A multi-stage planning method and system for active distribution network considering evolution-driven

Abstract

The present invention discloses a multi-stage planning method and system for active distribution networks considering evolutionary driving. It comprehensively considers the common driving factors, market driving factors, and innovative technology driving factors of the evolution of the power system form, establishes a multi-stage planning model for the distribution network including multi-stage constraint conditions, and uses the SGO algorithm to obtain the optimal solution of the multi-stage planning model for the distribution network, thereby obtaining an optimal distribution network planning scheme. While reducing the construction and renovation costs of the distribution network, it improves the utilization rate of new energy and the adaptability of the planning scheme in the way of collaborative planning of energy storage and new energy, and solves the technical problems that the existing multi-stage planning of active distribution networks considering evolutionary driving lacks consideration of the driving force factors for the evolution of the power system form and the adaptability of the planning scheme is weak.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network planning, and in particular, to a multi-stage planning method and system for active distribution network considering evolutionary drive. Background Art

[0002] Distribution network planning refers to determining when, where, and what types of lines, substations, or distributed generation (DG) and other equipment to build to meet the requirements of future annual load growth and grid development. The main goal of traditional distribution network planning is to determine the time, location, and capacity of distribution network investment at the minimum cost while ensuring the safe operation of the distribution network to meet the load growth demand and new load. The distribution network planning results directly affect the investment, revenue, and the safety, economy, and stability of the distribution network operation in future years.

[0003] The structure form of the power system refers to the connection organization form and interaction mode of the components of the power system and its participants. The driving forces for the evolution of the power system form come from three aspects: the driving factors of government-led incentive behaviors, the driving factors of market management behaviors, and the driving factors of innovative technologies. The evolution of the power system form plays a crucial role in distribution network planning, affecting the utilization rate and carrying capacity of new energy in the distribution network. However, in the existing multi-stage planning of active distribution network considering evolutionary drive, the driving force factors of power system form evolution are lacking, and there is room for improvement in the adaptability of the planning scheme. Summary of the Invention

[0004] The present invention provides a multi-stage planning method and system for active distribution network considering evolutionary drive, which is used to solve the technical problems that the existing multi-stage planning of active distribution network considering evolutionary drive lacks the driving force factors of power system form evolution and the adaptability of the planning scheme is weak.

[0005] In view of this, the first aspect of the present invention provides a multi-stage planning method for active distribution network considering evolutionary drive, including:

[0006] Constructing a multi-stage planning model for the distribution network, where the multi-stage planning model for the distribution network includes a distribution network planning objective function considering the driving factors of power system evolution and multi-stage constraint conditions, and the driving factors of power system evolution include public driving factors, market driving factors, and innovative technology driving factors;

[0007] Obtaining target parameters, where the target parameters include the predicted load of the distribution network within the planning stage and the network structure parameters of the current area to be planned;

[0008] According to the target parameters, using the SGO algorithm to solve the multi-stage planning model for the distribution network to obtain the optimal solution of the multi-stage planning model for the distribution network;

[0009] Plan the distribution network according to the distribution network planning scheme corresponding to the optimal solution of the multi-stage distribution network planning model.

[0010] Optionally, the objective function of the distribution network planning considering the driving factors of power system evolution is:

[0011] min(C line +C loss +C dg +C ess )

[0012] Among them, C line is the cost of grid framework upgrade, C dg is the investment cost of distributed power generation equipment, C loss is the annual cost of network loss, C ess is the installation cost of distributed energy storage equipment;

[0013]

[0014] Among them, is a 0-1 decision variable used to represent whether a new branch ij is built during the planning stage T h , T h is the planning stage, is the set of planning stages, R is the capital recovery factor, is the construction cost per unit length of the line, Ω line is the set of positions of the lines to be newly built, η line is the line operation and maintenance cost coefficient, l ij is the length of the newly built line;

[0015]

[0016] Among them, is a 0-1 decision variable used to represent whether a distributed power generation is installed at node i during the planning stage T h , is the installation cost per unit capacity of the distributed power generation, Ω dg is the set of nodes where the distributed power generation is installed, r is the inflation rate, T is the service life of the distributed power generation equipment, S dg is the technical maturity of the distributed power generation equipment, B dg is the photovoltaic subsidy coefficient, is the capacity of the distributed power generation installed at node i during the planning stage T h ;

[0017]

[0018] Among them, Ω br is the set of all branches in the distribution network, r ijis the resistance of line ij, and are the active and reactive powers flowing through line ij at time t within the planning period T, h λ is the electricity price at time t, Δt is the time interval, sc is the scenario day, and N t is the number of scenario days; sc

[0019]

[0020] where C e and C p are the investment costs per unit capacity and per unit power of the energy storage system, respectively. S ess is the technical maturity of the energy storage device, is a 0-1 decision variable used to indicate whether an energy storage is installed at node i within the planning period T h Ω is the set of energy storage installation nodes, ess is the capacity of the energy storage installed at node i during the planning period T h is the maximum charge and discharge power of the energy storage installed at node i during the planning period T h

[0021] Optionally, the multi-stage constraint conditions include:

[0022]

[0023]

[0024]

[0025] g ∈ G R

[0026]

[0027]

[0028]

[0029]

[0030]

[0031]

[0032]

[0033]

[0034] where P​​​​i,sc,t and Q i,sc,t are the active injection power and reactive injection power of node i at time t in scenario sc respectively, Ω0 is the set of nodes, G ij and B ij are the conductance and susceptance of branch ij respectively, θ ij,sc,t is the voltage phase angle difference between node i and node j at time t in scenario sc, I l,sc,t is the current of branch l at time t in scenario sc, is the upper limit of the current allowed to pass through branch l, U i,sc,t is the voltage of node i at time t in scenario sc, U j,sc,t is the voltage of node j at time t in scenario sc, is the lower limit of the voltage of node i, is the upper limit of the voltage of node i, g is the target network, G R is the set composed of all possible radial connected networks, ESS i,t is the stored electricity of the energy storage system of node i at time t, is the active power generated by the energy storage device of node i at time t, E ess,i is the capacity of the energy storage system of node i, SOC max and SOC min are the upper and lower limits of the state of charge of the energy storage system respectively, is the rated power of the energy storage system of node i, p dg is the active power of the distributed power source, and are the lower and upper limits of the active power of the distributed power source respectively, and are the maximum number of new lines allowed to be built, the number of distributed power sources, and the maximum number of energy storage systems during the entire planning stage respectively.

