Distributed photovoltaic bearing capacity assessment method and system

Abstract

The invention discloses a distributed photovoltaic bearing capacity assessment method and system, and the method comprises the steps: obtaining distributed photovoltaic historical data, and carrying out the first preprocessing of the historical data, and obtaining first historical data; performing feature extraction on the first historical data, calculating the distance of feature data, and constructing a first scene in combination with a first clustering algorithm; based on the first scene, different distributed photovoltaic access power distribution network schemes are obtained through simulation, a multi-target optimization model is constructed, the multi-target optimization model is solved, and photovoltaic bearing capacities under the different distributed photovoltaic access power distribution network schemes are obtained; based on the photovoltaic bearing capacity, a comprehensive evaluation index system for the distributed photovoltaic access power distribution network is established, the bearing capacity of the power distribution network under different distributed photovoltaic access power distribution network schemes is evaluated, a comprehensive evaluation result is obtained, and the solving efficiency, the solving quality and the algorithm adaptability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of distributed photovoltaic, and particularly to a method and system for evaluating the carrying capacity of distributed photovoltaic. Background Art

[0002] The usage ratio of distributed wind power and photovoltaic is increasing continuously, and the penetration rate of distributed power sources in the distribution network system is rising constantly. The large-scale access of distributed photovoltaic power sources and other loads has caused great changes in the structure and operation state of the traditional distribution network. Due to the increasingly large scale of distributed photovoltaic access to the distribution network, its intermittent and random characteristics will bring many adverse effects to the distribution network, such as changing the power flow distribution of the distribution network, voltage over-limit, sharp increase in harmonics, power flow reversal, increase in network loss, malfunction of relay protection, etc., which poses a severe challenge to its full consumption. Therefore, it is very necessary to study the evaluation of the carrying capacity of large-scale distributed photovoltaic access to the distribution network.

[0003] Regarding the evaluation of the carrying capacity of distributed photovoltaic in the distribution network, scholars at home and abroad have carried out a large number of studies, mainly including the evaluation of the access location, access capacity and maximum carrying capacity of distributed photovoltaic. Since the problem of the maximum access capacity of distributed photovoltaic is usually a non-linear optimization problem, most of the existing studies use the random scenario simulation method or intelligent algorithms to solve it, such as the simulated annealing algorithm, particle swarm algorithm, random scenario simulation method, etc. The above methods have problems such as large computational amount and insufficient stability. Moreover, the optimization objective is relatively single, mainly considering a large distributed access capacity, and less considering economy. In addition, most studies obtain the access capacity of distributed photovoltaic under certain constraint conditions, or separately analyze the impact of photovoltaic access or load change on the distribution network, without deeply studying the mutual influence of source-load changes and the constraint conditions of distributed photovoltaic access under the new power system. At the same time, most of the existing studies regard the impact of distributed photovoltaic power sources on the distribution network as one of the indicators, and the established evaluation index system for the carrying capacity of distributed photovoltaic access to the distribution network is not comprehensive, scientific and objective enough, and does not comprehensively consider the impact of each index on the performance of the distribution network from the subjective and objective levels, making the evaluation results lack reliability and having certain limitations. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method and system for evaluating the carrying capacity of distributed photovoltaic to solve the problems existing in the current existing methods, such as large computational amount, insufficient stability, relatively single optimization objective, the established evaluation index system for the carrying capacity of distributed photovoltaic access to the distribution network is not comprehensive, scientific and objective enough, and does not comprehensively consider the impact of each index on the performance of the distribution network from the subjective and objective levels, making the evaluation results lack reliability and having certain limitations.

[0006] To solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a method for evaluating the carrying capacity of distributed photovoltaic power generation, including:

[0008] Obtaining historical data of distributed photovoltaic power generation, performing a first preprocessing on the historical data to obtain first historical data;

[0009] Performing feature extraction on the first historical data, calculating the distances of the feature data, and constructing a first scenario in combination with a first clustering algorithm;

[0010] Based on the first scenario, simulating different distributed photovoltaic access distribution network schemes and constructing a multi-objective optimization model, solving the multi-objective optimization model to obtain the photovoltaic carrying capacity under different distributed photovoltaic access distribution network schemes;

[0011] Based on the photovoltaic carrying capacity, establishing a comprehensive evaluation index system for distributed photovoltaic access to the distribution network, evaluating the carrying capacity of the distribution network under different distributed photovoltaic access distribution network schemes, and obtaining a comprehensive evaluation result.

