Control parameter optimization method and system of power distribution system

Abstract

The invention provides a control parameter optimization method and system for a power distribution system, and the method comprises the steps: obtaining operation parameters of the power distribution system, and building a network topology structure of the power distribution system; based on the power distribution system, taking minimum real power loss, minimum busbar voltage deviation, highest busbar voltage stability and minimum real power variation on a feeder line as objective functions, and combining constraint conditions to construct a multi-objective power distribution system optimization model; taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind driven generator and the discharge power of the energy storage system as decision variables of the optimization model; and solving the optimization model by adopting an image group search algorithm, and obtaining an optimal decision variable. By solving the multi-target power distribution system optimization model, the optimal control parameters meeting multiple limiting conditions are obtained, the calculation efficiency and the calculation accuracy are improved, and optimal operation control over the power distribution system is achieved.

Description

Technical Field

[0001] The present invention relates to the field of power grid control, and in particular to a control parameter optimization method and system for a power distribution system and a computer-readable storage medium. Background Art

[0002] With the rapid development of economy and industry, how to reduce the harm of resources to the environment and achieve sustainable development of resources has become the main research direction of power grid technology. Therefore, green energy has been widely used, such as wind power generation, photovoltaic power generation, etc., which have been gradually incorporated into the power grid for civil or commercial use. However, the intelligent control of the green energy distribution system is relatively complex. Its structure includes green energy and energy storage system. On the one hand, it is necessary to ensure the power supply quality of the distribution system, and at the same time, it is necessary to reduce the power loss of the distribution system as much as possible and improve the voltage stability. There are many factors that need to be considered at the same time. Therefore, how to achieve the optimal operation control of the green energy distribution system and determine the optimal control parameters are technical problems that need to be solved in this field.

[0003] In order to determine the optimal control parameters of the distribution system, the traditional method is to calculate the control parameters based on empirical values. However, this method has many defects, because the control parameter optimization of the distribution system considers many objective functions and constraints, and relying solely on empirical values ​​cannot adapt to the solution of multi-objective optimization problems. The control parameters set based on empirical values ​​are not accurate; at the same time, when setting empirical values, relevant technicians are required to have rich experience in power grid optimization, and the results will also vary greatly due to the different levels of technicians. In addition, due to the large number of parameters, technicians need more time to calculate the control parameters when setting the control parameters. Therefore, the prior art has technical problems such as low calculation efficiency and inaccurate calculation results in the optimization process of the control parameters of the distribution system. Summary of the invention

[0004] The purpose of the present invention is to provide a control parameter optimization method for a green energy distribution system, which can solve the technical problems existing in the prior art of the control parameter optimization process of the distribution system, such as low calculation efficiency and inaccurate calculation results.

[0005] A first aspect of the present invention provides a control parameter optimization method for a green energy distribution system, comprising the following steps:

[0006] S1: Acquire the operating parameters of the power distribution system and build the network topology of the power distribution system; the power distribution system includes a bus, a photovoltaic power generation system, a wind power generation system, an energy storage system and a capacitor bank;

[0007] S2: Based on the distribution system, a multi-objective distribution system optimization model is constructed in combination with constraint conditions, taking the minimum total real power loss on the feeder, the minimum bus voltage offset, the highest bus voltage stability, and the minimum real power flow change as the objective function;

[0008] S3: taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind turbine generator and the discharge power of the energy storage system as decision variables of the optimization model;

[0009] S4: using an elephant group search algorithm to solve the optimization model and obtain optimal decision variables, and using the optimal decision variables as optimal control parameters of the power distribution system.

