OPTIMIZATION OF A DUAL REFRIGERATION SYSTEM NATURAL GAS LIQUID PLANT VIA EMPIRICAL EXPERIMENTAL METHOD

BACKGROUND OF THE INVENTION

1. Priority Application

[0001] This application is related to and claims priority and benefit of U. S. Provisional Patent Application Serial No. 60/793,111 filed April 19, 2006, which is incorporated herein by reference in its entirety.

2. Field of the Invention

[0002] This invention generally relates to the field of optimization of control variables to maximize production of Natural Gas Liquids ("NGL") in a gas plant while minimizing the refrigeration system power usage.

3. Description of the Related Art

[0003] Gas plants produce fuel gas, Natural Gas Liquids (NGL) and other solid components such as sulfur. Such plants typically include distillation columns, heat exchangers, and refrigeration systems. The NGL product must meet certain specifications in order to be a saleable product, but variation within these boundaries is acceptable. Early efforts to improve NGL quality have been directed toward maximizing the amount of refrigeration used to achieve better recovery of heavier components. As energy costs have increased, this approach is no longer economical.

[0004] U.S. Patent No. 6,332,336, issued to Mirsky, teaches a method and apparatus for maximizing the productivity of a natural gas liquids production plant using a method that specifically manipulates a turboexpander that drives a recompressor, with the objective of maximizing NGL production.

[0005] U.S. Patent No. 5,488,561, issued to Berkowitz, generally described a multivariable process control method and apparatus, which may be used in a fractionation process involving natural gas liquids or other process. It includes an algorithm which controls multiple variables during the process based upon control equations utilizing information from rigorous process simulations and actual plant performance. Historical data is used for calibration purposes through the routines.

[0006] U.S. Patent No. 5,791,160, issued to Mandler, discusses a dynamic method and apparatus for regulatory control of production and temperature in a mixed refrigerant liquefied natural gas facility utilizing a generic system modeling program and an optimization computer program which may use simulation techniques.

[0007] U.S. Patent No. 4,616,308, Morshedi et al., teaches a method of dynamically controlling a process having a plurality of independently controlled, manipulated variables and at least one controlled variable that is dependent upon the manipulated variables. This method applies generally to process control.

[0008] U.S. Patent Nos. 6,116,050 and 5,992,175, the Yao patents, discuss NGL recovery processes utilizing various controllers to control the recovery machinery. The patent teaches physically manipulating the temperature profile within the column to obtain desired separation results.

[0009] U.S. Patent No. 4,164,452, issued to Funk, discusses a pressure responsive fractionation control system and method for utilizing various controllers and simulation techniques. This is applies to a propane recovery system that includes a de-methanizer tower followed by a de-ethanizer tower.

[0010] It would be advantageous to develop a new method and apparatus that provides improvement in the recovery of the valuable NGL products while minimizing energy requirements. It would be advantageous to allow for the optimization of the process variables within allowable quality variations and equipment constraints while minimizing the electricity or energy usage.

SUMMARY OF THE INVENTION

[0011] The current invention is an empirical optimization method based on statistical modeling relating NGL plant process variables with the refrigeration system's electricity usage. The method identifies the key process control variables in an NGL plant to be optimized. This method is applicable to an NGL plant that uses dual refrigeration systems. It describes methods to calculate the key optimal targets for the process control settings. These key optimal targets can be fed to a multivariable controller algorithm that controls the NGL plants, or can be implemented directly by the NGL plant operators inputting the calculated optimal targets in the NGL plant's distributed control system (DCS).

[0012] The process in consideration is an NGL plant that uses dual refrigeration systems. The optimization is done by developing empirical statistical models relating NGL plant process variables with the compressors' electricity load. These models are then used to calculate optimum values for the process control settings. These optimal targets can be fed to a multivariable controller algorithm that controls the NGL plants, or can be implemented directly by the NGL plant operators inputting set-points in distributed control systems (DCS).

[0013] The method herein describes a strategy to optimize the dual refrigeration system compressors to minimize the electricity consumption while maintaining the NGL recovery at a desired level. This is done by modeling the plant using experimental design. The method is also applicable to systems where only one refrigeration system is available. In this case, the benefit is less because the number of degrees of freedom is reduced.

[0014] The methodology described herein optimizes the control of load distribution in the NGL chillers to minimize compressor power usage versus NGL recovery.

