Multi-objective optimization configuration method and system for active distribution network energy storage based on HTL-MOPSO algorithm

Abstract

The present invention discloses a multi-objective optimization configuration method and system for energy storage in an active distribution network based on the HTL-MOPSO algorithm, the method comprising: establishing a distributed energy storage model, wherein the ratio of the remaining capacity of the energy storage device to the capacity after full charge is expressed by the state of charge; establishing a multi-objective optimization configuration model for energy storage in an active distribution network with an economic index function, an active network loss function, a voltage deviation function, and a frequency deviation function as objective functions; solving the multi-objective optimization configuration model for energy storage in an active distribution network based on the HTL-MOPSO multi-objective optimization algorithm, and obtaining the optimal location and capacity determination scheme for distributed energy storage devices. The present invention has a fast convergence speed and realizes the optimal configuration of energy storage in an active distribution network.

Description

Technical Field

[0001] The present invention relates to the field of grid-connected consumption of distributed power sources in distribution networks and optimization management of distribution networks under the background of comprehensive energy, and specifically to a multi-objective optimization configuration method and system for active distribution network energy storage based on the HTL-MOPSO algorithm. Background Art

[0002] As society's demand for energy continues to increase, traditional energy sources are gradually exhausted, and the huge pollution problems caused by traditional fossil energy to the environment cannot be effectively solved. Renewable energy represented by photovoltaic and wind energy has received widespread attention. Distributed power sources that use wind and solar power generation have gradually become popular in power systems, effectively improving energy utilization and greatly reducing greenhouse gas emissions. However, due to the intermittent, random, and unpredictable characteristics of distributed power sources, large-scale grid connection will have a significant impact on the safe and stable operation of the power system, such as increasing the imbalance between supply and demand between sources and loads, causing drastic fluctuations in grid voltage, serious deviations in node voltage and system frequency, and increased line active power loss.

[0003] Energy storage is an important means of regulation to deal with the randomness, volatility and unpredictability of new energy, and is of great significance to the construction of new power systems. As an important means of regulation, energy storage can quickly adjust power and absorb and release excess electricity. It can store energy when the distributed power generation is sufficient and release energy when the power generation is insufficient, giving full play to the role of power transfer and improving the flexible regulation ability of new power systems. However, energy storage devices are currently expensive, so the optimal configuration of energy storage devices will directly affect the economic operation of the distribution network.

[0004] After searching the prior art field, it was found that the Chinese patent application number is 202211197912.3, the application publication number is CN115622101A, and the patent name is: A two-level planning method for energy storage optimization configuration to promote renewable energy consumption. In the inner optimization, the patent uses a quantum particle swarm optimization algorithm based on a chaotic search mechanism to optimize the load power of the power grid system and the output of various units. In the outer optimization, an optimization model is constructed with the location and capacity of the energy storage system as decision variables, with the goal of minimizing the amount of new energy abandoned, the full life cycle cost of the energy storage system, and the node frequency deviation. However, the particle swarm algorithm used in this patent is prone to fall into the problem of local optimality and the convergence result is poor.

[0005] The Chinese patent application number is 202211259700.3, the application publication number is CN115663854A, and the patent name is: A method for optimizing the configuration of power grid energy storage. This patent obtains the optimal energy storage configuration plan through the coordination of the lower-level optimization scheduling model of the power grid and the upper-level planning model of the power grid energy storage, and achieves maximum benefits at the planning and operation levels. However, this patent only considers the operating cost and benefit of the energy storage system when setting the objective function. In order to take into account the impact of energy storage equipment on factors such as power grid voltage and frequency, the energy storage configuration plan obtained is not the optimal solution. Summary of the invention

[0006] The purpose of the present invention is to provide a multi-objective optimization configuration method and system for active distribution network energy storage based on HTL-MOPSO algorithm, which has fast convergence speed and realizes the optimal configuration of active distribution network energy storage.

