Method and device for solving and optimizing self-healing control parameters of power distribution network, computer device and program product

Abstract

The application relates to a power distribution network self-healing regulation parameter solving optimization method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the following steps: randomly generating an initial population in a multi-dimensional solution space, dividing the initial population, and continuously performing iterative updating of the population; monitoring the stagnation state and diversity index of the population; in the case that the stagnation state and the diversity index meet preset conditions, triggering at least one improved strategy among a stagnation disturbance strategy based on golden sine segmentation, a dynamic variation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on congestion degree calculation and niche technology, and updating the population; when the population reaches a preset iteration number or a convergence condition, taking a global optimal solution as a power distribution network self-healing regulation parameter. The method can improve the improved optimization algorithm, and finally enables the power distribution network self-healing regulation, and solves a more optimal and reliable operation control scheme.

Description

Technical Field

[0001] This application relates to the field of parameter optimization technology, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for solving and optimizing the self-healing control parameters of a distribution network. Background Technology

[0002] A high proportion of distributed generation (DG) grid connection has become a prominent feature of new distribution networks. However, the strong volatility of DG output and the high level of power electronics significantly increase the potential disturbance power of the system, leading to a shortage of frequency regulation resources, increased fault probability, and decreased power supply reliability. To improve the self-healing capability and coordinated control level of the distribution network, it is necessary to simultaneously solve multiple tasks such as fault recovery, network reconfiguration, and reactive power optimization. Mathematically, this problem is modeled as a mixed-integer nonlinear programming (MINLP) problem with a large number of discrete-continuous mixed variables, nonlinearity, nonconvexity, high dimension, and multiple extrema.

[0003] Tornado Optimization with Coriolis Force (TOC) is an emerging metaheuristic algorithm that simulates the dynamic lifecycle of storms, thunderstorms, and tornadoes to achieve a balance between exploration and development within the solution space. It has demonstrated good performance in some low-dimensional continuous optimization scenarios. However, when directly applied to the aforementioned distribution network self-healing control scenario, TOC suffers from the following common problems: as the problem dimension increases (≥50 dimensions), the algorithm becomes extremely sensitive to initial values ​​and control parameters, easily getting trapped in local optima. Its convergence accuracy and stability decrease significantly, making it unable to stably obtain globally feasible solutions in a large-scale discrete-continuous coupled complex search space. Therefore, it struggles to meet the requirements of highly reliable and real-time distribution network self-healing control.

[0004] Therefore, overcoming the local optimum trap of TOC in high-dimensional mixed integer optimization has become a single core technical problem that urgently needs to be solved. Summary of the Invention

[0005] Therefore, it is necessary to provide a method, device, computer equipment, computer-readable storage medium, and computer program product for solving and optimizing the self-healing control parameters of distribution networks, which can improve the ability to solve complex parameter optimization problems, in order to address the above-mentioned technical problems.

[0006] Firstly, this application provides an optimization method for solving the self-healing control parameters of a distribution network, including:

[0007] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0008] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0009] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0010] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0011] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0012] In one embodiment, when the stagnation state and the diversity index meet preset conditions, triggering at least one improved strategy among the following: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology, includes:

[0013] When the population is in a stagnant state, the stagnation perturbation strategy based on the golden sine segmentation is used to improve the iterative update of the population.

[0014] When the population is stagnant or the diversity index is below the first threshold, the iterative update of the population is improved based on the dynamic mutation and selection strategy of adaptive distribution.

[0015] When the diversity index is below the second threshold, the iterative update of the population is improved based on the diversity maintenance strategy of crowding calculation and niche technology.

[0016] In one embodiment, the stagnation perturbation strategy based on the golden sine segmentation improves the iterative update of the population, including:

[0017] Within the current population, individuals to be disturbed are identified based on a preset proportion or fitness ranking.

[0018] For each individual to be disturbed, the position is updated based on the golden ratio and the sine function;

[0019] Calculate the new fitness value of the updated position of the individual to be disturbed. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be disturbed.