[0035] Optionally, according to the target parameters, the SGO algorithm is used to solve the multi-stage distribution network planning model to obtain the optimal solution of the multi-stage distribution network planning model, including:

[0036] The population is initialized based on the Logitic chaotic map, and the fitness value of each individual is calculated according to the multi-stage distribution network planning model to determine the initial population position. Among them, each individual represents a set of planning results, including the number of expanded lines, the location of distributed power sources, and the location of energy storage;

[0037] The population is updated based on the population update formula in the SGO algorithm improvement stage, and the new individuals are evaluated. If the new solution is better than the current solution, the new solution is replaced with the current solution; otherwise, no update operation is performed;

[0038] Determine the global optimal solution of the population in this iteration;

[0039] Update the population according to the population update formula in the SGO algorithm acquisition stage, evaluate the new individuals. If the new solution is better than the global optimal solution, replace the current solution with the new solution; otherwise, no update operation is performed.

[0040] Judge whether the maximum number of iterations is reached. If so, output the global optimal solution; otherwise, return to calculate the fitness value of each individual according to the multi-stage planning model of the distribution network.

[0041] Optionally, initialize the population based on the Logistic chaotic mapping, including:

[0042] Randomly generate a D-dimensional vector x1=(x 11 ,x 12 ,...,x 1D ) with numerical components between (0,1), and generate L + 2N vectors according to the expression of the Logistic chaotic mapping:

[0043] x m+1,n =f(x m,n ), m = 1, 2,..., L + 2N; n = 1, 2,..., D

[0044] where N is the number of individuals, L is a preset integer, and the expression of the Logistic chaotic mapping is:

[0045]

[0046] where x i is a random quantity between [0,1], μ is the Logistic parameter, and 3.5699 ≤ μ ≤ 4;

[0047] Take the last 2N vectors: x L+1 ,x L+2 ,...,x L+2N ;

[0048] According to z mn =min n +x mn (max n -min n ), map the nth component of x mn to the interval [min n ,max n , where z mn is the gene vector of the mth individual with dimension D after mapping;

[0049] Calculate the fitness value of each individual according to the multi-stage planning model of the distribution network, and select the N vectors with the smallest fitness value as the initial population positions.

[0050] In the second aspect of the present invention, a multi-stage planning system for active distribution network considering evolutionary drive is provided, including:

[0051] A modeling unit for constructing a multi-stage planning model of the distribution network. The multi-stage planning model of the distribution network includes a distribution network planning objective function and multi-stage constraint conditions considering the evolutionary drive factors of the power system. The evolutionary drive factors of the power system include common drive factors, market drive factors, and innovative technology drive factors;

[0052] An acquisition unit for acquiring target parameters, where the target parameters include the predicted load of the distribution network within the planning stage and the network structure parameters of the current area to be planned;

[0053] A model solving unit for solving the multi-stage planning model of the distribution network using the SGO algorithm according to the target parameters to obtain the optimal solution of the multi-stage planning model of the distribution network;

[0054] A planning unit for planning the distribution network according to the distribution network planning scheme corresponding to the optimal solution of the multi-stage planning model of the distribution network.

[0055] Optionally, the distribution network planning objective function considering the evolutionary drive factors of the power system is:

[0056] min(C line +C loss +C dg +C ess )

[0057] Where C line is the grid upgrade cost, C dg is the investment cost of distributed power generation equipment, C loss is the annual cost of network loss, C ess is the installation cost of distributed energy storage equipment;

[0058]

[0059] Where is a 0-1 decision variable used to represent whether to newly build a branch ij within the planning stage T h , T h is the planning stage, is the set of planning stages, R is the capital recovery factor, is the construction cost per unit length of the line, Ω line is the set of positions of the lines to be newly built, η line is the line operation and maintenance cost coefficient, l ij is the length of the newly built line;

[0060]

[0061] Among them, is the 0-1 decision variable used to represent whether to install a distributed power source at node i during the planning stage T h ; is the installation cost per unit capacity of the distributed power source, Ω dg is the set of nodes where the distributed power source is installed, r is the inflation rate, T is the service life of the distributed power source equipment, S dg is the technological maturity of the distributed power source equipment, B dg is the photovoltaic subsidy coefficient, is the capacity of the distributed power source installed at node i during the planning stage T h ;

[0062]

[0063] Among them, Ω br is the set of all branches in the distribution network, r ij is the resistance of line ij, and are the active and reactive powers flowing through line ij at time t during the planning stage T h respectively, λ t is the electricity price at time t, Δt is the time interval, sc is the scenario day, N sc is the number of scenario days;

[0064]

[0065] Among them, C e and C p are the unit capacity and unit power investment costs of the energy storage system respectively, S ess is the technological maturity of the energy storage equipment, is the 0-1 decision variable used to represent whether to install an energy storage at node i during the planning stage T h ; ess is the set of nodes where the energy storage is installed, is the capacity of the energy storage installed at node i during the planning stage T h ; is the maximum charge and discharge power of the energy storage installed at node i during the planning stage T h ;

[0066] Optionally, the multi-stage constraint conditions include:

[0067]

[0068]

[0069]

[0070] g ∈ GR

[0071]

[0072]

[0073]

[0074]

[0075]

[0076]

[0077]

[0078]

[0079] where P i,sc,t and Q i,sc,t are the active injection power and reactive injection power of node i at time t on scenario day sc respectively, Ω0 is the set of nodes, G ij and B ij are the conductance and susceptance of branch ij respectively, θ ij,sc,t is the voltage phase angle difference between node i and node j at time t on scenario day sc, I l,sc,t is the current of branch l at time t on scenario day sc, is the upper limit of the current allowed to pass through branch l, U i,sc,t is the voltage of node i at time t on scenario day sc, U j,sc,t is the voltage of node j at time t on scenario day sc, is the lower limit of the voltage of node i, is the upper limit of the voltage of node i, g is the target network, G R is the set composed of all possible radial connected networks, ESS i,t is the stored power of the energy storage system of node i at time t, is the active power generated by the energy storage device of node i at time t, E ess,i is the capacity of the energy storage system of node i, SOC max and SOC min are the upper and lower limits of the state of charge of the energy storage system respectively, is the rated power of the energy storage system of node i, p dg is the active power of the distributed power source, and are the lower and upper limits of the active power of the distributed power source respectively, and They are respectively the maximum number of new lines allowed during the entire planning stage, the number of distributed power sources, and the maximum number of energy storage systems.