[0015] The present invention advantageously includes a method to improve the efficiency of NGL product recovery by utilizing empirical mathematical representations of the NGL plant, capturing the behavior of the different process equipments.

[0016] The method herein describes optimizing the usage of dual refrigeration systems. The method manages the load distribution in the chill-down equipments to allow for increasing feed processing capacity, best recovery conditions, and optimizing the electrical energy. This method would allow the optimization of a dual refrigeration system to minimize the electricity consumption while maintaining the NGL recovery at desired level. This is done empirically by carrying out plant tests in operating NGL plants and is based on design of experiment.

[0017] A method is described for controlling the load distribution in the NGL plant chilldown equipments which will result in an optimal temperature profile to achieve the desired NGL recovery level with minimal power required by the refrigeration system.

BRIEF DESCRIPTION OF THE DRAWINGS

[0018] So that the manner in which the features, advantages and objects of the invention, as well as others which will become apparent, may be understood in more detail, more particular description of the invention briefly summarized above may be had by reference to the embodiment thereof which is illustrated in the appended drawings, which form a part of this specification. It is to be noted, however, that the drawings illustrate only a preferred embodiment of the invention and is therefore not to be considered limiting of the invention's scope as it may admit to other equally effective embodiments.

[0019] Fig. 1 shows one preferred embodiment of the gas plant of the invention. The refrigeration systems shown separately as part of Fig. 1 integrate into the entire plant as discussed in the following description.

[0020] Fig. 2 shows one preferred statistical optimization embodiment for implementing the optimization scheme discussed in the following description. The Statistical Optimizer 1 entitled RT-MSPC resides in a computer where optimization programs and process models are used to calculate. These targets are then communicated electronically to the NGL plant Decentralized Control System 3, also called "DCS", or to plant Multivariable Control System 2, also called "MPC", which in turn sends control signals to the NGL plant's actuators.

[0021] Fig. 3 shows one preferred statistical optimizer embodiment with steps for either online or offline use of the NGL Optimizer.

[0022] Fig 4 and Fig 5 show the process variables trends with normalized electricity cost and suction pressures been manipulated during the design of the experiment.

DETAILED DESCRIPTION

[0023] The invention includes a method for optimizing the production of NGL product stream 303 from a Natural Gas Liquids ("NGL") plant. The invention provides a method for optimizing the utilization of the available refrigeration capacity. The method honors process equipment and product quality constraints such as the NGL product specification, an upper limit of the percent of ethane and propane (mole percent) in the residue gases, a maximum pressure drop across the top section of the demethanizer column and a predetermined operating range for the refrigerant compressors suction pressures. Natural gas feed stream 9 is fed to first chilling unit 12 to produce chilled rich gas stream 37 and chilled liquid stream 36. Pressure and flow monitoring devices are useful for determining or controlling the pressure and flow of the feed stream. Residue gas stream 31, in combination with other residue from the demethanizer 200, is collected as sales gas. Pressure of stream 31 is measured and monitored. Valve 131 on stream 31 is used to control the unit pressure. Flow of stream 31 is measured, typically after valve 131. Chilled rich gas stream 37 and chilled liquid stream 36 have different compositions as a result of separation of natural gas feed stream 9. Natural gas feed stream 9 contains sweet gas that has been submitted to a sweetening process to remove hydrogen sulfide and carbon dioxide. Natural gas stream 9 is dehydrated in molecular sieve beds to reduce moisture levels. Natural gas feed stream 9 is preferably in a pressure range of 200-1000 psig or is compressed to reach this range. Chilled gas stream 37 is fed to second chilling unit 18 to produce second chilled gas stream 92 and second chilled liquid stream 91. The second chilled gas stream 92 is fed to the third chilling unit 22 to produce third chilled liquid stream 116.