[0020] In one embodiment, the adaptive distribution-based dynamic mutation and selection strategy improves the iterative update of the population, including:

[0021] Calculate the dynamic mutation probability of the current population, and based on the dynamic mutation probability, determine the individuals to be mutated from the current population;

[0022] For each individual to be mutated, a random perturbation is applied to determine the updated position of the individual to be mutated;

[0023] Calculate the new fitness value of the updated position. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be mutated.

[0024] In one embodiment, the diversity maintenance strategy based on crowding calculation and niche technology improves the iterative updating of the population, including:

[0025] Calculate the crowding degree of all individuals in the current population, and sort them according to the crowding degree to obtain the sorting result;

[0026] Based on the sorting results, a predetermined number of individuals are identified as highly crowded individuals;

[0027] A boundary perturbation is applied to the highly crowded individuals, and their positions are updated based on a nearest-neighbor competition strategy.

[0028] In one embodiment, the formula for calculating the dynamic mutation probability is:

[0029] ;

[0030] Where p is the dynamic mutation probability. The first preset value, The second preset value, This represents the number of iterations for the current population. This is the preset number of iterations.

[0031] Secondly, this application also provides an optimization device for solving the self-healing control parameters of a distribution network, comprising:

[0032] The partitioning module is used to randomly generate an initial population in the multidimensional solution space and divide the initial population into three subgroups: storm, thunderstorm, and tornado.

[0033] An update module is used to continuously iterate and update the population based on the subgroup;

[0034] The monitoring module is used to monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0035] An improvement module is used to, when the stagnation state and the diversity index meet preset conditions, trigger at least one of the following improved strategies to improve the iterative update of the population: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology.

[0036] The output module is used to output the final global optimal solution when a preset number of iterations or convergence conditions are reached, and to use the global optimal solution as the self-healing control parameter of the distribution network.

[0037] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0038] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0039] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0040] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0041] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0042] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0043] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0044] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0045] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0046] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0047] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0048] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0049] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0050] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0051] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0052] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0053] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0054] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0055] The aforementioned method, device, computer equipment, computer-readable storage medium, and computer program product for solving the self-healing control parameters of the distribution network firstly generate an initial population randomly in the multidimensional solution space, dividing the initial population into three subgroups: storm, thunderstorm, and tornado. Based on the subgroups, the population is continuously iteratively updated. The stagnation state and diversity index of the population are monitored during the iterative update process. When the stagnation state and diversity index meet preset conditions, at least one improved strategy is triggered among the following: a stagnation disturbance strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on congestion calculation and niche technology, to improve the iterative update of the population. When the preset number of iterations or convergence conditions are reached, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameters of the distribution network. In this way, an intelligent iterative mechanism is formed by the organic integration of the three strategies: golden sine segmentation, adaptive t-distribution mutation, and congestion niche. This mechanism collaboratively overcomes the three core defects of the original algorithm: "premature convergence," "imbalance between exploration and development," and "loss of diversity." As a result, a strongly improved optimization algorithm with significant improvements in convergence accuracy, robustness, and versatility is obtained. Ultimately, it empowers the self-healing regulation of the distribution network and solves a better and more reliable operation control scheme. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is an application environment diagram of the optimization method for solving the self-healing control parameters of the distribution network in one embodiment;

[0058] Figure 2 This is a flowchart illustrating the optimization method for solving the self-healing control parameters of a distribution network in one embodiment.

[0059] Figure 3 This is a structural block diagram of a device for solving and optimizing the self-healing control parameters of a distribution network in one embodiment;

[0060] Figure 4 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0061] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0062] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0063] The optimization method for solving the self-healing control parameters of the distribution network provided in this application embodiment can be applied to, for example... Figure 1In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located on the cloud or other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. Portable wearable devices can be smartwatches, smart bracelets, head-mounted devices, etc. Head-mounted devices can be virtual reality (VR) devices, augmented reality (AR) devices, smart glasses, etc. Server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0064] In one embodiment, the Tornado Optimization Algorithm (TOC) is a metaheuristic optimization algorithm inspired by the formation and dissipation of tornadoes in nature. This method simulates the dynamic lifecycle of storms and thunderstorms evolving into tornadoes under the influence of the Coriolis force. In the algorithm model, TOC divides the search agents into three levels: storms, thunderstorms, and tornadoes. The initial population is randomly generated, with the best individual selected as a tornado (usually assumed to be only one), secondary best individuals as thunderstorms, and the rest as storms. The core mechanism of the algorithm includes using mathematical models to simulate the evolution of storms into thunderstorms or tornadoes, and the process of thunderstorms evolving into tornadoes.