[0080] Optionally, the model solving unit includes:

[0081] An initialization subunit, configured to perform population initialization based on the Logitic chaotic map, calculate the fitness value of each individual according to the multi-stage distribution network planning model, and determine the initial population position, where each individual represents a set of planning results, including the number of expanded lines, the positions of distributed power sources, and the positions of energy storage;

[0082] A first update subunit, configured to update the population according to the population update formula in the improvement stage of the SGO algorithm, evaluate the new individuals, and if the new solution is better than the current solution, replace the current solution with the new solution, otherwise, no update operation is performed;

[0083] An optimal solution subunit, configured to determine the global optimal solution of the population in this iteration;

[0084] A second update subunit, configured to update the population according to the population update formula in the obtaining stage of the SGO algorithm, evaluate the new individuals, and if the new solution is better than the global optimal solution, replace the current solution with the new solution, otherwise, no update operation is performed;

[0085] An output subunit, configured to determine whether the maximum number of iterations is reached. If so, output the global optimal solution, otherwise, return the fitness value of each individual calculated according to the multi-stage distribution network planning model.

[0086] Optionally, the initialization subunit is specifically configured to:

[0087] Randomly generate a D-dimensional vector x1=(x 11 ,x 12 ,...,x 1D ) with numerical components between (0,1), and generate L + 2N vectors according to the expression of the Logitic chaotic map:

[0088] x m+1,n =f(x m,n ), m = 1, 2,..., L + 2N; n = 1, 2,..., D

[0089] where N is the number of individuals, L is a preset integer, and the expression of the Logitic chaotic map is:

[0090]

[0091] where x i is a random quantity between [0,1], μ is the Logitic parameter, and 3.5699 ≤ μ ≤ 4;

[0092] Take the last 2N vectors: x L+1 , x L+2 ,..., x L+2N ;

[0093] According to z mn = min n + x mn (max n - min n ), map the n-th component of x mn to the interval [min n , max n , where z mn is the m-th individual gene vector of dimension D after mapping;

[0094] Calculate the fitness value of each individual according to the multi-stage planning model of the distribution network, and select the N vectors with the smallest fitness value as the initial population positions.

[0095] As can be seen from the above technical solutions, the multi-stage planning method and system for active distribution network considering evolutionary drive provided by the present invention have the following advantages:

[0096] The multi-stage planning method and system for active distribution network considering evolutionary drive provided by the present invention comprehensively consider the common driving factors, market driving factors and innovative technology driving factors of the evolution of the power system form, establish a multi-stage planning model of the distribution network including multi-stage constraint conditions, and use the SGO algorithm to obtain the optimal solution of the multi-stage planning model of the distribution network, and obtain the optimal distribution network planning scheme. While reducing the construction and transformation costs of the distribution network, in the way of coordinated planning of energy storage and new energy, it improves the utilization rate of new energy and the adaptability of the planning scheme, and solves the technical problems that the existing multi-stage planning of active distribution network considering evolutionary drive lacks consideration of the driving force factors of the evolution of the power system form and the adaptability of the planning scheme is weak.

[0097] The multi-stage planning method and system for active distribution network considering evolutionary drive provided by the present invention simultaneously consider power balance constraints, branch current constraints, node voltage constraints, radial topology constraints, energy storage system operation constraints, distributed power output constraints and multi-stage planning constraints, improve the rationality and adaptability of the distribution network planning scheme, and can effectively improve the energy utilization efficiency of the distribution network.

[0098] At the same time, the multi-stage planning method and system for active distribution network considering evolutionary drive provided by the present invention use the properties of Logitic chaotic motion to initialize the population. The randomness, ergodicity and sensitivity to initial conditions of chaotic motion improve the diversity of the initial population and provide a guarantee for solving the global optimal distribution network planning scheme. Description of the Drawings

[0099] To more clearly illustrate the technical solutions in the embodiments of the present invention or in the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0100] Figure 1 It is a schematic flow chart of a multi-stage planning method for an active distribution network considering evolutionary drive provided in the present invention;

[0101] Figure 2 It is a schematic structural diagram of a multi-stage planning model for an active distribution network considering evolutionary drive provided in the present invention;

[0102] Figure 3 It is a schematic flow chart of using the SGO algorithm to solve the multi-stage planning model of the distribution network provided in the present invention;

[0103] Figure 4 It is a schematic structural diagram of a multi-stage planning system for an active distribution network considering evolutionary drive provided in the present invention. Specific embodiments

[0104] In order to enable those skilled in the art to better understand the solution of the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0105] For ease of understanding, please refer to Figure 1 , an embodiment of a multi-stage planning method for an active distribution network considering evolutionary drive provided in the present invention includes:

[0106] Step 101: Construct a multi-stage planning model for the distribution network. The multi-stage planning model for the distribution network includes a distribution network planning objective function considering the evolutionary drive factors of the power system and multi-stage constraint conditions. The evolutionary drive factors of the power system include common drive factors, market drive factors, and innovative technology drive factors.

[0107] It should be noted that the driving factors for the evolution of the power system include public driving factors, market driving factors, and innovative technology driving factors. The public driving factors are mainly incentive behaviors led by the government, corresponding to the subsidy coefficient for the construction of distributed power sources. The market driving factors correspond to the time-of-use electricity price coefficient, and the innovative technology driving factors correspond to the technology maturity coefficient and the capital recovery coefficient of distributed generation (DG) and energy storage systems (ESS).