[0012] As a preferred solution of the method for evaluating the carrying capacity of distributed photovoltaic power generation according to the present invention, wherein: calculating the distances of the feature data includes:

[0013] Calculating a first distance between the feature data, and constructing a first distance matrix according to the calculation result;

[0014] Based on the first distance matrix, discretizing the continuous probability distribution of the feature data to obtain an optimal discrete scenario set and probability values, and constructing a distribution scenario.

[0015] As a preferred solution of the method for evaluating the carrying capacity of distributed photovoltaic power generation according to the present invention, wherein: constructing a first scenario in combination with a first clustering algorithm includes:

[0016] Using the first clustering algorithm to perform iterative clustering on the scenarios in the optimal discrete scenario set;

[0017] Determining the optimal number of clusters through a first clustering evaluation method to obtain a first scenario.

[0018] As a preferred solution of the method for evaluating the carrying capacity of distributed photovoltaic power generation according to the present invention, wherein: constructing a multi-objective optimization model includes:

[0019] The objective function of the multi-objective optimization model is to maximize the distributed photovoltaic capacity included in the distribution network and optimize the economic quality;

[0020] The constraint conditions include equality constraints and inequality constraints.

[0021] As a preferred solution of the distributed photovoltaic carrying capacity evaluation method described in the present invention, wherein: solving the multi-objective optimization model includes:

[0022] Solving the multi-objective optimization model by using a first solving algorithm;

[0023] Using a constraint handling method to perform linearization and convex relaxation processing on the multi-objective optimization model, and dynamically adjusting the crossover and mutation probabilities, and calculating the photovoltaic carrying capacity under different distributed photovoltaic access distribution network schemes.

[0024] As a preferred solution of the distributed photovoltaic carrying capacity evaluation method described in the present invention, wherein: evaluating different distributed photovoltaic access distribution network schemes includes:

[0025] Constructing a comprehensive evaluation model to evaluate the carrying capacity of the distribution network under different distributed photovoltaic access distribution network schemes;

[0026] Based on the calculation results of the distributed photovoltaic access capacity under different distributed photovoltaic access distribution network schemes, obtaining index values, and calculating the subjective weight and objective weight of the index values to obtain a comprehensive weight.

[0027] As a preferred solution of the distributed photovoltaic carrying capacity evaluation method described in the present invention, wherein: it further includes:

[0028] Fuzzifying the index values and constructing a fuzzy relation matrix;

[0029] Combining the analytic hierarchy process and the entropy weight method to determine the evaluation index weight to obtain a fuzzy evaluation result;

[0030] Based on the comprehensive weight and the fuzzy evaluation result, calculating the closeness of different schemes, evaluating and sorting the carrying capacity of the distribution network under different distributed photovoltaic access distribution network schemes, and obtaining the evaluation results of different distributed photovoltaic access distribution network schemes.

[0031] In a second aspect, the present invention provides a distributed photovoltaic carrying capacity evaluation system, including:

[0032] A preprocessing module, configured to obtain historical data of distributed photovoltaic, and perform first preprocessing on the historical data to obtain first historical data;

[0033] A scenario construction module, configured to extract features from the first historical data, calculate the distance of the feature data, and construct a first scenario by combining a first clustering algorithm;

[0034] A model construction module, configured to simulate different distributed photovoltaic access to distribution network schemes based on a first scenario, construct a multi-objective optimization model, solve the multi-objective optimization model, and obtain the photovoltaic carrying capacity under different distributed photovoltaic access to distribution network schemes;

[0035] An evaluation module, configured to establish a comprehensive evaluation index system for distributed photovoltaic access to a distribution network based on the photovoltaic carrying capacity, evaluate the carrying capacity of the distribution network under different distributed photovoltaic access to distribution network schemes, and obtain a comprehensive evaluation result.

[0036] In a third aspect, the present invention provides a computing device, including:

[0037] A memory and a processor;

[0038] The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the distributed photovoltaic carrying capacity evaluation method are implemented.

