Medium-voltage distribution network reclosing configuration optimization method based on grey wolf optimizer

Abstract

The invention belongs to the technical field of power distribution automation, and particularly relates to a medium-voltage power distribution network reclosing configuration optimization method based on a grey wolf optimizer, which comprises the following steps: analyzing the past fault history, identifying the fault generation mode, frequency and interruption severity, and optimizing the reclosing configuration of a medium-voltage power distribution network; original data input is provided for subsequent calculation of a system average power failure duration index and a system average power failure frequency index; the correlation between the average power failure duration index and the average power failure frequency index of the system is calculated, and a basis is provided for construction of a subsequent objective function and selection of an optimization method; performing objective function modeling, parameterizing a grey wolf optimization algorithm, exploring optimal configuration through experimental setting, and providing various possible schemes and data support for finally determining proper recloser configuration; and a test scene is defined, different scenes in the network are tested, it is ensured that the optimized recloser configuration scheme can be effective in different actual network environments, and finally optimization of recloser configuration is achieved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid configuration, and in particular relates to a medium voltage distribution network reclosing configuration optimization method based on a Grey Wolf optimizer. Background Art

[0002] With the continuous growth of electricity demand, the reliability and power quality requirements of distribution systems are becoming increasingly stringent. As a critical link, the stable operation of medium-voltage distribution networks is directly related to users' electricity experience and the normal operation of the economy and society. Genetic algorithms (GAs) iteratively evolve populations through operations such as selection, crossover, and mutation to find the optimal solution. In large-scale distribution networks, these algorithms require processing a large number of individuals and complex genetic operations, resulting in long computational times and reduced efficiency. Simulated annealing algorithms (SAs) simulate the solid annealing process, accepting inferior solutions with a certain probability, escaping local optima and approaching the global optimum. In complex distribution network environments, these algorithms require a large number of random walks, resulting in long computational times and slow convergence, making them difficult to meet the demand for rapid decision-making. Particle swarm optimization algorithms (PSOs) simulate the behavior of flocks of birds or schools of fish, with particles representing potential solutions. The optimal solution is found by updating their speed and position. In complex distribution network environments, it is difficult to escape local optima, which affects the optimization of recloser configuration and makes it easy to get stuck in local optima. Therefore, to effectively reduce the impact of faults and improve power supply reliability, medium-voltage distribution network recloser configuration optimization technology based on the Gray Wolf Optimizer is imperative.

[0003] When solving the recloser configuration problem in medium-voltage distribution networks, the computational complexity of genetic algorithms increases exponentially with increasing problem size, resulting in low computational efficiency and a tendency to get stuck in local optima. Particle swarm optimization algorithms lack global search capabilities, making it difficult to conduct a comprehensive search, thus affecting the optimality of recloser configuration. Simulated annealing algorithms are sensitive to parameters and have slow convergence, making them unable to meet the high-reality requirements of grid operations. Summary of the Invention

[0004] In order to solve the above problems, the present invention provides a medium voltage distribution network reclosing configuration optimization method based on the Gray Wolf optimizer. The specific technical solution is as follows:

[0005] A method for optimizing reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer comprises the following steps:

[0006] S1. Analyze past fault history to identify fault occurrence patterns, frequency, and interruption severity, providing raw data input for the subsequent calculation of the system average outage duration index (SAIDI) and the system average outage frequency index (SAIFI);

[0007] S2. Calculate the correlation between the system average outage duration index SAIDI and the system average outage frequency index SAIFI to provide a basis for the subsequent construction of the objective function and the selection of the optimization method;

[0008] S3. Model the objective function, parameterize the Grey Wolf optimization algorithm, explore the optimal configuration through different experimental settings, and provide multiple possible solutions and data support for the final determination of the appropriate recloser configuration;

[0009] S4. Define test scenarios and test different scenarios in the network to ensure that the optimized recloser configuration solution can be effective in different actual network environments, and ultimately achieve optimization of the recloser configuration.

[0010] Preferably, in step S1, by analyzing past fault history, the pattern of fault occurrence is identified, the frequency of events is determined and the severity of power outages is assessed, thereby providing a realistic baseline for evaluating the impact of recloser installation in subsequent simulations and analyses; and by introducing reclosers and dividing the system into blocks defined by reclosers, each block is treated as an independent entity.