[0024] In an embodiment of the invention, a method is provided for optimizing the production of NGL outlet stream 303. The invention includes maximizing the utilization of

the available refrigeration capacity while minimizing the operating column pressure and temperature of the demethanizer column. Similarly, propane losses to the overhead product from the demethanizer are minimized. This is achieved within the constraints of the NGL bottom product specification, an upper constraint of the percent of propane (mole percent) in the residue gases, a maximum pressure drop across the top section of the demethanizer column and a predetermined operating range for the propane and ethane refrigerant compressors suction pressure. Natural gas feed stream 9 is fed to first chilling unit 12 to produce chilled rich gas stream 37 and chilled liquid stream 36. Pressure and flow monitoring devices are useful for determining or controlling the pressure and flow of the feed stream. Lean gas stream 31 can also be removed and, alone or in combination with other residue from the demethanizer, be collected. Pressure of stream 31 is measured through a pressure monitor. Valve 131 on stream 31 is useful to control the unit pressure. Flow of stream 31 is measured, typically after valve 131. Chilled rich gas stream 37 and chilled liquid stream 36 have different compositions as a result of separation of natural gas feed stream 9. Natural gas feed stream 9 preferably contains sweet gas, such as gas that has been submitted to a sweetening process to remove hydrogen sulfide or carbon dioxide. Natural gas stream 9 is preferably dehydrated in molecular sieve beds to reduce moisture levels to less than 1 ppm water. Natural gas feed stream 9 is preferably in a pressure range of 800-1000 psia or is compressed to reach this range. Chilled rich gas stream 37 is fed to second chilling unit 18 to produce second chilled rich gas stream 92 and second chilled liquid stream 91, also having different compositions. Second chilled rich gas stream 92 is fed to third chilling unit 22 to produce third chilled liquid stream 116.

[0025] The three liquid streams, namely, chilled liquid stream 36, second chilled liquid stream 91 and third chilled liquid stream 116, are fed to the demethanizer column 200. Demethanizer column 200 produces overhead stream 201 and bottoms stream 202.

Demethanizer column 200 is a trayed column. Bottoms stream 202 is controlled to a specified bottoms product specification. The overhead stream 201 is characterized by an overhead ethane and propane concentration. The overhead valve 32 can be described by the percent open of the valve, with 100 % being fully open. It can also be described by pressure drop, such as PSI

[0026] In an embodiment of the invention, the three liquid streams, namely, chilled liquid stream 36, second chilled liquid stream 91 and third chilled liquid stream 116, are fed to demethanizer column 200. Demethanizer column 200 produces overhead stream 201 and bottoms stream 202. Demethanizer column 200 has a top tray 33 in an upper section of the demethanizer column and mid-tray 44 in a middle section of the demethanizer column. The top tray defines a top tray temperature and the demethanizer defines a column operating pressure. Bottoms stream 202 is characterized by having a bottom ratio defined by methane concentration of the bottom stream divided by ethane concentration of the bottom stream. The overhead stream is characterized by having an overhead propane concentration.

[0027] Demethanizer column 200 overhead stream 201 is fed through an overhead valve 32. This valve 32 is used to control the operating pressure of the Demethanizer column 200.

[0028] In an embodiment of the invention, overhead stream 201 is fed through an overhead valve 32 at overhead valve outlet pressure or operating pressure. The overhead valve 32 can be described by the percent open of the valve, with 100 % being fully open. It can also be described by pressure drop, such as PSI.

[0029] First propane refrigeration system 34 is operated to provide cooling to first chiller 30, second chiller 70 and third chiller 80. The first chilling unit 12 includes first chiller 30 and first chill down separator 38. The second chilling unit 18 includes second chiller 70, third

chiller 80 and separator 90. The third chilling unit 22 includes fourth chiller 105 and separator 115. The fourth chiller is refrigerated by ethane refrigeration system 64.

[0030] First propane refrigeration system 34 includes propane compressor 800A. The propane compressor defines a propane compressor power output and the propane compressor suction pressures.

[0031] Second propane refrigeration system 54 is operated to provide cooling to the same equipment as system 34. It can be implemented in parallel with first propane refrigeration system 34 that can be operated independently, or it can be used as a backup system when the first propane refrigeration system 34 is out of service. Second propane refrigeration system 54 includes a second propane compressor 800B. Second propane compressor 800B defines a second propane compressor power output and a second propane compressor suction pressure.

[0032] Second chilling unit 18 includes second chiller 70, second chill down separator 90 and the third chiller 80. In an embodiment, the second chill down separator 90 defines a second chill down separator temperature, and the subsequent second chiller 80 defines a subsequent second chiller output level. Level instruments are installed in second chiller 70 and subsequent second chiller 80.

[0033] Ethane refrigeration system 64 provides heat exchange to fourth chiller 105. Ethane refrigeration system 64 includes ethane compressor 900. Ethane compressor 900 defines an ethane compressor suction pressure. One preferred embodiment includes controlling the compressor suction pressure of the ethane compressor 900, which in turn controls the heat exchange to the chilled gas stream 92.