[0010] As a further improvement of the present invention, the decision variables are mapped to elephant group members of the elephant group search algorithm, and the algorithm specifically includes the following steps:

[0011] S41: Initialize algorithm parameters: Set parameters α, β, and γ, where α is the influence factor of the elephant group leader on the group members, β is the influence factor that determines the movement of elephant group members to the center of the elephant group, and γ is the influence factor of the random movement of elephant group members. β, γ∈(0, 1), the number of elephant groups n c and the number of members of each elephant group n ci , the maximum number of iterations T, the current number of iterations TI, α = cos(π·TI / 2T), the positions of the group members are randomly initialized, and the objective function of the algorithm is set to f = {f i}, i = 1 to 4 and fitness function AF;

[0012] S42: calculating the fitness values ​​of the fitness functions of the elephant herd members in each elephant herd and sorting them to obtain the elephant herd member with the best fitness value;

[0013] S43: optimizing the objective function values ​​corresponding to the members of each herd using the Pareto optimization algorithm to obtain a Pareto optimal solution set, calculating the clustering distance of the solution set, and determining the herd member with the largest value in the result as the herd leader of the herd;

[0014] S44: updating the position of each elephant group member, and correcting the updated position using a correction algorithm, and finally replacing the elephant group member with the worst fitness function;

[0015] S45: calculating the function value of the fitness function of each elephant group member after correction, and sorting them to obtain the elephant group member with the best fitness function;

[0016] S46: Optimizing the objective function value of each herd member again using the Pareto optimization algorithm to obtain a Pareto optimal solution set, calculating the clustering distance of the solution set, and determining the herd member with the largest value in the result as the herd leader;

[0017] S47: Determine whether the number of iterations reaches the maximum value. If not, repeat steps S44-S46 until the stop condition is met;

[0018] S48: Output the optimal result.

[0019] As a further improvement of the present invention, the objective functions are f1-f4, specifically including:

[0020]

[0021] Where f1 is the actual power loss on the feeder per day, N b is the number of busbars, R ij is the feeder resistance from busbar i to busbar j, P ih is the real power flowing out of bus i at hour h, Q ih is the virtual power flowing out of bus i at hour h, V ih is the voltage of bus i at hour h, f2 is the voltage deviation of the bus every day, V ref is the reference voltage, f3 is the inverse of the bus voltage stability per day, X ij is the feeder reactance from busbar i to busbar j, f4 is the actual power change per day, P 1,h is the real power injected into the substation at hour h, P 1,ref is the reference real power.

[0022] As a further improvement of the present invention, the constraint conditions include: power conservation: Energy conservation of energy storage system: E Bk,h =E Bk,h-1 +ρ B,ch P Bk,ch,h -P Bk,dis,h / ρ B,dis ,k=1,2,…N BESS ; Transformer connection switching times limit: Capacitor bank switching times limit: Among them, P jh is the real power flowing out of bus j at hour h, Q jh is the virtual power flowing out of bus j at hour h, P Gjh is the real power injected by the generator of bus j at hour h, P Bj,dis,h is the energy storage system discharge power of bus j at hour h, PBj,ch,h is the energy storage system charging power of bus j at hour h, P Djh is the actual power load demand of bus j at hour h, Q wjh is the virtual power injected by the wind turbine of bus j at hour h, Q Cjh is the virtual power injected into the capacitor of bus j at hour h, Q Djh is the virtual power load demand of bus j at hour h, Tap h is the transformer load tap position at hour h, C s,h is the on / off state of the sth capacitor in the hth hour, 1 is on, 0 is off, N s,max is the maximum switching times of the sth capacitor, E Bk,h is the power of the kth energy storage system in the hth hour, P Bk,dis,h is the kth energy storage system discharge power at the hth hour, P Bk,ch,h is the charging power of the kth energy storage system at the hth hour, ρ B,ch is the charging efficiency of the energy storage system, ρ B,dis is the discharge efficiency of the energy storage system, N BESS is the total number of energy storage systems.

[0023] As a further improvement of the present invention, step S43 includes: the clustering distance calculation method is: where u s is the sth solution of the Pareto optimal solution set, cd s for u s The aggregation distance, f n (u s+1 ) is u s+1 Input to the nth objective function f n The resulting function value, f n min The value to minimize the nth objective function, N sol is the total number of all solutions.

