Natural gas pipeline network scheduling optimization method considering flow distribution and carbon emission

Abstract

The invention provides a natural gas pipeline network scheduling optimization method considering flow distribution and carbon emission, and the method comprises the steps: obtaining basic data of a target natural gas pipeline network, and constructing a constraint condition and a target function of a natural gas pipeline network scheduling optimization model considering flow distribution and carbon emission; and solving the natural gas pipeline network scheduling optimization mathematical model considering flow distribution and carbon emission, and generating a pipeline network flow distribution scheme and a compressor station operation scheme. On the basis of pipeline hydraulic characteristics of the natural gas pipeline network and pressurization characteristics of the compressor station, pipeline path flow distribution and compressor station carbon emission control are combined, a natural gas pipeline network scheduling optimization method is provided, a pipeline network flow distribution scheme and a compressor station operation scheme are reasonably decided, and on the premise that the natural gas flow requirement of a user is met, the user experience is improved. The operation cost of a multi-path complex natural gas pipe network system can be greatly reduced, and the economic and environmental benefits in the natural gas conveying process are improved.

Description

Technical Field

[0001] The present invention relates to the field of natural gas pipeline network system scheduling optimization, and in particular to a natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions. Background Art

[0002] With the development of natural gas pipeline network and the interconnection between different trunk pipelines, there are often different pipeline paths in the pipeline network system, which makes natural gas transportation face a variety of path choices. The change of pipeline path flow will have a key impact on the pipeline hydraulic conditions, facility operation status and system transportation cost. Therefore, it is necessary to take optimization measures to optimize the flow distribution of different pipeline paths to ensure that the natural gas pipeline network achieves the optimal flow state and transportation economy while meeting the user's natural gas transportation needs. However, most of the previous studies on natural gas pipeline network scheduling optimization were based on single-path linear or branch pipeline networks, with constant natural gas flow in the pipeline network. The flow distribution decisions of different pipeline paths were rarely considered, and the impact of changes in flow distribution schemes on pipeline network operating conditions was not deeply explored.

[0003] When natural gas flows in the pipeline, the friction between the gas and the inner wall of the pipeline will cause pressure loss. Therefore, a large number of compressor stations are usually built along the pipeline to pressurize the natural gas to ensure the long-distance transportation of natural gas. As the main energy-consuming equipment of the natural gas pipeline network, the compressor station will consume fuel or electricity in the process of pressurizing natural gas and contribute a large amount of carbon dioxide emissions. By optimizing and controlling this part of carbon emissions, the environmental benefits of the natural gas transportation link can be effectively improved. However, most previous studies have focused on the economic goals of the natural gas pipeline network and lack systematic research on environmental goals. Therefore, in the context of "dual carbon", in order to improve the environmental benefits of the natural gas transportation link, it is necessary to optimize and control the operating parameters of the compressor station in the process of formulating the pipeline network scheduling plan to reduce the carbon emission pollution of the system. Summary of the invention

[0004] Based on the overall optimization strategy, the present invention proposes a natural gas pipeline network scheduling optimization method that considers flow distribution and carbon emissions. It combines pipeline path flow distribution and compressor station carbon emission control to make reasonable decisions on pipeline network flow distribution plans and compressor station operation plans. On the premise of meeting the user's natural gas flow demand, it can significantly reduce the operating cost of the multi-path complex natural gas pipeline system and improve the economic and environmental benefits of the natural gas transportation process.

[0005] The present invention provides a natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions, the method comprising:

[0006] S1: Obtain basic data of the target pipeline network, and divide the target pipeline network into areas with supply nodes, transmission nodes, demand nodes, and compressor station nodes;

[0007] S2: Based on the basic data of the target pipeline network, the constraints of the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions are constructed;

[0008] S3: Generate a natural gas pipeline network scheduling optimization model that considers flow distribution and carbon emissions based on constraints and the minimum total cost objective function of the pipeline network;

[0009] S4: Solve the mathematical model of natural gas pipeline network scheduling optimization considering flow distribution and carbon emissions, and generate a natural gas pipeline network scheduling optimization plan.

[0010] Furthermore, the basic data in step S1 includes gas source and user data, pipeline network parameters, and compressor station parameters;

[0011] The gas source and user data include: gas source natural gas supply flow, supply pressure and user natural gas demand flow;

[0012] The pipeline network parameters include: pipeline network topology, pipeline length, pipeline diameter, pipeline resistance coefficient, natural gas physical parameters, pipeline pressure boundary and pipeline flow boundary;

[0013] The compressor station parameters include: the number of compressors in the compressor station, the boundary of the compressor working feasible region, the pressure ratio boundary and the power boundary.

[0014] Further, the constraints described in step S2 include node constraints, pipeline constraints, and compressor station constraints;

[0015] The node constraints include: node flow constraints and pressure constraints;

[0016] The flow rate flowing into the node at any node is equal to the flow rate flowing out of the node. The flow rate flowing into the node includes the upstream edge flow rate and the gas source input flow rate, and the flow rate flowing out of the node includes the downstream edge flow rate and the user output flow rate. The input and output flows of the node should meet the gas source supply capacity and user demand restrictions respectively. The node flow constraint relationship is:

[0017]

[0018] In the formula, s i is the gas source input flow; d i Outflow traffic for users; ij It is the flow rate at the pipeline or compressor station;

[0019] The node pressure should be subject to the node minimum and maximum pressure constraints. The node pressure constraint relationship is:

[0020]

[0021] In the formula, p i is the node pressure;

[0022] The pipeline constraints include: pipeline hydraulic constraints and flow direction constraints;

[0023] The pipeline hydraulic pressure drop represents the difference between the pressure at the end point and the pressure at the start point under the pipeline flow condition calculated by the pipeline hydraulic equation. The pipeline hydraulic constraint relationship is:

[0024]

