Method, device and equipment for accelerating convex decentralized optimization of data in electric-gas interconnected system
Through the combination of nonlinear non-convex constraints and Benders decomposition algorithm, the power network and natural gas network are alternately iterated, solving the problems of low data processing efficiency and large transmission volume in the electrical-gas interconnection system, and realizing the electrical-gas interconnection system Optimized calculation and information sharing.
Patent Information
- Application Number
- CN202111320158.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-09
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2041-11-09
AI Technical Summary
In the existing electrical-gas interconnection systems, there are problems such as low computing efficiency and large data transmission volume. Especially after the scale of the power grid and gas network has expanded, problems such as insufficient computer memory and data transmission bottlenecks have gradually emerged, and because the network belongs to two different systems. In individual departments, grid parameters and private information cannot be fully shared, making it difficult to achieve decentralized optimization.
Using an accelerated convex dispersion optimization method for data from electrical-gas interconnection systems, the punishment function and the punishment function are obtained by obtaining parameter data and energy coupling data of the power network and natural gas network, and non-linear non-convex constraint processing and Benders decomposition algorithm are used to perform iterative processing, and the punishment function and The penalty factor, if the convergence conditions are met, the optimization results will be output, and the decentralized solution to the power network and the natural gas network will be achieved.
The calculation efficiency of the data of the electrical-gas interconnected system is improved, the data transmission amount is reduced, the problems of low efficiency and large transmission amount in the data processing of power grids and gas networks are solved, and the optimization calculation of the power network and natural gas network is realized.
Smart Images

Figure CN113988453B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of electrical data processing, and in particular to a method, device and equipment for accelerating convex dispersion optimization of electrical-pneumatic interconnection system data. Background Art
[0002] The integrated energy system is a new form of system derived from the traditional power system. The dispatching data of the existing electricity-gas interconnected integrated energy system is generally solved by centralized optimization. Under the centralized optimization framework, all data information of the power grid and gas grid needs to be uploaded to the upper center for unified calculation, and the data transmission volume is large; in addition, the gas pipeline airflow equation is a non-convex constraint, which has a certain impact on the convergence of the decentralized solution of data in the electricity-gas interconnected system.
[0003] As the scale of power grids and gas grids continues to grow, problems such as insufficient computer memory, data transmission bottlenecks, and information privacy gradually arise. In addition, the power grid and gas grid belong to two different departments, and grid parameters and private information cannot be fully shared. Therefore, it is very necessary to study the decentralized optimization method of the power-gas interconnection system. Summary of the invention
[0004] The embodiments of the present invention provide a method, device and equipment for accelerating convex decentralized optimization of data in an electric-gas interconnected system, which are used to solve the technical problems of low computational efficiency and large data transmission volume in the existing processing of data of both the electric power grid and the gas grid.
[0005] In order to achieve the above purpose, the embodiment of the present invention provides the following technical solutions:
[0006] An accelerated convex dispersion optimization method for data of an electric-gas interconnected system, wherein a test platform is connected to a converter valve to be tested, wherein the converter valve to be tested includes a power module and a bypass switch connected to the power module, and the accelerated convex dispersion optimization method includes the following steps:
[0007] S1. Acquire the first parameter data of the power network, the second parameter data of the natural gas network, and the energy coupling data of the power network and the natural gas network;
[0008] S2. Applying nonlinear non-convex constraint processing to the second parameter data to obtain slack variables of the natural gas network;
[0009] S3. According to the slack variable and using the Benders decomposition algorithm, the first parameter data, the second parameter data and the energy coupling data are iterated n times to obtain a penalty function and a penalty factor;
[0010] S4. If both the slack variable and the penalty function satisfy the convergence condition, the optimization result is output; wherein the convergence condition is:
[0011]
[0012] In the formula, ε 1 and ε 2 are convergence tolerances, δ is the slack variable, W k is the penalty function for the kth iteration, W k-1 is the penalty function for k-1 iterations.
[0013] Preferably, the accelerated convex dispersion optimization method for the electrical-gas interconnection system data comprises: S5. If the slack variable and / or the penalty function does not meet the convergence condition, the penalty factor is updated and step S3 is executed again.
[0014] Preferably, the second parameter data is processed by nonlinear non-convex constraints to obtain slack variables of the natural gas network including:
[0015] The second parameter data is processed using a natural gas pipeline gas flow formula to obtain a transmission characteristic constant of the natural gas pipeline;
[0016] The second parameter data and the transmission characteristic constant are processed by using non-convex constraints to obtain a constraint inequality of natural gas;
[0017] The constrained inequality is subjected to a linearization process of the reduced terms by using a first-order Taylor expansion method to obtain a linearized constrained inequality;
[0018] The second parameter data and the transmission characteristic constant are iterated multiple times to obtain relaxation variables that satisfy the linearized constraint inequality.
[0019] Preferably, the natural gas pipeline gas flow formula is:
[0020]
[0021] The constraint inequality is:
[0022]
[0023] The linearized constraint inequality is:
[0024]
[0025] In the formula, Δx ij is the spatial step length between natural gas pipeline node i and node j, M 2 is the transmission characteristic constant, is the gas pressure at position d between nodes i and j in the natural gas pipeline at time t, is the gas pressure at position d+1 between nodes i and j in the natural gas pipeline at time t, is the gas flow between nodes i and j in the natural gas pipeline at position d at time t, is the gas pressure between nodes i and j of the natural gas pipeline at position d after k-1 iterations at time t, is the gas flow between nodes i and j in the natural gas pipeline at position d after k-1 iterations at time t, is the slack variable between the natural gas pipeline node i and node j at position d at time t, that is, the slack variable of the natural gas network, and k is a natural number greater than zero.