[0039] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, and when the computer-executable instructions are executed by a processor, the steps of the distributed photovoltaic carrying capacity evaluation method are implemented.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention uses the Wasserstein distance and the improved K-Means clustering method to generate the joint probability scenario of photovoltaic-load in the distribution network, obtains the typical scenario set and its probability value, and improves the efficiency of subsequent carrying capacity optimization calculation while ensuring a certain accuracy. Based on the joint probability scenario of photovoltaic-load in the distribution network and the actual situation, a comprehensive evaluation index system for the carrying capacity of large-scale distributed photovoltaic access to the distribution network is constructed, which comprehensively, scientifically and objectively reflects the comprehensive carrying capacity of the distribution network for distributed photovoltaic from multiple levels. At the same time, considering the capacity and economic quality of distributed photovoltaic access to the distribution network, a carrying capacity model for distributed photovoltaic access to the distribution network is established; a linearization and relaxation process is performed on the multi-objective optimization model by using linearization and convex relaxation methods. Considering constraint conditions such as node voltage over-limit, harmonic pollution, and distributed power source capacity, an improved NSGA-II algorithm is used to solve the distributed photovoltaic carrying capacity under different typical scenarios, improving the solution efficiency, quality and algorithm adaptability. The evaluation method based on the combined weight and TOPSIS method fully considers subjective and objective factors, calculates the subjective and objective weights of each index by using the analytic hierarchy process and the entropy weight method respectively, calculates the combined weight by using the evidence theory, and the process of weighting each index for different scenarios and different access schemes by using this method is more objective and reasonable, and finally obtains the evaluation ranking results of each photovoltaic access scheme. Description of the Drawings

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the description of the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0042] Figure 1 It is a schematic diagram of the overall process logic of the distributed photovoltaic carrying capacity evaluation method described in an embodiment of the present invention. Specific embodiments

[0043] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the accompanying drawings of the specification. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the scope of protection of the present invention.

[0044] Embodiment 1

[0045] Referring to Figure 1 , an embodiment of the present invention provides a distributed photovoltaic carrying capacity evaluation method, including:

[0046] S100: Obtain the historical data of the distributed photovoltaic, perform a first preprocessing on the historical data to obtain the first historical data;

[0047] In the embodiments of the present application, the historical data of the distributed photovoltaic includes the distributed photovoltaic power generation data connected to the power grid, the load data of the power grid nodes, relevant economic quality parameters, and relevant meteorological data;

[0048] Specifically, the distributed photovoltaic power generation data includes the number, capacity, and output data of the distributed photovoltaic power sources; the load data of the power grid nodes includes the number and load magnitude of the power grid nodes; the relevant quality parameters include the installation cost per unit capacity of the distributed photovoltaic, the maintenance cost per unit capacity, the economic service life of the distributed photovoltaic, the discount rate, the annual loss cost of the distribution network, the annual active management cost, the unit power loss cost, the active management cost per unit output, the annual surplus power grid connection income of the distributed photovoltaic, the unit power sales income, etc.; the meteorological data includes the light intensity, temperature, humidity, etc.;

[0049] In an optional embodiment, the first preprocessing includes outlier processing. An outlier is a data point that significantly deviates from other observations in the dataset. Outliers are identified through statistical methods or distribution-based methods, and the outliers are deleted or replaced with a specific value or estimated using interpolation methods;

[0050] In an alternative embodiment, the first preprocessing further includes missing value imputation, identifying missing values in the dataset and marking them, and using imputation methods for imputation. The imputation methods can be deleting observations containing missing values, using statistics for imputation, using interpolation methods, and using machine learning methods;

[0051] In an alternative embodiment, the first preprocessing further includes normalization processing. The data is scaled proportionally so that it falls within a specific range, and a normalization method is selected and applied to each feature in the dataset.

[0052] Specifically, by performing outlier processing, missing value imputation, normalization processing, etc. on the historical data, the first historical data, i.e., the preprocessed historical data, is obtained.

[0053] It should be noted that by obtaining multi-dimensional historical data including distributed photovoltaic power generation data, grid node load data, economic quality parameters, and meteorological data, the operating state of the distributed photovoltaic system and external environmental factors can be more comprehensively reflected. Preprocessing the historical data can effectively improve the accuracy of the data.

[0054] S200: Extract features from the first historical data, calculate the distances of the feature data, and construct the first scenario in combination with the first clustering algorithm;

[0055] S300: Based on the first scenario, simulate different distributed photovoltaic access to the distribution network schemes and construct a multi-objective optimization model, and solve the multi-objective optimization model to obtain the photovoltaic carrying capacity under different distributed photovoltaic access to the distribution network schemes;

[0056] S400: Based on the photovoltaic carrying capacity, establish a comprehensive evaluation index system for distributed photovoltaic access to the distribution network, evaluate the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes, and obtain a comprehensive evaluation result;

[0057] It should be noted that this application is not only applicable to the current distribution network structure, but also can adapt to the development and changes of the future distribution network, with strong adaptability and foresight, and has remarkable effects in improving the evaluation efficiency, ensuring the reliability and objectivity of the evaluation results, optimizing model solving, having strong adaptability, and promoting the utilization of renewable energy.