[0024] As a further improvement of the present invention: the updating of the position of each elephant group member is specifically as follows:

[0025] P(c,j,T+1)=P(c,j,T)(1-α-β)+αP(c,b,T)+βP(c,ct,T)+γr

[0026] Where c is the index of the herd, P(c,j,T+1) is the updated position of member j of herd c, P(c,j,T) is the current position of member j of herd c, P(c,b,T) is the position of the leader of the herd, P(c,ct,T) is the center position of herd c, r = (2rand-1)(Pmax -P min ), rand is a random number between 0 and 1, P max and P min are the upper and lower limits of the positions of the elephant group members; the correction algorithm is: Q(c,j,T+1)=θP(c,j,T+1)+ω1Q(c,j1,T)+ω2Q(c,j2,T-1)+ω1Q(c,j3,T-2), wherein Q(c,j,T+1) is the corrected position of the elephant group member j after the T+1th iteration, Q(c,j1,T) is the corrected position of the elephant group member j1 after the Tth iteration, Q(c,j2,T-1) is the corrected position of the elephant group member j2 after the T-1th iteration, Q(c,j3,T-2) is the corrected position of the elephant group member j3 after the T-2th iteration, θ, ω1~ω3 are correction factors, θ is a random number between 0 and 1, j1~j3 are individuals randomly selected from the elephant group,

[0027] They are the fitness values ​​of the corrected positions after the Tth, T-1th, and T-2th iterations of individuals j1~j3 respectively; the replacement of the group members in step S44 is specifically: P(c,wt,T)=μP(c,b,T), where P(c,wt,T) is the position of the group member with the worst fitness function value, and μ is a random number between 0.95 and 1.05.

[0028] As a further improvement of the present invention: the step S48 comprises: calculating the total belonging function of each solution in the Pareto optimal solution set, and determining the solution with the largest total belonging function value as the optimal solution, wherein: in is the membership function of the nth objective function of the sth solution, are the upper and lower limits of the nth objective function, is the nth objective function value of the sth solution, L p is the total membership function value of the sth solution.

[0029] As a further improvement of the present invention: the fitness function in step S41 is: AF=b1f1+b2f2+b3f3+b4f4, where b1-b4 are proportional factors, 0<{b1~b4]<1.

[0030] The second aspect of the present invention provides a control parameter optimization system for a power distribution system, characterized in that it includes a parameter acquisition module, a model construction module, a decision variable determination module, and a model optimal solution module, wherein:

[0031] A parameter acquisition module, used to acquire the operating parameters of the power distribution system and build a network topology structure of the power distribution system; the power distribution system includes a bus, a photovoltaic power generation system, a wind power generation system, an energy storage system and a capacitor bank;

[0032] A model building module is used to build a multi-objective distribution system optimization model based on the distribution system, taking the minimum real power loss on the feeder, the minimum bus voltage deviation, the highest bus voltage stability, and the minimum real power change as the objective function, combined with constraint conditions;

[0033] A decision variable determination module, used for taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind turbine generator and the discharge power of the energy storage system as decision variables of the optimization model;

[0034] The model optimal solution module is used to solve the optimization model using an elephant group search algorithm and obtain optimal decision variables, and use the optimal decision variables as optimal control parameters of the power distribution system.

[0035] A third aspect of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement the control parameter optimization method of the power distribution system as described above.

[0036] Compared with the prior art, the present invention has the following beneficial effects: by solving the multi-objective distribution system optimization model, the present invention can quickly obtain the optimal control parameters that meet multiple constraints, thereby improving the calculation efficiency and calculation accuracy, and can better achieve the optimal operation control of the distribution system; at the same time, the elephant search algorithm (Elephant Search Algorithm) adopted by the present invention is an efficient multi-objective search algorithm, which has stronger search capabilities and faster convergence speed than the prior art, and is not easy to fall into local solutions. It has a good guiding role in the optimization of control parameters of the distribution system and provides a better reference for decision makers of the distribution system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a structural block diagram of a power distribution system provided in Example 1 of the present invention.

[0038] Figure 2 A flow chart of the steps for optimizing control parameters of a power distribution system provided in Example 1 of the present invention.

[0039] Figure 3 This is a flowchart of an elephant group search algorithm provided in Example 1 of the present invention.