[0025] In the formula, q ij is the pipeline flow rate; p i is the starting pressure of the pipeline; p j is the pipeline end point pressure; It is a 0-1 binary variable for the forward flow of the pipeline; is a binary variable of 0-1 for pipeline reverse flow; M is the maximum value; is the pipeline flow resistance coefficient; D ij is the pipe diameter; ij is the pipeline friction coefficient;

[0026] There can only be one flow direction in the pipeline during the same period, and the two flow direction variables must satisfy the flow direction uniqueness constraint. The flow direction constraint relationship is:

[0027]

[0028] The compressor station constraints include: compressor station operating status, feasible domain and power constraints;

[0029] The compressor station operation status includes shutdown, startup and bypass. Shutdown means that all valves of the compressor station are closed, the natural gas flow through the compressor station is zero, and the pressure p between the upstream and downstream nodes is i 、p j No correlation; startup means that the natural gas is pressurized by the compressor unit, and the upstream and downstream node pressures p i 、p j The suction and exhaust pressure of the compressor unit The bypass means that the natural gas only flows through the bypass valve in the station without pressurization, and the pressure of the upstream and downstream nodes is directly equal; the switch state of the compressor unit and the bypass valve is subject to the switch state of the compressor station; at the same time, only one device can be turned on in the compressor unit and the bypass valve, and the constraint relationship is:

[0030]

[0031] In the formula, It is a 0-1 variable for the switch status of the compressor unit; It is the bypass valve switch state 0-1 variable; 0-1 variable for the compressor station switch status;

[0032] There is a flow balance relationship between the compressor station flow, the compressor unit flow and the bypass valve flow, and the constraint relationship is:

[0033]

[0034] In the formula, q ij is the compressor station flow; is the flow rate of the compressor unit; is the bypass valve flow;

[0035] When natural gas flows through the bypass valve, the pressure of the upstream and downstream nodes of the compressor station is equal, and the maximum value M is introduced to determine whether the constraint is established. The constraint relationship is:

[0036]

[0037] In the formula, p i is the node pressure upstream of the compressor station; p j is the node pressure downstream of the compressor station;

[0038] When the compressor station is pressurized, the upstream node pressure is associated with the compressor inlet pressure, and the downstream node pressure is associated with the compressor exhaust pressure. The constraint relationship is:

[0039]

[0040]

[0041] In the formula, is the compressor inlet pressure; is the compressor exhaust pressure;

[0042] When natural gas is pressurized through a compressor station, the compressor inlet and exhaust pressures should meet the maximum and minimum boost pressure ratio constraints of the compressor. The constraint relationship is:

[0043]

[0044] In the formula, ε min is the minimum pressure ratio of the compressor; ε max is the maximum pressure ratio of the compressor;

[0045] A compressor station is usually composed of multiple compressor devices. The flow rate of a single device can be calculated by the number of compressor devices that are turned on, and the number of compressors turned on should meet the equipment configuration constraints of the compressor station. The constraint relationship is:

[0046]

[0047] In the formula, The number of compressors turned on; is the flow rate of a single compressor;

[0048] The pressure head indicates the energy provided by the compressor to pressurize a unit mass of natural gas. The relationship is:

[0049]

[0050] In the formula, H ij is the compressor pressure head; m is the natural gas expansion index;

[0051] The compressor characteristic diagram defines the feasible combination of compressor speed, flow rate and pressure head, which can determine the feasible working range of the equipment, and the boundary line of the characteristic diagram is represented by the speed constraint and surge stagnation constraint. The compressor feasible domain can be obtained by fitting the characteristic diagram, and the coefficient A H , B H , C H , D H All are determined by least squares regression, and the constraint relationship is:

[0052]

[0053] In the formula, ω ij is the compressor speed; surge is the compressor surge boundary coefficient; stonewall is the compressor stagnation boundary coefficient;

[0054] The compressor station power can be calculated from the compressor head, the number of compressors on, the flow rate and the efficiency, and the constraint relationship is:

[0055]

[0056] Where PW ij is the compressor power; η is the compressor efficiency.

[0057] Furthermore, the objective function described in step S3 is to minimize the total operating cost of the natural gas pipeline network;

[0058] The total operating cost of the natural gas pipeline network consists of pipeline transportation costs and carbon emission costs, and the relationship is:

[0059] min f=f tran +f carb

[0060] Where, f is the total operating cost of the natural gas pipeline network; tran is the pipeline transportation cost; f carbfor the cost of carbon emissions;

[0061] The pipeline transportation cost is the cost paid by the shipper when natural gas occupies the pipeline transportation capacity when flowing in the pipeline network. The relationship is:

[0062]

[0063] In the formula, C tran is the unit cost coefficient of pipeline transportation; q ij is the pipeline flow rate; L ij is the length of the pipeline;

[0064] The carbon emission cost is calculated by the unit carbon tax price, the carbon emission intensity coefficient of electricity and the power of the compressor station, and the relationship is:

[0065]

[0066] In the formula, C car is the unit carbon tax price; e car PW is the carbon emission intensity coefficient of electricity; ij is the compressor station power;

[0067] According to the constraints and the minimum total cost objective function of the pipeline network, the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions is generated through the convex MILP model.

