A county power distribution network multi-dimensional planning index flexible construction and evaluation method, system, device and medium

Abstract

The present application relates to power distribution network planning evaluation technical field, disclose a kind of county power distribution network multidimensional planning index flexible construction and evaluation method, system, equipment and medium, comprising: analysis and fusion the operating characteristics of county power distribution network of multiple flexibility resources, in combination with the energy endowment of different county power distribution network, flexible resource configuration, digital degree characteristics, carry out space-time resource fusion county power distribution network planning scene division, construct the multidimensional planning evaluation index of different scene;Adopt data-driven method, the importance and correlation between planning scheme and index are considered, adopt game theory to comprehensively sort index, construct multidimensional planning evaluation index based on data-driven, the planning scheme is evaluated by entropy weight method.The present application can be aimed at different planning scene, construct multidimensional planning evaluation index system based on data-driven, evaluate using entropy weight method, finally realize the evaluation of county power distribution network planning scene, difference, flexibility.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network planning and evaluation, and in particular to a method, system, equipment and medium for flexibly constructing and evaluating multi-dimensional planning indicators of a county distribution network. Background Art

[0002] A scientific and reasonable distribution network planning and evaluation method can effectively evaluate the impact of distributed power access on the system power supply reliability, optimize the power supply layout, optimize the distribution network structure and equipment configuration, evaluate the system flexibility and new energy absorption capacity, evaluate the system's digital intelligence level and safe and stable control capabilities, identify weak links in the system, provide a scientific decision-making basis for distribution network planning and construction, ensure the feasibility and effectiveness of the planning scheme, and provide a comprehensive reference for decision makers.

[0003] Against this backdrop, county-level distribution network planning must not only comprehensively reflect technical, economic, environmental, and social characteristics, but also be highly flexible and digitally intelligent to adapt to the dynamic development needs of different counties. The necessity of a flexible planning indicator system is primarily reflected in the following aspects: first, it must be able to comprehensively assess the operating characteristics and planning effectiveness of the distribution network; second, it must ensure that the indicator system can be differentiated and dynamically adjusted according to different scenarios to meet the specific distribution network needs of different counties and at different stages of development. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed. Therefore, the present invention provides a flexible construction and evaluation method of multi-dimensional planning indicators of county distribution networks to solve the above problems.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In the first aspect, the present invention provides a flexible construction and evaluation method for multi-dimensional planning indicators of county distribution networks, including: analyzing the operating characteristics of county distribution networks that integrate multiple flexibility resources, combining the energy endowment, flexible resource configuration, and digitalization characteristics of different county distribution networks, and dividing the planning scenarios of county distribution networks that integrate spatiotemporal resources;

[0007] Based on the results of different planning scenarios and the complete indicator system for county distribution network planning evaluation, multi-dimensional planning evaluation indicators for different scenarios are constructed;

[0008] Based on the multidimensional planning evaluation indicators of different scenarios, a data-driven approach is adopted, taking into account the importance and correlation between planning schemes and indicators, and using game theory to comprehensively rank the indicators to construct a data-driven multidimensional planning evaluation indicator;

[0009] Based on the data-driven multi-dimensional planning evaluation indicators, the planning scheme is evaluated using the entropy weight method.

[0010] As a preferred solution of the flexible construction and evaluation method of multi-dimensional planning indicators of a county distribution network described in the present invention, the division of the planning scenarios of the county distribution network with spatiotemporal resource integration includes:

[0011] The scenario division of the county-level distribution network planning for the integration of spatiotemporal resources includes the division of energy endowments, flexible resource allocation, and digital intelligence levels in different power supply areas or at different development stages of power supply areas;

[0012] The energy endowment divides energy resources into energy input type and energy output type. The energy input type indicates that the installed capacity of centralized and distributed energy resources in the county distribution network planning scheme is small, while the energy output type indicates that the installed capacity of centralized and distributed energy resources in the county distribution network planning scheme is large.

[0013] The flexible resource configuration represents the installed capacity of the flexible resources in the county distribution network planning scheme;

[0014] The degree of digitalization represents the degree of digitalization configuration in the county distribution network planning scheme.