[0058] In the embodiment of this application, the above step S200 includes the following sub-steps A1 - A2;

[0059] In A1: Calculate the first distance between the feature data, and construct the first distance matrix according to the calculation result;

[0060] In A2: Based on the first distance matrix, discretize the continuous probability distribution of the feature data to obtain the optimal discrete scenario set and probability values, and construct the distribution scenario;

[0061] Specifically, the first distance is the Wassertein distance, and the first distance matrix is the Wassertein distance matrix;

[0062] Extract the characteristics of distributed photovoltaic power sources and loads from the preprocessed historical data, including periodicity, seasonality, trends, etc.; describe the uncertainties of photovoltaic output and load with Beta distribution and normal distribution respectively, and conduct correlation analysis on them;

[0063] Calculate the Wassertein distance between the feature data, and construct the Wassertein distance matrix;

[0064] Use the Wassertein distance to measure the differences between different common scenarios, obtain the optimal quantiles and corresponding probability values of the continuous probability distributions of photovoltaic power generation and load power at each moment, that is, the optimal discrete scenario set and its probability values, discretize the continuous probability distributions of photovoltaic output and load power into several probabilities, and construct the photovoltaic-load joint probability distribution scenario.

[0065] In the embodiment of the present application, after completing steps A1 - A2 in the above step S200, the following steps A3 - A4 are further included;

[0066] In A3: Use the first clustering algorithm to iteratively cluster the scenarios in the optimal discrete scenario set;

[0067] In A4: Determine the optimal number of clusters through the first clustering evaluation method to obtain the first scenario.

[0068] In an alternative embodiment, the first clustering algorithm may include DBSCAN clustering. For each point, find all the points (density-reachable points) within its eps neighborhood. If the number of density-reachable points of a point is greater than or equal to minPts, then the point is a core point. Form clusters by connecting the core points and their density-reachable points. Non-core points are assigned to the corresponding clusters if they are within the neighborhood of a core point; points that do not belong to any cluster are regarded as noise. The first clustering algorithm may also include hierarchical clustering. Consider each data point as a separate cluster, calculate the distances between all clusters, merge the two closest clusters to form a new cluster, update the distances between clusters, and repeat the steps until all points are merged into one cluster or the required number of clusters is reached. The final number of clusters can be determined by cutting the dendrogram.

[0069] In an alternative embodiment, the first clustering algorithm may further include spectral clustering. A similarity matrix is constructed to represent the similarity between data points. The graph Laplacian matrix of the similarity matrix is calculated. The eigenvectors and eigenvalues of the graph Laplacian matrix are calculated. The eigenvectors corresponding to the smallest k non-zero eigenvalues are selected to form an eigenvector matrix. The eigenvector matrix is used as input, and the K-Means or other clustering algorithms are applied for clustering. According to the clustering results, the original data points are assigned to different clusters.

[0070] In the embodiment of the present application, the first clustering algorithm includes improved K-Means clustering.

[0071] The K-Means++ algorithm is used to select the initial centroids to improve the clustering effect. The Wasserstein distance is used as the distance metric in the clustering process to better reflect the differences between scenarios. In each iteration, the data points are assigned to the nearest centroid according to the Wasserstein distance, and the centroid positions are updated. When the change in the centroid positions is less than a certain threshold or the maximum number of iterations is reached, the algorithm converges.

[0072] It should be noted that the Wasserstein distance and the improved K-Means clustering method are used to generate the joint probability scenarios of photovoltaic-load in the distribution network, and the typical scenario set and its probability values are obtained. Under the condition of ensuring a certain accuracy, the efficiency of the subsequent bearing capacity optimization calculation is improved.

[0073] In an alternative embodiment, the first clustering evaluation method includes the elbow method, which is applicable to algorithms that require specifying the number of clusters, such as K-means and spectral clustering. By drawing the graph of the sum of squared errors (SSE) within the clusters corresponding to different numbers of clusters, the "elbow" point is selected as the number of clusters. The first clustering evaluation method may also include the silhouette coefficient, which is applicable to all clustering algorithms. By calculating the ratio of the average distance from each point to other points within its cluster to the average distance to the nearest other cluster, the compactness and separation of the clusters are evaluated.

[0074] In an alternative embodiment, the first clustering evaluation method further includes the DBI index, which is applicable to all clustering algorithms. By calculating the average similarity between all clusters, the clustering effect is determined. The smaller the DBI value, the better the clustering effect.