[0068] Furthermore, the mathematical model for optimizing the scheduling of the natural gas pipeline network considering flow distribution and carbon emissions described in step S4 is a complex non-convex MINLP model. In view of the nonlinear constraints and non-convex feasible region boundaries existing in the model, the MILP relaxation method is required to relax and solve the model. The solution steps include:

[0069] S401: The relaxation process uses piecewise linear approximation to linearize one-dimensional nonlinear functions, mainly for pipeline hydraulic equations; uses spatial grid approximation to linearize high-dimensional nonlinear functions, mainly for compressor pressure head equations and compressor power equations; uses linear external approximation to perform convex relaxation on the non-convex nonlinear feasible domain of the compressor in three-dimensional space, so as to transform it into a linear feasible domain with convex properties;

[0070] For the pipeline hydraulic equation, the one-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear flow square term and pressure square term in the original equation, transform the original equation into a linear equation, and generate a set of flow square and pressure square nonlinear functions. By constructing a piecewise linear function and using the convex combination method to calculate the approximate value of each square term, the piecewise linear approximation of the flow square and pressure square nonlinear functions is achieved. The relationship is:

[0071]

[0072] Where P i is the pressure p i Square term; P j is the pressure p j Square term; Q ij is the flow rate q ij square term;

[0073] For the compressor station pressure head equation, the high-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear intake and exhaust pressure quotient term in the original equation, transform the original equation into a linear equation, and generate a high-dimensional nonlinear function of the exhaust pressure and the intake pressure quotient m-th power. By constructing a spatial unit plane function and using the extended convex combination method to calculate the approximate value of the quotient term, the spatial grid linear approximation of the high-dimensional nonlinear function is realized. The relationship is:

[0074]

[0075] In the formula, Φ ij It is the mth power term of the quotient of exhaust pressure and intake pressure;

[0076] For the compressor station power equation, the high-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear flow and pressure head product terms in the original equation, so that the original equation is converted into a linear equation, and a high-dimensional nonlinear function of the compressor station flow and pressure head product is generated. By constructing a spatial unit plane function and using the extended convex combination method to calculate the approximate value of the product term, the spatial grid approximation of the nonlinear function is realized. The relationship is:

[0077]

[0078] Ψ ij =q ij H ij

[0079] In the formula, Ψ ij is the product term of compressor station flow and pressure head;

[0080] For the non-convex nonlinear bounded feasible domain of the compressor, the convex relaxation process is: in three-dimensional space, six spatial planes are used to relax the feasible domain of the compressor through a linear external approximation method, so that the non-convex nonlinear feasible domain is converted into a convex linear feasible domain; after relaxation, the boundary of the convex linear feasible domain of the compressor is the function equation of the six spatial planes, and the relationship is:

[0081]

[0082] In the formula, a k 、b k is the coefficient of the spatial plane equation;

[0083] After relaxation, the original non-convex MINLP model is transformed into a convex MILP model;

[0084] S402: The transformed MILP model is iteratively solved using a branch and bound algorithm. The branch and bound algorithm is a commonly used algorithm for solving integer programming problems. The algorithm combines search and iteration strategies to find the optimal solution that satisfies the constraints, generates new sub-problems through branches, and prunes through bounding to narrow the search range and ultimately find the global optimal solution.

[0085] Furthermore, the pipeline scheduling optimization plan described in step S4 includes a pipeline flow distribution plan and a compressor station operation plan; the pipeline flow distribution plan is a distribution and optimization strategy for the flow of natural gas in the pipeline network; the compressor station operation plan is a strategy for distributing the number of compressor stations operating in the pipeline network and the compressor station boost ratio.

[0086] In summary, the present invention provides a method for optimizing natural gas pipeline scheduling that takes flow distribution and carbon emissions into consideration, the method comprising: obtaining basic data of the target natural gas pipeline network, constructing constraints and objective functions of a natural gas pipeline scheduling optimization model that takes flow distribution and carbon emissions into consideration; solving a mathematical model for optimizing natural gas pipeline scheduling that takes flow distribution and carbon emissions into consideration, and generating a pipeline network flow distribution plan and a compressor station operation plan. Based on the pipeline hydraulic characteristics of the natural gas pipeline network and the boost characteristics of the compressor station, the present invention combines pipeline path flow distribution and compressor station carbon emission control, proposes a method for optimizing natural gas pipeline scheduling, rationally decides on pipeline network flow distribution plans and compressor station operation plans, and can significantly reduce the operating costs of the natural gas pipeline system and improve the economic and environmental benefits of the natural gas transportation process while meeting the user's natural gas flow demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a flow chart of the present invention;

[0088] Figure 2 This is a schematic diagram of the natural gas pipeline network structure;

[0089] Figure 3 Iterative convergence process for solution method;

[0090] Figure 4 Assign results to traffic;

[0091] Figure 5 Analyze compressor station operating parameters for carbon emission optimization;

[0092] Figure 6 Analyze pipeline hydraulic pressure drop results for carbon emissions optimization. DETAILED DESCRIPTION

[0093] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0094] like Figure 1 As shown, the present invention provides a natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions, comprising the following steps:

[0095] S1: Obtain basic data of the target pipeline network, and divide the target pipeline network into areas with supply nodes, transmission nodes, demand nodes, and compressor station nodes;

[0096] S2: Based on the basic data of the target pipeline network, the constraints of the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions are constructed;

[0097] S3: Generate a natural gas pipeline network scheduling optimization model that considers flow distribution and carbon emissions based on constraints and the minimum total cost objective function of the pipeline network;

[0098] S4: Solve the mathematical model of natural gas pipeline network scheduling optimization considering flow distribution and carbon emissions, and generate a natural gas pipeline network scheduling optimization plan.

[0099] In one embodiment, the basic data described in step S1 includes gas source and user data, pipeline network parameters, and compressor station parameters;

[0100] The gas source and user data include: gas source natural gas supply flow, supply pressure and user natural gas demand flow;

[0101] The pipeline network parameters include: pipeline network topology, pipeline length, pipeline diameter, pipeline resistance coefficient, natural gas physical parameters, pipeline pressure boundary and pipeline flow boundary;

[0102] The compressor station parameters include: the number of compressors in the compressor station, the boundary of the compressor working feasible region, the pressure ratio boundary and the power boundary.