[0007] In order to solve the above technical problems, the technical solution of the present invention is:

[0008] In a first aspect, a multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm is provided, comprising:

[0009] Step S100: Establishing a distributed energy storage model, the ratio of the remaining capacity of the energy storage device to the capacity after full charge is expressed by the state of charge;

[0010] Step S200: Establishing a multi-objective optimization configuration model for active distribution network energy storage using an economic index function, an active network loss function, a voltage deviation function, and a frequency deviation function as objective functions;

[0011] Step S300: Solve the multi-objective optimization configuration model of active distribution network energy storage based on the HTL-MOPSO multi-objective optimization algorithm to obtain the optimal location and capacity solution for distributed energy storage equipment.

[0012] Furthermore, in step S100, the distributed energy storage model is:

[0013]

[0014] In formula (1), is the state of charge of the energy storage installed at node i at time t; when the energy storage device is charging, α=1, β=0; when the energy storage device is discharging, α=0, β=1; P E i SS,t is the output power of the energy storage installed at node i at time t; Install energy storage capacity for node i; η i is the efficiency of the energy storage device installed at node i; Δt is the charging and discharging time interval.

[0015] Furthermore, the constraints of the distributed energy storage model are:

[0016]

[0017]

[0018] In formulas (2) and (3), P ESS,min , P ESS,max are the lower and upper limits of the rated power of the energy storage device installed at node i, E ESS,min 、E ESS,max are the lower and upper limits of the capacity of the energy storage device installed at node i, respectively.

[0019] Furthermore, in step S200, the economic indicator function f1 is:

[0020] f1=f inp +f sta (4)

[0021]

[0022]

[0023] In formula (4), formula (5) and formula (6), f inp is the investment cost of the energy storage device, f sta is the total operating cost of the system; C E is the price related to capacity, C P is the price related to power; n ESS The number of energy storage devices installed; n DG Number of distributed generators; C i (t) and P DG (t) are the output cost and output power of the i-th distributed generator at time t;

[0024] Active network loss function is:

[0025]

[0026] In formula (7), M L is the active distribution network branch, G ij is the conductance between nodes i and j in the active distribution network, U i,t , U j,t is the voltage of the active distribution network node i, j at time t, δ ij,t is the phase angle difference between nodes i and j in the active distribution network at time t;

[0027] The voltage deviation function is:

[0028]

[0029] In formula (8), N is the number of load nodes, Ui (t) is the voltage of node i at time t, U iexp is the reference voltage of node i, ΔU i,max (t) is the maximum allowable voltage deviation of node i at time t;

[0030] The frequency deviation function is:

[0031]

[0032]

[0033] In formula (9) and formula (10), f(t) is the system frequency at time t, f exp is the system reference frequency, f COI (t) is the system inertia center frequency under the disturbance of active distribution network power electronic equipment at time t, M i is the inertia constant of generator i, f i (t) is the frequency of generator i at time t, and m is the number of generators.

[0034] Furthermore, in step S200, the active distribution network energy storage multi-objective optimization configuration model is a three-layer hierarchical structure, which is divided into three layers: outer layer, middle layer, and inner layer; wherein,

[0035] The outer optimization objective function is the economic index function f1 and the voltage deviation function f3;

[0036] The middle-level optimization objective function is the total system operating cost f sta , active network loss function f2 and system inertia center frequency f considering the disturbance of active distribution network power electronic equipment COI ;

[0037] The inner optimization objective function is the frequency deviation function f4.

[0038] Furthermore, in step S300, the method for solving the multi-objective optimization configuration model of active distribution network energy storage based on the HTL-MOPSO multi-objective optimization algorithm is:

[0039] Step S310: reading the active distribution network parameters at the start stage, including the distribution network system structure, distributed energy storage, distributed power sources, node voltages, system and generator frequency parameters and their respective constraints;

[0040] Step S320: Set the parameters of the HTL-MOPSO algorithm, and initialize the outer, middle, and inner optimizations in turn: use the installation location and capacity of the energy storage as control variables to initialize the position, speed, and position of the optimal individual of the outer population P1; use the total system operating cost, active network loss, and inertia center frequency as control variables to initialize the position, speed, and position of the optimal individual of the middle population P2; use the system frequency deviation value as the control variable to initialize the position, speed, and position of the optimal individual of the inner population P3;

[0041] Step S330: Update the middle population P2 and the inner population P3 in sequence: to minimize the objective function f sta , f2, f COI To optimize the target, the multi-objective HTL-MOPSO algorithm is used to update the population P2 until the middle layer optimization is completed, and the optimal compromise operation scheme is selected to return to the inner layer; to minimize the objective function f4 as the optimization target, the multi-objective HTL-MOPSO algorithm is used to update the population P3 until the inner layer optimization is completed, and the optimal compromise operation scheme is selected to return to the outer layer through the middle layer;

[0042] Step S340: Use the multi-objective HTL-MOPSO algorithm to update the population P1 and output the Pareto solution of the optimal energy storage configuration plan.