[0026] Preferably, the first parameter data, the second parameter data and the energy coupling data are iterated n times according to the slack variables and the Benders decomposition algorithm is used to obtain the penalty function and the penalty factor, which include:
[0027] The first parameter data and the energy coupling data are relaxed using a convex quadratic constraint inequality to obtain a natural gas network variable and a natural gas flow rate;
[0028] If the objective function of the natural gas network has a solution, a solution vector is obtained and a coupling constraint formula is used to process the natural gas network variables and the natural gas flow to obtain a Benders cut constraint inequality;
[0029] If the objective function of the natural gas network has no solution, add a gas slack variable of the coupling constraint, use the natural gas network constraint of the Benders decomposition algorithm to process the natural gas network variable, the natural gas flow and the energy coupling data, and obtain a feasible Benders cut constraint inequality;
[0030] The power network constraint of the Benders decomposition algorithm is used to iterate the Benders cut constraint inequality and the feasibility Benders cut constraint inequality n times to obtain the upper bound and the lower bound of the objective function;
[0031] If the upper boundary and the lower boundary meet the optimal solution condition, a penalty factor is obtained;
[0032] A penalty function is formed by the penalty factor, the optimization target of the energy coupling data and the relaxation variable.
[0033] Preferably, the convex quadratic constraint inequality is: GT,a,t ≥h 2,a (P GT,a,t ) 2 +h 1,a P GT,a,t +h 0,a , a∈Ω GT , a∈Ω GT ; In the formula, f GT,a,t is the natural gas flow consumed by gas generator set a at time t, h 0,a is the first consumption characteristic coefficient of gas unit a, h 1,a is the second consumption characteristic coefficient of gas unit a, h 2,a is the third consumption characteristic coefficient of gas unit a, a is the number of the gas unit, Ω GT is the collection of gas units, P GT,a,t is the active power of gas generator set a at time t, is the natural gas network variable.
[0034] Preferably, the objective function is:
[0035]
[0036] The coupling constraint formula is:
[0037] st
[0038]
[0039]
[0040] The Benders cut constraint inequality is: gas (g n )+F power (p)+(λ n ) T E(pp n )≤μ;
[0041] In the formula, F gas For gas network targets, is the slack variable between the natural gas pipeline node i and node j at position d at time t, Ω d is the set of natural gas pipeline locations, Ω pipe is the set between different nodes i and j, T is the set of time, ρ k is the penalty factor after the kth iteration, a is the number of the gas unit, Ω GT is the set of gas units, b is the number of the power-to-gas unit, Ω P2G is a collection of power-to-gas units, is the natural gas flow consumed by gas generator set a after the nth iteration at time t, is the natural gas network variable, h P2G is the conversion factor of electricity to gas, is the natural gas flow converted by the power-to-gas unit b after the nth iteration at time t, f P2G,b,tis the active power consumed by the power-to-gas unit b after the nth iteration at time t, μ is a real number variable, g n is the solution vector of the nth iteration, E is the coefficient matrix of the coupling constraint, F power is the power network target, p is the decision vector of the power network target, is the result after the nth iteration, λ n is the multiplier vector of the coupling constraints in the nth iteration.
[0042] Preferably, the natural gas network constraints are:
[0043] min v
[0044] Gas network constraints
[0045]
[0046]
[0047] The feasibility Benders cut constraint inequality is:
[0048] The power network constraints are:
[0049] minμ
[0050] st power network constraints
[0051] F gas (g n )+F power (p)+(λ n ) T E(pp n )≤μ
[0052]
[0053] The optimal solution condition is: B -L B |≤ε|L B |;
[0054] Where η c is the gas relaxation variable of the cth coupling constraint, v is the sum of all gas relaxation variables, U B is the upper boundary of the objective function, L B is the lower bound of the objective function, ε is the convergence parameter of the optimal solution, is the natural gas network variable, is the natural gas flow consumed by gas generator set a after the nth iteration at time t, is the gas relaxation variable of the cth coupling constraint of gas generator set a, is the gas relaxation variable of the cth coupling constraint of the power-to-gas unit b, is the natural gas flow converted by the power-to-gas unit b after the nth iteration at time t, f P2G,b,t is the active power consumed by the power-to-gas unit b after the nth iteration at time t, h P2G is the conversion factor of electricity to gas, is the multiplier vector of the coupling constraint formula under the natural gas network constraint in the nth iteration, E is the coefficient matrix of the coupling constraint, p is the decision vector of the power network target, is the result after the nth iteration, μ is a real variable, λ n is the multiplier vector of the coupling constraint in the nth iteration, F power is the power network target, F gas For the gas network target, g n is the solution vector of the nth iteration.
[0055] The present invention also provides an accelerated convex dispersion optimization device for electric-gas interconnected system data, comprising a data processing module, a constraint processing module, a decomposition processing module and an optimization output module;
[0056] The data processing module is used to obtain first parameter data of the power network, second parameter data of the natural gas network, and energy coupling data of the power network and the natural gas network;
[0057] The constraint processing module is used to process the second parameter data using nonlinear non-convex constraints to obtain slack variables of the natural gas network;
[0058] The decomposition processing module is used to perform n-times iterative processing on the first parameter data, the second parameter data and the energy coupling data according to the slack variables and adopt the Benders decomposition algorithm to obtain a penalty function and a penalty factor;
[0059] The optimization output module is used to output the optimization result when both the slack variable and the penalty function meet the convergence condition;
[0060] Wherein, the convergence condition is:
[0061]
[0062] In the formula, ε 1 and ε 2 are convergence tolerances, δ is the slack variable, W k is the penalty function for the kth iteration, W k-1 is the penalty function for k-1 iterations.