[0040] Figure 4A control parameter optimization system for a power distribution system is provided in Example 2 of the present invention.

[0041] Figure 5 A computer-readable storage medium provided for Embodiment 3 of the present invention. DETAILED DESCRIPTION

[0042] The present invention will be further described below in conjunction with the embodiments.

[0043] Embodiment 1:

[0044] Figure 1 A structural block diagram of a power distribution system provided in an embodiment of the present invention adopts IEEE33-Bus, including 33 buses, and parallel capacitor banks are arranged on the feeders. The power distribution system also includes a wind turbine (WG), a photovoltaic generator (PV), and an energy storage system (BESS). A parallel capacitor bank is arranged on the feeder of the substation, and the voltage limit value of the bus is 0.95pu-1.05pu.

[0045] like Figure 2 As shown, this embodiment provides a control parameter optimization method for a green energy distribution system, comprising the following steps:

[0046] S1: Acquire the operating parameters of the power distribution system and build the network topology of the power distribution system; the power distribution system includes a bus, a photovoltaic power generation system, a wind power generation system, an energy storage system and a capacitor bank;

[0047] S2: Based on the distribution system, a multi-objective distribution system optimization model is constructed in combination with constraint conditions, taking the minimum total real power loss on the feeder, the minimum bus voltage offset, the highest bus voltage stability, and the minimum real power flow change as the objective function;

[0048] S3: taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind turbine generator and the discharge power of the energy storage system as decision variables of the optimization model;

[0049] S4: using an elephant group search algorithm to solve the optimization model and obtain optimal decision variables, and using the optimal decision variables as optimal control parameters of the power distribution system.

[0050] By solving the multi-objective distribution system optimization model, the present invention can quickly obtain the optimal control parameters that meet multiple constraints, improve the calculation efficiency and calculation accuracy, and better realize the optimal operation control of the distribution system; at the same time, the elephant search algorithm (Elephant Search Algorithm) adopted by the present invention is an efficient multi-objective search algorithm. Compared with the existing technology, it has stronger search ability and faster convergence speed, and is not easy to fall into local solutions. It has a good guiding role in the optimization of control parameters of the distribution system and provides a better reference for decision makers of the distribution system.

[0051] As a further improvement of the present invention, the decision variables are mapped to elephant group members of the elephant group search algorithm, and the algorithm specifically includes the following steps:

[0052] S41: Initialize algorithm parameters: Set parameters α, β, and γ, where α is the influence factor of the elephant group leader on the group members, β is the influence factor that determines the movement of elephant group members to the center of the elephant group, and γ is the influence factor of the random movement of elephant group members. β, γ∈(0, 1), the number of elephant groups n c and the number of members of each elephant group n ci , the maximum number of iterations T, the current number of iterations TI, α = cos(π·TI / 2T), the positions of the group members are randomly initialized, and the objective function of the algorithm is set to f = {f i}, i = 1 to 4 and fitness function AF;

[0053] S42: calculating the fitness values ​​of the fitness functions of the elephant herd members in each elephant herd and sorting them to obtain the elephant herd member with the best fitness value;

[0054] S43: optimizing the objective function values ​​corresponding to the members of each herd using the Pareto optimization algorithm to obtain a Pareto optimal solution set, calculating the clustering distance of the solution set, and determining the herd member with the largest value in the result as the herd leader of the herd;

[0055] S44: updating the position of each elephant group member, and correcting the updated position using a correction algorithm, and finally replacing the elephant group member with the worst fitness function;

[0056] S45: calculating the function value of the fitness function of each elephant group member after correction, and sorting them to obtain the elephant group member with the best fitness function;

[0057] S46: Optimizing the objective function value of each herd member again using the Pareto optimization algorithm to obtain a Pareto optimal solution set, calculating the clustering distance of the solution set, and determining the herd member with the largest value in the result as the herd leader;

[0058] S47: Determine whether the number of iterations reaches the maximum value. If not, repeat steps S44-S46 until the stop condition is met;

[0059] S48: Output the optimal result.