[0034] The method of the invention includes optimizing refrigeration load while maintaining the bottom product specification. This optimization is also performed while

staying within a prescribed range of the overhead ethane and propane concentration. The refrigeration load is defined as the electricity required, or similar energy requirements, to operate the first propane refrigeration system 34, the second propane refrigeration system 54 and the ethane refrigeration system 64. The current invention is an empirical optimization method based on statistical modeling relating NGL plant process variables with the refrigeration system's electricity usage. The method identifies the key process control variables in an NGL plant to be optimized. This method is applicable to an NGL plant that uses dual refrigeration systems. It describes methods to calculate the key optimal targets for the process control settings. These key optimal targets can be fed to a multivariate controller algorithm (such as MPC) that controls the NGL plants, or can be implemented directly by the NGL plant operators inputting the calculated optimal targets in the NGL plant's distributed control system (DCS).

[0035] Model Predictive Control, ("MPC), is an advanced control method for process industries that improves on standard feedback control by predicting how a process, such as distillation, will react to inputs such as heat input. This means that reliance on feedback can be reduced since the effects of inputs will be derived from mathematical empirical models. Feedback can still used to correct for model inaccuracies. The MPC controller relies on an empirical model of a process obtained, for example, by plant testing to predict the future behavior of dependent variables of a dynamic system based on past moves of independent variables. It usually relies on linear models of the process. Commercial suppliers of MPC software useful in this invention include AspenTech (DMC+), Honeywell (RMPCT) and Shell Global Solutions (SMOC).

[0036] In an embodiment of the invention, the method further includes maintaining the propane compressor power output within a predetermined propane compressor power output

range while continuing to minimize the refrigeration load and maintain the bottom ratio within a predetermined bottom ratio range. Thus, the minimum refrigeration load can vary from the embodiment where the propane compressor power output is not limited given that the propane compressor power output is an additional constraint on the system. As with all constraints to the optimization, the embodiment is preferably calculated using rigorous simulation techniques in combination with optimization algorithms. Alternately, simulation using historical data can be used. A combination of rigorous simulation corrected by historical data information can also be used. The simulation solution is preferably implemented using control systems. In a preferred embodiment, real time dynamic simulation is used. These models are then used in conjunction with an optimization package to perform real time calculations of the process variables targets. Advantageously, the invention allows for the increase of NGL production using a real time optimization of the liquefied natural gas plant. This can be accomplished by developing mathematical models of all the process units from plants, tests, and laboratory experiments. This simulation can be used in real time to receive information from monitoring equipment in the liquefied natural gas plant which is then used to simulate the plant and calculate the optimization with manipulation of the identified control variables. Upon determination of the setting of the control variables in order to optimize the conditions, the plant can be modified preferably through the use of dynamic controllers. The calculations can be repeated to verify optimization with the new operating parameters. Such repetition can be as needed or on a regular basis. One example would be gather dynamic data, perform the simulation calculations and optimizations and provide instructions to controller means every four minutes.

[0037] The current invention is also applicable to an NGL plant with a single refrigeration system by using the same empirical optimization method based on statistical modeling

relating NGL plant process variables with the refrigeration system's electricity usage. The method identifies the key process control variables in an NGL plant to be optimized.

[0038] In a preferred embodiment of the invention, the method includes optimizing the suction pressures of the compressors which in turn impact the power usage of the compressors. This preferred embodiment of the invention allows maintaining the constraints of the compressors and the associated process variables within allowable ranges while at the same time continues to minimize the refrigeration load and maintain the Demethanizer Column bottom products specifications.

[0039] In a preferred embodiment of the invention, the method includes maintaining the propane compressor power output within a predetermined propane compressor power output range while continuing to minimize the refrigeration load and maintain the Demethanizer Column bottom products specifications. Thus, the minimum refrigeration load can vary from the embodiment where the propane compressor power output is not limited given that the propane compressor power output is an additional constraint on the system. All target process variable values are preferably determined using experimental techniques in combination with optimization algorithms. The resulting solution is preferably implemented using modern electronic control systems. Advantageously, the invention allows for the increase of NGL production using a real time optimization of the NGL plant. The invention also advantageously minimizes the energy use of the refrigeration system. This can be accomplished by developing mathematical models of all the process units from data collected via plants tests. The plant tests are performed by varying the process operating conditions and collecting the operating data in electronic media for analysis and modeling.