[0103] In one embodiment, the constraints described in step S2 include node constraints, pipeline constraints, and compressor station constraints;

[0104] The node constraints include: node flow constraints and pressure constraints;

[0105] The flow rate flowing into the node at any node is equal to the flow rate flowing out of the node. The flow rate flowing into the node includes the upstream edge flow rate and the gas source input flow rate, and the flow rate flowing out of the node includes the downstream edge flow rate and the user output flow rate. The input and output flows of the node should meet the gas source supply capacity and user demand restrictions respectively. The node flow constraint relationship is:

[0106]

[0107] In the formula, s i is the gas source input flow; d i Outflow traffic for users; ij It is the flow rate at the pipeline or compressor station;

[0108] The node pressure should be subject to the node minimum and maximum pressure constraints. The node pressure constraint relationship is:

[0109]

[0110] In the formula, p i is the node pressure;

[0111] The pipeline constraints include: pipeline hydraulic constraints and flow direction constraints;

[0112] The pipeline hydraulic pressure drop represents the difference between the pressure at the end point and the pressure at the start point under the pipeline flow condition calculated by the pipeline hydraulic equation. The pipeline hydraulic constraint relationship is:

[0113]

[0114] In the formula, q ij is the pipeline flow rate; p i is the starting pressure of the pipeline; p j is the pipeline end point pressure; It is a 0-1 binary variable for the forward flow of the pipeline; is a binary variable of 0-1 for pipeline reverse flow; M is the maximum value; is the pipeline flow resistance coefficient; D ij is the pipe diameter; ij is the pipeline friction coefficient;

[0115] There can only be one flow direction in the pipeline during the same period, and the two flow direction variables must satisfy the flow direction uniqueness constraint. The flow direction constraint relationship is:

[0116]

[0117] The compressor station constraints include: compressor station operating status, feasible domain and power constraints;

[0118] The compressor station operation status includes shutdown, startup and bypass. Shutdown means that all valves of the compressor station are closed, the natural gas flow through the compressor station is zero, and the pressure p between the upstream and downstream nodes is i 、p j No correlation; startup means that the natural gas is pressurized by the compressor unit, and the upstream and downstream node pressures p i 、p j The suction and exhaust pressure of the compressor unit The bypass means that the natural gas only flows through the bypass valve in the station without pressurization, and the pressure of the upstream and downstream nodes is directly equal; the switch state of the compressor unit and the bypass valve is subject to the switch state of the compressor station; at the same time, only one device can be turned on in the compressor unit and the bypass valve, and the constraint relationship is:

[0119]

[0120] In the formula, It is a 0-1 variable for the switch status of the compressor unit; It is the bypass valve switch state 0-1 variable; 0-1 variable for the compressor station switch status;

[0121] There is a flow balance relationship between the compressor station flow, the compressor unit flow and the bypass valve flow, and the constraint relationship is:

[0122]

[0123] In the formula, q ij is the compressor station flow; is the flow rate of the compressor unit; is the bypass valve flow;

[0124] When natural gas flows through the bypass valve, the pressure of the upstream and downstream nodes of the compressor station is equal, and the maximum value M is introduced to determine whether the constraint is established. The constraint relationship is:

[0125]

[0126] In the formula, p i is the node pressure upstream of the compressor station; p j is the node pressure downstream of the compressor station;

[0127] When the compressor station is pressurized, the upstream node pressure is associated with the compressor inlet pressure, and the downstream node pressure is associated with the compressor exhaust pressure. The constraint relationship is:

[0128]

[0129] In the formula, is the compressor inlet pressure; is the compressor exhaust pressure;

[0130] When natural gas is pressurized through a compressor station, the compressor inlet and exhaust pressures should meet the maximum and minimum boost pressure ratio constraints of the compressor. The constraint relationship is:

[0131]

[0132] In the formula, ε min is the minimum pressure ratio of the compressor; ε max is the maximum pressure ratio of the compressor;

[0133] A compressor station is usually composed of multiple compressor devices. The flow rate of a single device can be calculated by the number of compressor devices that are turned on, and the number of compressors turned on should meet the equipment configuration constraints of the compressor station. The constraint relationship is:

[0134]

[0135] In the formula, The number of compressors turned on; is the flow rate of a single compressor;

[0136] The pressure head indicates the energy provided by the compressor to pressurize a unit mass of natural gas. The relationship is:

[0137]

[0138] In the formula, H ij is the compressor pressure head; m is the natural gas expansion index;

[0139] The compressor characteristic diagram defines the feasible combination of compressor speed, flow rate and pressure head, which can determine the feasible working range of the equipment, and the boundary line of the characteristic diagram is represented by the speed constraint and surge stagnation constraint. The compressor feasible domain can be obtained by fitting the characteristic diagram, and the coefficient A H , B H , C H , D H All are determined by least squares regression, and the constraint relationship is:

[0140]

[0141]

[0142] In the formula, ω ij is the compressor speed; surge is the compressor surge boundary coefficient; stonewall is the compressor stagnation boundary coefficient;

[0143] The compressor station power can be calculated from the compressor head, the number of compressors on, the flow rate and the efficiency, and the constraint relationship is:

[0144]

[0145] Where PW ij is the compressor power; η is the compressor efficiency.

[0146] Furthermore, the objective function described in step S3 is to minimize the total operating cost of the natural gas pipeline network;

[0147] The total operating cost of the natural gas pipeline network consists of pipeline transportation costs and carbon emission costs, and the relationship is:

[0148] minf=f tran f carb

[0149] Where, f is the total operating cost of the natural gas pipeline network; tran is the pipeline transportation cost; f carb for the cost of carbon emissions;

[0150] The pipeline transportation cost is the cost paid by the shipper when natural gas occupies the pipeline transportation capacity when flowing in the pipeline network. The relationship is:

[0151]

[0152] In the formula, C tran is the unit cost coefficient of pipeline transportation; q ij is the pipeline flow rate; L ij is the length of the pipeline;

[0153] The carbon emission cost is calculated by the unit carbon tax price, the carbon emission intensity coefficient of electricity and the power of the compressor station, and the relationship is:

[0154]

[0155] In the formula, C car is the unit carbon tax price; e car PW is the carbon emission intensity coefficient of electricity; ij is the compressor station power;

[0156] According to the constraints and the minimum total cost objective function of the pipeline network, the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions is generated through the convex MILP model.