[0060] The elephant group search algorithm is a group intelligence optimization algorithm, which simulates the social structure and behavior of the elephant group, including clan update and clan separation operations, to achieve the solution of multi-objective optimization problems. In the basic elephant group search algorithm, the clan represents the local search, and the male elephants leaving the clan perform the global search, which balances the global search ability and local search accuracy of the algorithm. However, the original elephant group search algorithm does not use the optimal information of the individuals of the previous generation to guide the current and subsequent search processes when updating the individual positions, which will cause the convergence speed of the algorithm to slow down. The present invention is improved by the object group search algorithm, and the individual information of the previous generation is fully utilized when the position is updated, and the updated position is corrected by the correction algorithm, which can increase the communication between individuals of different offspring generations, improve the search ability of the algorithm, avoid falling into the local optimal solution, and improve the convergence speed of the algorithm.

[0061] As a further improvement of the present invention, the objective functions are f1-f4, specifically including:

[0062]

[0063] Where f1 is the actual power loss on the feeder per day, N b is the number of busbars, R ij is the feeder resistance from busbar i to busbar j, P ih is the real power flowing out of bus i at hour h, Q ih is the virtual power flowing out of bus i at hour h, V ih is the voltage of bus i at hour h, f2 is the voltage deviation of the bus every day, V ref is the reference voltage, f3 is the inverse of the bus voltage stability per day, X ij is the feeder reactance from busbar i to busbar j, f4 is the actual power change per day, P 1,h is the real power injected into the substation at hour h, P 1,ref is the reference real power.

[0064] As a further improvement of the present invention, the constraint conditions include: power conservation: Energy conservation of energy storage system: E Bk,h =E Bk,h-1 +ρ B,ch P Bk,ch,h -P Bk,dis,h / ρ B,dis ,k=1,2,…N BESS ; Transformer connection switching times limit: Capacitor bank switching times limit: Among them, P jh is the real power flowing out of bus j at hour h, Q jh is the virtual power flowing out of bus j at hour h, P Gjh is the real power injected by the generator of bus j at hour h, P Bj,dis,h is the energy storage system discharge power of bus j at hour h, P Bj,ch,h is the energy storage system charging power of bus j at hour h, P Djh is the actual power load demand of bus j at hour h, Q wjh is the virtual power injected by the wind turbine of bus j at hour h, Q Cjh is the virtual power injected into the capacitor of bus j at hour h, Q Djh is the virtual power load demand of bus j at hour h, Tap h is the transformer load tap position at hour h, C s,h is the on / off state of the sth capacitor in the hth hour, 1 is on, 0 is off, N s,max is the maximum switching times of the sth capacitor, E Bk,h is the power of the kth energy storage system in the hth hour, P Bk,dis,h is the kth energy storage system discharge power at the hth hour, P Bk,ch,h is the charging power of the kth energy storage system at the hth hour, ρ B,ch is the charging efficiency of the energy storage system, ρ B,dis is the discharge efficiency of the energy storage system, N BESS is the total number of energy storage systems.

[0065] As a further improvement of the present invention, step S43 includes: the clustering distance calculation method is: where u s is the sth solution of the Pareto optimal solution set, cd s for u s The aggregation distance, f n (u s+1 ) is u s+1 Input to the nth objective function f n The resulting function value, f n min The value to minimize the nth objective function, N sol is the total number of all solutions.

[0066] In step S43, the objective function values ​​of each group member need to be optimized using the Pareto optimization algorithm to obtain the Pareto optimal solution set. The Pareto optimization algorithm is a commonly used algorithm for solving multi-objective optimization. The algorithm is an optimization method based on the Pareto principle, including concepts such as non-dominated relations, Pareto optimal solutions, and Pareto frontiers. Pareto proposed the concept of non-dominated solutions for multiple objectives in 1986, which is defined as: Assuming that any two solutions s1 and s2 are better than s2 for all objective functions, we call s1 dominates s2. If s1 is not dominated by other solutions, s1 is called a non-dominated solution, also known as the Pareto optimal solution. When there are multiple objectives, due to conflicts and incomparable phenomena between objectives, a solution is the best in a certain objective and may be the worst in other objectives. These will inevitably weaken the solution of at least one other objective function as a non-dominated solution while improving any objective function. Making at least one objective better without making any objective worse is a Pareto improvement. The Pareto optimal state is a state where there can be no more Pareto improvements. A set of optimal solutions to an objective function is called a Pareto optimal set. The surface formed by the optimal set in space is called the Pareto frontier.