[0040] The method of the invention includes minimizing refrigeration load while maintaining the bottom ratio within a predetermined bottom ratio range to meet specification.

This minimization is also performed while maintaining the overhead propane concentration within a predetermined overhead propane concentration range. The refrigeration load is defined as the electricity required, or similar energy requirements, to operate the first propane refrigeration system 34, the second propane refrigeration system 54 and the ethane refrigeration system 64. This minimization is calculated through simulation of the LNG plant. The simulation is then optimized using numerical methods known in the art to obtain the minimum refrigeration load while maintaining the product streams with the appropriate characteristics such that specifications are met. The method described herein uses mathematical models obtained from process testing and optimization method to drive the process to the best economic conditions. Simulation can be used to determine the optimum levels and control systems operable to implement such levels can be used.

[0041] hi a preferred embodiment of the invention, the method further includes maintaining the propane compressor power output within a predetermined propane compressor power output range while continuing to minimize the refrigeration load and maintain the bottom ratio within a predetermined bottom ratio range. Thus, the minimum refrigeration load can vary from the embodiment where the propane compressor power output is not limited given that the propane compressor power output is an additional constraint on the system. As with all constraints to the optimization, the embodiment is preferably calculated using rigorous simulation techniques in combination with optimization algorithms. Alternately, simulation using historical data can be used. A combination of rigorous simulation corrected by historical data information can also be used. The simulation solution is preferably implemented using control systems. In a preferred embodiment, real time dynamic simulation is used. These models are then used in conjunction with an optimization package to perform real time calculations of the process variables targets. Advantageously, the invention allows for the increase of NGL production using a real time optimization of the

liquefied natural gas plant. This can be accomplished by developing mathematical models of all the process units from plants, tests, and laboratory experiments. This simulation can be used in real time to receive information from monitoring equipment in the liquefied natural gas plant which is then used to simulate the plant and calculate the optimization with manipulation of the identified control variables. Upon determination of the setting of the control variables in order to optimize the conditions, the plant can be modified preferably through the use of dynamic controllers. The calculations can be repeated to verify optimization with the new operating parameters. Such repetition can be as needed or on a regular basis. One example would be gather dynamic data, perform the simulation calculations and optimizations and provide instructions to controller means every four minutes.

[0042] In a preferred embodiment, the optimization variables in the NGL plant are to be selected via statistical design of experiment (DOE) which allows the selection of the variables that have the most impact on the energy use of the refrigeration systems.

[0043] In another embodiment, the experimental data obtained using DOE are analyzed using multivariate statistical process control (MVSPC) techniques to determine the principal components analysis (PCA) models. Several commercially available tools of PCA are available such as MATLAB from MATHWORKS, Inc. or SIMCA-P from UMETRICS, Inc.

[0044] In this embodiment, the energy load of the refrigeration system utilized in the NGL process described earlier is optimized using the statistical modeling techniques can be minimized while maintaining product qualities.

[0045] In another preferred embodiment of the invention, the method further includes maintaining the refrigeration compressors suction pressures within the predetermined

refrigeration compressors suction pressures ranges while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. The refrigeration compressor suction pressure ranges are typically the manufacturers' recommended ranges defined to avoid surging, stonewalling or mechanical damage to the compressors. In yet another alternate embodiment, the compressor power usage can also be maintained within predetermined limits as described above.

[0046] In another preferred embodiment of the invention, the method further includes maintaining the propane compressor suction pressure within a predetermined propane compressor suction pressure range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. The propane compressor suction pressure range is typically the manufacturers' recommended range defined to avoid surging, stonewalling or mechanical damage to the compressors. In yet another alternate embodiment, the propane compressor power output can also be maintained within predetermined limits as described above. In yet another alternate embodiment, the constraints discussed below can also be used in conjunction with the restraints of this embodiment. It is recognized that the larger the number of constraints, the more difficult and time consuming it is to solve the optimization algorithm.. It is also recognized that it is possible to provide constraints that allow for BO solution set. Ranges and limits for constraints for all embodiments of the current invention must allow for a possible solution while taking into account the constraints.