[0011] Preferably, the specific method of step S2 is: grouping the fault duration information and fault number of each feeder line, and calculating the Pearson correlation, and the calculation formula is:

[0012]

[0013] Here, x and y represent variables, that is, the features to be observed, and n is the total number of lines in the dataset.

[0014] Preferably, the specific method of step S3 is as follows: the prediction of future power outages will be reflected by the historical fault database, and the impact of each power outage on the relevant load blocks configured in the system will be calculated respectively. These impacts are determined from the two aspects of power outage duration and frequency, as shown below:

[0015] Impact SAIDI =D fault ×N[h]

[0016] Impact SAIFI =N[customers]

[0017] Among them, D fault is the duration of the failure being evaluated, and N is the number of customers in the affected block.

[0018] Preferably, through this method, the total number of customers who need to evaluate the service in step S3 is

[0019]

[0020] where N feeder is the total number of users in the evaluated feeder, z is the total number of blocks generated, and i represents the block index;

[0021] After obtaining the total number of customers, construct the objective function model for the current problem as follows:

[0022]

[0023] Among them, w1 and w2 are the weights of each comprehensive indicator. To maintain the equal importance of the indicators, they are set to 0.5. ref and SAIFI ref These two items are indices calculated in the scenario without recloser, and the purpose is to normalize the indices because they have different dimensions;

[0024] Finally, the optimization problem to be solved is presented in its complete form as follows:

[0025]

[0026] Among them: Impact SAIDI =D fault ×N、Impact SAIFI =N、

[0027]

[0028] The constraints are: w1=0.5 and w2=0.5.

[0029] Preferably, the specific steps of the gray wolf optimization include:

[0030] S31. Initialize parameters, set the number of individuals in the wolf pack, define the algorithm termination condition, determine the optimization variable dimension, define the objective function to be minimized, the number of reclosers is a natural number, and the recloser locations must be mapped to valid nodes in the distribution network;

[0031] S32. Initialize the wolf pack positions, randomly generate the initial positions of the wolf pack, and the position vector of each wolf represents a possible coincidencer configuration;

[0032] S33. Calculate the fitness value. For each wolf position configuration, perform the following operations: Network segmentation: Divide the distribution network into multiple independent blocks based on the recloser locations; Fault impact assessment: Calculate the SAIDI and SAIFI of each block based on the historical fault database; Objective function value: Calculate the weighted normalized objective function value, where a smaller value indicates a higher fitness;

[0033] S34. Determine the social hierarchy, sort the wolves according to their fitness values, select the three best wolves as α wolf, β wolf, and δ wolf, and the remaining individuals as ω wolf to follow the first three to update their positions;

[0034] S35. Update the position of the wolf pack. Update the positions of other wolves based on the positions of α, β, and δ. The specific formula is:

[0035]

[0036] in and is the coefficient vector, and the calculation formula is:

[0037]

[0038] a is a parameter that decreases linearly with the number of iterations, and is a random vector in [0,1];

[0039] S36. Constraint processing, mapping continuous positions to valid nodes of the distribution network to ensure that recloser positions are not repeated;

[0040] S37. Iterative optimization, repeating steps S33 to S36 until the maximum number of iterations is reached or the objective function converges;

[0041] S38. Output the optimal solution. The final position of α wolf is the optimal recloser configuration, corresponding to the minimized SAIDI and SAIFI values.

[0042] The beneficial effects of the present invention are as follows: the gray wolf optimization algorithm provided by the present invention can perform a comprehensive and efficient search, is not easily trapped in a local optimal solution, can find a more optimal medium-voltage distribution network recloser configuration scheme, and improves power supply reliability;

[0043] The Gray Wolf Optimization Algorithm has relatively few parameters and is less sensitive to parameters, which reduces the difficulty of algorithm application and improves the applicability and stability of the algorithm;

[0044] The Gray Wolf optimization algorithm has a fast convergence speed and can obtain a recloser configuration result that meets the requirements in a relatively short time. When the grid operating status changes, it can quickly adjust the recloser configuration to reduce the power outage time and the scope of the fault impact.

[0045] The Grey Wolf Optimization Algorithm has good adaptability and can better adapt to the complex environment and dynamic changes of the medium voltage distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following briefly describes the drawings required for the specific embodiments or the description of the prior art. Similar elements or parts are generally identified by similar reference numerals throughout the drawings. Elements or parts in the drawings are not necessarily drawn to scale.