[0108] As Figure 2 shown, in one embodiment, the objective function of the distribution network planning considering the driving factors for the evolution of the power system is as follows:

[0109] min(C line +C loss +C dg +C ess )

[0110] where C line is the cost of grid framework upgrade, C dg is the investment cost of distributed power source equipment, C loss is the annual cost of network losses, C ess is the installation cost of distributed energy storage equipment;

[0111]

[0112] where is a 0-1 decision variable used to represent whether a new branch ij is built during the planning stage T h , T h is the planning stage, is the set of planning stages, R is the capital recovery coefficient, is the construction cost per unit length of the line, Ω line is the set of locations of lines to be newly built, η line is the line operation and maintenance cost coefficient, l ij is the length of the newly built line;

[0113]

[0114] where is a 0-1 decision variable used to represent whether a distributed power source is installed at node i during the planning stage T h , is the installation cost per unit capacity of the distributed power source, Ω dg is the set of nodes where the distributed power source is installed, r is the inflation rate, T is the service life of the distributed power source equipment, S dg is the technology maturity of the distributed power source equipment, B dgis the photovoltaic subsidy coefficient, is the planning stage T h The capacity of the distributed power source installed at node i;

[0115] According to the season and the time period in the typical day, the load time series characteristics are divided into 4 typical day scenarios, representing the four seasons of spring, summer, autumn, and winter respectively. The annual cost of network loss can be expressed as:

[0116]

[0117] Among them, Ω br is the set of all branches in the distribution network, r ij is the resistance of line ij, and are the active and reactive powers flowing through line ij at time t during the planning stage T h respectively, λ t is the electricity price at time t, Δt is the time interval, and in the present invention, the time interval Δt is set to 1h, sc is the scenario day, N sc is the number of scenario days;

[0118]

[0119] Among them, C e and C p are the unit capacity and unit power investment costs of the energy storage system respectively, S ess is the technical maturity of the energy storage device, is used to represent whether an energy storage is installed at node i during the planning stage T h The 0-1 decision variable, Ω ess is the set of energy storage installation nodes, is the planning stage T h The capacity of the energy storage installed at node i, is the planning stage T h The maximum charge and discharge power of the energy storage installed at node i.

[0120] As Figure 2 shown, the multi-stage constraint conditions include: power balance constraint, branch current constraint, node voltage constraint, radial topology constraint, energy storage system operation constraint, distributed power output constraint, and multi-stage planning constraint.

[0121] Power balance constraint:

[0122]

[0123] Among them, P i,sc,t and Q i,sc,t are the active injection power and reactive injection power of node i at time t of scenario day sc respectively, Ω0 is the set of nodes, Gij and B ij are the conductance and susceptance of branch ij, respectively, and θ ij,sc,t is the voltage phase angle difference between nodes i and j at time t on scenario day sc, and U i,sc,t is the voltage of node i at time t on scenario day sc, and U j,sc,t is the voltage of node j at time t on scenario day sc.

[0124] Branch current constraint:

[0125]

[0126] where I l,sc,t is the current of branch l at time t on scenario day sc, is the upper limit of the current allowed to pass through branch l.

[0127] Node voltage constraint:

[0128]

[0129] where is the lower limit of the voltage of node i, is the upper limit of the voltage of node i.

[0130] Radial topology constraint:

[0131] g ∈ G R

[0132] where g is the target network, and G R is the set composed of all possible radial connected networks.

[0133] Energy storage system (ESS) operation constraint:

[0134] [[ID=�4]]

[0135] where ESS i,t is the stored power of the energy storage system at node i at time t, is the active power output by the energy storage device at node i at time t. A positive value indicates that the ESS is discharging, and a negative value indicates that the ESS is charging. E ess,i is the capacity of the energy storage system at node i, and SOC max and SOC min are the upper and lower limits of the state of charge of the energy storage system, respectively, is the rated power of the energy storage system at node i.

[0136] Distributed power output constraint:

[0137]

[0138] Among them, p dg is the active power of the distributed power source, and are respectively the lower limit and upper limit of the active power of the distributed power source.

[0139] Assume that the construction or renovation type of all lines is unique during the entire planning stage. The lines need to meet the constraints of being constructed first and then operated, and not being renovated after new construction. At the same time, there should also be certain constraints on the number of newly built distribution lines, distributed power sources, and energy storage during the planning period. The specific limiting values can be determined according to the actual distribution network planning requirements. By adding multi-stage planning constraints, the rationality and adaptability of the planning scheme can be improved.

[0140] Multi-stage planning constraints:

[0141]

[0142]

[0143]

[0144]

[0145]

[0146]

[0147] Among them, and are respectively the maximum number of newly built lines, the number of distributed power sources, and the maximum number of energy storage systems allowed during the entire planning stage.

[0148] Step 102, obtain the target parameters. The target parameters include the predicted load of the distribution network during the planning stage and the network structure parameters of the current area to be planned.

[0149] It should be noted that the predicted load of the distribution network during the planning stage and the network structure parameters of the current area to be planned are parameters related to solving the optimal distribution network planning scheme. Therefore, it is necessary to obtain the predicted load of the distribution network during the planning stage and the network structure parameters of the current area to be planned.

[0150] Step 103, according to the target parameters, use the SGO algorithm to solve the multi-stage distribution network planning model and obtain the optimal solution of the multi-stage distribution network planning model.

[0151] It should be noted that the problem of solving the multi-stage planning model of the distribution network in the present invention is a complex multi-objective solving problem. Solving complex problems requires the participation of multiple factors. When an individual is not sufficient to solve a complex problem, it is necessary to learn from surrounding individuals or those with such capabilities in order to improve its own problem-solving ability. Therefore, in the present invention, the SGO (social group optimization) algorithm is used to solve the multi-stage planning model of the distribution network. The principle flow chart of using the SGO algorithm to solve the multi-stage planning model of the distribution network is as Figure 3 shown.

[0152] Meanwhile, in order to improve the exploration efficiency of the SGO algorithm, the present invention uses chaos theory to improve its population initialization method. Chaos has the characteristics of randomness, ergodicity, and sensitivity to initial conditions, and can traverse all states within a certain range without repetition according to its own laws. Through these properties of chaotic motion, the population can be initialized, improving the diversity of the initial population.

[0153] Specifically, the process of using the SGO algorithm to solve the multi-stage planning model of the distribution network includes:

[0154] S1. Initialize the population based on the Logitic chaotic map, calculate the fitness value of each individual according to the multi-stage planning model of the distribution network, and determine the position of the initial population. Among them, each individual represents a set of planning results, including the number of expanded lines, the location of distributed power sources, and the location of energy storage.