[0043] Furthermore, in step S300, the HTL-MOPSO algorithm is divided into two parts: the PSO algorithm and the TLBO algorithm. In the PSO optimization algorithm, the iteration formula of each particle in the population is:

[0044] v new,i =w×v i +c1*r1×(pbest i -x i )+c2×r2(gbest-x i ) (11)

[0045] x new,i =x i +v new,i (12)

[0046] In formula (11) and formula (12), v i 、x i represent the velocity and position of the ith particle, v new,i 、x new,i They represent the speed and position of the iteratively updated i-th particle, w is the inertia coefficient, c1 and c2 are the speed coefficients, r1 and r2 are random constants between 0 and 1, pbest and gbest are the individual optimal particle and the global optimal particle respectively;

[0047] In the TLBO algorithm, the joint stage expression of “teaching” and “learning” is:

[0048] x new,i =x i +D1*diff_T i +D2*diff_L i (13)

[0049] diff_T i =r3*(M_new-TF*M) (14)

[0050]

[0051] In formula (13), formula (14), and formula (15), D1 and D2 are the “teaching” and “learning” coefficients, respectively, and diff_T i 、diff_L i are the differences in the objective function values ​​of the “teaching” and “learning” stages, r3, r4, and r5 are random constants between 0 and 1, TF is the teaching factor, M is the center position of the class, M_new is the center position of the class after iterative update, and f(x i )、f(x j ) are the particle positions x i 、x j The corresponding objective function value.

[0052] Furthermore, in step S300, the method of updating the population using the multi-objective HTL-MOPSO algorithm is as follows:

[0053] (1) Set the parameters required by the algorithm, including the total number of individuals N in the population P, the size of the elite archive set N_arv, and the parameters in the PSO and TLBO algorithms;

[0054] (2) Initialize the position and velocity of each particle in the population P;

[0055] (3) According to the concept of non-dominated solution, obtain the non-dominated solution in the population P, and store the solution in the elite archive set arv;

[0056] (4) Use the PSO and TLBO algorithms to update the positions of all individuals in the population P, obtain the updated non-dominated solution nd_x of the population P, and determine whether it needs to be stored in the elite archive set arv;

[0057] (5) Determine whether arv needs to be screened. If it does, go to (6); if not, go to (7);

[0058] (6) Screening arvs through cyclic crowding sorting strategy (CCS);

[0059] (7) Determine whether the speed of all particles in population P is less than the threshold V limt , if it is less than , then perform polynomial mutation on the population P, otherwise, go to (8);

[0060] (8) Determine whether the number of iterations has reached the maximum value. If so, output the individuals in arv. Otherwise, return to (4) and re-use the PSO and TLBO algorithms to update the positions of all individuals in the population P.

[0061] In a second aspect, a multi-objective optimization configuration system for energy storage in an active distribution network based on an HTL-MOPSO algorithm is provided, comprising a memory and a processor;

[0062] The memory stores a computer program, and when the program is executed by the processor, it can implement the above-mentioned multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm.

[0063] In a third aspect, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm is implemented.

[0064] The present invention has the following beneficial effects: the present invention proposes a method and system for multi-objective optimization configuration of energy storage in an active distribution network based on the HTL-MOPSO algorithm, which comprehensively considers the operating economy of energy storage equipment, active power loss of the active distribution network, and system voltage and frequency deviation, and uses the HTL-MOPSO multi-objective optimization algorithm based on particle swarm and teaching theory to solve the multi-objective optimization configuration model of energy storage in the active distribution network, thereby achieving the optimal configuration of energy storage in the active distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 is a flow chart of the steps of the method of the present invention;

[0066] Figure 2 It is a schematic diagram of the multi-objective optimization configuration model of active distribution network energy storage of the present invention;