[0047] Figure 1 Design a flow chart for the reclosing configuration optimization method based on the Grey Wolf Optimizer (GWO);

[0048] Figure 2 It is the correlation distribution diagram between the total number of faults and the total duration of faults;

[0049] Figure 3 Flowchart of the optimization algorithm for gray wolves;

[0050] Figure 4 This is the area division diagram of the IEEE 37-node network;

[0051] Figure 5 The figure shows the changes in the average power outage duration indicator of the system in each scenario based on the number of configured reclosers. DETAILED DESCRIPTION

[0052] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0053] like Figure 1 As shown, a specific embodiment of the present invention provides a method for optimizing reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer, comprising the following steps:

[0054] S1. Analyze past fault history to identify fault occurrence patterns, frequency, and interruption severity, providing raw data input for subsequent calculations of indicators such as the System Average Interruption Duration Index (SAIDI) and the System Average Interruption Frequency Index (SAIFI).

[0055] By analyzing past fault history, fault patterns can be identified, event frequencies can be determined, and the severity of outages can be assessed. This retrospective analysis helps understand critical points and weak links in the system. This analysis helps identify areas prone to recurring faults, allowing for strategic configuration of protective devices to minimize the impact of these outages on the power supply. Therefore, an initial study scenario was constructed based on the feeder's fault history. This initial scenario represents a situation without reclosers installed, reflecting the prevailing conditions in distribution networks without reclosers deployed. The fault history provides a realistic baseline for evaluating the impact of recloser installation in subsequent simulations and analyses.

[0056] By introducing reclosers and segmenting the system into zones defined by reclosers, greater granularity in fault control and response is expected. Each zone can be treated as a separate entity, where reclosers provide flexibility in isolating the affected portion of the system.

[0057] In the event of a fault in one block, the entire block is shut down. This measure is intended to prevent the problem from spreading to other parts of the system. This is particularly important for minimizing the impact of localized faults and ensuring the integrity of the rest of the power system. This final condition allows for load transfer to other feeders in the block not affected by the fault. This is particularly important for ensuring operational efficiency and system resiliency. This allows unaffected areas to continue receiving power from alternative sources. In this situation, the presence of reclosers is crucial, enabling a quick and coordinated response, isolating the affected area, and facilitating load transfer to other feeders. This strategy helps maximize energy supply and shorten outage duration.

[0058] To simulate scenarios without reclosers, the positive effects of installed reclosers inherent in the original database were intentionally excluded. This adjustment ensured consistency with the research objectives and provided a suitable fault dataset for analysis. Because the original data came from real distribution feeders whose topology was unavailable, adjustments were made to the feeders used in the experiment. Consequently, the location of each fault was randomly selected, resulting in a variety of simulated scenarios. Furthermore, the number of customers associated with each busbar on the test feeder was uniquely and randomly determined.

[0059] S1. Calculate the correlation between SAIDI and SAIFI to provide a basis for the subsequent construction of the objective function and the selection of the optimization method.

[0060] First, the fault duration and number of faults for each feeder line were grouped. Using this information, a method called "corr" in the Pandas library was used. According to its documentation, this method is used to calculate the Pearson correlation. This calculation method is shown below:

[0061]

[0062] Here, x and y represent the variables, or the features to be observed, and n is the total number of lines in the dataset. This allows us to calculate the correlation between this information to determine the best method for our study. Based on this, the Pearson method considers correlations close to 1 or -1 to be strong. Figure 2 The relationship between the system average outage duration index (SAIDI) and the system average outage frequency index (SAIFI) calculated when two reclosers are installed in the first scenario is shown. Since these attributes are essential information for evaluating reliability indicators, a single-objective approach focusing on a single criterion is sufficient to analyze the current problem.

[0063] S3. Model the objective function, parameterize the GWO algorithm, explore the optimal configuration through different experimental settings, and provide multiple possible solutions and data support for the final determination of the appropriate recloser configuration.