[0155] Specifically:

[0156] (1) Randomly generate a D-dimensional vector x1=(x 11 ,x 12 ,...,x 1D ) with the numerical components between (0,1), and generate L + 2N vectors according to the expression of the Logitic chaotic map:

[0157] x m+1,n =f(x m,n ), m = 1, 2,..., L + 2N; n = 1, 2,..., D

[0158] where N is the number of individuals, L is a preset integer, and the expression of the Logitic chaotic map is:

[0159]

[0160] Take the last 2N vectors: x L+1 ,x L+2 ,...,x L+2N ;

[0161] (2) According to z mn= min n + x mn (max n - min n ), map the n-th component of x mn to the interval [min n , ma n x], where z mn is the m-th individual gene vector of dimension D after mapping;

[0162] (3) Calculate the fitness value of each individual according to the multi-stage distribution network planning model. If the individual meets the constraints of the multi-stage distribution network planning model, calculate the individual fitness according to the distribution network planning objective function considering the driving factors of power system evolution. Otherwise, set the individual fitness value to positive infinity, and select the N vectors with the smallest fitness values as the initial population positions.

[0163] Each individual X in the population represents a set of planning results, including the number of expanded lines, the locations of distributed power sources, and the locations of energy storage:

[0164]

[0165] Among them, |dg| is the total number of DGs to be installed, and |ess| is the total number of ESSs to be installed. In X, the variables at the DG installation locations are integers. If x l+1 = 0, it means no DG is installed. Otherwise, the installed capacity of the DG is which is the unit capacity of the DG. In X, the variables at the energy storage installation locations are integers. If x l+|dg|+1 = 0, it means no ESS is installed. Otherwise, the installed capacity of the ESS is which is the unit capacity of the ESS.

[0166] S2. After initializing the population, update the population based on the population update formula in the improvement stage of the SGO algorithm, and evaluate the new individuals. If the new solution is better than the current solution, replace the current solution with the new solution. Otherwise, no update operation is performed. Specifically, the population update formula in the improvement stage of the SGO algorithm is:

[0167] X new m,n = c × Xold m,n + v × (gbest m - Xold m,n )

[0168] where Xold m,n and Xnew m,n are the n-th genes of the m-th individual before and after update respectively, and gbest mIt is the m-th dimensional gene of the optimal individual in the current iteration stage. c is the self-reflection coefficient, and its value usually ranges from 0 to 1. When c takes 0.2, the best effect is achieved. v is a random number in the range of [0, 1].

[0169] S3. Determine the global optimal solution of the population in this iteration.

[0170] S4. Update the population based on the population update formula obtained by the SGO algorithm for the stage, evaluate the new individuals. If the new solution is better than the global optimal solution, replace the current solution with the new solution; otherwise, no update operation is performed.

[0171] S5. Determine whether the maximum number of iterations is reached. If so, output the global optimal solution; otherwise, return to the step of calculating the fitness value of each individual according to the multi-stage planning model of the distribution network in step S1 and continue to execute downward.

[0172] Step 104. Plan the distribution network according to the distribution network planning scheme corresponding to the optimal solution of the multi-stage planning model of the distribution network.

[0173] It should be noted that after determining the finally output global optimal solution, the optimal distribution network planning scheme is determined, that is, the corresponding optimal X includes the number of expansion lines, the locations of distributed power sources, and the locations of energy storage.

[0174] The multi-stage planning method for active distribution network considering evolutionary drive provided by the present invention comprehensively considers the common drive factors, market drive factors, and innovative technology drive factors of the evolution of the power system form, establishes a multi-stage planning model for the distribution network including multi-stage constraint conditions, and uses the SGO algorithm to obtain the optimal solution of the multi-stage planning model of the distribution network, obtaining the optimal distribution network planning scheme. While reducing the construction and renovation costs of the distribution network, in the way of collaborative planning of energy storage and new energy, it improves the utilization rate of new energy and the adaptability of the planning scheme, and solves the technical problems that the existing multi-stage planning of active distribution network considering evolutionary drive lacks consideration of the driving force factors of the evolution of the power system form and the adaptability of the planning scheme is weak.

[0175] The multi-stage planning method for active distribution network considering evolutionary drive provided by the present invention simultaneously considers power balance constraints, branch current constraints, node voltage constraints, radial topology constraints, energy storage system operation constraints, distributed power output constraints, and multi-stage planning constraints, improving the rationality and adaptability of the distribution network planning scheme and effectively improving the energy utilization efficiency of the distribution network.

[0176] Meanwhile, the multi-stage planning method for active distribution networks considering evolutionary drive provided by the present invention initializes the population by utilizing the properties of Logistic chaotic motion. The randomness, ergodicity, and sensitivity to initial conditions of chaotic motion improve the diversity of the initial population, providing a guarantee for solving the global optimal distribution network planning scheme.

[0177] For ease of understanding, please refer to Figure 4 , an embodiment of a multi-stage planning system for active distribution networks considering evolutionary drive provided by the present invention includes:

[0178] A modeling unit for constructing a multi-stage planning model of the distribution network. The multi-stage planning model of the distribution network includes a distribution network planning objective function considering power system evolutionary drive factors and multi-stage constraint conditions. The power system evolutionary drive factors include common drive factors, market drive factors, and innovative technology drive factors;

[0179] An acquisition unit for acquiring target parameters, where the target parameters include the predicted load of the distribution network within the planning stage and the network structure parameters of the current area to be planned;

[0180] A model solving unit for solving the multi-stage planning model of the distribution network using the SGO algorithm according to the target parameters to obtain the optimal solution of the multi-stage planning model of the distribution network;

[0181] A planning unit for planning the distribution network according to the distribution network planning scheme corresponding to the optimal solution of the multi-stage planning model of the distribution network.