[0065] In the TOC algorithm model, the search agent is divided into three levels: storm, thunderstorm, and tornado. The algorithm uses specific physical equations to simulate the movement, evolution, and random reorganization of these meteorological units in the search space, and seeks the optimal solution of the objective function through a balanced exploration and exploitation mechanism. Although the TOC algorithm has shown good performance in handling some low-dimensional continuous optimization problems, its original mechanism still faces challenges such as limited convergence accuracy and susceptibility to local optima when facing complex optimization scenarios with high dimensions, multimodalities, and discrete variable coupling in new power distribution networks. Targeted improvements and optimizations are urgently needed.

[0066] In one exemplary embodiment, such as Figure 2 As shown, an optimization method for solving the self-healing control parameters of a distribution network is provided, and this method is applied to... Figure 1 Taking terminal 102 as an example, the explanation includes the following steps 202 to 210. Wherein:

[0067] Step 202: Randomly generate an initial population in the multidimensional solution space and divide the initial population into three subgroups: storm, thunderstorm, and tornado.

[0068] For example, based on the specific problems of self-healing regulation of distribution networks, a tornado optimization model is determined. In the multidimensional solution space, an initial population is randomly generated, and based on the fitness of individuals in the population, the initial population is divided into three subgroups: storm, thunderstorm, and tornado.

[0069] Step 204: Based on the subgroup, continuously iterate and update the population.

[0070] Optionally, the population can be iteratively updated continuously based on the three subgroups of storms, thunderstorms and tornadoes obtained from the initial population division.

[0071] Step 206: Monitor the stagnation state and diversity indicators of the population during the iterative update process.

[0072] For example, during the iterative update of a population, the stagnation state and diversity indicators of the population are monitored.

[0073] Step 208: When the stagnation state and diversity index meet the preset conditions, trigger at least one improved strategy among the following: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0074] For example, during the iterative update of the population, if the stagnation state and diversity indicators meet preset conditions, at least one improved strategy among the following can be triggered to improve the iterative update of the population: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology.

[0075] In one embodiment, all three improvement strategies are triggered throughout the entire iteration of the population, but different strategies are triggered depending on the specific iteration of the population. For example, the stagnation perturbation strategy based on the golden sine division is aimed at the problem of the population getting stuck in local optima; the dynamic mutation and selection strategy based on adaptive distribution is aimed at the problem of algorithm parameter sensitivity and insufficient later development capabilities; and the diversity maintenance strategy based on crowding degree calculation and niche technology is aimed at the problem of premature convergence caused by the loss of population diversity.

[0076] Step 210: If the preset number of iterations or convergence conditions are met, output the final global optimal solution and use the global optimal solution as the self-healing control parameter of the distribution network.

[0077] Optionally, when the number of iterations of the population reaches a preset number or the iterations of the population reach the convergence condition, the final global optimal solution is output, and the final global optimal solution is used as the self-healing control parameter of the distribution network, that is, the optimal parameter solution of the distribution network electrical engineering problem.

[0078] In the aforementioned optimization method for solving the self-healing control parameters of the distribution network, an initial population is randomly generated in the multidimensional solution space and divided into three subgroups: storm, thunderstorm, and tornado. Based on these subgroups, the population is continuously iterated and updated. The stagnation state and diversity indicators of the population are monitored during the iterative update process. When the stagnation state and diversity indicators meet preset conditions, at least one improved strategy is triggered among the following: a stagnation disturbance strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on congestion calculation and niche technology. This improves the iterative update of the population. Upon reaching a preset number of iterations or convergence conditions, the final global optimal solution is output and used as the self-healing control parameters of the distribution network. Thus, through the organic integration of the three strategies—golden sine segmentation, adaptive t-distribution mutation, and congestion-based niche—an intelligent iterative mechanism is formed. This mechanism collaboratively overcomes the three core defects of the original algorithm: "premature convergence," "imbalance between exploration and development," and "loss of diversity." As a result, a strongly improved optimization algorithm with significant improvements in convergence accuracy, robustness, and versatility is obtained. Ultimately, it empowers the self-healing regulation of the distribution network and solves a better and more reliable operation control scheme.