[0075] In the embodiment of the present application, the optimal number of clusters is determined by combining methods such as the elbow method, the silhouette coefficient, and the DBI index. For a series of k values, the WCSS, silhouette coefficient, and DBI of the elbow method are calculated respectively. Considering the k value of the elbow position, the maximum silhouette coefficient, and the minimum DBI comprehensively, a suitable number of clusters is selected to obtain the first scenario, that is, the reduced typical scenario set and its probability values.

[0076] It should be noted that by determining the number of clusters through scientific methods, it can ensure that the selected scenarios can more accurately reflect the true distribution of distributed photovoltaics and loads in the distribution network. The reduced set of typical scenarios reduces the unnecessary number of scenarios, thereby improving the calculation efficiency and reducing the consumption of computing resources in subsequent carrying capacity assessments.

[0077] In the embodiment of the present application, the above step S300 includes the following sub-steps B1 - B2;

[0078] In B1: The objective function of the multi-objective optimization model is to maximize the distributed photovoltaic capacity included in the distribution network and optimize the economic quality;

[0079] In B2: The constraint conditions include equality constraints and inequality constraints.

[0080] Specifically, based on the reduced typical scenarios, considering the purpose of distributed photovoltaic development at the same time, a random scenario simulation method is used to randomly simulate multiple distributed photovoltaic access schemes to the distribution network, including the number of distributed photovoltaic accesses and the access locations;

[0081] Specifically, considering the photovoltaic access capacity, reverse power flow, node voltage violation, harmonic pollution, and economy, etc., a multi-objective optimization model for the distributed photovoltaic carrying capacity is constructed. The multi-objective optimization model for the distributed photovoltaic carrying capacity is expressed as:

[0082]

[0083] Among them, x is the decision variable vector, f s (x)(i = 1, 2, …, m) is the s-th objective function, m is the number of objective functions, h d (x) and g k (x) are the equality constraint and the inequality constraint respectively, and p and q are the numbers of the equality constraint and the inequality constraint respectively;

[0084] The multi-objective optimization model includes two objective functions, which are to maximize the distributed photovoltaic capacity included in the distribution network and optimize the economic quality respectively, and are expressed as:

[0085]

[0086] f2 = min(C M + C L - B O )

[0087] Among them, P DPV,i is the distributed photovoltaic access amount of the i-th node, N is the number of nodes where the distributed photovoltaic accesses the power grid, C M is the installation, operation and maintenance cost of the distributed photovoltaic in one year, C L is the active management and loss cost of the distribution network in one year, BO The annual comprehensive operation income of distributed photovoltaic includes annual power generation income, income from selling surplus electricity to the grid, and government subsidy income;

[0088] The constraint conditions of the distributed photovoltaic carrying capacity model include equality constraints and inequality constraints;

[0089] The equality constraints include branch power flow constraints;

[0090] The inequality constraints include distributed photovoltaic output constraints, node voltage constraints, line current-carrying capacity constraints, total voltage harmonic distortion rate constraints, transformer reverse load rate constraints, and prohibited power reverse transmission constraints;

[0091] Specifically, the power flow equality constraint is expressed as:

[0092]

[0093] Among them, U i is the voltage of any node, Q DPV,i is the reactive power output of the distributed photovoltaic connected to node i, G ij , B ij are the line susceptances between nodes i and j respectively, and θ ij is the power angle between nodes i and j;

[0094] The distributed photovoltaic output constraint is expressed as:

[0095]

[0096] Among them, Q DPV,i is the reactive power output of the distributed photovoltaic connected to node i, P DPV,i,max , Q DPV,i,max are the upper limits of the active and reactive power of the distributed photovoltaic on node i respectively;

[0097] The node voltage constraint is expressed as:

[0098] U min ≤U i ≤U max

[0099] Among them, U min , U max are the lower and upper limit values of the node voltage under specific voltage levels respectively;

[0100] The line current-carrying capacity constraint is expressed as:

[0101] I l ≤I l,max

[0102] Among them, I l is the current on line l, I l,maxIt is the maximum current value that the line l allows to transmit;

[0103] The total voltage harmonic constraint is expressed as:

[0104] THD U ≤δ

[0105] Where, THD U is the total voltage harmonic distortion rate, and δ is the limit value of the total voltage harmonic distortion rate under a specific voltage level;

[0106] The reverse load rate constraint of the transformer is expressed as:

[0107]

[0108] Where, S e is the actual operation limit of the transformer, and λ max is the maximum value of the reverse load rate of the transformer;

[0109] The constraint against power reverse transmission is expressed as:

[0110] P re ≤0

[0111] Where, P re is the power transmitted from the low-voltage side node to the high-voltage side node of the distribution network;

[0112] For inequality constraint conditions, methods such as the penalty function method, the Lagrange multiplier method, and the sequential quadratic programming method can be used to transform them, so as to improve the solution quality and efficiency and reduce computing resources.

