly during the winter as the recovered heat from seawater decreases.
Figure: 14 – Economic Analysis of different Seawater Temperature (In-Soo Chun 2008) The results showed that, for selection of optimum vaporization method, design constraints with unfavourable seawater temperature have been successfully solved with heating up seawater. This results in heat recovery increment from cold seawater. Unless otherwise, the seawater cannot be used when its temperature is lower than the design temperature.
22
CHAPTER 3 Problem Formulation and Work Done so Far In this thesis, we consider the problem of optimizing LNG regasification terminal processes. The scope of the optimization that we seek spans from the upstream operations of unloading LNG cargos at the jetty to the downstream system that sends out natural gas to customers. Given a schedule of LNG carriers to the terminal and demand of downstream customers, a typical LNG regasification terminal has unloading arms, storage tanks, Low pressure pumps, BOG compressor, Recondenser, High pressure pumps and vaporizers as its main unit operations. The key decisions for optimization are related to recirculation operation, BOG maintenance and handling, Recondensing BOG, vaporization of LNG. After going through, several research papers in field of optimization of LNG process, it is seen that no work has been done to optimize the whole plant in an integrated manner. The works are done in individual sub-systems such as optimization of storage tank BOG handling, optimization of recirculation operation and optimization of recondenser operation. This work will attempt an integrated look at the entire plant, based on mass & energy balances, and will also consider minimizing external work that may be necessary to maintain phase of the LNG at appropriate points. As such therefore, this work proposes to optimize the entire LNG regasification plant by formulating an optimization problem having objective function as minimization of power consumption and maximization of profit.
In the first phase of the work, we look at optimizing the wait time of LNG cargos given a schedule of their arrival to the terminal. According to demand, constant discharge of LNG from each storage tank is assumed. The terminal consists of two jetties and four LNG storage tanks. A cargo can be assigned to any one of the jetty based on the availability of space there and total storage tank volume. Once a ssigned to a jetty, a cargo will unload its contents to 4 tanks assuming constant discharge flow rate. There can be three cases for this operation. First, both the jetties have cargos. Second, one jetty has cargo and one does not. Third, both the jetties do not have any cargo at a time. The total unloading time of a cargo will depend on the amount of LNG in it, constant unloading rate from the cargo and total volume of the storage tanks. It is essential to minimize unloading time of a LNG cargo because the profit of
23
operating a LNG terminal is based on the demand of Natural Gas from customers and demand cannot be fulfilled without supply. Secondly, there is a maximum time limit in which a cargo has to be unloaded otherwise there will be a huge penalty on terminal for every minute of delay.
There are two approaches used in the literature for scheduling problems formulation: continuous and discrete. There are uniform slots for a job in continuous approach and variable length slots for discrete time approach. For the LNG cargo scheduling formulation, we will adopt a continuous time approach. The benefit of using this approach is, it has fewer time slots as compared to the discrete time formulation and less variables. Having said that, nonlinearities in the system is also introduced and we need to accommodate that in the solution procedure. The following figure shows the structure of jetty and storage tanks.
Figure 15: Block diagram of LNG unloading operation
I ndex notations & constraints:
i = jetty (1,2) j= cargos (1,.... , N) k = Number of tanks (1,2,3,4) m = Number of slots (1,2,3...........,M)
24
, If jth cargo is assigned to ith jetty in its mth slot. At a time, one jetty can have only one cargo. So,
(3-1)
The same constraint can be written for ship. At a time, a ship can be boarded to only one of the jetties. So,
(3-2)
Now, after completion of unloading of a cargo another cargo should be assigned to the jetty. For that,
=Start time of the unloading of cargo at ith jetty in its mth slot. End time of the unloading of cargo at ith jetty in its mth slot Processing time of the jth cargo at ith jetty in its mth slot.
(3-3)
Then,
Where, =Processing time in the unloading
(3-4) (3-5)
All of the tanks can be filled from both of the jetties or from the single one which will be based on the availability of cargos. Given there is no tank at its highest level, The feed to tank coming from each jetty will be divided into four parts otherwise, that tank will not be filled.