[0079] In an exemplary embodiment, when the stagnation state and diversity index meet preset conditions, at least one improved strategy among the following is triggered: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology. This includes: when the population is in a stagnation state, improving the iterative update of the population using the stagnation perturbation strategy based on golden sine segmentation; when the population is in a stagnation state or the diversity index is below a first threshold, improving the iterative update of the population using the dynamic mutation and selection strategy based on adaptive distribution; and when the diversity index is below a second threshold, improving the iterative update of the population using the diversity maintenance strategy based on crowding calculation and niche technology.

[0080] In practice, when the population is in a stagnant state, the stagnation perturbation strategy based on the golden sine segmentation is used to improve the iterative update of the population; when the population is in a stagnant state or the diversity index is below the first threshold, the dynamic mutation and selection strategy based on adaptive distribution is used to improve the iterative update of the population; when the diversity index is below the second threshold, the diversity maintenance strategy based on crowding calculation and niche technology is used to improve the iterative update of the population.

[0081] Among them, the diversity index reflects the distribution breadth of all candidate solutions (individuals) in the solution space (decision space) or objective function space in the population. It can be calculated by Euclidean distance between all individuals in the population or by the average distance based on the population center, or by other calculation methods. It can be set by the user according to different situations, and this application does not limit it. The first threshold and the second threshold are set by the user according to different situations, and this application does not limit them.

[0082] In the above embodiments, different strategies are executed according to different population conditions during the iteration process, resulting in higher convergence accuracy, stronger robustness, and better computational efficiency when solving complex engineering optimization problems.

[0083] In an exemplary embodiment, a stagnation perturbation strategy based on the golden sine segmentation is used to improve the iterative update of the population, including: determining individuals to be perturbed in the current population based on a preset ratio or fitness ranking; updating the position of each individual to be perturbed based on the golden ratio and a sine function; calculating the new fitness value of the updated position of the individual to be perturbed, and retaining the updated position of the individual to be perturbed if the new fitness value is higher than the position before the update.

[0084] In practice, when the population is in a stagnant state, it indicates that the population may be in a local optimum. A stagnant perturbation strategy based on the golden sine partition, which has strong global optimization capabilities, is introduced. In the current population, individuals to be perturbed are determined based on a preset proportion or fitness ranking. For any individual i to be perturbed in the population, its coordinates in the d-dimensional solution space at the t-th iteration can be represented as... . Let i be the historical best position of individual i in the t-th iteration, and its position update rule is shown in formulas (1)-(3):

[0085] (1)

[0086] (2)

[0087] (3)

[0088] Where R1 is [0,2] A random number within [0, 1] whose value affects the displacement magnitude of the individual in subsequent iterations; R2 is [0, 1] A random number within [ ], used to control the adjustment of the search direction; constant The golden ratio is denoted as .

[0089] Calculate the new fitness value of the updated position of the individual to be disturbed. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be disturbed.

[0090] In the above embodiments, the algorithm using a sine function for iterative optimization performs a fine-grained search of the solution space through the golden ratio coefficient, effectively coordinating the balance between global exploration and local development. The golden ratio coefficient narrows the search range of the solution space, and combined with the periodic oscillation characteristics of the sine function, a golden sine position update operator is constructed. This mechanism, when the algorithm gets stuck in a local optimum, forces the population to escape the current local optimal attraction region through the synergistic effect of compressing the search space and guiding oscillations, achieving rapid approximation to the global optimum.