[0113] It should be noted that by maximizing the distributed photovoltaic capacity and optimizing the economic quality, it helps to more effectively allocate and utilize the grid resources. By considering constraint conditions such as node voltage over-limit and harmonic pollution, it helps to improve the stability and reliability of the grid operation.

[0114] In the embodiment of the present application, after completing steps B1 - B2 in the above step S300, the following steps B3 - B4 are further included;

[0115] In B3: Use the first solution algorithm to solve the multi-objective optimization model;

[0116] In B4: Use the constraint handling method to perform linearization and convex relaxation processing on the multi-objective optimization model, and dynamically adjust the crossover and mutation probabilities, and calculate the photovoltaic carrying capacity under different distributed photovoltaic access distribution network schemes;

[0117] In an alternative embodiment, the first solution algorithm includes a multi-objective particle swarm algorithm. A group of particles (solutions) is randomly initialized, where each particle represents a potential solution. The fitness value of each particle, i.e., the values of multiple objective functions, is calculated. The velocity and position of the particles are updated based on the individual best and global best positions. If a particle violates the constraint conditions, correction or penalty is performed. The evaluation and update steps are repeated until the termination condition is met. The first solution algorithm also includes a multi-objective simulated annealing algorithm. An initial solution is selected, and the initial temperature and cooling schedule are set. A new solution is randomly selected within the neighborhood of the current solution, and the fitness value of the new solution is calculated. Whether to accept the new solution is determined according to the acceptance criterion of simulated annealing. This usually involves a probability function that takes into account the improvement of the objective function and the temperature. The temperature is gradually reduced, and the neighborhood search and acceptance criterion steps are repeated until the termination condition is met.

[0118] In an alternative embodiment, the first solution algorithm may also include a decomposition-based multi-objective evolutionary algorithm. The multi-objective optimization problem is decomposed into multiple single-objective optimization sub-problems. A group of solutions is randomly initialized for each sub-problem. The sub-problems interact with each other by sharing information to guide the search process. The solutions of each sub-problem are updated according to the interaction information. The solutions of all sub-problems are combined to form a set of approximate Pareto optimal solutions. The interaction and update steps are repeated until the termination condition is met.

[0119] In the embodiment of the present application, the first solution algorithm includes an improved NSGA-II algorithm. The improved NSGA-II algorithm is used to solve the multi-objective optimization model. The big M method is used for its linearization and convex relaxation processing, and the search efficiency and adaptability of the algorithm are improved by dynamically adjusting the crossover and mutation probabilities. The photovoltaic carrying capacity under different access schemes is calculated.

[0120] Specifically, the multi-objective optimization model of distributed photovoltaic carrying capacity belongs to a non-convex and non-linear problem and needs to be linearized. Methods such as Taylor series expansion, piecewise linearization, variable substitution, and cone relaxation can be used for linearization and convex relaxation processing to convert it into a linear programming problem, and existing commercial solvers such as CPLEX and Gurobi can be used for rapid solution.

[0121] It should be noted that considering the capacity and economic quality of distributed photovoltaic access to the distribution network, a distributed photovoltaic access to the distribution network carrying capacity model is established. The multi-objective optimization model is linearized and relaxed by using linearization and convex relaxation methods. At the same time, constraint conditions such as node voltage over-limit, harmonic pollution, and distributed power source capacity are considered. The improved NSGA-II algorithm is used to solve the distributed photovoltaic carrying capacity under different typical scenarios, improving the solution efficiency, quality, and adaptability of the algorithm.