[0182] The distribution network planning objective function considering power system evolutionary drive factors is:

[0183] min(C line +C loss +C dg +C ess )

[0184] Wherein, C line is the grid upgrade cost, C dg is the investment cost of distributed power generation equipment, C loss is the annual network loss cost, C ess is the installation cost of distributed energy storage equipment;

[0185]

[0186] Wherein, is a 0-1 decision variable used to represent whether a new branch ij is built within the planning stage T h , T h is the planning stage, is the set of planning stages, R is the capital recovery factor, is the construction cost per unit length of the line, Ω line is the set of locations of the lines to be newly built, η line is the coefficient of line operation and maintenance cost, l ij is the length of the newly built line;

[0187]

[0188] Among them, is used to represent the 0-1 decision variable indicating whether a distributed power source is installed at node i in the planning stage T h is the installation cost per unit capacity of the distributed power source, Ω is the set of nodes where the distributed power source is installed, r is the inflation rate, T is the service life of the distributed power source equipment, S dg is the technological maturity of the distributed power source equipment, B dg is the photovoltaic subsidy coefficient, dg is the capacity of the distributed power source installed at node i in the planning stage T is the planning stage T h ;

[0189]

[0190] Among them, Ω br is the set of all branches in the distribution network, r ij is the resistance of line ij, and are the active and reactive powers flowing through line ij at time t in the planning stage T h respectively, λ t is the electricity price at time t, Δt is the time interval, sc is the scenario day, N sc is the number of scenario days;

[0191]

[0192] Among them, C e and C p are the unit capacity and unit power investment costs of the energy storage system respectively, S ess is the technological maturity of the energy storage equipment, is used to represent the 0-1 decision variable indicating whether an energy storage is installed at node i in the planning stage T h Ω ess is the set of nodes where the energy storage is installed, is the planning stage T h is the capacity of the energy storage installed at node i, is the planning stage T h is the maximum charge and discharge power of the energy storage installed at node i.

[0193] The multi-stage constraint conditions include:

[0194]

[0195]

[0196]

[0197] g ∈ G R

[0198]

[0199]

[0200]

[0201]

[0202]

[0203]

[0204]

[0205]

[0206] where P i,sc,t and Q i,sc,t are the active injection power and reactive injection power of node i at time t of scenario day sc respectively, Ω0 is the set of nodes, G ij and B ij are the conductance and susceptance of branch ij respectively, θ ij,sc,t is the voltage phase angle difference between node i and node j at time t of scenario day sc, I l,sc,t is the current of branch l at time t of scenario day sc, is the upper limit of the current allowed to pass through branch l, U i,sc,t is the voltage of node i at time t of scenario day sc, U j,sc,t is the voltage of node j at time t of scenario day sc, is the lower limit of the voltage of node i, is the upper limit of the voltage of node i, g is the target grid, G R is the set composed of all possible radial connected networks, ESS i,t is the stored power of the energy storage system of node i at time t, is the active power generated by the energy storage device of node i at time t, E ess,i is the capacity of the energy storage system of node i, SOC max and SOC min are the upper and lower limits of the state of charge of the energy storage system respectively, is the rated power of the energy storage system for node i, p dg is the active power of the distributed power source, and are the lower and upper limits of the active power of the distributed power source respectively, and are the maximum number of new lines allowed to be built, the number of distributed power sources, and the maximum number of energy storage systems respectively during the entire planning stage.

[0207] The model solving unit includes:

[0208] An initialization subunit, which is used to initialize the population based on the Logitic chaotic mapping, calculate the fitness value of each individual according to the multi-stage distribution network planning model, and determine the initial population position. Among them, each individual represents a set of planning results, including the number of expanded lines, the location of the distributed power source, and the location of the energy storage;

[0209] A first update subunit, which is used to update the population according to the population update formula in the improvement stage of the SGO algorithm, evaluate the new individual. If the new solution is better than the current solution, the new solution is used to replace the current solution; otherwise, no update operation is performed;

[0210] An optimal solution subunit, which is used to determine the global optimal solution of the population in this iteration;

[0211] A second update subunit, which is used to update the population according to the population update formula in the obtaining stage of the SGO algorithm, evaluate the new individual. If the new solution is better than the global optimal solution, the new solution is used to replace the current solution; otherwise, no update operation is performed;

[0212] An output subunit, which is used to determine whether the maximum number of iterations is reached. If so, the global optimal solution is output; otherwise, the fitness value of each individual calculated according to the multi-stage distribution network planning model is returned.

[0213] Specifically, the initialization subunit is used for:

[0214] Randomly generate a D-dimensional vector x1=(x 11 ,x 12 ,...,x 1D ) with the numerical component between (0,1), and generate L + 2N vectors according to the expression of the Logitic chaotic mapping:

[0215] x m+1,n = f(x m,n ), m = 1, 2,..., L + 2N; n = 1, 2,..., D

[0216] Among them, N is the number of individuals, L is a preset integer, and the expression of the Logitic chaotic mapping is:

[0217]

[0218] where x i is a random quantity between [0, 1], μ is a Logistic parameter, and 3.5699 ≤ μ ≤ 4;

[0219] Take the last 2N vectors: x L+1 , x L+2 ,..., x L+2N ;

[0220] According to z mn = min n + x mn (max n - min n ), map the nth component of x mn to the interval [min n , max n , where z mn is the mth individual gene vector of dimension D after mapping;

[0221] Calculate the fitness value of each individual according to the multi-stage planning model of the distribution network, and select the N vectors with the smallest fitness value as the initial population positions.

[0222] The multi-stage planning system for active distribution network considering evolutionary drive provided by the present invention comprehensively considers the common drive factors, market drive factors, and innovative technology drive factors of the evolution of the power system form, establishes a multi-stage planning model of the distribution network including multi-stage constraint conditions, and uses the SGO algorithm to obtain the optimal solution of the multi-stage planning model of the distribution network, obtaining the optimal distribution network planning scheme. While reducing the construction and transformation costs of the distribution network, in the way of collaborative planning of energy storage and new energy, it improves the utilization rate of new energy and the adaptability of the planning scheme, and solves the technical problems that the existing multi-stage planning of active distribution network considering evolutionary drive lacks consideration of the driving force factors of the evolution of the power system form and the weak adaptability of the planning scheme.

[0223] The multi-stage planning system for active distribution network considering evolutionary drive provided by the present invention simultaneously considers power balance constraints, branch current constraints, node voltage constraints, radial topology constraints, energy storage system operation constraints, distributed power output constraints, and multi-stage planning constraints, improving the rationality and adaptability of the distribution network planning scheme and effectively improving the energy utilization efficiency of the distribution network.

[0224] At the same time, the multi-stage planning system for active distribution network considering evolutionary drive provided by the present invention uses the Logistic chaotic motion property to initialize the population. The randomness, ergodicity, and sensitivity to initial conditions of chaotic motion improve the diversity of the initial population and provide a guarantee for solving the global optimal distribution network planning scheme.