[0157] In one embodiment, the mathematical model for optimizing the scheduling of a natural gas pipeline network considering flow distribution and carbon emissions described in step S4 is a complex non-convex MINLP model. For the nonlinear constraints and non-convex feasible region boundaries existing in the model, the MILP relaxation method is required to relax and solve the model. The solving steps include:

[0158] S401: The relaxation process uses piecewise linear approximation to linearize one-dimensional nonlinear functions, mainly for pipeline hydraulic equations; uses spatial grid approximation to linearize high-dimensional nonlinear functions, mainly for compressor pressure head equations and compressor power equations; uses linear external approximation to perform convex relaxation on the non-convex nonlinear feasible domain of the compressor in three-dimensional space, so as to transform it into a linear feasible domain with convex properties;

[0159] For the pipeline hydraulic equation, the one-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear flow square term and pressure square term in the original equation, transform the original equation into a linear equation, and generate a set of flow square and pressure square nonlinear functions. By constructing a piecewise linear function and using the convex combination method to calculate the approximate value of each square term, the piecewise linear approximation of the flow square and pressure square nonlinear functions is achieved. The relationship is:

[0160]

[0161]

[0162] Where P i is the pressure p i Square term; P j is the pressure p j Square term; Q ij is the flow rate q ij square term;

[0163] For the compressor station pressure head equation, the high-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear intake and exhaust pressure quotient term in the original equation, transform the original equation into a linear equation, and generate a high-dimensional nonlinear function of the exhaust pressure and the intake pressure quotient m-th power. By constructing a spatial unit plane function and using the extended convex combination method to calculate the approximate value of the quotient term, the spatial grid linear approximation of the high-dimensional nonlinear function is realized. The relationship is:

[0164]

[0165] In the formula, Φ ij It is the mth power term of the quotient of exhaust pressure and intake pressure;

[0166] For the compressor station power equation, the high-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear flow and pressure head product terms in the original equation, so that the original equation is converted into a linear equation, and a high-dimensional nonlinear function of the compressor station flow and pressure head product is generated. By constructing a spatial unit plane function and using the extended convex combination method to calculate the approximate value of the product term, the spatial grid approximation of the nonlinear function is realized. The relationship is:

[0167]

[0168] In the formula, Ψ ij is the product term of compressor station flow and pressure head;

[0169] For the non-convex nonlinear bounded feasible domain of the compressor, the convex relaxation process is: in three-dimensional space, six spatial planes are used to relax the feasible domain of the compressor through a linear external approximation method, so that the non-convex nonlinear feasible domain is converted into a convex linear feasible domain; after relaxation, the boundary of the convex linear feasible domain of the compressor is the function equation of the six spatial planes, and the relationship is:

[0170]

[0171] In the formula, a k 、b k is the coefficient of the spatial plane equation;

[0172] After relaxation, the original non-convex MINLP model is transformed into a convex MILP model;

[0173] S402: The transformed MILP model is iteratively solved using a branch and bound algorithm. The branch and bound algorithm is a commonly used algorithm for solving integer programming problems. The algorithm combines search and iteration strategies to find the optimal solution that satisfies the constraints, generates new sub-problems through branches, and prunes through bounding to narrow the search range and ultimately find the global optimal solution.

[0174] The symbols are shown in Table 1, Table 2 and Table 3.

[0175] Table 1 Index and set of natural gas pipeline network scheduling optimization models considering flow distribution and carbon emissions

[0176]

[0177]

[0178] Table 2 Known parameters of the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions

[0179]

[0180] Table 3 Decision variables of the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions

[0181]

[0182]

[0183] In one embodiment, the pipeline scheduling optimization plan described in step S4 includes a pipeline flow distribution plan and a compressor station operation plan; the pipeline flow distribution plan is a distribution and optimization strategy for the flow of natural gas in the pipeline network; the compressor station operation plan is a strategy for distributing the number of compressor stations operating in the pipeline network and the compressor station boost pressure ratio.

[0184] In order to verify the effectiveness of the established mathematical model and solution method, the three-path large-scale natural gas pipeline network example is optimized and solved to obtain the optimal decision-making plan under different working conditions, and a comparative analysis is performed. A three-path large-scale natural gas pipeline network example is established to mainly verify the impact of flow distribution optimization on pipeline transportation costs in large-scale natural gas pipeline networks and explore the impact of compressor station carbon emission optimization on the environmental benefits of the pipeline network.

[0185] A large-scale natural gas pipeline network with three routes, which contains two gas sources, 28 users, 11 compressor stations and 35 pipelines, such as Figure 2 The detailed pipe network structure parameters are shown in Table 4, and the compressor station parameters are shown in Table 5. Four scenarios are set in the example. Scenario 1 means that both flow distribution optimization and carbon emission optimization are considered, scenario 2 means that flow distribution optimization is not considered, scenario 3 means that flow distribution optimization is considered, but carbon emission optimization is not considered, and scenario 4 means that only carbon emission optimization is considered, without considering flow distribution optimization.

[0186] Table 4 Pipeline network structure parameter data

[0187]

[0188]

[0189] Table 5 Compressor station parameters

[0190]

[0191] Specific application example 1: Traffic distribution optimization analysis

[0192] Scenario 1 and Scenario 2 are solved, and the optimization results are shown below.