Volume of tanks to be taken as Volume of cargos to be taken as
25
Let the fraction of feed stream divided to each tank is
and
(3-6) (3-7)
Now mass balance equation for a single tank can be written as,
(3-8) Additional Constraints:
Let where, k≠l and
(3-9) (3-10) (3-11)
(3-12)
(3-13)
Objective Function: minimization of standing time of the LNG carrier which is
(3-14)
26
Where, Tss(j) = standing time of jth cargo Tar(j) = arrival time of jth cargo It is essential to maintain the same level in each tank during the operation of terminal because in the case of outlet pump failure of one or two tanks, we can send LNG out from the rest of the tanks. Similarly, If there is a problem in upstream of the tank, we can take LNG into rest of the tanks. So for that the objective function should be
Minimize
+
27
LIST OF FIGURES Figure No.
Title
Page No.
1
Emission of combustion by-products from fossil fuels
5
2
Country wise Natural Gas Reserves in trillion cubic meters
6
3
LNG Supply Chain
7
4
Dahej LNG regasification terminal of Petronet LNG Ltd.
9
5
Total Boil-off gas flowrate and pressure evolutions during tank
11
top filling in a 120,000 m 3 LNG tank 6
Characterization of LNG stratification
11
7
Pressure optimization of BOG rate during tank filling of heavy
12
3
LNG under/over light heel LNG at a filling rate of 10,000 m /h,
obtained using the GDF SUEZ “LNG MASTER ®” software 8
9
procedure used to infer HTC and MTC from the real time LTD data profile
13
Diagram of the unloading procedure of a LNG regasification
13
terminal 10
Existing control system to BOG recondenser
17
11
Optimized control system to BOG recondenser
18
12
Multistage and Single Stage High Pressure Cryogenic Pumps
19
13
Schematic of an Open Rack Vaporizer
21
14
Economic Analysis of different Seawater Temperature
22
15
Block diagram of LNG unloading operation
24
28
REFERENCES: 1. Ahuja M.M. (2005). A general presentation on Dahej LNG terminal, (personal communication). 2. Aspelund, a., Gundersen, T., Myklebust, J., Nowak, M. P., & Tomasgard, a. (2010). An optimization-simulation model for a simple LNG process. Computers & Chemical Engineering , 34(10), 1606 – 1617. 3. Chun, I., Lee, S., & Meeting, A. S. (2008). Optimized Vaporization Conditions Process with Unfavorable Design, 61 – 75. 4. Chun, I. S., & Kim, C. H. (2006). Practical optimization of LNG terminal process. In AIChE Annual Meeting, Conference Proceedings . 5. Deybach F. (2012). Background information on Liquefied Natural Gas, cfafg, GDF Suez LNG. 6. Deshpande, K. B., Zimmerman, W. B., Tennant, M. T., Webster, M. B., & Lukaszewski, M. W. (2011). Optimization methods for the real-time inverse problem posed by modelling of liquefied natural gas storage. Chemical Engineering Journal , 170(1), 44 – 52. 7. Egashira, S. (2013). LNG Vaporizer for LNG Re-gasification Terminal, (32), 64 – 69. 8. http://www.cleanandaffordable.ca/#clean 9. http://thomaspmbarnett.com/globlogization/2012/7/20/chart-of-the-dayeconomists-listing-of-top-15-natural-gas-re.html 10. Kim, H., Kim, I. H., & Yoon, E. S. (2010). Multi-objective Design of Calorific Value Adjustment Process using Process Simulators. Industrial & Engineering Chemistry Research, 49(6), 2841 – 2848. 11. Li, Y., Chen, X., & Chein, M.-H. (2012). Flexible and cost-effective optimization of BOG (boil-off gas) recondensation process at LNG receiving terminals. Chemical Engineering Research and Design, 90(10), 1500 – 1505. 12. Lim, G., & Park, L. (2005). LNG Terminal Operator ’ s Design Feedbacks and Technical Challenges. 13. Lovelady, J., Wong, J., & Minton, B. (2008). LNG Pump applications with variable speed motor controls at LNG regasification terminals. In ACS National Meeting Book of Abstracts .
29