[0091] In an exemplary embodiment, an improved iterative update of the population is achieved based on an adaptive distribution-based dynamic mutation and selection strategy, including: calculating the dynamic mutation probability of the current population; determining individuals to be mutated from the current population based on the dynamic mutation probability; applying a random perturbation to each individual to be mutated to determine the updated position of the individual; calculating the new fitness value of the updated position; and retaining the updated position of the individual to be mutated if the new fitness value is higher than the position before the update.

[0092] In practice, when the population is stagnant or the diversity index is below the first threshold, it indicates that there may be problems with the sensitivity of algorithm parameters and insufficient development capabilities in the later stages. Therefore, a dynamic mutation and selection strategy based on adaptive distribution is selected to improve the iterative update of the population.

[0093] In one embodiment, the dynamic mutation probability of the current population is calculated, and based on the dynamic mutation probability, individuals to be mutated are determined from the current population; for each individual to be mutated, a random perturbation is applied to determine the updated position of the individual to be mutated. The t-distribution, also known as the Student distribution, is a unimodal structure symmetric about the origin. When its degrees of freedom approach infinity, the t-distribution almost coincides with the normal distribution; while when the degrees of freedom are small, its probability density function exhibits a flatter distribution shape, and the tail probability increases significantly. The formula for calculating its probability density function is shown in formula (4):

[0094] (4)

[0095] Where n represents the degrees of freedom.

[0096] A dynamic perturbation strategy based on the t-distribution is adopted, using the current iteration number as the degree of freedom parameter to adaptively adjust the individual position. This method gives the algorithm strong global exploration characteristics in the early stages of optimization when the degree of freedom parameter is small; in the later iteration stages (when the degree of freedom increases), it exhibits a local fine-tuning search capability similar to Gaussian mutation, effectively improving the convergence speed of the algorithm. The position update formula under this method is shown in formula (5):

[0097] (5)

[0098] in, The location of the individual after the mutation; The position of individual i at the t-th iteration, where iter is the iteration number.

[0099] Calculate the new fitness value of the updated position. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be mutated.

[0100] In the above embodiments, a dynamic mutation and selection strategy based on an adaptive t-distribution is constructed to balance the contradiction between "global exploration" and "local exploitation." To enhance the algorithm's fine-grained optimization capability in the later stages, this case introduces a t-distribution mutation operator whose degrees of freedom dynamically evolve with the iteration process. By controlling the nonlinear evolution of the degrees of freedom from low (heavy-tailed distribution, focusing on global exploration) to high (Gaussian distribution, focusing on local exploitation), and combining it with an adaptive mutation probability P linked to the stagnant state... t Applying dynamic random perturbations to the optimal individual or random individuals in the population effectively enhances the algorithm's ability to deeply mine complex solution spaces.

[0101] In an exemplary embodiment, the iterative update of the population is improved based on a diversity maintenance strategy that combines crowding calculation and niche technology. This includes: calculating the crowding of all individuals in the current population and sorting them according to the crowding to obtain a sorting result; determining a preset number of individuals as highly crowded individuals based on the sorting result; applying boundary perturbation to the highly crowded individuals and updating the position of the highly crowded individuals based on a nearest neighbor competition strategy.

[0102] In practice, when the diversity index is below the second threshold, it indicates that there may be a problem of premature convergence due to loss of population diversity. Based on the diversity maintenance strategy of crowding degree calculation and microhabitat technology, the iterative update of the population is improved. In order to address the problem of population diversity loss in the later stage when the tornado optimization algorithm deals with complex high-dimensional problems, a diversity maintenance strategy based on crowding degree calculation and microhabitat technology is introduced.

[0103] In one embodiment, population sorting is performed on the k-th objective function f. k Sort all individuals in the population according to their objective function values ​​from smallest to largest, and set the crowding distance between the first and last individuals in the sorted population to infinity. For individual i in the middle of the sorted population, calculate its distance contribution in the k-th objective direction, as shown in formula (6):

[0104] (6)

[0105] in, and These are the k-th objective function values ​​of the adjacent individuals after sorting; and These are the maximum and minimum values ​​of the current population on the k-th target, respectively.