[0064] In optimization problems, the objective function plays a central role, serving as a metric for evaluating the efficiency and quality of an algorithm. As previously mentioned, predictions of future power outages are made using a database of historical faults. Based on this information, it is necessary to separately calculate the impact of each outage on the relevant load blocks configured in the system. To do this, these impacts are determined in terms of both outage duration and frequency. This is illustrated below:

[0065] Impact SAIDI =D fault ×N[h]

[0066] Impact SAIFI =N[customers]

[0067] Among them, D fault is the duration of the failure being evaluated (in hours), and N is the number of customers in the affected block.

[0068] This approach validates the aforementioned assumptions, and only the affected sections are considered when compiling the evaluation metric. After assessing the impact of each documented outage, a comprehensive evaluation metric for continuity of supply is calculated. At this stage, a comprehensive understanding of the total number of customers served by the feeder is essential.

[0069]

[0070] where N feeder is the total number of users in the evaluated feeder, z is the total number of blocks generated, and i represents the block index.

[0071] With this information, we can evaluate the efficiency of each configuration proposed by the algorithm and build a model of the objective function for the problem at hand, as follows:

[0072]

[0073] Among them, w1 and w2 are the weights of each comprehensive indicator, which are set to 0.5 to maintain the equal importance of the indicators. The reason for setting equal weights is to avoid bias towards any specific indicator, as the priority factors may vary. ref and SAIFI ref These two items are indices calculated in the scenario without recloser. The purpose is to normalize the indices because they have different dimensions.

[0074] Finally, the optimization problem to be solved is presented in its complete form as follows:

[0075]

[0076] Among them: Impact SAIDI =D fault ×N、Impact SAIFI =N、

[0077]

[0078] The constraints are: w1=0.5 and w2=0.5.

[0079] To solve the optimal recloser configuration problem, it is necessary to create an algorithm that can interpret the physical characteristics of the distribution network, such as its topology. Furthermore, to address this issue, the Grey Wolf Optimizer (GWO) algorithm must be modified to add constraints and calculate the function to be optimized. Finally, this algorithm must be integrated into the network interpretation algorithm. The GWO algorithm follows these steps:

[0080] S31. Initialization parameters. ① Pack size: Set the number of individuals in the pack (e.g., 50 or 20 wolves). ② Maximum number of iterations: Define the termination condition of the algorithm (e.g., 2000 iterations). ③ Search space dimension: Determine the dimension of the optimization variable (e.g., if 1 to 5 reclosers are installed, the dimension is the corresponding position combination). ④ Objective function: Define the objective function to be minimized (weighted SAIDI and SAIFI). ⑤ Constraints: The number of reclosers is a natural number, and the recloser locations must be mapped to valid nodes of the distribution network (e.g., physical nodes of the IEEE 37 bus system).

[0081] S32. Initialize the wolf pack positions. Randomly generate the initial positions of the wolf pack. Each wolf's position vector represents a possible recloser configuration (e.g., if three reclosers are to be installed, the position is a combination of three nodes).

[0082] S33. Calculate the fitness value. For each wolf position configuration, perform the following operations: ① Network segmentation: Divide the distribution network into multiple independent blocks based on the recloser position. ② Fault impact assessment: Calculate the SAIDI and SAIFI of each block based on the historical fault database. ③ Objective function value: Calculate the weighted normalized objective function value. Smaller values indicate higher fitness.

[0083] S34. Determine the social hierarchy. Rank the wolves according to their fitness values and select the three best wolves as: α wolf (the current global optimal solution), β wolf (the second-best solution), and δ wolf (the third-best solution). The remaining individuals, ω wolves, follow the first three to update their positions.

[0084] S35. Update the wolf pack's position. Update the positions of other wolves based on the positions of α, β, and δ. The specific formula is:

[0085]

[0086] in and is the coefficient vector, and the calculation formula is:

[0087]

[0088] a is a parameter that decreases linearly with the number of iterations (from 2 to 0), and is a random vector in [0,1].

[0089] S36. Constraint processing: Map the consecutive positions to valid nodes of the distribution network (such as rounding or selecting the nearest neighbor node) to ensure that the recloser positions are not repeated (if multiple reclosers are configured).

[0090] S37. Iterative optimization. Repeat steps 3 to 6 until the maximum number of iterations is reached or the objective function converges (e.g., no significant improvement after several consecutive iterations).

[0091] S38. Output the optimal solution. The final position of α wolf is the optimal recloser configuration, corresponding to the minimized SAIDI and SAIFI values.