[0063] The present invention also provides an accelerated convex dispersion optimization device for electrical-pneumatic interconnection system data, comprising a processor and a memory;
[0064] The memory is used to store program codes and transmit the program codes to the processor;
[0065] The processor is used to execute the accelerated convex dispersion optimization method for the electric-gas interconnection system data according to the instructions in the program code.
[0066] It can be seen from the above technical scheme that the embodiment of the present invention has the following advantages: the accelerated convex decentralized optimization method, device and equipment of the data of the electric-gas interconnected system, the steps of the accelerated convex decentralized optimization method include: obtaining the first parameter data of the electric power network, the second parameter data of the natural gas network and the energy coupling data of the electric power network and the natural gas network; using nonlinear non-convex constraint processing on the second parameter data to obtain the slack variable of the natural gas network; according to the slack variable and using the Benders decomposition algorithm to iterate the first parameter data, the second parameter data and the energy coupling data for n times, to obtain the penalty function and the penalty factor; if the slack variable and the penalty function both meet the convergence condition, then the optimization result is output. The accelerated convex decentralized optimization method combines the nonlinear non-convex constraint and the Benders decomposition algorithm to perform alternating iterative processing on the first parameter data, the second parameter data and the energy coupling data, realizes the decentralized solution of the centralized optimization objective function and constraint of the electric power network and the natural gas network, thereby completing the optimization calculation of the electric-gas interconnected system data, improving the calculation efficiency of the electric-gas interconnected system data, and reducing the amount of transmitted data, solving the technical problems of low calculation efficiency and large data transmission volume in the existing processing of the data of both the power grid and the gas grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0067] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0068] Figure 1 A flowchart of the steps of the accelerated convex dispersion optimization method for electrical-gas interconnection system data according to an embodiment of the present invention;
[0069] Figure 2 A flowchart of an accelerated convex decentralized optimization method for electrical-gas interconnected system data according to an embodiment of the present invention;
[0070] Figure 3A schematic diagram of gas relaxation variables and the number of iterations of the accelerated convex decentralized optimization method for electricity-gas interconnection system data according to an embodiment of the present invention;
[0071] Figure 4 A schematic diagram of gas relaxation variables and the number of iterations of the accelerated convex decentralized optimization method for electricity-gas interconnection system data according to an embodiment of the present invention;
[0072] Figure 5 A framework diagram of an accelerated convex dispersion optimization device for electrical-gas interconnection system data according to an embodiment of the present invention. DETAILED DESCRIPTION
[0073] In order to make the purpose, features and advantages of the present invention more obvious and easy to understand, the technical solutions 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 embodiments described below are only 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.
[0074] The embodiments of the present application provide a method, device and equipment for accelerating convex decentralized optimization of data of an electric-gas interconnected system, which are used to solve the technical problems of low computational efficiency and large data transmission volume in the existing processing of data of both the electric power grid and the gas grid.
[0075] Embodiment 1:
[0076] Figure 1 This is a flowchart of the steps of the accelerated convex dispersion optimization method for electrical-gas interconnection system data according to an embodiment of the present invention. Figure 2 This is a flow chart of the accelerated convex decentralized optimization method for electrical-gas interconnected system data according to an embodiment of the present invention.
[0077] like Figure 1 and Figure 2 As shown, an embodiment of the present invention provides an accelerated convex dispersion optimization method for electrical-gas interconnection system data, comprising the following steps:
[0078] S1. Acquire first parameter data of the power network, second parameter data of the natural gas network, and energy coupling data of the power network and the natural gas network.
[0079] It should be noted that the main purpose is to obtain the parameter data and energy coupling data of the initialized power network and natural gas network. The first parameter data of the power network includes the power network topology, transmission line resistance, reactance, power network node load, generator set output limit, generator set power generation cost coefficient, unit climbing rate, power, etc. The second parameter data of the natural gas network includes the natural gas network topology, natural gas network node gas pressure limit, natural gas pipeline flow limit, natural gas network node gas load, gas source output limit, gas source supply cost, gas tank capacity, etc. In this embodiment, the first parameter data mainly includes the power network target, the decision vector of the power network target, etc.; the second parameter data mainly includes the natural gas network target, pipeline location, node, natural gas flow, power consumption, gas unit, etc.
[0080] In the embodiment of the present invention, the power grid and the natural gas grid realize mutual conversion of two energy sources through gas generators and power-to-gas technology (power-to-gas, P2G). The coupling constraints of the two energy flows of the power grid and the natural gas grid are realized through the consumption characteristic equation of the gas generator and the P2G power-to-gas conversion equation. The consumption characteristic equation of the gas generator is: GT,a,t =h 2,a (P GT,a,t ) 2 +h 1,a P GT,a,t +h 0,a a∈Ω GT ; The P2G electric-gas conversion equation is: P2G,b,t =h P2G P P2G,b,t b∈Ω P2G , where: Ω GT is the collection of gas units, Ω P2G is the set of power-to-gas units, f GT,a,t is the natural gas flow consumed by gas generator set a at time t, P GT,a,t is the active power of gas generator set a at time t, h 0,a is the first consumption characteristic coefficient of gas unit a, h 1,a is the second consumption characteristic coefficient of gas unit a, h 2,a is the third consumption characteristic coefficient of gas unit a, f P2G,b,t is the active power consumed by the power-to-gas unit b after the nth iteration at time t, is the natural gas flow converted by the power-to-gas unit b after the nth iteration at time t, a is the number of the gas unit, b is the number of the power-to-gas unit, and h P2G is the conversion factor of electricity to gas.