[0067] Figure 3 This is the specific calculation flow chart of the multi-objective HTL-MOPSO algorithm. DETAILED DESCRIPTION

[0068] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0069] Please refer to Figure 1 The present invention is a multi-objective optimization configuration method for active distribution network energy storage based on HTL-MOPSO algorithm, which includes:

[0070] Step S100: Establishing a distributed energy storage model, the ratio of the remaining capacity of the energy storage device to the capacity after full charge is expressed as the state of charge;

[0071] Step S200: Establishing a multi-objective optimization configuration model for active distribution network energy storage using an economic index function, an active network loss function, a voltage deviation function, and a frequency deviation function as objective functions;

[0072] Step S300: Solve the active distribution network energy storage multi-objective optimization configuration model based on the HTL-MOPSO multi-objective optimization algorithm to obtain the optimal location and capacity solution for distributed energy storage equipment.

[0073] The following are Figure 1 Each step in is described in detail.

[0074] In step S100, a distributed energy storage model is established, and the ratio of the remaining capacity of the energy storage device to the capacity after full charge is represented by the state of charge SOC;

[0075] The distributed energy storage model is expressed as:

[0076]

[0077] In formula (1), is the state of charge of the energy storage installed at node i at time t; when the energy storage device is charging, α=1, β=0; when the energy storage device is discharging, α=0, β=1; is the output power of the energy storage installed at node i at time t; Install energy storage capacity for node i; η i is the efficiency of the energy storage device installed at node i; Δt is the charging and discharging time interval.

[0078] The constraints of the distributed energy storage model are:

[0079]

[0080]

[0081] In formulas (2) and (3), P ESS,min , P ESS,max are the lower and upper limits of the rated power of the energy storage device installed at node i, E ESS,min 、E ESS,max are the lower and upper limits of the capacity of the energy storage device installed at node i, respectively.

[0082] In step S200, an active distribution network energy storage multi-objective optimization configuration model is established with the economic index function, active network loss function, voltage deviation function, and frequency deviation function as objective functions.

[0083] The economic indicator function f1 is:

[0084] f1=f inp +f sta (4)

[0085]

[0086]

[0087] In formula (4), formula (5) and formula (6), f inp is the investment cost of the energy storage device, f sta is the total operating cost of the system; C E is the price related to capacity, C P is the price related to power; n ESS The number of energy storage devices installed; n DG Number of distributed generators; C i (t) and P DG (t) are the output cost and output power of the i-th distributed generator at time t;

[0088] Active network loss function is:

[0089]

[0090] In formula (7), M L is the active distribution network branch, G ij is the conductance between nodes i and j in the active distribution network, U i,t , U j,t is the voltage of the active distribution network node i, j at time t, δ ij,t is the phase angle difference between nodes i and j in the active distribution network at time t;

[0091] The voltage deviation function is:

[0092]

[0093] In formula (8), N is the number of load nodes, U i (t) is the voltage of node i at time t, U iexp is the reference voltage of node i, ΔU i,max (t) is the maximum allowable voltage deviation of node i at time t;

[0094] The frequency deviation function is:

[0095]

[0096]

[0097] In formula (9) and formula (10), f(t) is the system frequency at time t, fexp is the system reference frequency, f COI (t) is the system inertia center frequency under the disturbance of active distribution network power electronic equipment at time t, M i is the inertia constant of generator i, f i (t) is the frequency of generator i at time t, and m is the number of generators.

[0098] In step S200, the multi-objective optimization configuration model of active distribution network energy storage is as follows: Figure 2 As shown in the figure, the multi-objective optimization configuration model of active distribution network energy storage is a three-layer hierarchical structure, which is divided into three layers: outer layer, middle layer and inner layer.

[0099] The outer optimization objective function is the economic index function f1 and the voltage deviation function f3;

[0100] The middle-level optimization objective function is the total system operating cost f sta , active network loss function f2 and system inertia center frequency f considering the disturbance of active distribution network power electronic equipment COI ;

[0101] The inner optimization objective function is the frequency deviation function f4.