[0122] In the embodiment of the present application, the above step S400 includes the following sub-steps C1 - C2;

[0123] In C1: Construct a comprehensive evaluation model to evaluate the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes;

[0124] In C2: Based on the calculation results of the distributed photovoltaic access capacity under different distributed photovoltaic access to the distribution network schemes, obtain the index values, and calculate the subjective weight and objective weight of the index values to obtain the comprehensive weight;

[0125] Specifically, considering the safety, economy of the distribution network operation and the grid connection characteristics of distributed photovoltaics, establish a comprehensive evaluation index system for distributed photovoltaics accessing the distribution network;

[0126] The comprehensive evaluation index system for distributed photovoltaics accessing the distribution network includes three first-level indicators and eleven second-level indicators; the three first-level indicators include distribution network safety, distribution network economy and distribution network economy; the distribution network safety includes three second-level indicators: voltage deviation index, total voltage harmonic distortion rate, and voltage qualification rate; the distribution network economy includes three second-level indicators: annual average load rate of the line, average loss degree, and loss change degree; the grid connection characteristics of distributed photovoltaics include five second-level indicators: voltage over-limit risk degree, comprehensive node vulnerability, static penetration rate of distributed photovoltaics, annual power generation and consumption rate of distributed photovoltaics, and matching rate of distributed photovoltaic output and load normalization form.

[0127] Construct a comprehensive evaluation model for the carrying capacity of distributed photovoltaics, evaluate the carrying capacity of the distribution network under different distributed photovoltaic access schemes, determine the index values based on the calculation results of the distributed photovoltaic access capacity under each access scheme, calculate the subjective weight and objective weight of each index by using the analytic hierarchy process and the entropy weight method respectively, and combine the subjective weight and objective weight by using the evidence theory to form the comprehensive weight;

[0128] In the embodiment of the present application, after completing steps C1 - C2 in the above step S400, the following steps C3 - C5 are further included;

[0129] In C3: Fuzzify the index values and construct a fuzzy relation matrix;

[0130] In C4: Combine the analytic hierarchy process and the entropy weight method to determine the weight of the evaluation index and obtain the fuzzy evaluation result;

[0131] In C5: Based on the comprehensive weight and the fuzzy evaluation result, calculate the closeness of different schemes, evaluate and rank the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes, and obtain the evaluation results of different distributed photovoltaic access to the distribution network schemes.

[0132] Specifically, the index data is fuzzified using the fuzzy set theory to construct a fuzzy relation matrix. The analytic hierarchy process and entropy weight method are combined to determine the weights of each evaluation index, and a fuzzy evaluation result is obtained. The TOPSIS method is used to calculate the closeness degree of each scheme, and based on this, the carrying capacity of the distribution network under different distributed photovoltaic access distribution network schemes is evaluated and ranked to obtain the comprehensive evaluation results of different distributed photovoltaic access schemes.

[0133] It should be noted that the evaluation method based on the combined weight and TOPSIS method fully considers subjective and objective factors. The subjective and objective weights of each index are calculated using the analytic hierarchy process and entropy weight method respectively, and the combined weight is calculated using the evidence theory. The process of assigning weights to each index under different scenarios and different access schemes using this method is more objective and reasonable, and finally the evaluation ranking results of each photovoltaic access scheme are obtained.

[0134] The above is a schematic scheme of a distributed photovoltaic carrying capacity evaluation method in this embodiment. It should be noted that the technical solution of this distributed photovoltaic carrying capacity evaluation system belongs to the same concept as the technical solution of the above distributed photovoltaic carrying capacity evaluation method. For the details not described in detail in the technical solution of this distributed photovoltaic carrying capacity evaluation system in this embodiment, reference can be made to the description of the technical solution of the above distributed photovoltaic carrying capacity evaluation method.

[0135] The distributed photovoltaic carrying capacity evaluation system in this embodiment includes:

[0136] A preprocessing module, configured to obtain historical data of distributed photovoltaic, perform first preprocessing on the historical data to obtain first historical data;

[0137] A scenario construction module, configured to extract features from the first historical data, calculate the distance of the feature data, and construct a first scenario in combination with a first clustering algorithm;

[0138] A model construction module, configured to simulate different distributed photovoltaic access distribution network schemes based on the first scenario, construct a multi-objective optimization model, and solve the multi-objective optimization model to obtain the photovoltaic carrying capacity under different distributed photovoltaic access distribution network schemes;

[0139] An evaluation module, configured to establish a comprehensive evaluation index system for distributed photovoltaic access to the distribution network based on the photovoltaic carrying capacity, evaluate the carrying capacity of the distribution network under different distributed photovoltaic access distribution network schemes, and obtain comprehensive evaluation results.

[0140] This embodiment also provides a computing device applicable to the evaluation of distributed photovoltaic carrying capacity, including:

[0141] A memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the distributed photovoltaic carrying capacity evaluation method as proposed in the above embodiments.

[0142] This embodiment also provides a storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the distributed photovoltaic carrying capacity evaluation method as proposed in the above embodiments.