[0081] It should be noted that the energy coupling data includes the consumption characteristic equation coefficient of the gas unit, the power-to-gas conversion coefficient, the power-to-gas unit, the power consumption, the conversion coefficient, etc.
[0082] S2. Apply nonlinear non-convex constraints to the second parameter data to obtain the slack variables of the natural gas network.
[0083] It should be noted that in step S2, the second parameter data such as gas pressure, gas flow, time step, and space step of the natural gas network are mainly processed by nonlinear non-convex constraints to obtain the slack variables of the natural gas network. Among them, the nonlinear non-convex constraints in step S2 are outer layer processing methods.
[0084] S3. Based on the relaxation variables and using the Benders decomposition algorithm, the first parameter data, the second parameter data and the energy coupling data are iterated n times to obtain a penalty function and a penalty factor.
[0085] It should be noted that in step S3, the objective functions in the electrical network and the natural gas network are alternately iterated by the Benders decomposition algorithm, and the data conversion between the electrical network and the natural gas network is converted into an iterative process of gradually tightening the relaxation domain of the constraint conditions to obtain the optimal penalty function and penalty factor. Among them, the Benders decomposition algorithm in step S3 is an inner layer processing method.
[0086] In the embodiment of the present application, the penalty function is:
[0087]
[0088] Where F is the power-to-gas target, W k is the penalty function for the kth iteration, is the slack variable between the natural gas pipeline node i and node j at position d at time t, ρ k is the penalty factor after the kth iteration, Ω d is the set of natural gas pipeline locations, Ω pipe is the set between different nodes i and j, and T is the set of time.
[0089] S4. If both the slack variables and the penalty function meet the convergence conditions, the optimization results are output.
[0090] In the embodiment of the present invention, the convergence condition is:
[0091]
[0092] In the formula, ε 1 and ε 2 are convergence tolerances, δ is the slack variable, W kis the penalty function for the kth iteration, W k-1 is the penalty function for k-1 iterations, k∈n, and n is a natural number greater than 0.
[0093] It should be noted that in step S4, the slack variables obtained in step S2 and the penalty function in step S3 are mainly judged according to the convergence conditions. Only when the convergence conditions are met, the natural gas flow and active power required for switching the gas generator set between the natural gas network and the electrical network are output; if the convergence conditions are not met, step S5 is executed.
[0094] S5. If the slack variable and / or penalty function does not meet the convergence condition, update the penalty factor and execute step S3 again.
[0095] It should be noted that the slack variable obtained in step S2 and the penalty function in step S3 are mainly judged according to the convergence condition. If one of them does not meet the convergence condition, the penalty factor needs to be updated and step S3 is executed again.
[0096] In the embodiment of the present invention, the rule for updating the penalty factor is: k+1 =min(v c ρ k ,ρ max ), where v c is the dynamic adjustment coefficient of the penalty factor, and v c >1,ρ max is the maximum value of the penalty factor.
[0097] It should be noted that the maximum values of the dynamic adjustment coefficient and the penalty factor can be set according to needs and are not limited here.
[0098] The invention provides an accelerated convex decentralized optimization method for data of an electric-gas interconnected system, the steps of which include: obtaining first parameter data of an electric power network, second parameter data of a natural gas network, and energy coupling data of the electric power network and the natural gas network; processing the second parameter data with nonlinear nonconvex constraints to obtain slack variables of the natural gas network; iterating the first parameter data, the second parameter data, and the energy coupling data n times according to the slack variables and using the Benders decomposition algorithm to obtain a penalty function and a penalty factor; if both the slack variables and the penalty function meet the convergence conditions, the optimization result is output. The accelerated convex decentralized optimization method alternately iterates the first parameter data, the second parameter data, and the energy coupling data by combining nonlinear nonconvex constraints and the Benders decomposition algorithm, realizes decentralized solution of the centralized optimization objective function and constraints of the electric power network and the natural gas network, thereby completing the optimization calculation of the electric-gas interconnected system data, improving the calculation efficiency of the electric-gas interconnected system data, and reducing the amount of transmitted data, solving the technical problems of low calculation efficiency and large data transmission volume in the existing processing of data of both the power grid and the gas grid.
[0099] In one embodiment of the present invention, the second parameter data is processed by nonlinear non-convex constraints to obtain slack variables of the natural gas network including:
[0100] The second parameter data is processed using the gas flow formula of the natural gas pipeline to obtain the transmission characteristic constant of the natural gas pipeline;
[0101] The second parameter data and the transmission characteristic constant are processed by non-convex constraints to obtain the constraint inequality of natural gas;
[0102] The constrained inequality is linearized by using the first-order Taylor expansion method to obtain the linearized constrained inequality.
[0103] The second parameter data and the transmission characteristic constant are processed by multiple iterations to obtain the relaxation variables satisfying the linearized constraint inequality.
[0104] In the embodiment of the invention, the gas flow formula of the natural gas pipeline is:
[0105]
[0106] The constraint inequality is: The constraint inequality is:
[0107]
[0108] The linearized constraint inequality is:
[0109]
[0110] In the formula, Δx ijis the spatial step length between natural gas pipeline node i and node j, M 2 is the transmission characteristic constant, is the gas pressure at position d between nodes i and j in the natural gas pipeline at time t, is the gas pressure at position d+1 between nodes i and j in the natural gas pipeline at time t, is the gas flow between nodes i and j in the natural gas pipeline at position d at time t, is the gas pressure between nodes i and j of the natural gas pipeline at position d after k-1 iterations at time t, is the gas flow between nodes i and j in the natural gas pipeline at position d after k-1 iterations at time t, is the slack variable between the natural gas pipeline node i and node j at position d at time t, that is, the slack variable of the natural gas network, and k is a natural number greater than zero.