[0225] The multi-stage planning system for active distribution network considering evolutionary drive provided by the embodiment of the present invention is used to execute the multi-stage planning method for active distribution network considering evolutionary drive in the foregoing embodiment of the multi-stage planning method for active distribution network considering evolutionary drive. Its principle is the same as that of the multi-stage planning method for active distribution network considering evolutionary drive in the foregoing embodiment of the multi-stage planning method for active distribution network considering evolutionary drive, and will not be elaborated herein.

[0226] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. An active distribution network multi-stage planning method considering evolution-driven, characterized in that Including: Construct a multi-stage distribution network planning model, which includes a distribution network planning objective function considering the driving factors of power system evolution and multi-stage constraint conditions. The driving factors of power system evolution include common driving factors, market driving factors, and innovative technology driving factors; Obtain target parameters, where the target parameters include the predicted load of the distribution network within the planning stage and the network structure parameters of the current area to be planned; According to the target parameters, use the SGO algorithm to solve the multi-stage distribution network planning model to obtain the optimal solution of the multi-stage distribution network planning model; Plan the distribution network according to the distribution network planning scheme corresponding to the optimal solution of the multi-stage distribution network planning model; The distribution network planning objective function considering the driving factors of power system evolution is: min(C line +C loss +C dg +C ess ) Among them, C line is the cost of grid upgrade, C dg is the investment cost of distributed power generation equipment, C loss is the annual cost of network loss, C ess is the installation cost of distributed energy storage equipment; Among them, is a 0-1 decision variable used to represent whether a new branch ij is to be built during the planning stage T h , T h is the planning stage, is the set of planning stages, R is the capital recovery factor, is the construction cost per unit length of the line, Ω line is the set of locations of the lines to be newly built, η line is the line operation and maintenance cost factor, l ij is the length of the newly built line; Among them, is a 0-1 decision variable used to represent whether to install a distributed power source at node i during the planning stage T h ; is the installation cost of the distributed power source per unit capacity, Ω dg is the set of nodes where the distributed power source is installed, r is the inflation rate, T is the service life of the distributed power source equipment, S dg is the technological maturity of the distributed power source equipment, B dg is the photovoltaic subsidy coefficient, is the capacity of the distributed power source installed at node i during the planning stage T h ; where, Ω br is the set of all branches in the distribution network, r ij is the resistance of line ij, and are the active and reactive powers flowing through line ij at time t within the planning period T h respectively, λ t is the electricity price at time t, Δt is the time interval, sc is the scenario day, N sc is the number of scenario days, is the voltage at node i at time t within the planning period T h ; Among them, C e and C p are respectively the investment costs per unit capacity and per unit power of the energy storage system, S ess is the technical maturity of the energy storage device, is a 0-1 decision variable used to represent whether to install an energy storage at the internal node i during the planning stage T h , Ω ess is the set of energy storage installation nodes, is the capacity of the energy storage installed at node i during the planning stage T h , is the maximum charge and discharge power of the energy storage installed at node i during the planning stage T h ; The multi-stage constraint conditions include: g ∈ G R Among them, P i,sc,t and Q i,sc,t are the active injection power and reactive injection power of node i at time t of scenario day sc respectively. Ω0 is the set of nodes, G ij and B ij are the conductance and susceptance of branch ij respectively. θ ij,sc,t is the voltage phase angle difference between node i and node j at time t of scenario day sc. I l,sc,t is the current of branch l at time t of scenario day sc. is the upper limit of the current allowed to pass through branch l. U i,sc,t is the voltage of node i at time t of scenario day sc. U j,sc,t is the voltage of node j at time t of scenario day sc. is the lower limit of the voltage of node i. is the upper limit of the voltage of node i. g is the target grid, G R is the set composed of all possible radial connected networks. ESS i,t is the stored electricity of the energy storage system of node i at time t. is the active power generated by the energy storage device of node i at time t. E ess,i is the capacity of the energy storage system of node i. SOC max and SOC min are the upper and lower limits of the state of charge of the energy storage system respectively. P i N is the rated power of the energy storage system of node i. p dg is the active power of the distributed power source. and are the lower and upper limits of the active power of the distributed power source respectively. and are the maximum number of new lines allowed to be built, the number of distributed power sources, and the maximum number of energy storage systems during the entire planning stage respectively.

2. The multi-stage planning method for an active distribution network considering evolutionary drive according to claim 1, wherein According to the target parameters, use the SGO algorithm to solve the multi-stage distribution network planning model to obtain the optimal solution of the multi-stage distribution network planning model, including: Perform population initialization based on the Logitic chaotic map, calculate the fitness value of each individual according to the multi-stage distribution network planning model, and determine the initial population position. Among them, each individual represents a set of planning results, including the number of expanded lines, the location of distributed power sources, and the location of energy storage; Update the population based on the population update formula in the SGO algorithm improvement stage, evaluate the new individuals. If the new solution is better than the current solution, replace the new solution with the current solution; otherwise, do not perform the update operation; Determine the global optimal solution of the current iteration population; Update the population based on the population update formula in the SGO algorithm acquisition stage, evaluate the new individuals. If the new solution is better than the global optimal solution, replace the new solution with the current solution; otherwise, do not perform the update operation; Judge whether the maximum number of iterations is reached. If so, output the global optimal solution; otherwise, return to calculate the fitness value of each individual according to the multi-stage distribution network planning model.

3. The multi-stage planning method for active distribution network considering evolutionary drive according to claim 2, characterized in that Perform population initialization based on the Logitic chaotic map, including: Randomly generate a D-dimensional vector \(x_1=(x 11 ,x 12 ,\cdots,x 1D )\) with a numerical component between \((0,1)\). Generate \(L + 2N\) vectors according to the expression of the Logistic chaotic map: x m+1,n = f(x m,n ), m = 1, 2, ..., L + 2N; n = 1, 2, ..., D Among them, N is the number of individuals, L is a preset integer, and the expression of the Logitic chaotic map is: where x i is a random variable between [0, 1], μ is a Logistic parameter, and 3.5699 ≤ μ ≤ 4; Take the last 2N vectors: x L+1 , x L+2 ,..., x L+2N ; According to z mn = min n + x mn (max n - min n ), the n-th component of x mn is mapped to the interval [min n , max n , where z mn is the m-th individual gene vector of dimension D after mapping; Calculate the fitness value of each individual according to the multi-stage distribution network planning model, and select the N vectors with the smallest fitness value as the initial population position.