[0106] Repeat the above steps until all objective functions are traversed, and finally the total crowding distance L of individual i is obtained. i That is, the crowding degree is the sum of the distances across all target dimensions, and the specific calculation formula is shown in formula (7):

[0107] (7)

[0108] Where M is the number of objective functions. When two individuals are on the same Pareto front, the individual with a larger crowding distance is preferred to ensure a uniform distribution of the solution set.

[0109] The individuals are sorted according to their crowding level. Based on the sorting results, a preset number of individuals are identified as highly crowded individuals.

[0110] Niche techniques are an effective method for promoting population diversity by partitioning the solution space. By dividing the entire population into multiple sub-regions, each containing particles with similar characteristics, and optimizing each region independently, population diversity is maintained.

[0111] First, define the Euclidean distance between any two individuals i and j in the population as d. ij Assume the solution space has dimension D, and the individual solution vector is X = (x1, x2, ..., xn). D If ), then the Euclidean distance d ij The calculation formula is shown in formula (8):

[0112] (8)

[0113] Set a small habitat radius Build shared functions This is used to quantify the similarity between individuals, and the specific calculation formula is shown in formula (9):

[0114] (9)

[0115] in, The parameters control the shape of the shared function. When the distance between individuals is less than the radius, the closer the distance, the larger the shared function value and the heavier the penalty. When the distance exceeds the radius, there is no mutual influence.

[0116] Microhabitat technology for calculating highly crowded individual i i That is, the sum of the population density effects on an individual, and the specific calculation formula is shown in formula (10):

[0117] (10)

[0118] In the formula, N represents the population size.

[0119] The shared fitness of highly crowded individual i is calculated as shown in formula (11):

[0120] (11)

[0121] In the formula, The initial fitness value for highly crowded individuals. This represents the final shared fitness after adjustment. The above steps reduce the probability of solutions with higher fitness being retained, guiding the population to search towards sparser regions.

[0122] In the above embodiments, by monitoring the population diversity index in real time, when the population aggregation degree exceeds the threshold, a niche evolution strategy is triggered to apply strong boundary perturbation to individuals in highly crowded areas to force diffusion. At the same time, a k-nearest neighbor competition strategy is adopted during the update process to ensure that the new solution only competes with the original solution in the spatial neighborhood, thereby effectively protecting the potential excellent solutions with different peak values ​​and maintaining the richness and activity of the population throughout the entire evolutionary cycle.

[0123] In an exemplary embodiment, a dynamic selection probability p is introduced as a regulation mechanism to regulate the application of the t distribution factor. The formula for calculating the dynamic mutation probability is shown in formula (12):

[0124] (12)

[0125] Where p is the dynamic mutation probability. The first preset value, The second preset value, This represents the number of iterations for the current population. This is the preset number of iterations.

[0126] in, Determines the upper limit of the dynamic selection probability. This determines the magnitude of change in the dynamic selection probability, for example, when =0.5, The adjustment effect is optimal when the value is 0.1.

[0127] In the above embodiments, the introduction of the dynamic selection mechanism enables the algorithm to utilize the adaptive mutation operator based on the t-distribution with a high probability in the early stages of optimization to effectively perturb the individual positions, thereby preventing premature convergence.

[0128] To illustrate in detail the method for solving and optimizing the self-healing control parameters of the distribution network in this application, an embodiment is provided below. For example, this application describes the method for solving and optimizing the self-healing control parameters of the distribution network in a specific scenario.

[0129] First, based on the specific problems of self-healing regulation of distribution networks, a tornado optimization model is determined. In the multidimensional solution space, an initial population is randomly generated, and based on the fitness of individuals in the population, the initial population is divided into three subgroups: storm, thunderstorm, and tornado.

[0130] Based on the initial population division into three subgroups—storms, thunderstorms, and tornadoes—the population is continuously iterated and updated. During the iterative update process, the stagnation state and diversity indicators of the population are monitored.