[0111] In the embodiment of the present invention, the gas flow formula of the natural gas pipeline may also be:
[0112]
[0113] Where M 1 is another transmission characteristic constant, and Δt is the time step between nodes i and j of the natural gas pipeline.
[0114] It should be noted that regarding the constraint inequality is a standard second-order cone constraint, which is convex itself; and the constraint inequality It is a difference constraint of convex functions, that is, it is a non-convex constraint.
[0115] In one embodiment of the present invention, the first parameter data, the second parameter data and the energy coupling data are iterated n times according to the slack variables and the Benders decomposition algorithm is used to obtain the penalty function and the penalty factor, which include:
[0116] The first parameter data and energy coupling data are relaxed using convex quadratic constraint inequalities to obtain natural gas network variables and natural gas flow;
[0117] If the objective function of the natural gas network has a solution, the solution vector is obtained and the coupling constraint formula is used to process the natural gas network variables and the natural gas flow to obtain the Benders cut constraint inequality;
[0118] If the objective function of the natural gas network has no solution, add the gas slack variable of the coupling constraint, use the natural gas network constraint of the Benders decomposition algorithm to process the natural gas network variables, natural gas flow and energy coupling data, and obtain the feasibility Benders cut constraint inequality;
[0119] The power network constraints of Benders decomposition algorithm are used to iterate the Benders cut constraint inequality and feasibility Benders cut constraint inequality n times to obtain the upper and lower bounds of the objective function;
[0120] If the upper and lower boundaries meet the optimal solution conditions, the penalty factor is obtained;
[0121] The penalty function is composed of the penalty factor, the optimization target of the energy coupling data and the slack variable.
[0122] In the embodiment of the present invention, the convex quadratic constraint inequality is: GT,a,t ≥h 2,a (P GT,a,t ) 2 +h 1,a P GT,a,t +h 0,a , a∈Ω GT , a∈Ω GT ; In the formula, f GT,a,t is the natural gas flow consumed by gas generator set a at time t, h 0,a is the first consumption characteristic coefficient of gas unit a, h 1,a is the second consumption characteristic coefficient of gas unit a, h 2,a is the third consumption characteristic coefficient of gas unit a, a is the number of the gas unit, Ω GT is the collection of gas units, P GT,a,t is the active power of gas generator set a at time t, is the natural gas network variable.
[0123] In the embodiment of the present invention, the objective function is:
[0124]
[0125] The coupling constraint is:
[0126] st
[0127]
[0128]
[0129] The Benders cut constraint inequality is: gas (g n )+F power (p)+(λ n ) T E(pp n )≤μ;
[0130] In the formula, Fgas For gas network targets, is the slack variable between the natural gas pipeline node i and node j at position d at time t, Ω d is the set of natural gas pipeline locations, Ω pipe is the set between different nodes i and j, T is the set of time, ρ k is the penalty factor after the kth iteration, a is the number of the gas unit, Ω GT is the set of gas units, b is the number of the power-to-gas unit, Ω P2G is a collection of power-to-gas units, is the natural gas flow consumed by gas generator set a after the nth iteration at time t, is the natural gas network variable, h P2G is the conversion factor of electricity to gas, is the natural gas flow converted by the power-to-gas unit b after the nth iteration at time t, f P2G,b,t is the active power consumed by the power-to-gas unit b after the nth iteration at time t, μ is a real number variable, g n is the solution vector of the nth iteration, E is the coefficient matrix of the coupling constraint, F power is the power network target, p is the decision vector of the power network target, is the result after the nth iteration, λ n is the multiplier vector of the coupling constraints in the nth iteration.
[0131] In the embodiment of the present invention, the natural gas network constraints are:
[0132] min v
[0133] Gas network constraints
[0134]
[0135]
[0136] The feasibility Benders cut constraint inequality is:
[0137] The power network constraints are:
[0138] minμ
[0139] st power network constraints
[0140] F gas (g n )+F power (p)+(λ n ) T E(pp n )≤μ
[0141]
[0142] The optimal solution condition is: |U B -L B |≤ε|L B |;
[0143] Where η c is the gas relaxation variable of the cth coupling constraint and is greater than 0, v is the sum of all gas relaxation variables, U B is the upper boundary of the objective function, L B is the lower bound of the objective function, ε is the convergence parameter of the optimal solution, is the natural gas network variable, is the natural gas flow consumed by gas generator set a after the nth iteration at time t, is the gas relaxation variable of the cth coupling constraint of gas generator set a, is the gas relaxation variable of the cth coupling constraint of the power-to-gas unit b, is the natural gas flow converted by the power-to-gas unit b after the nth iteration at time t, f P2G,b,t is the active power consumed by the power-to-gas unit b after the nth iteration at time t, h P2G is the conversion factor of electricity to gas, is the multiplier vector of the coupling constraint formula under the natural gas network constraint in the nth iteration, E is the coefficient matrix of the coupling constraint, p is the decision vector of the power network target, is the result after the nth iteration, μ is a real variable, λ n is the multiplier vector of the coupling constraint in the nth iteration, F power is the power network target, F gas For the gas network target, g n is the solution vector of the nth iteration.