[0067] As a further improvement of the present invention: the updating of the position of each elephant group member is specifically as follows:

[0068] P(c,j,T+1)=P(c,j,T)(1-α-β)+αP(c,b,T)+βP(c,ct,T)+γr

[0069] Where c is the index of the herd, P(c,j,T+1) is the updated position of member j of herd c, P(c,j,T) is the current position of member j of herd c, P(c,b,T) is the position of the leader of the herd, P(c,ct,T) is the center position of herd c, r = (2rand-1)(P max -P min ), rand is a random number between 0 and 1, P max and P min are the upper and lower limits of the positions of the elephant group members; the correction algorithm is:

[0070] Q(c,j,T+1)=θP(c,j,T+1)+ω1Q(c,j1,T)+ω2Q(c,j2,T-1)+ω1Q(c,j3,T-2), where Q(c,j,T+1) is the corrected position of herd member j after the T+1th iteration, Q(c,j1,T) is the corrected position of herd member j1 after the Tth iteration, Q(c,j2,T-1) is the corrected position of herd member j2 after the T-1th iteration, Q(c,j3,T-2) is the corrected position of herd member j3 after the T-2th iteration, θ, ω1~ω3 are correction factors, θ is a random number between 0 and 1, j1~j3 are individuals randomly selected from the herd,

[0071] They are the fitness values ​​of the corrected positions after the Tth, T-1th, and T-2th iterations of individuals j1~j3 respectively; the replacement of the group members in step S44 is specifically: P(c,wt,T)=μP(c,b,T), where P(c,wt,T) is the position of the group member with the worst fitness function value, and μ is a random number between 0.95 and 1.05.

[0072] As a further improvement of the present invention: the step S48 comprises: calculating the total belonging function of each solution in the Pareto optimal solution set, and determining the solution with the largest total belonging function value as the optimal solution, wherein: in is the membership function of the nth objective function of the sth solution, are the upper and lower limits of the nth objective function, is the nth objective function value of the sth solution, L p is the total membership function value of the sth solution.

[0073] As a further improvement of the present invention: the fitness function in step S41 is: AF=b1f1+b2f2+b3f3+b4f4, where b1-b4 are proportional factors, 0<{b1~b4}<1.

[0074] Embodiment 2:

[0075] A second aspect of the present invention provides a control parameter optimization system for a power distribution system, see Figure 4 As shown, it is characterized by comprising a parameter acquisition module, a model construction module, a decision variable determination module, and a model optimal solution module, wherein:

[0076] A parameter acquisition module, used to acquire the operating parameters of the power distribution system and build a network topology structure of the power distribution system; the power distribution system includes a bus, a photovoltaic power generation system, a wind power generation system, an energy storage system and a capacitor bank;

[0077] A model building module is used to build a multi-objective distribution system optimization model based on the distribution system, taking the minimum real power loss on the feeder, the minimum bus voltage deviation, the highest bus voltage stability, and the minimum real power change as the objective function, combined with constraint conditions;

[0078] A decision variable determination module, used for taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind turbine generator and the discharge power of the energy storage system as decision variables of the optimization model;

[0079] The model optimal solution module is used to solve the optimization model using an elephant group search algorithm and obtain optimal decision variables, and use the optimal decision variables as optimal control parameters of the power distribution system.

[0080] Embodiment 3:

[0081] A third aspect of the present invention provides a computer readable storage medium. Figure 5 As shown, it is characterized in that the computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement the control parameter optimization method of the power distribution system as described above.