[0102] In step S300, the multi-objective optimization configuration model of active distribution network energy storage is solved based on the HTL-MOPSO multi-objective optimization algorithm to obtain the optimal location and capacity solution of distributed energy storage equipment; including:

[0103] Step S310: reading the active distribution network parameters at the start stage, including the distribution network system structure, distributed energy storage, distributed power sources, node voltage, system and generator frequencies and other parameters and their respective constraints;

[0104] Step S320: Set the parameters of the HTL-MOPSO algorithm, such as population size, elite file set size, inertia coefficient and other parameters, and initialize the outer, middle and inner optimization in turn: use the installation location and capacity of the energy storage as control variables to initialize the position, speed and position of the optimal individual of the outer population P1; use the total system operating cost, active network loss and inertia center frequency as control variables to initialize the position, speed and position of the optimal individual of the middle population P2; use the system frequency deviation value as the control variable to initialize the position, speed and position of the optimal individual of the inner population P3;

[0105] Step S330: Update the middle population P2 and the inner population P3 in sequence: to minimize the objective function f sta , f2, f COITo optimize the target, the multi-objective HTL-MOPSO algorithm is used to update the population P2 until the middle layer optimization is completed, and the optimal compromise operation scheme is selected to return to the inner layer; to minimize the objective function f4 as the optimization target, the multi-objective HTL-MOPSO algorithm is used to update the population P3 until the inner layer optimization is completed, and the optimal compromise operation scheme is selected to return to the outer layer through the middle layer;

[0106] Step S340: Use the multi-objective HTL-MOPSO algorithm to update the population P1 and output the Pareto solution of the optimal energy storage configuration plan.

[0107] In step S300, the HTL-MOPSO algorithm is divided into two parts: the PSO algorithm and the TLBO algorithm. In the PSO optimization algorithm, the iteration formula of each particle in the population is:

[0108] v new,i =w×v i +c1*r1×(pbest i -x i )+c2×r2(gbest-x i ) (11)

[0109] x new,i =x i +v new,i (12)

[0110] In formula (11) and formula (12), v i 、x i represent the velocity and position of the ith particle, v new,i 、x new,i They represent the speed and position of the iteratively updated i-th particle, w is the inertia coefficient, c1 and c2 are the speed coefficients, r1 and r2 are random constants between 0 and 1, pbest and gbest are the individual optimal particle and the global optimal particle respectively;

[0111] In the TLBO algorithm, the joint stage expression of “teaching” and “learning” is:

[0112] x new,i =x i +D1*diff_T i +D2*diff_L i (13)

[0113] diff_T i =r3*(M_new-TF*M) (14)

[0114]

[0115] In formula (13), formula (14), and formula (15), D1 and D2 are the “teaching” and “learning” coefficients, respectively, and diff_T i 、diff_L i are the differences in the objective function values ​​of the “teaching” and “learning” stages, r3, r4, and r5 are random constants between 0 and 1, TF is the teaching factor, M is the center position of the class, M_new is the center position of the class after iterative update, and f(x i )、f(x j ) are the particle positions x i 、x j The corresponding objective function value.

[0116] In step S300, refer to Figure 3 , the method of updating the population using the multi-objective HTL-MOPSO algorithm is as follows:

[0117] (1) Set the parameters required by the algorithm, including the total number of individuals N in the population P, the size of the elite archive set N_arv, and the parameters in the PSO and TLBO algorithms;

[0118] (2) Initialize the position and velocity of each particle in the population P;

[0119] (3) According to the concept of non-dominated solution, obtain the non-dominated solution in the population P, and store the solution in the elite archive set arv;

[0120] (4) Use the PSO and TLBO algorithms to update the positions of all individuals in the population P, obtain the updated non-dominated solution nd_x of the population P, and determine whether it needs to be stored in the elite archive set arv;

[0121] (5) Determine whether arv needs to be screened. If it does, go to (6); if not, go to (7);

[0122] (6) Screening arvs through cyclic crowding sorting strategy (CCS);

[0123] (7) Determine whether the speed of all particles in population P is less than the threshold V limt , if it is less than , then perform polynomial mutation on the population P, otherwise, go to (8);

[0124] (8) Determine whether the number of iterations has reached the maximum value. If so, output the individuals in arv. Otherwise, return to (4) and re-use the PSO and TLBO algorithms to update the positions of all individuals in the population P.