[0144] In an embodiment of the present invention, since unnecessary natural gas consumption will inevitably lead to an increase in the operating cost and carbon emissions of the electricity-gas interconnection system, a convex quadratic constraint inequality with a tightening function is used to perform nonlinear non-convex constraints on the first parameter data, the second parameter data and the energy coupling data to obtain natural gas network variables and natural gas flow; then, based on the natural gas network variables and the natural gas flow, the centralized optimization of data in the electricity-gas interconnection system is divided into the main problem of optimal power flow of the power network and the sub-problem of optimal power flow of the natural gas network through the Benders decomposition algorithm, and the centralized optimization of data in the electricity-gas interconnection system is achieved by alternating the main problem and the sub-problem; finally, in the sub-problem of optimal power flow of the natural gas network, the penalty factor is obtained by different methods based on whether the objective function of the natural gas network has a solution, and the penalty function is composed of the penalty factor, the optimization target of the energy coupling data and the slack variable; in steps S4 and S5, the optimization results that pass the convergence conditions are obtained.
[0145] It should be noted that the natural gas target network in the objective function of the natural gas network refers to the sum of the gas source cost, the natural gas storage cost and the penalty term for the slack variable of the natural gas pipeline flow equation. In the objective function of the natural gas network, g is the decision vector of the optimal power flow subproblem of the natural gas network, also known as the solution vector of the objective function. It mainly refers to the result after the nth iteration of the main problem of optimal power flow in the power network. The sub-problem of optimal power flow in the natural gas network is expressed by the objective function, and the alternating iteration of the main problem and sub-problem is realized by coupling constraints and Benders cut constraint inequalities or natural gas network constraints and feasibility Benders cut constraint inequalities. In the iterative process, the generated Benders cut constraint inequalities and feasibility Benders cut constraint inequalities are substituted into the power network constraints of the main problem for solution, and the upper bound of the objective function iteration n times and the lower bound of iteration n+1 times are obtained; in the iterative processing process, the main problem and sub-problems are also repeatedly solved, that is, the upper bound and lower bound of the objective function are gradually corrected. Only when the upper bound and the lower bound meet the optimal solution conditions, the optimal solution of the data concentration optimization in the electric-gas interconnection system can be obtained, and the penalty factor of the optimal solution can be obtained.
[0146] In the embodiment of the present invention, the accelerated convex decentralized optimization method of the electric-gas interconnected system data is illustrated by taking the electric-gas interconnected system coupled with the IEEE 39-node power network and the Belgian 20-node natural gas network as a case study, including: the parameters affecting the convergence effect mainly include the initial value of the penalty factor ρ 0 , dynamically adjust the coefficient v c (referring to the above-mentioned gas relaxation variable), convergence tolerance ε 1 and ε 2 , set the convergence tolerance to ε 1 =10-4 , ε 2 =10 -3 To ensure the convergence accuracy, the convergence condition ε=10 in step S4 -5 Based on the initial value of the penalty factor ρ 0 , dynamically adjust the coefficient v c The impact on the convergence performance of the inner and outer layer algorithms (i.e., nonlinear non-convex constraints and Benders decomposition algorithms).
[0147] For the initial value of the penalty factor ρ 0 , target value and initial value of penalty factor ρ 0 The ratio between them will affect the final effect of the algorithm. In this paper, the target value is fixed at the thousand level, and ρ is set 0 In 1×10 -4 The convergence effect of the inner and outer layer algorithms is shown in Table 1.
[0148] Table 1
[0149]
[0150] From Table 1 above, we can see that for PCCP-Benders, the smaller the initial value, the better the quality of the solution. 0 ≥1, the target value result is slightly worse; when ρ 0 =10, the total number of iterations is large, and the computation time increases accordingly; the above results show that the initial penalty of the nonlinear non-convex constraint on the relaxation domain cannot be too heavy, so that the algorithm searches in a more ideal area at the beginning, thereby ensuring the convergence effect. In this embodiment, considering the target value and the number of iterations, ρ 0 Set to 10 -3 .
[0151] Figure 3 It is a schematic diagram of the gas relaxation variables and the number of iterations of the accelerated convex decentralized optimization method for the electric-gas interconnected system data according to an embodiment of the present invention, Figure 4 A schematic diagram of gas relaxation variables and the number of iterations of the accelerated convex decentralized optimization method for electricity-gas interconnection system data according to an embodiment of the present invention.
[0152] For the dynamic adjustment coefficient v c , the penalty factor of nonlinear non-convex constraints accelerates the convergence speed in the later iteration through dynamic adjustment, such as Figure 3 As shown. For PCCP-Benders, the dynamic adjustment coefficient v c A large value will cause the penalty to increase too quickly, the number of iterations will increase, and the convergence result will be unsatisfactory. Figure 3 It can be seen that a value between 2 and 4 can achieve better convergence results and fewer iterations.
[0153] Therefore, the dynamic adjustment coefficient v c Uniformly take a suitable value v c = 2.5, the carbon trading price changes from 0 to 60 USD / t, the iteration process and time consumption of PCCP-Benders are as follows Figure 4 As shown, from Figure 4 It can be seen that the PCCP-Benders algorithm has relatively stable iteration characteristics and a short convergence time. In the first iteration, when solving the initial point, the number of iterations of the inner electrical decoupling (Benders decomposition algorithm) is relatively large, but the number of electrical decoupling is stable to 2 in subsequent iterations; the time spent on solving the initial point in the early stage of PCCP-Benders is more than 75% of the total calculation time, but the subsequent iterations can be completed in only 5 to 7 seconds. PCCP-Benders is superior in computational efficiency because the natural gas system has sufficient carrying capacity for the changes in the gas consumption of the gas unit and the electricity-to-gas output, and the objective function of the optimization problem is a linear function, so the subproblem is usually solvable and can generate more optimality cut constraints. The optimality cut constraints are superimposed in the main problem, which can quickly guide PCCP-Benders to converge. Then, the existing interior point method (IPOPT), second-order cone programming method (SCOP), mixed integer linear programming method (MILP), and PCCP centralized algorithm are used to centrally solve the multi-period dynamic power flow model of the power-gas interconnection system, and compared with the PCCP-Benders decentralized solution. The comparison results are shown in Table 2.