[0193] (1) Iteration convergence analysis

[0194] Taking scenario 1 as an example, the iterative convergence analysis of the solution method in a large-scale natural gas pipeline network is carried out. The iterative convergence process of this scenario is as follows: Figure 3As shown in the figure. Based on the MILP relaxation method, the initial convergence error of the upper and lower bounds of the solution process was 4.09%, and after 800 iterations, it quickly converged to 1.88%. After 6000 iterations, the convergence error was reduced to 0.81% and maintained for a long time. Finally, after 12921 iterations, the optimal solution was obtained and the pipeline network scheduling plan was obtained. The iteration time was 212.57s and the convergence error accuracy was 0.087%.

[0195] (2) Analysis of economic optimization results

[0196] The pipeline transportation cost results of scenario 1 and scenario 2 obtained by solving. In scenario 1, the transportation cost of path a is 1.499 million yuan, the transportation cost of path b is 8.622 million yuan, the transportation cost of path c is 2.003 million yuan, and the total transportation cost is 12.125 million yuan. In scenario 2, the transportation cost of path a is 333,000 yuan, the transportation cost of path b is 3.907 million yuan, the transportation cost of path c is 8.905 million yuan, and the total transportation cost is 13.145 million yuan. Therefore, compared with scenario 2 without considering the optimization of pipeline network flow distribution, the optimized scenario 1 reduces the pipeline transportation cost by 1.020 million yuan, a reduction of 7.76%. The example further verifies the important value of optimizing the pipeline network flow distribution scheme in improving the economic efficiency of pipeline network transportation for complex natural gas pipeline networks with multiple pipeline paths.

[0197] (3) Traffic distribution plan analysis

[0198] In scenario 1, the traffic of path a is 862.67×10 4 m 3 / d, the flow rate of path b is 4263.88×10 4 m 3 / d, the flow rate of path c is 825.17×10 4 m 3 / d, pipeline path b undertakes the main transportation task, accounting for 71.64% of the flow. In scenario 2, the flow of path a is 274.3×10 4 m 3 / d, the flow rate of path b is 2095.93×10 4 m 3 / d, the flow rate of path c is 3581.51×10 4 m 3 / d, pipeline path c shares more natural gas flow, accounting for 60.18%. The flow distribution results are as follows: Figure 4As shown. Through the pipeline network structure parameters, it can be seen that the length of path a is 606.3km, the length of path b is 725.5km, and the length of path c is 892.6km. Therefore, compared with scenario 2, the optimized scenario 1 transports more natural gas through the shorter path b, effectively reducing the pipeline transportation cost. It can be seen that the mathematical model and solution method of the present invention can reasonably decide the optimal pipeline flow distribution plan when facing a multi-path complex large-scale natural gas pipeline network, and has good scalability.

[0199] Specific application example 2: Carbon emission optimization analysis

[0200] Scenarios 1 and 3 are solved to explore the impact of compressor station carbon emission optimization on the environmental benefits of pipeline transportation. The optimization results are shown below.

[0201] (1) Carbon emission analysis

[0202] The carbon emission cost results of scenario 1 and scenario 3 are obtained after solving. In scenario 1, the total carbon emission cost of the compressor station is 284,000 yuan, and according to the carbon emission calculation formula, the carbon emissions are 1,893.5 tons. In scenario 3, the total carbon emission cost of the compressor station is 390,000 yuan, and according to the carbon emission calculation formula, the carbon emissions are 2,597.0 tons. Compared with scenario 3 without considering the carbon emission optimization of the compressor station, the optimized scenario 1 reduces the carbon emission cost by 106,000 yuan, directly reduces carbon dioxide emissions by 703.5 tons, and the reduction is as high as 27.09%. It can be seen that reasonable optimization of carbon emissions of natural gas pipeline compressor stations can effectively improve environmental benefits while meeting the requirements of natural gas transportation.

[0203] (2) Analysis of compressor station operation plan

[0204] The compressor station operation schemes for scenario 1 and scenario 3 are as follows Figure 5 From the perspective of the number of compressor stations in operation, Figure 5 As shown in (a), in scenario 1, a total of 8 compressor stations were started in the natural gas pipeline network, and the number of compressors in operation was 11. In scenario 3, a total of 9 compressor stations were started in the natural gas pipeline network, and the number of compressors in operation was 14. Therefore, compared with scenario 3 without considering carbon emission optimization, the optimized scenario 1 has reduced both the number of started compressor stations and the number of compressors in operation. From the perspective of the compressor station boosting pressure ratio, Figure 5As shown in (b), under scenario 1, the maximum pressure ratios of the compressor stations of the three pipeline paths all appear at the upstream compressor station, indicating that this scenario is more inclined to use the upstream compressor station to undertake the main boosting task. On the contrary, under scenario 2, the maximum pressure ratio appears more frequently at the downstream compressor station, indicating that this scenario is more inclined to use the downstream compressor station to undertake the main boosting task. Therefore, the optimized scenario 1 effectively reduces the boosting power of the compressor stations on each pipeline path by reducing the number of operating compressor stations and changing the boosting pressure ratio allocation strategy of the compressor stations, as shown in Figure 2. Figure 5 As shown in (c), the carbon emissions of the compressor are reduced, which effectively promotes the green development of the natural gas transmission link.

[0205] (3) Comparative analysis of hydraulic pressure drop

[0206] The pipeline hydraulic pressure drop results for scenario 1 and scenario 3 are as follows: Figure 6 As shown. The results show that the hydraulic pressure drop of pipelines under different compressor station operation schemes is significantly different. In scenario 1, the maximum pressure appears at the starting end of pipeline Pb2 in path b, which is 9.19MPa, and the minimum pressure appears at the terminal section of the pipeline at the end of the three paths, which is 4.5MPa. In scenario 3, the maximum pressure appears at the starting end of pipeline Pb5 in path b, which is 9.14MPa, and the minimum pressure appears at the terminal end of pipeline Pa11 in path a, which is 4.5MPa. It can be seen that both scenarios can meet the operating pressure requirements of pipelines of 3.35-10MPa. It is worth noting that in scenario 1, the terminal pressures of the three pipeline paths are all 4.5MPa, which meets the gas transmission pressure requirements of the terminal users. In scenario 3, only the terminal pressure of pipeline path a is 4.5MPa, and the terminal pressures of pipeline paths b and pipeline paths c are 5.27MPa and 5.48MPa, respectively, which are higher than the gas transmission pressure requirements of the terminal users, and there is energy waste, which is another important reason why the carbon emissions of the compressor station in scenario 3 are higher than those in scenario 1. Therefore, on the premise of meeting the natural gas transportation needs of users, reasonable optimization of the operation plan of the compressor station can effectively avoid energy waste and reduce environmental pollution, and promote the low-carbon development of enterprises under the background of "dual carbon".