4. An active distribution network multi-stage planning system considering evolution-driven, characterized in that, Including: A modeling unit for constructing a multi-stage distribution network planning model, which includes a distribution network planning objective function considering the driving factors of power system evolution and multi-stage constraint conditions. The driving factors of power system evolution include common driving factors, market driving factors, and innovative technology driving factors; An acquisition unit for obtaining target parameters, where the target parameters include the predicted load of the distribution network within the planning stage and the network structure parameters of the current area to be planned; A model solving unit for using the SGO algorithm to solve the multi-stage distribution network planning model according to the target parameters to obtain the optimal solution of the multi-stage distribution network planning model; A planning unit for planning the distribution network according to the distribution network planning scheme corresponding to the optimal solution of the multi-stage distribution network planning model; The distribution network planning objective function considering the driving factors of power system evolution is: min(C line +C loss +C dg +C ess ) Among them, C line is the cost of grid upgrade, C dg is the investment cost of distributed power generation equipment, C loss is the annual cost of network loss, C ess is the installation cost of distributed energy storage equipment; Among them, is a 0-1 decision variable used to represent whether a new branch ij is to be built within the planning stage T h , T h is the planning stage, is the set of planning stages, R is the capital recovery factor, is the construction cost per unit length of the line, Ω line is the set of positions of the lines to be newly built, η line is the line operation and maintenance cost factor, l ij is the length of the newly built line; Among them, is a 0-1 decision variable used to represent whether to install a distributed power source at node i during the planning stage T h ; is the installation cost per unit capacity of the distributed power source, Ω dg is the set of nodes where the distributed power source is installed, r is the inflation rate, T is the service life of the distributed power source equipment, S dg is the technological maturity of the distributed power source equipment, B dg is the photovoltaic subsidy coefficient, is the capacity of the distributed power source installed at node i during the planning stage T h ; Among them, Ω br is the set of all branches in the distribution network, r ij is the resistance of line ij, and are the active and reactive powers flowing through line ij at time t within the planning period T h respectively, λ t is the electricity price at time t, Δt is the time interval, sc is the scenario day, N sc is the number of scenario days, is the voltage at node i at time t within the planning period T h ; Among them, C e and C p are the investment costs per unit capacity and per unit power of the energy storage system respectively, S ess is the technology maturity of the energy storage device, is a 0-1 decision variable used to represent whether to install energy storage at the internal node i during the planning stage T h , Ω ess is the set of energy storage installation nodes, is the capacity of the energy storage installed at node i during the planning stage T h , is the maximum charge-discharge power of the energy storage installed at node i during the planning stage T h ; The multi-stage constraint conditions include: g ∈ G R where P i,sc,t and Q i,sc,t are the active and reactive injection powers of node i at time t of scenario day sc respectively, Ω0 is the set of nodes, G ij and B ij are the conductance and susceptance of branch ij respectively, θ ij,sc,t is the voltage phase angle difference between node i and node j at time t of scenario day sc, I l,sc,t is the current of branch l at time t of scenario day sc, is the upper limit of the current allowed to pass through branch l, U i,sc,t is the voltage of node i at time t of scenario day sc, U j,sc,t is the voltage of node j at time t of scenario day sc, is the lower limit of the voltage of node i, is the upper limit of the voltage of node i, g is the target network, G R is the set composed of all possible radial connected networks, ESS i,t is the stored electricity of the energy storage system of node i at time t, is the active power generated by the energy storage device of node i at time t, E ess,i is the capacity of the energy storage system of node i, SOC max and SOC min are the upper and lower limits of the state of charge of the energy storage system respectively, P i N is the rated power of the energy storage system of node i, p dg is the active power of the distributed power source, and are the lower and upper limits of the active power of the distributed power source respectively, and are the maximum number of new lines allowed to be built, the number of distributed power sources, and the maximum number of energy storage systems during the entire planning stage respectively.

5. The active distribution network multi-stage planning system considering evolutionary drive according to claim 4, characterized in that The model solving unit includes: An initialization subunit, which is used to initialize the population based on the Logistic chaotic map, calculate the fitness value of each individual according to the multi-stage planning model of the distribution network, and determine the initial population position, where each individual represents a set of planning results, including the number of expanded lines, the positions of distributed power sources, and the positions of energy storage; A first update subunit, which is used to update the population according to the population update formula in the improvement stage of the SGO algorithm, evaluate the new individuals, and if the new solution is better than the current solution, replace the current solution with the new solution, otherwise, no update operation is performed; An optimal solution subunit, which is used to determine the global optimal solution of the population in this iteration; A second update subunit, which is used to update the population according to the population update formula in the obtaining stage of the SGO algorithm, evaluate the new individuals, and if the new solution is better than the global optimal solution, replace the current solution with the new solution, otherwise, no update operation is performed; An output subunit, which is used to judge whether the maximum number of iterations is reached. If so, output the global optimal solution, otherwise, return to calculate the fitness value of each individual according to the multi-stage planning model of the distribution network.

6. The active distribution network multi-stage planning system considering evolutionary drive according to claim 5, characterized in that The initialization subunit is specifically used for: Randomly generate a D-dimensional vector x1=(x 11 ,x 12 ,...,x 1D ) within the range (0,1) for the numerical component, and generate L + 2N vectors according to the expression of the Logitic chaotic map: x m+1,n = f(x m,n ), m = 1, 2, ..., L + 2N; n = 1, 2, ..., D where N is the number of individuals, L is a preset integer, and the expression of the Logistic chaotic map is: where x i is a random quantity between [0, 1], μ is a Logistic parameter, and 3.5699 ≤ μ ≤ 4; Take the last 2N vectors: x L+1 , x L+2 ,..., x L+2N ; According to z mn = min n + x mn (max n - min n ), map the n-th component of x mn to the interval [min n , max n , where z mn is the m-th individual gene vector of dimension D after mapping; Calculate the fitness value of each individual according to the multi-stage planning model of the distribution network, and select the N vectors with the smallest fitness values as the initial population positions.

Images