[0154] Table 2
[0155]
[0156] From the centralized optimization data in Table 2, it can be seen that except for the MILP method, the other three methods all obtained calculation results. Among them, the IPOPT algorithm did not find the optimal solution, and its result was a compromise solution. The optimization target value of the SOCP algorithm is the smallest, but the dynamic airflow constraint of the natural gas pipeline is only the constraint inequality. Convex relaxation is not compact, and the calculation results do not meet all constraints. For decentralized optimization of electric-pneumatic decoupling, PCCP-Benders can obtain results similar to centralized optimization, and the calculation time is not much different from that of centralized PCCP. It can be seen that the use of appropriate decomposition algorithms and acceleration strategies can obtain better results and calculation speed while protecting the privacy of the subject's internal information.
[0157] In an embodiment of the present invention, the accelerated convex decentralized optimization method of the electricity-gas interconnection system data can accurately calculate the natural gas network flow and obtain a high-quality solution for the operation of the electricity-gas interconnection system, namely, the slack variables and penalty factors, on the basis of ensuring the privacy of the power network and the natural gas network information and decentralized autonomy.
[0158] Embodiment 2:
[0159] Figure 5 A framework diagram of an accelerated convex dispersion optimization device for electrical-gas interconnection system data according to an embodiment of the present invention.
[0160] like Figure 5 As shown, an embodiment of the present invention further provides an accelerated convex dispersion optimization device for electric-gas interconnection system data, comprising: a data acquisition module 10, a constraint processing module 20, a decomposition processing module 30 and an optimization output module 40;
[0161] A data acquisition module 10, for acquiring first parameter data of the power network, second parameter data of the natural gas network, and energy coupling data of the power network and the natural gas network;
[0162] A constraint processing module 20, for applying nonlinear non-convex constraint processing to the second parameter data to obtain a slack variable of the natural gas network;
[0163] A decomposition processing module 30 is used to perform n-times iterative processing on the first parameter data, the second parameter data and the energy coupling data according to the slack variables and using the Benders decomposition algorithm to obtain a penalty function and a penalty factor;
[0164] The optimization output module 40 is used to output the optimization result when both the slack variables and the penalty function meet the convergence conditions;
[0165] The convergence condition is:
[0166]
[0167] In the formula, ε 1 and ε 2 are convergence tolerances, δ is the slack variable, W k is the penalty function for the kth iteration, W k-1 is the penalty function for k-1 iterations.
[0168] It should be noted that the modules in the device of the second embodiment correspond to the steps in the method of the first embodiment. The steps in the method of the first embodiment have been described in detail in the first embodiment, and the contents of the modules in the device will not be described in detail in the second embodiment.
[0169] Embodiment three:
[0170] An embodiment of the present invention provides an accelerated convex dispersion optimization device for electrical-pneumatic interconnection system data, including a processor and a memory;
[0171] A memory, used for storing program codes and transmitting the program codes to a processor;
[0172] A processor is used to execute the above-mentioned accelerated convex dispersion optimization method for electric-gas interconnection system data according to instructions in the program code.
[0173] It should be noted that the processor is used to execute the steps in the above-mentioned embodiment of the accelerated convex dispersion optimization method for electric-gas interconnected system data according to the instructions in the program code. Alternatively, the processor implements the functions of each module / unit in the above-mentioned system / device embodiments when executing the computer program.
[0174] Exemplarily, the computer program may be divided into one or more modules / units, one or more modules / units are stored in a memory and executed by a processor to complete the present application. One or more modules / units may be a series of computer program instruction segments capable of completing a specific function, and the instruction segments are used to describe the execution process of the computer program in a terminal device.
[0175] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that this does not constitute a limitation on the terminal device, and may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal device may also include input and output devices, network access devices, buses, etc.
[0176] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0177] The memory may be an internal storage unit of the terminal device, such as a hard disk or memory of the terminal device. The memory may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device. Furthermore, the memory may also include both an internal storage unit of the terminal device and an external storage device. The memory is used to store computer programs and other programs and data required by the terminal device. The memory may also be used to temporarily store data that has been output or is to be output.
[0178] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.
[0179] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.
[0180] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
[0181] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.
[0182] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.