[0082] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0083] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A control parameter optimization method for a power distribution system, comprising the following steps: S1: Acquire the operating parameters of the power distribution system and build the network topology of the power distribution system; the power distribution system includes a bus, a photovoltaic power generation system, a wind power generation system, an energy storage system and a capacitor bank; S2: Based on the distribution system, a multi-objective distribution system optimization model is constructed in combination with constraint conditions, taking the minimum real power loss on the feeder, the minimum bus voltage deviation, the highest bus voltage stability, and the minimum real power variation as the objective function; S3: taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind turbine generator and the discharge power of the energy storage system as decision variables of the optimization model; S4: using an elephant group search algorithm to solve the optimization model and obtain optimal decision variables, and using the optimal decision variables as optimal control parameters of the power distribution system.

2. The method according to claim 1, characterized in that: The decision variables are mapped to elephant group members of the elephant group search algorithm, and the algorithm specifically includes the following steps: S41: Initialize algorithm parameters: Set parameters α, β, and γ, where α is the influence factor of the elephant group leader on the group members, β is the influence factor that determines the movement of elephant group members to the center of the elephant group, and γ is the influence factor of the random movement of elephant group members. β, γ∈(0, 1), the number of elephant groups n c and the number of members of each elephant group n ci , the maximum number of iterations T, the current number of iterations TI, α = cos(π·TI / 2T), the positions of the group members are randomly initialized, and the objective function of the algorithm is set to f = {f i }, i = 1 to 4 and fitness function AF; S42: calculating the fitness values ​​of the fitness functions of the elephant herd members in each elephant herd and sorting them to obtain the elephant herd member with the best fitness value; S43: optimizing the objective function values ​​corresponding to the members of each herd using the Pareto optimization algorithm to obtain a Pareto optimal solution set, calculating the clustering distance of the solution set, and determining the herd member with the largest value in the result as the herd leader of the herd; S44: updating the position of each elephant group member, and correcting the updated position using a correction algorithm, and finally replacing the elephant group member with the worst fitness function; S45: calculating the function value of the fitness function of each elephant group member after correction, and sorting them to obtain the elephant group member with the best fitness function; S46: Optimizing the objective function value of each herd member again using the Pareto optimization algorithm to obtain a Pareto optimal solution set, calculating the clustering distance of the solution set, and determining the herd member with the largest value in the result as the herd leader; S47: Determine whether the number of iterations reaches the maximum value. If not, repeat steps S44-S46 until the stop condition is met; S48: Output the optimal result.

3. The method according to claim 2, characterized in that: The objective functions are f1 to f4, specifically including: Where f1 is the actual power loss on the feeder per day, N b is the number of busbars, R ij is the feeder resistance from busbar i to busbar j, P ih is the real power flowing out of bus i at hour h, Q ih is the virtual power flowing out of bus i at hour h, V ih is the voltage of bus i at hour h, f2 is the voltage deviation of the bus every day, V ref is the reference voltage, f3 is the inverse of the bus voltage stability per day, X ij is the feeder reactance from busbar i to busbar j, f4 is the actual power change per day, P 1,h is the real power injected into the substation at hour h, P 1,ref is the reference real power.

4. The method according to claim 3, characterized in that: The constraints include: Conservation of power: Energy conservation of energy storage system: E Bk,h =E Bk,h-1 +ρ B,ch P Bk,ch,h -P Bk,dis,h / ρ B,dis ,k=1,2,…N BESS ; Transformer connection switching times limit: Capacitor bank switching times limit: Among them, P jh is the real power flowing out of bus j at hour h, Q jh is the virtual power flowing out of bus j at hour h, P Gjh is the real power injected by the generator of bus j at hour h, P Bj,dis,h is the energy storage system discharge power of bus j at hour h, P Bj,ch,h is the energy storage system charging power of bus j at hour h, P Djh is the actual power load demand of bus j at hour h, Q wjh is the virtual power injected by the wind turbine of bus j at hour h, Q Cjh is the virtual power injected into the capacitor of bus j at hour h, Q Djh is the virtual power load demand of bus j at hour h, Tap h is the transformer load tap position at hour h, C s,h is the on / off state of the sth capacitor in the hth hour, 1 is on, 0 is off, N s,max is the maximum switching times of the sth capacitor, E Bk,h is the power of the kth energy storage system in the hth hour, P Bk,dis,h is the kth energy storage system discharge power at the hth hour, P Bk,ch,h is the charging power of the kth energy storage system at the hth hour, ρ B,ch is the charging efficiency of the energy storage system, ρ B,dis is the discharge efficiency of the energy storage system, N BESS is the total number of energy storage systems.