[0047] In an alternate embodiment, the method further includes maintaining a propane compressor scraper output pressure within a predetermined propane compressor output pressure range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0048J In another alternate embodiment, the method further includes maintaining the first tray temperature within a predetermined first tray temperature range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0049J In another alternate embodiment, the method further includes maintaining the overhead valve output pressure of the overhead stream within a predetermined overhead valve output pressure range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. It is known in the art of multivariable control to use multivariable control so that the constraints can be maintained. In a preferred embodiment, the multivariable controllers are the preferred means to maintain the process constraints within the specified engineering ranges. The empirical optimization described in this invention provides the optimization techniques for the refrigeration compressors and provide the optimal targets for the optimization variables and these targets can be sent to the multivariable controllers which drive the process to reach these targets.

[0050] In another alternate embodiment, the method further includes maintaining the first chiller output level within a predetermined first chiller output level range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0051] In yet another embodiment, the efficiency of the NGL process is calculated and named the Coefficient of Performance (COP). Statistical models relating the optimization variables and the coefficient of performance (COP) are developed. The optimization tools in the solution are then used to calculate the optimal targets for the optimization variables in order to optimize the COP. One formulation of COP for the NGL plant is the ratio of the amount of heat removed from the feed gas to the total power input.

[0052] In yet another embodiment, the heat duty for each stage of the cooling process can be computed and used for the overall optimization. Then the models relating the statistical optimizer variables to these heat duties are developed and used to calculate the targets for the optimization solution. The heat duty for each stage of the cooling process can be computed either from the process side or the refrigeration side.

[0053] In yet another embodiment, an alternative method to actually carry out the experiments obtained from the design of experiment (DOE) can be substituted using NGL plant simulators. Examples of this type of simulators are rigorous steady-state or dynamic or operator training simulators.

[0054] In a preferred embodiment, the statistical optimizer disclosed in the present invention can be used as an on-line optimizer, which continuously sends targets to the process control systems.

[0055] In a preferred embodiment, the statistical optimizer disclosed in the present invention can be used as an off-line advisory statistical optimizer, which computes the optimal targets of the NGL plant and advises the operators on the best settings in order to achieve the optimal economic benefits.

[0056] In a preferred embodiment, the statistical optimizer disclosed in the present invention can be used as a design tool to compute the optimal targets for design purposes using a high fidelity simulator.

[0057] In another alternate embodiment, the method further includes maintaining the ethane compressor suction pressure within a predetermined ethane compressor suction pressure range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

-1.8-

[0058] In yet another alternate embodiment, the method further includes maintaining the ethane compressor power output within a predetermined ethane compressor power output range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. In another preferred alternate embodiment, the method further includes maintaining an overhead temperature of the overhead stream within a predetermined overhead temperature range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0059] In yet another alternate embodiment, the method further includes maintaining the first chill down separator temperature within a predetermined first chill down separator temperature range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. In yet another alternate embodiment, the method further includes maintaining a second propane exchanger temperature of a second propane exchanger 58 within a predetermined second propane exchanger temperature range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0060] In yet another alternate embodiment, the method further includes maintaining a second propane compressor suction pressure within a predetermined second propane compressor suction pressure range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. In yet another alternate embodiment, the method further includes maintaining the second propane compressor power output within a predetermined second propane power output range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0061] In yet another alternate embodiment, the method further includes maintaining the primary second chiller output level within a predetermined primary second chiller output level while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range. In yet another alternate embodiment, the method further includes maintaining the subsequent second chiller output level within a predetermined subsequent second chiller output level range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0062] In yet another alternate embodiment, the method further includes maintaining the second chill down separator 90 within a second chill down separator temperature range while minimizing the refrigeration load and maintaining the bottom ratio within a predetermined bottom ratio range.

[0063] The following examples are meant to illustrate in detail manipulated variables and controlled variables according to embodiments of the invention, but are in no way meant to limit the scope of the invention.

[0064] The present invention also includes an apparatus corresponding to the method of the invention. A preferred embodiment of the apparatus is shown in Figure 1. First propane refrigeration system 34, second propane refrigeration system 54 and ethane refrigeration system 64 are shown separately for clarity. These refrigeration systems integrate with the process equipment of the invention to provide heat exchange. The apparatus is a liquefied natural gas plant for maximizing the production of NGL from an inlet gas feed stream and includes means for controlling specific portions of the plant within constraints.