[0131] During the iterative update of the population, when the stagnation state and diversity indicators meet the preset conditions, at least one of the following improved strategies is triggered: the stagnation perturbation strategy based on golden sine segmentation, the dynamic mutation and selection strategy based on adaptive distribution, and the diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0132] When the number of iterations of the population reaches the preset number or the population reaches the convergence condition, the final global optimal solution is output. The final global optimal solution is used as the self-healing control parameter of the distribution network, that is, the optimal parameter solution of the distribution network electrical engineering problem.

[0133] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0134] Based on the same inventive concept, this application also provides a device for solving and optimizing the self-healing control parameters of a distribution network, used to implement the aforementioned method for solving and optimizing the self-healing control parameters of a distribution network. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations in one or more embodiments of the device for solving and optimizing the self-healing control parameters of a distribution network provided below can be found in the limitations of the method for solving and optimizing the self-healing control parameters of a distribution network described above, and will not be repeated here.

[0135] In one exemplary embodiment, such as Figure 3 As shown, a device for solving and optimizing the self-healing control parameters of a distribution network is provided, comprising: a partitioning module 301, an updating module 302, a monitoring module 303, an improvement module 304, and an output module 305, wherein:

[0136] The partitioning module is used to randomly generate an initial population in the multidimensional solution space and divide the initial population into three subgroups: storm, thunderstorm, and tornado.

[0137] The update module is used to continuously iterate and update the population based on the subgroup.

[0138] The monitoring module is used to monitor the stagnation state and diversity indicators of the population during iterative updates.

[0139] An improvement module is used to trigger at least one of the following improved strategies when the stagnation state and the diversity index meet preset conditions: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0140] The output module is used to output the final global optimal solution when a preset number of iterations or convergence conditions are reached, and to use the global optimal solution as the self-healing control parameter of the distribution network.

[0141] In one exemplary embodiment, the improved module described above is further configured to:

[0142] When the population is in a stagnant state, the stagnation perturbation strategy based on the golden sine segmentation is used to improve the iterative update of the population.

[0143] When the population is stagnant or the diversity index is below the first threshold, the iterative update of the population is improved based on the dynamic mutation and selection strategy of adaptive distribution.

[0144] When the diversity index is below the second threshold, the iterative update of the population is improved based on the diversity maintenance strategy of crowding calculation and niche technology.

[0145] In one exemplary embodiment, the improved module described above is further configured to:

[0146] Within the current population, individuals to be disturbed are identified based on a preset proportion or fitness ranking.

[0147] For each individual to be disturbed, the position is updated based on the golden ratio and the sine function;

[0148] Calculate the new fitness value of the updated position of the individual to be disturbed. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be disturbed.

[0149] In one exemplary embodiment, the improved module described above is further configured to:

[0150] Calculate the dynamic mutation probability of the current population, and based on the dynamic mutation probability, identify the individuals to be mutated from the current population;

[0151] For each individual to be mutated, apply a random perturbation to determine the updated position of the individual to be mutated;

[0152] Calculate the new fitness value of the updated position. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be mutated.

[0153] In one exemplary embodiment, the improved module described above is further configured to:

[0154] Calculate the crowding degree of all individuals in the current population, and sort them according to the crowding degree to obtain the sorting result;

[0155] Based on the ranking results, a predetermined number of individuals are identified as highly crowded individuals.

[0156] Apply boundary perturbations to highly crowded individuals and update their positions based on a nearest-neighbor competition strategy.

[0157] In one exemplary embodiment, the above-described apparatus further includes a calculation module for calculating the dynamic mutation probability:

[0158] ;

[0159] Where p is the dynamic mutation probability. The first preset value, The second preset value, This represents the number of iterations for the current population. This is the preset number of iterations.

[0160] Each module in the aforementioned optimization device for solving the self-healing control parameters of the distribution network can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0161] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 4 As shown, the computer device includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage media. The input / output interface is used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When the computer program is executed by the processor, it implements an optimization method for solving self-healing control parameters of a power distribution network.

[0162] The display unit of this computer device is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of this computer device can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad set on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0163] Those skilled in the art will understand that Figure 4 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0164] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0165] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0166] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0167] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0168] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0169] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0170] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0171] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0172] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0173] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0174] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0175] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0176] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps:

[0177] An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado.