[0092] S4. Finally, define the test scenarios and test different scenarios in the network to ensure that the optimized recloser configuration scheme can be effective in different actual network environments, and ultimately achieve the optimization of the recloser configuration.

[0093] like Figure 3 As shown, the IEEE 37-node standard feeder is subdivided into different zones. In each simulation scenario, these zones represent the parts of the feeder where faults are most concentrated. These scenarios are generated by assigning weights to each line in the feeder. This results in lines within the highlighted zone receiving higher weights, leading to the identification of more faults in that zone. In each scenario, the highlighted zone experiences 30% more faults than other zones. The numbers within the blue circles represent nodes, and the red numbers represent line identifiers; the substation is located at node 0. Zone 1 is the focus of the first scenario. It is a small end zone of the feeder, making it easy to observe the algorithm's performance in this scenario. Zone 2 is the focus of the second scenario. It is also an end zone of the feeder, but located at the opposite end and with a larger area than the previous scenario. The third scenario proposes concentrating faults in Zone 3, another end zone of the feeder, but located lower down. The final scenario involves Zone 4, located in the center of the feeder. Fault concentration zones are set based on regional characteristics. The algorithm can be tested under various conditions to enhance its applicability and reliability, ensuring that the optimized recloser configuration scheme is effective in different actual network environments, and ultimately achieving optimization of the recloser configuration.

[0094] GWO can conduct a comprehensive and efficient search, is not prone to falling into local optimal solutions, and can find a better medium-voltage distribution network recloser configuration solution, thereby improving power supply reliability. GWO has relatively few parameters and is less sensitive to parameters, which reduces the difficulty of algorithm application and improves the applicability and stability of the algorithm. GWO has a fast convergence speed and can obtain a recloser configuration result that meets the requirements in a relatively short period of time. When the operating status of the power grid changes, the recloser configuration can be adjusted quickly to reduce the power outage time and the scope of the fault. GWO has good adaptability and can better adapt to the complex environment and dynamic changes of the medium-voltage distribution network. Figure 4 and Figure 5 As shown in the data, the average reduction in the system average outage duration index (SAIDI) value ranged from 176.72 hours (when one recloser was configured) to 265.55 hours (when five reclosers were configured), and the average reduction in the system average outage frequency index (SAIFI) value ranged from 52.61 faults (when one recloser was configured) to 80.25 faults (when five reclosers were configured).

[0095] This embodiment provides a method for optimizing recloser configuration in a medium-voltage distribution network based on the Gray Wolf Optimizer. This method ensures the accuracy and representativeness of the fault database by filtering and processing the data, removing irrelevant information and eliminating the influence of existing reclosers, providing a reliable foundation for subsequent analysis. By analyzing the occurrence patterns, frequency, and severity of historical faults, high-prone areas are identified, which facilitates precise recloser configuration and improves distribution network reliability.

[0096] By comprehensively considering the SAIDI and SAIFI indicators, a rational objective function is constructed. Through precise calculation and weight setting, the objective function accurately reflects the operating status of the distribution network, guiding the Gray Wolf optimizer to find the optimal recloser configuration solution and optimize the power supply continuity indicators.

[0097] The Grey Wolf Optimizer algorithm was adapted to incorporate constraints tailored to the physical characteristics of distribution networks, such as network topology and load demand. Appropriate parameters were set in different experimental scenarios, including varying numbers of reclosers and wolf pack sizes, to ensure the algorithm's ability to effectively search for optimal solutions and improve its application to the recloser configuration problem in medium-voltage distribution networks.

[0098] By setting up multiple different fault scenarios and conducting tests under these diverse scenarios, the effectiveness and applicability of the optimization method are fully verified, ensuring that the method can cope with different actual distribution network operation conditions and provide reliable protection for practical applications.

[0099] Finally, it should be noted that 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 above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some or all of the technical features therein. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and description of the present invention.