[0207] The above is only an example of the embodiment of the present specification and is not intended to limit the embodiment of the present specification. For those skilled in the art, the embodiment of the present specification may have various changes and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiment of the present specification shall be included in the scope of the claims of the embodiment of the present specification.

Claims

1. A natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions, characterized in that: The following steps are involved: S1: Obtain basic data of the target pipeline network, and divide the target pipeline network into areas with supply nodes, transmission nodes, demand nodes, and compressor station nodes; S2: Based on the basic data of the target pipeline network, the constraints of the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions are constructed; S3: Generate a natural gas pipeline network scheduling optimization model that considers flow distribution and carbon emissions based on constraints and the minimum total cost objective function of the pipeline network; S4: Solve the mathematical model of natural gas pipeline network scheduling optimization considering flow distribution and carbon emissions, and generate a natural gas pipeline network scheduling optimization plan.

2. A natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions as claimed in claim 1, characterized in that: The basic data described in step S1 includes gas source and user data, pipeline network parameters, and compressor station parameters; The gas source and user data include: gas source natural gas supply flow, supply pressure and user natural gas demand flow; The pipeline network parameters include: pipeline network topology, pipeline length, pipeline diameter, pipeline resistance coefficient, natural gas physical parameters, pipeline pressure boundary and pipeline flow boundary; The compressor station parameters include: the number of compressors in the compressor station, the boundary of the compressor working feasible region, the pressure ratio boundary and the power boundary.

3. A natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions as claimed in claim 1, characterized in that: The constraints described in step S2 include node constraints, pipeline constraints, and compressor station constraints; The node constraints include: node flow constraints and pressure constraints; The flow rate flowing into the node at any node should be equal to the flow rate flowing out of the node. The flow rate flowing into the node includes the upstream edge flow rate and the gas source input flow rate, and the flow rate flowing out of the node includes the downstream edge flow rate and the user output flow rate. The input and output flows of the node should meet the gas source supply capacity and user demand restrictions respectively. The node flow constraint relationship is: In the formula, s i is the gas source input flow; d i Outflow traffic for users; ij It is the flow rate at the pipeline or compressor station; The node pressure should be subject to the node minimum and maximum pressure constraints. The node pressure constraint relationship is: In the formula, p i is the node pressure; The pipeline constraints include: pipeline hydraulic constraints and flow direction constraints; The pipeline hydraulic pressure drop represents the difference between the pressure at the end point and the pressure at the start point under the pipeline flow condition calculated by the pipeline hydraulic equation. The pipeline hydraulic constraint relationship is: In the formula, q ij is the pipeline flow rate; p i is the starting pressure of the pipeline; p j is the pipeline end point pressure; It is a 0-1 binary variable for the forward flow of the pipeline; is a binary variable of 0-1 for pipeline reverse flow; M is the maximum value; is the pipeline flow resistance coefficient; D ij is the pipe diameter; ij is the pipeline friction coefficient; There can only be one flow direction in the pipeline during the same period, and the two flow direction variables must satisfy the flow direction uniqueness constraint. The flow direction constraint relationship is: The compressor station constraints include: compressor station operating status, feasible domain and power constraints; The compressor station operation status includes shutdown, startup and bypass. Shutdown means that all valves of the compressor station are closed, the natural gas flow through the compressor station is zero, and the pressure p between the upstream and downstream nodes is i 、p j No correlation; startup means that the natural gas is pressurized by the compressor unit, and the upstream and downstream node pressures p i 、p j The suction and exhaust pressure of the compressor unit The bypass means that the natural gas only flows through the bypass valve in the station without pressurization, and the pressure of the upstream and downstream nodes is directly equal; the switch state of the compressor unit and the bypass valve is subject to the switch state of the compressor station; at the same time, only one device can be turned on in the compressor unit and the bypass valve, and the constraint relationship is: In the formula, It is a 0-1 variable for the switch status of the compressor unit; It is the bypass valve switch state 0-1 variable; 0-1 variable for the compressor station switch status; There is a flow balance relationship between the compressor station flow, the compressor unit flow and the bypass valve flow, and the constraint relationship is: In the formula, q ij is the compressor station flow; is the flow rate of the compressor unit; is the bypass valve flow; When natural gas flows through the bypass valve, the pressure of the upstream and downstream nodes of the compressor station is equal, and the maximum value M is introduced to determine whether the constraint is established. The constraint relationship is: In the formula, p i is the node pressure upstream of the compressor station; p j is the node pressure downstream of the compressor station; When the compressor station is pressurized, the upstream node pressure is associated with the compressor inlet pressure, and the downstream node pressure is associated with the compressor exhaust pressure. The constraint relationship is: In the formula, is the compressor inlet pressure; is the compressor exhaust pressure; When natural gas is pressurized through a compressor station, the compressor inlet and exhaust pressures should meet the maximum and minimum boost pressure ratio constraints of the compressor. The constraint relationship is: In the formula, ε min is the minimum pressure ratio of the compressor; ε max is the maximum pressure ratio of the compressor; A compressor station is usually composed of multiple compressor devices. The flow rate of a single device can be calculated by the number of compressor devices that are turned on, and the number of compressors turned on should meet the equipment configuration constraints of the compressor station. The constraint relationship is: In the formula, The number of compressors turned on; is the flow rate of a single compressor; The pressure head indicates the energy provided by the compressor to pressurize a unit mass of natural gas. The relationship is: In the formula, H ij is the compressor pressure head; m is the natural gas expansion index; The compressor characteristic diagram can determine the feasible operating range of the equipment and define the feasible combination of compressor speed, flow and pressure head. The boundary line of the characteristic diagram is represented by the speed constraint and surge stagnation constraint. The compressor feasible domain is obtained by fitting the characteristic diagram. The coefficient A H , B H , C H , D H All are determined by least squares regression, and the constraint relationship is: In the formula, ω ij is the compressor speed; surge is the compressor surge boundary coefficient; stonewall is the compressor stagnation boundary coefficient; The compressor station power can be calculated from the compressor head, the number of compressors on, the flow rate and the efficiency, and the constraint relationship is: Where PW ij is the compressor power; η is the compressor efficiency.