[0125] The present invention is based on the multi-objective optimization configuration method of active distribution network energy storage based on the HTL-MOPSO algorithm. Aiming at the characteristics of randomness, volatility, and unpredictability after the new energy is connected to the grid, the HTL-MOPSO multi-objective optimization algorithm based on particle swarm and teaching theory is used to solve the multi-objective optimization configuration model of active distribution network energy storage, and the optimal strategy for site selection and sizing of distributed energy storage system in active distribution network is proposed, so as to realize the voltage and frequency quality support of active distribution network based on active power optimization, and the effective regulation of load peak-valley difference for real-time balance of energy production and consumption.

[0126] In another embodiment, a multi-objective optimization configuration system for active distribution network energy storage based on the HTL-MOPSO algorithm is provided, comprising a memory and a processor; wherein the memory stores a computer program, and when the program is executed by the processor, the multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm can be implemented.

[0127] In another embodiment, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm is implemented.

[0128] The parts not involved in the present invention are the same as the prior art or are implemented by using the prior art.

[0129] The above contents are further detailed descriptions of the present invention in combination with specific implementation methods, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the protection scope of the present invention.

Claims

1. A multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm, characterized by: include Step S100: Establishing a distributed energy storage model, the ratio of the remaining capacity of the energy storage device to the capacity after full charge is expressed by the state of charge; Step S200: Establishing a multi-objective optimization configuration model for active distribution network energy storage using an economic index function, an active network loss function, a voltage deviation function, and a frequency deviation function as objective functions; The economic indicator function f1 is: f1=f inp +f sta (4) In formula (4), formula (5) and formula (6), f inp is the investment cost of the energy storage device, f sta is the total operating cost of the system; C E is the price related to capacity, C P is the price related to power; n ESS The number of energy storage devices installed; n DG Number of distributed generators; C i (t) and P DG (t) are the output cost and output power of the i-th distributed generator at time t; Active network loss function is: In formula (7), M L is the active distribution network branch, G ij is the conductance between nodes i and j in the active distribution network, U i,t , U j,t is the voltage of the active distribution network node i, j at time t, δ ij,t is the phase angle difference between nodes i and j in the active distribution network at time t; The voltage deviation function is: In formula (8), N is the number of load nodes, U i (t) is the voltage of node i at time t, is the reference voltage of node i, ΔU i,max (t) is the maximum allowable voltage deviation of node i at time t; The frequency deviation function is: In formula (9) and formula (10), f(t) is the system frequency at time t, f exp is the system reference frequency, f COI (t) is the system inertia center frequency under the disturbance of active distribution network power electronic equipment at time t, M i is the inertia constant of generator i, f i (t) is the frequency of generator i at time t, and m is the number of generators; The multi-objective optimization configuration model of active distribution network energy storage is a three-layer hierarchical structure, which is divided into three layers: outer layer, middle layer and inner layer. The outer optimization objective function is the economic index function f1 and the voltage deviation function f3; The middle-level optimization objective function is the total system operating cost f sta , active network loss function f2 and system inertia center frequency f considering the disturbance of active distribution network power electronic equipment COI ; The inner optimization objective function is the frequency deviation function f4; Step S300: Solve the multi-objective optimization configuration model of active distribution network energy storage based on the HTL-MOPSO multi-objective optimization algorithm to obtain the optimal location and capacity solution of distributed energy storage equipment; the solution method is: Step S310: reading the active distribution network parameters at the start stage, including the distribution network system structure, distributed energy storage, distributed power sources, node voltages, system and generator frequency parameters and their respective constraints; Step S320: Set the parameters of the HTL-MOPSO algorithm, and initialize the outer, middle, and inner optimizations in turn: use the installation location and capacity of the energy storage as control variables to initialize the position, speed, and position of the optimal individual of the outer population P1; use the total system operating cost, active network loss, and inertia center frequency as control variables to initialize the position, speed, and position of the optimal individual of the middle population P2; use the system frequency deviation value as the control variable to initialize the position, speed, and position of the optimal individual of the inner population P3; Step S330: Update the middle population P2 and the inner population P3 in sequence: to minimize the objective function f sta , f2, f COI To optimize the target, the multi-objective HTL-MOPSO algorithm is used to update the population P2 until the middle layer optimization is completed, and the optimal compromise operation scheme is selected to return to the inner layer; to minimize the objective function f4 as the optimization target, the multi-objective HTL-MOPSO algorithm is used to update the population P3 until the inner layer optimization is completed, and the optimal compromise operation scheme is selected to return to the outer layer through the middle layer; Step S340: Use the multi-objective HTL-MOPSO algorithm to update the population P1 and output the Pareto solution of the optimal energy storage configuration plan.