[0065] The liquefied natural gas plant includes first chilling unit 12 for cooling at least a portion of the inlet gas feed stream by heat exchange contact with first and second expanded

refrigerants to produce chilled rich gas stream 37 and chilled liquid stream 36 from first chilling unit 12. Chilled rich gas stream 37 and chilled liquid stream 36 have different compositions as a result of the separation. First chilling unit 12 includes first chiller 30 and first chill down separator 38. First chiller 30 defines a first chiller output level, and first chill down separator 38 defines a first chill down separator temperature.

[0066] The liquefied natural gas plant also includes second chilling unit 18 that receives chilled rich gas stream 37. Second chilling unit 18 further chills the chilled rich gas stream 37 to produce second chilled rich gas stream 92 and second chilled liquid stream 91. Second chilled rich gas stream 92 and second chilled liquid stream 91 have different compositions. Second chilling unit 18 includes second chill down separator 90, primary second chiller 70 and subsequent second chiller. Second chill down separator 90 defining a second chill down separator temperature. Subsequent second chiller defines a subsequent second chiller output level.

[0067] The liquefied natural gas plant includes third chilling unit 22, which receives second chilled rich gas stream 92. Third chilling unit 22 includes third chiller 105 and is operable to further chill second chilled rich gas 92 to produce third chilled liquid stream 116.

[0068] Demethanizer column 200 receives chilled liquid stream 36, second chilled liquid stream 91, and third chilled liquid stream 116 as feed streams to the column. Demethanizer column 200 producing overhead stream 201 and bottoms stream 202. Demethanizer column 200 has top tray 33 in an upper section of the demethanizer column 200 and mid-tray 44 in a middle section of demethanizer column 200. The top tray has a top tray temperature that can be monitored. The bottoms stream has a bottom ratio defined by methane concentration of the bottom stream divided by ethane concentration of the bottom stream. This bottom ratio can also be monitored. The overhead stream defines an overhead propane concentration, which

can also be monitored, as can the other measured or constrained properties. Overhead valve 32 receives overhead stream 201. The overhead valve has an overhead valve outlet pressure and thereby sets the pressure of the overhead stream.

[0069] First propane refrigeration system 34 is operable to provide heat exchange with first chilling unit 12. First propane refrigeration system 34 includes propane compressor 800 and can be a typical propane refrigeration cycle. The propane compressor defines a propane compressor power output and a propane compressor suction pressure.

[0070] Second propane refrigeration system 54 is operable to provide heat exchange to second chilling unit 18. Second propane refrigeration system 54 includes second propane compressor 56.

[0071] Ethane refrigeration system 64 is operable to provide heat exchange to third chilling unit 22. The ethane refrigeration system has an ethane compressor 900. The ethane compressor defines an ethane compressor suction pressure.

[0072] The liquefied natural gas plant includes means for minimizing a refrigeration load while maintaining the bottom ratio within a predetermined bottom ratio range and while maintaining the overhead propane concentration within a predetermine overhead propane concentration range, hereinafter referred to simply as "means for minimizing refrigeration load". This means for minimizing refrigeration load can include one or more controllers. The refrigeration load is the electricity required to operate first propane refrigeration system 34, second propane refrigeration system 54 and ethane refrigeration system 64.

[0073] In another preferred embodiment, the liquefied natural gas plant also includes means for maintaining the propane compressor power output within a predetermined propane compressor power output range. This is in addition to the means for minimizing refrigeration

load, which includes controlling the bottom ratio and controlling the overhead propane concentration as described above. This can also be in addition to any additional controlling means discussed below or in the absence of such other controlling means. Each of the ^' controlling means discussed can be used alone or in conjunction with each other.

[0074] In another preferred embodiment, the liquefied natural gas plant also includes means for maintaining the propane compressor suction pressure within a predetermined propane compressor suction pressure range in addition to the means for minimizing the refrigeration load. In yet another preferred embodiment, the liquefied natural gas plant includes means for maintaining a propane compressor scraper output pressure within a predetermined propane compressor output pressure range.

[0075] In an alternate preferred embodiment, the liquefied natural gas plant includes means for maintaining the first tray temperature within a predetermined first tray temperature range. In another embodiment, the liquefied natural gas plant includes means for maintaining the overhead valve output pressure of the overhead stream within a predetermined overhead valve output pressure range.

[0076] In another embodiment, the liquefied natural gas plant includes means for maintaining the first chiller output level within a predetermined first chiller output level range. In yet another embodiment, the liquefied natural gas plant includes means for maintaining the ethane compressor suction pressure within a predetermined ethane compressor suction pressure range.