4. A natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions as claimed in claim 1, characterized in that: The objective function described in step S3 is to minimize the total operating cost of the natural gas pipeline network; The total operating cost of the natural gas pipeline network consists of pipeline transportation costs and carbon emission costs, and the relationship is: minf=f tran +f carb Where, f is the total operating cost of the natural gas pipeline network; tran is the pipeline transportation cost; f carb for the cost of carbon emissions; The pipeline transportation cost is the cost paid by the shipper when natural gas occupies the pipeline transportation capacity when flowing in the pipeline network. The relationship is: In the formula, C tran is the unit cost coefficient of pipeline transportation; q ij is the pipeline flow rate; L ij is the length of the pipeline; The carbon emission cost is calculated by the unit carbon tax price, the carbon emission intensity coefficient of electricity and the power of the compressor station, and the relationship is: In the formula, C car is the unit carbon tax price; e car PW is the carbon emission intensity coefficient of electricity; ij is the compressor station power; According to the constraints and the minimum total cost objective function of the pipeline network, the natural gas pipeline network scheduling optimization model considering flow distribution and carbon emissions is generated through the convex MILP model.

5. A natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions as claimed in claim 1, characterized in that: The mathematical model for optimizing the scheduling of the natural gas pipeline network considering flow distribution and carbon emissions described in step S4 is a complex non-convex MINLP model. In view of the nonlinear constraints and non-convex feasible domain boundaries in the model, the MILP relaxation method is required to relax and solve the model. The solution steps include: S401: The relaxation process uses piecewise linear approximation to linearize one-dimensional nonlinear functions, mainly for pipeline hydraulic equations; uses spatial grid approximation to linearize high-dimensional nonlinear functions, mainly for compressor pressure head equations and compressor power equations; uses linear external approximation to perform convex relaxation on the non-convex nonlinear feasible domain of the compressor in three-dimensional space, so as to transform it into a linear feasible domain with convex properties; For the pipeline hydraulic equation, the one-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear flow square term and pressure square term in the original equation, transform the original equation into a linear equation, and generate a set of flow square and pressure square nonlinear functions. By constructing a piecewise linear function and using the convex combination method to calculate the approximate value of each square term, the piecewise linear approximation of the flow square and pressure square nonlinear functions is achieved. The relationship is: Where P i is the pressure p i Square term; P j is the pressure p j Square term; Q ij is the flow rate q ij square term; For the compressor station pressure head equation, the high-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear intake and exhaust pressure quotient term in the original equation, transform the original equation into a linear equation, and generate a high-dimensional nonlinear function of the exhaust pressure and the intake pressure quotient m-th power. By constructing a spatial unit plane function and using the extended convex combination method to calculate the approximate value of the quotient term, the spatial grid linear approximation of the high-dimensional nonlinear function is realized. The relationship is: In the formula, Φ ij It is the mth power term of the quotient of exhaust pressure and intake pressure; For the compressor station power equation, the high-dimensional nonlinear function linearization process is as follows: introduce new variables to replace the nonlinear flow and pressure head product terms in the original equation, so that the original equation is converted into a linear equation, and a high-dimensional nonlinear function of the compressor station flow and pressure head product is generated. By constructing a spatial unit plane function and using the extended convex combination method to calculate the approximate value of the product term, the spatial grid approximation of the nonlinear function is realized. The relationship is: Ψ ij =q ij H ij In the formula, Ψ ij is the product term of compressor station flow and pressure head; For the non-convex nonlinear bounded feasible domain of the compressor, the convex relaxation process is: in three-dimensional space, six spatial planes are used to relax the feasible domain of the compressor through a linear external approximation method, so that the non-convex nonlinear feasible domain is converted into a convex linear feasible domain; after relaxation, the boundary of the convex linear feasible domain of the compressor is the function equation of the six spatial planes, and the relationship is: In the formula, a k 、b k is the coefficient of the spatial plane equation; After relaxation, the original non-convex MINLP model is transformed into a convex MILP model; S402: The transformed MILP model is iteratively solved using a branch and bound algorithm. The branch and bound algorithm is a commonly used algorithm for solving integer programming problems. The algorithm combines search and iteration strategies to find the optimal solution that satisfies the constraints, generates new sub-problems through branches, and prunes through bounding to narrow the search range and ultimately find the global optimal solution.

6. A natural gas pipeline network scheduling optimization method considering flow distribution and carbon emissions as claimed in claim 1, characterized in that: The pipeline network scheduling optimization plan described in step S4 includes a pipeline network flow distribution plan and a compressor station operation plan; the pipeline network flow distribution plan is a distribution and optimization strategy for the flow of natural gas in the pipeline network; The compressor station operation plan is the number of compressor stations operating in the pipeline network and the compressor station boost pressure ratio allocation strategy.

Images