[0183] As described above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. An accelerated convex decentralized optimization method for electrical-gas interconnected system data, characterized in that: The following steps are involved: S1. Obtaining first parameter data of the power network, second parameter data of the natural gas network, and energy coupling data of the power network and the natural gas network; S2. Applying nonlinear non-convex constraint processing to the second parameter data to obtain slack variables of the natural gas network; S3. According to the slack variable and using the Benders decomposition algorithm, the first parameter data, the second parameter data and the energy coupling data are iterated n times to obtain a penalty function and a penalty factor; S4. If both the slack variable and the penalty function satisfy the convergence condition, output the optimization result; Wherein, the convergence condition is: ; In the formula, and are convergence tolerances, is the slack variable, W k is the penalty function for the kth iteration, W k-1 is the penalty function for k-1 iterations; According to the slack variables and using the Benders decomposition algorithm, the first parameter data, the second parameter data and the energy coupling data are iterated n times to obtain a penalty function and a penalty factor including: The first parameter data and the energy coupling data are relaxed using a convex quadratic constraint inequality to obtain a natural gas network variable and a natural gas flow rate; If the objective function of the natural gas network has a solution, a solution vector is obtained and a coupling constraint formula is used to process the natural gas network variables and the natural gas flow to obtain a Benders cut constraint inequality; If the objective function of the natural gas network has no solution, add a gas slack variable of the coupling constraint, use the natural gas network constraint of the Benders decomposition algorithm to process the natural gas network variable, the natural gas flow and the energy coupling data, and obtain a feasible Benders cut constraint inequality; The power network constraint of the Benders decomposition algorithm is used to iterate the Benders cut constraint inequality and the feasibility Benders cut constraint inequality n times to obtain the upper bound and the lower bound of the objective function; If the upper boundary and the lower boundary meet the optimal solution condition, a penalty factor is obtained; A penalty function is formed by the penalty factor, the optimization target of the energy coupling data and the relaxation variable; The convex quadratic constraint inequality is: , ; The objective function is: ; The coupling constraint formula is: ; The Benders cut constraint inequality is: ; The natural gas network constraints are: ; The feasibility Benders cut constraint inequality is: ; The power network constraints are: ; The optimal solution condition is: ; In the formula, is the natural gas flow consumed by gas generator set a at time t, h 0,a is the first consumption characteristic coefficient of gas unit a, h 1,a is the second consumption characteristic coefficient of gas unit a, h 2,a is the third consumption characteristic coefficient of gas unit a, a is the number of the gas unit, For the collection of gas units, is the active power of gas generator set a at time t, is the natural gas network variable; F gas For gas network targets, is the slack variable between the natural gas pipeline node i and node j at position d at time t, is the set of natural gas pipeline locations, is the set between different nodes i and j, T is the set of moments, is the penalty factor after the kth iteration, b is the number of the power-to-gas unit, is a collection of power-to-gas units, is the natural gas flow consumed by gas generator set a after the nth iteration at time t, h P2G is the conversion factor of electricity to gas, is the natural gas flow converted by the power-to-gas unit b after the nth iteration at time t, is the active power consumed by the power-to-gas unit b after the nth iteration at time t, μ is a real number variable, g n is the solution vector of the nth iteration, E is the coefficient matrix of the coupling constraint, F power is the power network target, p is the decision vector of the power network target, is the result after the nth iteration, is the multiplier vector of the coupling constraint in the nth iteration; η c is the gas relaxation variable of the cth coupling constraint, v is the sum of all gas relaxation variables, U B is the upper boundary of the objective function, L B is the lower bound of the objective function, ε is the convergence parameter of the optimal solution, is the gas relaxation variable of the cth coupling constraint of gas generator set a, is the gas relaxation variable of the cth coupling constraint of the power-to-gas unit b, is the multiplier vector of the coupling constraint formula under the natural gas network constraint in the nth iteration, E is the coefficient matrix of the coupling constraint, and F power is the power network target, F gas For gas network targets.
2. The accelerated convex dispersion optimization method for electrical-gas interconnection system data according to claim 1, characterized in that: include: S5. If the slack variable and / or the penalty function does not satisfy the convergence condition, update the penalty factor and execute step S3 again.
3. The accelerated convex dispersion optimization method for electrical-gas interconnection system data according to claim 1, characterized in that: The second parameter data is processed by nonlinear non-convex constraints, and the slack variables of the natural gas network are obtained, including: The second parameter data is processed using a natural gas pipeline gas flow formula to obtain a transmission characteristic constant of the natural gas pipeline; The second parameter data and the transmission characteristic constant are processed by using non-convex constraints to obtain a constraint inequality of natural gas; The constrained inequality is subjected to a linearization process of the reduced terms by using a first-order Taylor expansion method to obtain a linearized constrained inequality; The second parameter data and the transmission characteristic constant are iterated multiple times to obtain relaxation variables that satisfy the linearized constraint inequality.
4. The accelerated convex dispersion optimization method for electrical-gas interconnection system data according to claim 3, characterized in that: The gas flow formula of the natural gas pipeline is: ; The constraint inequality is: ; ; The linearized constraint inequality is: ; In the formula, is the spatial step length between nodes i and j of the natural gas pipeline, M2 is the transmission characteristic constant, is the gas pressure at position d between nodes i and j in the natural gas pipeline at time t, is the gas pressure at position d+1 between nodes i and j in the natural gas pipeline at time t, is the gas flow between nodes i and j in the natural gas pipeline at position d at time t, is the gas pressure between nodes i and j of the natural gas pipeline at position d after k-1 iterations at time t, is the gas flow between nodes i and j in the natural gas pipeline at position d after k-1 iterations at time t, is the slack variable between the natural gas pipeline node i and node j at position d at time t, that is, the slack variable of the natural gas network, and k is a natural number greater than zero.
5. An accelerated convex dispersion optimization device for electric-gas interconnection system data using the method according to any one of claims 1 to 4, characterized in that: include: Data acquisition module, constraint processing module, decomposition processing module and optimization output module; The data acquisition module is used to acquire first parameter data of the power network, second parameter data of the natural gas network, and energy coupling data of the power network and the natural gas network; The constraint processing module is used to process the second parameter data using nonlinear non-convex constraints to obtain slack variables of the natural gas network; The decomposition processing module is used to perform n-times iterative processing on the first parameter data, the second parameter data and the energy coupling data according to the slack variables and adopt the Benders decomposition algorithm to obtain a penalty function and a penalty factor; The optimization output module is used to output the optimization result when both the slack variable and the penalty function meet the convergence condition; Wherein, the convergence condition is: ; In the formula, and are convergence tolerances, is the slack variable, W k is the penalty function for the kth iteration, W k-1 is the penalty function for k-1 iterations.
6. An accelerated convex dispersion optimization device for electrical-gas interconnection system data, characterized in that: including a processor and a memory; The memory is used to store program codes and transmit the program codes to the processor; The processor is used to execute the accelerated convex dispersion optimization method for electrical-gas interconnection system data as described in any one of claims 1 to 4 according to the instructions in the program code.