Claims

1. A method for optimizing reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer, characterized in that: The following steps are involved: S1. Analyze past fault history to identify fault occurrence patterns, frequency, and interruption severity, providing raw data input for the subsequent calculation of the system average outage duration index (SAIDI) and the system average outage frequency index (SAIFI); S2. Calculate the correlation between the system average outage duration index SAIDI and the system average outage frequency index SAIFI to provide a basis for the subsequent construction of the objective function and the selection of the optimization method; S3. Model the objective function, parameterize the Grey Wolf optimization algorithm, explore the optimal configuration through different experimental settings, and provide multiple possible solutions and data support for the final determination of the appropriate recloser configuration; S4. Define test scenarios and test different scenarios in the network to ensure that the optimized recloser configuration solution can be effective in different actual network environments, and ultimately achieve optimization of the recloser configuration.

2. A medium voltage distribution network reclosing configuration optimization method based on the Gray Wolf optimizer according to claim 1, characterized in that: In step S1, by analyzing past fault history, the pattern of fault occurrence is identified, the event frequency is determined, and the severity of the power outage is assessed, providing a realistic baseline for evaluating the impact of recloser installation in subsequent simulations and analyses; and by introducing reclosers and dividing the system into blocks defined by the reclosers, each block is treated as an independent entity.

3. The method for optimizing the reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer according to claim 1, characterized in that: The specific method of step S2 is: grouping the fault duration information and fault number of each feeder line, and calculating the Pearson correlation. The calculation formula is: Here, x and y represent variables, that is, the features to be observed, and n is the total number of lines in the dataset.

4. A method for optimizing reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer according to claim 1, characterized in that: The specific method of step S3 is as follows: the prediction of future power outages will be reflected in the historical fault database, and the impact of each power outage on the relevant load blocks configured in the system will be calculated respectively. These impacts are determined from the two aspects of power outage duration and frequency, as shown below: Impact SAIDI =D fault ×N[h] Impact SAIFI =N[customers] Among them, D fault is the duration of the failure being evaluated, and N is the number of customers in the affected block.

5. A method for optimizing reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer according to claim 4, characterized in that: By this method, the total number of customers who need to be evaluated for service in step S3 is where N feeder is the total number of users in the evaluated feeder, z is the total number of blocks generated, and i represents the block index; After obtaining the total number of customers, construct the objective function model for the current problem as follows: Among them, w1 and w2 are the weights of each comprehensive indicator. To maintain the equal importance of the indicators, they are set to 0.

5. ref and SAIFI ref These two items are indices calculated in the scenario without recloser, and the purpose is to normalize the indices because they have different dimensions; Finally, the optimization problem to be solved is presented in its complete form as follows: Among them: Impact SAIDI =D fault ×N、Impact SAIFI =N、 The constraints are: n reclosers ∈N *+ 、SAIDI∈R *+ 、SAIFI∈R *+ , w1=0.5 and w2=0.

5.

6. A method for optimizing reclosing configuration of a medium voltage distribution network based on a Gray Wolf optimizer according to claim 4, characterized in that: The specific steps of the gray wolf optimization include: S31. Initialize parameters, set the number of individuals in the wolf pack, define the algorithm termination condition, determine the optimization variable dimension, define the objective function to be minimized, the number of reclosers is a natural number, and the recloser locations must be mapped to valid nodes in the distribution network; S32. Initialize the wolf pack positions, randomly generate the initial positions of the wolf pack, and the position vector of each wolf represents a possible coincidencer configuration; S33. Calculate the fitness value. For each wolf position configuration, perform the following operations: Network segmentation: Divide the distribution network into multiple independent blocks based on the recloser locations; Fault impact assessment: Calculate the SAIDI and SAIFI of each block based on the historical fault database; Objective function value: Calculate the weighted normalized objective function value, where a smaller value indicates a higher fitness; S34. Determine the social hierarchy, sort the wolves according to their fitness values, select the three best wolves as α wolf, β wolf, and δ wolf, and the remaining individuals as ω wolf to follow the first three to update their positions; S35. Update the position of the wolf pack. Update the positions of other wolves based on the positions of α, β, and δ. The specific formula is: in and is the coefficient vector, and the calculation formula is: a is a parameter that decreases linearly with the number of iterations, and is a random vector in [0,1]; S36. Constraint processing, mapping continuous positions to valid nodes of the distribution network to ensure that recloser positions are not repeated; S37. Iterative optimization, repeating steps S33 to S36 until the maximum number of iterations is reached or the objective function converges; S38. Output the optimal solution. The final position of α wolf is the optimal recloser configuration, corresponding to the minimized SAIDI and SAIFI values.

Images