5. The method according to claim 2, characterized in that: Step S43 includes: the clustering distance calculation method is: where u s is the sth solution of the Pareto optimal solution set, cd s for u s The aggregation distance, f n (u s+1 ) is u s+1 Input to the nth objective function f n The obtained function value is The value to minimize the nth objective function, N sol is the total number of all solutions.

6. The method according to claim 5, wherein the updating of the position of each elephant group member is specifically: P(c,j,T+1)=P(c,j,T)(1-α-β)+αP(c,b,T)+βP(c,ct,T)+γr Where c is the index of the herd, P(c,j,T+1) is the updated position of member j of herd c, P(c,j,T) is the current position of member j of herd c, P(c,b,T) is the position of the leader of the herd, P(c,ct,T) is the center position of herd c, r = (2rand-1)(P max -P min ), rand is a random number between 0 and 1, P max and P min are the upper and lower limits of the positions of the elephant group members; the correction algorithm is: Q(c,j,T+1)=θP(c,j,T+1)+ω1Q(c,j1,T)+ω2Q(c,j2,T-1)+ω1Q(c,j3,T-2), where Q(c,j,T+1) is the corrected position of herd member j after the T+1th iteration, Q(c,j1,T) is the corrected position of herd member j1 after the Tth iteration, Q(c,j2,T-1) is the corrected position of herd member j2 after the T-1th iteration, Q(c,j3,T-2) is the corrected position of herd member j3 after the T-2th iteration, θ, ω1~ω3 are correction factors, θ is a random number between 0 and 1, j1~j3 are individuals randomly selected from the herd, They are the fitness values ​​of the corrected positions after the Tth, T-1th, and T-2th iterations of individuals j1~j3 respectively; the replacement of the group members in step S44 is specifically: P(c,wt,T)=μP(c,b,T), where P(c,wt,T) is the position of the group member with the worst fitness function value, and μ is a random number between 0.95 and 1.

05.

7. The method according to claim 6, characterized in that: The step S48 comprises: calculating the total belonging function of each solution in the Pareto optimal solution set, and determining the solution with the largest total belonging function value as the optimal solution, wherein: in is the membership function of the nth objective function of the sth solution, are the upper and lower limits of the nth objective function, is the nth objective function value of the sth solution, L p is the total membership function value of the sth solution.

8. The method according to claim 3, characterized in that: The fitness function in step S41 is: AF=b1f1+b2f2+b3f3+b4f4, where b1-b4 are proportional factors. 0<{b1~b4}<1.

9. A control parameter optimization system for a power distribution system, characterized in that: It includes parameter acquisition module, model construction module, decision variable determination module and model optimal solution module, among which: A parameter acquisition module, used to acquire the operating parameters of the power distribution system and build a network topology structure of the power distribution system; the power distribution system includes a bus, a photovoltaic power generation system, a wind power generation system, an energy storage system and a capacitor bank; A model building module is used to build a multi-objective distribution system optimization model based on the distribution system, taking the minimum real power loss on the feeder, the minimum bus voltage deviation, the highest bus voltage stability, and the minimum real power change as the objective function, combined with constraint conditions; A decision variable determination module, used for taking the transformer tap position, the virtual power of the input capacitor bank, the virtual power output by the wind turbine generator and the discharge power of the energy storage system as decision variables of the optimization model; The model optimal solution module is used to solve the optimization model using an elephant group search algorithm and obtain optimal decision variables, and use the optimal decision variables as optimal control parameters of the power distribution system.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and the computer program can be executed by a processor to implement the control parameter optimization method of the power distribution system as described in claims 1-8.

Images