2. The multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm according to claim 1 is characterized in that: In step S100, the distributed energy storage model is: In formula (1), is the state of charge of the energy storage installed at node i at time t; when the energy storage device is charging, α=1, β=0; when the energy storage device is discharging, α=0, β=1; is the output power of the energy storage installed at node i at time t; Install energy storage capacity for node i; η i is the efficiency of the energy storage device installed at node i; Δt is the charging and discharging time interval.

3. The multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm according to claim 2 is characterized in that: The constraints of the distributed energy storage model are: In formulas (2) and (3), P ESS,min , P ESS,max are the lower and upper limits of the rated power of the energy storage device installed at node i, E ESS,min 、E ESS,max are the lower and upper limits of the capacity of the energy storage device installed at node i, respectively.

4. The multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm according to claim 1 is characterized in that: In step S300, the HTL-MOPSO algorithm is divided into two parts: the PSO algorithm and the TLBO algorithm. In the PSO optimization algorithm, the iteration formula of each particle in the population is: v new,i =w×v i +c1*r1×(pbest i -x i )+c2×r2(gbest-x i ) (11) x new,i =x i +v new,i (12) In formula (11) and formula (12), v i 、x i represent the velocity and position of the ith particle, v new,i 、x new,i They represent the speed and position of the iteratively updated i-th particle, w is the inertia coefficient, c1 and c2 are the speed coefficients, r1 and r2 are random constants between 0 and 1, pbest and gbest are the individual optimal particle and the global optimal particle respectively; In the TLBO algorithm, the joint stage expression of "teaching" and "learning" is: x new,i =x i +D1*diff_T i +D2*diff_L i (13) diff_T i =r3*(M_new-TF*M) (14) In formula (13), formula (14), and formula (15), D1 and D2 are the "teaching" and "learning" coefficients, respectively, and diff_T i 、diff_L i are the differences in the objective function values ​​between the "teaching" and "learning" stages, r3, r4, and r5 are random constants between 0 and 1, TF is the teaching factor, M is the center position of the class, M_new is the center position of the class after iterative update, and f(x i )、f(x j ) are the particle positions x i 、x j The corresponding objective function value.

5. The multi-objective optimization configuration method for active distribution network energy storage based on HTL-MOPSO algorithm according to claim 1 is characterized in that: In step S300, the method of updating the population using the multi-objective HTL-MOPSO algorithm is as follows: (1) Set the parameters required by the algorithm, including the total number of individuals N in the population P, the size of the elite archive set N_arv, and the parameters in the PSO and TLBO algorithms; (2) Initialize the position and velocity of each particle in the population P; (3) According to the concept of non-dominated solution, obtain the non-dominated solution in the population P, and store the solution in the elite archive set arv; (4) Use the PSO and TLBO algorithms to update the positions of all individuals in the population P, obtain the updated non-dominated solution nd_x of the population P, and determine whether it needs to be stored in the elite archive set arv; (5) Determine whether arv needs to be screened. If it does, go to (6); if not, go to (7); (6) Screening arvs through cyclic crowding sorting strategy (CCS); (7) Determine whether the speed of all particles in population P is less than the threshold V limt , if it is less than , then perform polynomial mutation on the population P, otherwise, go to (8); (8) Determine whether the number of iterations has reached the maximum value. If so, output the individuals in arv. Otherwise, return to (4) and re-use the PSO and TLBO algorithms to update the positions of all individuals in the population P.

6. The active distribution network energy storage multi-objective optimization configuration system based on the HTL-MOPSO algorithm is characterized by: including memory and processor; Wherein, the memory stores a computer program, and when the program is executed by the processor, it can implement the multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm as described in any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the multi-objective optimization configuration method for active distribution network energy storage based on the HTL-MOPSO algorithm as described in any one of claims 1 to 5 is implemented.

Images