[0143] The storage medium proposed in this embodiment and the distributed photovoltaic carrying capacity evaluation method proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be referred to in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0144] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented by means of software and necessary general-purpose hardware, and of course, it can also be implemented by hardware. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (FLASH), a hard disk, or an optical disc of a computer, and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of the various embodiments of the present invention.

Claims

1. A distributed photovoltaic carrying capacity evaluation method, characterized in that Including: Obtain historical data of distributed photovoltaics, perform first preprocessing on the historical data to obtain first historical data; Extract features from the first historical data, calculate the distances of the feature data, and construct a first scenario in combination with a first clustering algorithm; Based on the first scenario, simulate different distributed photovoltaic access to the distribution network schemes and construct a multi-objective optimization model, solve the multi-objective optimization model to obtain the photovoltaic carrying capacity under different distributed photovoltaic access to the distribution network schemes; Based on the photovoltaic carrying capacity, establish a comprehensive evaluation index system for distributed photovoltaic access to the distribution network, evaluate the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes, and obtain a comprehensive evaluation result.

2. The distributed photovoltaic carrying capacity evaluation method according to claim 1, wherein Calculating the distances of the feature data includes: Calculate the first distance between the feature data, and construct a first distance matrix according to the calculation result; Based on the first distance matrix, discretize the continuous probability distribution of the feature data to obtain an optimal discrete scenario set and probability values, and construct a distribution scenario.

3. The distributed photovoltaic carrying capacity evaluation method according to claim 2, wherein Constructing the first scenario in combination with the first clustering algorithm includes: Use the first clustering algorithm to iteratively cluster the scenarios in the optimal discrete scenario set; Determine the optimal number of clusters through the first clustering evaluation method to obtain the first scenario.

4. The distributed photovoltaic load-carrying capacity evaluation method according to claim 3, wherein Constructing the multi-objective optimization model includes: The objective function of the multi-objective optimization model is to maximize the distributed photovoltaic capacity included in the distribution network and optimize the economic quality; The constraint conditions include equality constraints and inequality constraints.

5. The distributed photovoltaic carrying capacity evaluation method according to claim 4, wherein Solving the multi-objective optimization model includes: Use the first solution algorithm to solve the multi-objective optimization model; Use the constraint handling method to linearize and convexly relax the multi-objective optimization model, and dynamically adjust the crossover and mutation probabilities to calculate the photovoltaic carrying capacity under different distributed photovoltaic access to the distribution network schemes.

6. The distributed photovoltaic carrying capacity evaluation method according to claim 5, wherein Evaluating different distributed photovoltaic access to the distribution network schemes includes: Construct a comprehensive evaluation model to evaluate the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes; Based on the calculation results of the distributed photovoltaic access capacity under different distributed photovoltaic access to the distribution network schemes, obtain index values, and calculate the subjective weight and objective weight of the index values to obtain a comprehensive weight.

7. The distributed photovoltaic carrying capacity evaluation method according to claim 5 or 6, characterized in that, Also including: Fuzzify the index values and construct a fuzzy relation matrix; Combine the analytic hierarchy process and the entropy weight method to determine the evaluation index weights to obtain a fuzzy evaluation result; Based on the comprehensive weight and the fuzzy evaluation result, calculate the closeness of different schemes, evaluate and rank the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes to obtain the evaluation results of different distributed photovoltaic access to the distribution network schemes.

8. A system applying the distributed photovoltaic carrying capacity evaluation method as described in any one of claims 1-7, characterized in that, Including: A preprocessing module for obtaining historical data of distributed photovoltaics, performing first preprocessing on the historical data to obtain first historical data; A scenario construction module for extracting features from the first historical data, calculating the distances of the feature data, and constructing a first scenario in combination with a first clustering algorithm; A model construction module for simulating different distributed photovoltaic access to the distribution network schemes based on the first scenario and constructing a multi-objective optimization model, solving the multi-objective optimization model to obtain the photovoltaic carrying capacity under different distributed photovoltaic access to the distribution network schemes; An evaluation module, configured to establish a comprehensive evaluation index system for distributed photovoltaic access to a distribution network based on the photovoltaic carrying capacity, evaluate the carrying capacity of the distribution network under different distributed photovoltaic access to the distribution network schemes, and obtain a comprehensive evaluation result.

9. An electronic device, comprising: a memory and a processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions. When the computer-executable instructions are executed by the processor, the steps of the distributed photovoltaic carrying capacity evaluation method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, the steps of the distributed photovoltaic carrying capacity evaluation method according to any one of claims 1 to 7 are implemented.

Images