[0178] Based on the aforementioned subgroups, the population is continuously iteratively updated;

[0179] Monitor the stagnation state and diversity indicators of the population during the iterative update process;

[0180] When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population.

[0181] If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

[0182] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0183] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0184] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0185] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for solving and optimizing the self-healing control parameters of a distribution network, characterized in that, The method includes: An initial population is randomly generated in the multidimensional solution space, and the initial population is divided into three subgroups: storm, thunderstorm and tornado. Based on the aforementioned subgroups, the population is continuously iteratively updated; Monitor the stagnation state and diversity indicators of the population during the iterative update process; When the stagnation state and the diversity index meet the preset conditions, at least one improved strategy among the following is triggered: stagnation perturbation strategy based on golden sine segmentation, dynamic mutation and selection strategy based on adaptive distribution, and diversity maintenance strategy based on crowding calculation and niche technology, to improve the iterative update of the population. If the preset number of iterations or convergence conditions are met, the final global optimal solution is output, and the global optimal solution is used as the self-healing control parameter of the distribution network.

2. The method according to claim 1, characterized in that, When the stagnation state and the diversity index meet preset conditions, at least one improved strategy is triggered among the following: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology. This includes: When the population is in a stagnant state, the stagnation perturbation strategy based on the golden sine segmentation is used to improve the iterative update of the population. When the population is stagnant or the diversity index is below the first threshold, the iterative update of the population is improved based on the dynamic mutation and selection strategy of adaptive distribution. When the diversity index is below the second threshold, the iterative update of the population is improved based on the diversity maintenance strategy of crowding calculation and niche technology.

3. The method according to claim 2, characterized in that, The stagnation perturbation strategy based on the golden sine segmentation improves the iterative update of the population, including: Within the current population, individuals to be disturbed are identified based on a preset proportion or fitness ranking. For each individual to be disturbed, the position is updated based on the golden ratio and the sine function; Calculate the new fitness value of the updated position of the individual to be disturbed. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be disturbed.

4. The method according to claim 2, characterized in that, The adaptive distribution-based dynamic mutation and selection strategy improves the iterative update of the population, including: Calculate the dynamic mutation probability of the current population, and based on the dynamic mutation probability, determine the individuals to be mutated from the current population; For each individual to be mutated, a random perturbation is applied to determine the updated position of the individual to be mutated; Calculate the new fitness value of the updated position. If the new fitness value is higher than the position before the update, retain the updated position of the individual to be mutated.

5. The method according to claim 2, characterized in that, The diversity maintenance strategy based on crowding calculation and niche technology improves the iterative update of the population, including: Calculate the crowding degree of all individuals in the current population, and sort them according to the crowding degree to obtain the sorting result; Based on the sorting results, a predetermined number of individuals are identified as highly crowded individuals; A boundary perturbation is applied to the highly crowded individuals, and their positions are updated based on a nearest-neighbor competition strategy.

6. The method according to claim 4, characterized in that, The formula for calculating the dynamic mutation probability is: ； Where p is the dynamic mutation probability. The first preset value, The second preset value, This represents the number of iterations for the current population. This is the preset number of iterations.

7. A device for solving and optimizing the self-healing control parameters of a distribution network, characterized in that, The device includes: The partitioning module is used to randomly generate an initial population in the multidimensional solution space and divide the initial population into three subgroups: storm, thunderstorm, and tornado. An update module is used to continuously iterate and update the population based on the subgroup; The monitoring module is used to monitor the stagnation state and diversity indicators of the population during the iterative update process; An improvement module is used to, when the stagnation state and the diversity index meet preset conditions, trigger at least one of the following improved strategies to improve the iterative update of the population: a stagnation perturbation strategy based on golden sine segmentation, a dynamic mutation and selection strategy based on adaptive distribution, and a diversity maintenance strategy based on crowding calculation and niche technology. The output module is used to output the final global optimal solution when a preset number of iterations or convergence conditions are reached, and to use the global optimal solution as the self-healing control parameter of the distribution network.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images