[0077] In an embodiment, the liquefied natural gas plant includes means for maintaining ethane compressor power output within a predetermined ethane compressor power output range. In another embodiment, the liquefied natural gas plant includes means for maintaining

an overhead temperature of the overhead stream within a predetermined overhead temperature range. In still another embodiment of the liquefied natural gas plant, the plant includes means for maintaining the first chill down separator temperature within a predetermined first chill down separator temperature range. In an alternate embodiment, the liquefied natural gas plant includes means for maintaining a second propane exchanger temperature of a second propane exchanger 58 within a predetermined second propane exchanger temperature range.

[0078] In another embodiment of the invention, the liquefied natural gas plant includes means for maintaining a second propane compressor suction pressure within a predetermined second propane compressor suction pressure range. Another embodiment of the liquefied natural gas plant includes means for maintaining the second propane compressor power output within a predetermined second propane compressor power output range. The liquefied natural gas plant of claim includes means for maintaining the primary second chiller output level within a predetermined primary second chiller output level in another embodiment. The liquefied natural gas plant includes means for maintaining the subsequent second chiller output level within a predetermined subsequent second chiller output level range in yet another embodiment.

[0079] In an alternate embodiment, the liquefied natural gas plant includes means for maintaining the second chill down separator 90 within a second chill down separator temperature range.

[0080] Natural gas stream 9 is preferably a sweet gas, such as one that has been submitted to a sweetening process. Dehydration of the natural gas is also a common and desirable treatment. Molecular sieve beds are commonly used to reduce moisture levels. The moisture is preferably reduced to less than 1 ppm H ₂ O by volume.

[0081] The natural feed gas is cooled and chilled as described above to a preselected temperature range, with a preferred preselected temperature range being —80 to -120 degrees F. This cooling/chilling can be accomplished by using refrigerants of propane and ethane in the first propane refrigeration system, second propane refrigeration system and ethane refrigeration system. Multiple levels of propane and ethane temperatures are appropriate. For example, propane refrigerants can be at 66 degrees F and 12 degrees F for the first propane refrigeration system and second refrigeration system respectively while ethane can be at —39 degrees F. This results in a high pressure residue gas and three liquid streams of different compositions as described herein.

[0082] Overhead stream 201 is compressed to become residue gas stream 42, which is a sales gas stream. In another embodiment (not shown), the overhead stream 201 can be split, with compression before or after the split, to produce the residue gas stream and a recycle stream that is recycled into the demethanizer or other unit. In an alternate embodiment, the overhead stream of the column is low pressure residue gas, which can be combined with the high pressure residue gas to produce a sales gas.

[0083] Bottom stream 202 can be split to provide NGL outlet stream 303. When alternate heat sources are available to the bottom of the demethanizer and/or a stream containing at least partial vapor is fed to the bottom of the demethanizer, then the entire bottom stream 202 can be removed as NGL product. The three liquid streams provide feed stream for the demethanizer column from which the NGL product is drawn from the bottom.

[0084] Implementation of the invention advantageously includes commercially available multivariable controllers. AspenTech DMCplus™ is one multivariate controller useful in implementation of the invention. Process models are developed and the control strategy

implemented. The current invention results in maximized incremental feed rate to the apparatus and maximized yield of the valuable NGL product.

[0085] As an advantage of the present invention, the method and apparatus of the invention allow for the optimization of the usage of the propane and ethane refrigeration systems. The invention also advantageously allows for managing the natural feed gas distribution in between the chill-down units to allow for maximum feed processing and best recovery conditions. The present invention allows for the optimizing of the recovered NGL qualities so there is less quality "give-away". In turn, this enhances the product quality. The invention also advantageously allows for optimizing of the electrical energy used by the refrigeration systems. The invention also can allow for a greater throughput and decreased over all power consumption. Furthermore, the method when using automatic control feedback loop requires less intervention by a console operator.

[0086] While the invention has been shown or described in only some of its forms, it should be apparent to those skilled in the art that it is not so limited, but is susceptible to various changes without departing from the scope of the invention. For example, this invention may be used in process design but is also useful in conjunction with an existing process plant. This invention is useful as a steady state tool and also for real time optimization.

[0087] For example, splitters can be added to redirect amounts of flow or to allow for control of amounts of flow. Recycle streams can be used to enhance recovery or as a heat since for heat exchangers. Other variation can also be made.