[0015] As a preferred solution of the flexible construction and evaluation method of multi-dimensional planning indicators for county distribution networks described in the present invention, the construction of multi-dimensional planning evaluation indicators for different scenarios includes:

[0016] The comprehensive indicator system for county-level distribution network planning and evaluation includes multiple first-level, second-level, and third-level indicators in terms of clean and low-carbon, safe and sufficient, economical and efficient, flexible and intelligent, and supply-demand coordination.

[0017] According to the complete indicator system for planning and evaluation of county distribution network, combined with the characteristic data of different planning scenarios, the planning evaluation indicators for different scenarios are differentiated and constructed, and the planning evaluation indicators for different scenarios are screened out from the planning evaluation indicator system.

[0018] As a preferred solution of the flexible construction and evaluation method of multi-dimensional planning indicators for county-level distribution networks described in the present invention, the construction of data-driven multi-dimensional planning evaluation indicators includes:

[0019] A data-driven approach is used to preliminarily screen potential indicators for each scenario using the maximum correlation minimum redundancy algorithm to maintain the maximum correlation between the indicators and the target variable and minimize the redundant information between the indicators.

[0020] The importance of the selected indicators is evaluated through the random forest algorithm to deal with the nonlinear relationship between the indicators;

[0021] By combining relevance and importance through the game theory optimization model, planning indicators are sorted and screened, and indicator weights are allocated and dynamically adjusted.

[0022] As a preferred solution of the flexible construction and evaluation method of multi-dimensional planning indicators of county distribution network described in the present invention, the potential indicators of each scenario are preliminarily screened by the maximum correlation minimum redundancy algorithm to maintain the maximum correlation between the indicators and the target variables and minimize the redundant information between the indicators, including:

[0023] Input dataset for indicator and target variable c, perform data standardization and missing value processing;

[0024] Calculate each feature With the target variable Mutual information between , expressed as:

[0025] ,

[0026] in, represents the joint probability distribution, and represents the marginal probability distribution;

[0027] Calculate mutual information between features , evaluate the redundancy between features, expressed as:

[0028] ,

[0029] Perform MIQ criterion to calculate the correlation between features and target variables , expressed as:

[0030] ,

[0031] Perform redundancy measurement calculation and calculate the average mutual information between the feature and the features in the selected feature set S , expressed as:

[0032] ,

[0033] in, Indicates the size of the selected feature set;

[0034] In each iteration, the feature with the maximum MIQ score is selected and added to the set of selected features to obtain the mRMR importance score. , expressed as:

[0035] ,

[0036] in, Represents the total number of indicators that are effectively ranked by mRMR, and is the indicator The above steps are repeated until the preset number of features is reached or the MIQ score is lower than the set threshold.

[0037] As a preferred solution of the flexible construction and evaluation method of multi-dimensional planning indicators of a county distribution network described in the present invention, the importance of the selected indicators is evaluated by a random forest algorithm, and the nonlinear relationship between the indicators is processed, including:

[0038] Bootstrap sampling is performed on the training set to randomly generate k sample subsets Si and the corresponding decision tree Ti;

[0039] When a node splits, m features are randomly selected from m features. try features as candidate feature sets;

[0040] Calculate the optimal splitting point of the candidate feature for each node, calculate the mean square error of the parent node using different splitting thresholds, and calculate the mean square error of the left and right child nodes at the same time. The calculation split gain is expressed as:

[0041] ,

[0042] Select the feature with the largest split gain and the corresponding split point. Select the feature with the largest mean square error reduction for splitting, and then calculate the importance of each feature in a single tree, which is expressed as:

[0043] ,

[0044] Calculate the average importance of a feature in the entire forest, expressed as:

[0045] ,

[0046] The importance of features is obtained through normalization, and the feature importance ranking is output.

[0047] As a preferred solution of the flexible construction and evaluation method of multi-dimensional planning indicators of a county distribution network described in the present invention, the evaluation of the planning scheme by the entropy weight method includes:

[0048] Setting indicator types, including benefit-based indicators, cost-based indicators, and intermediate indicators, dividing planning evaluation indicators according to the definitions of each indicator type, and performing data preprocessing on the cost-based indicators and intermediate indicators;

[0049] The probability matrix is ​​introduced through the entropy weight method to calculate the information entropy value of the information contained in each indicator. Based on the information entropy value, the redundancy and indicator weight of each level of indicators are calculated.

[0050] In a second aspect, the present invention provides a flexible construction and evaluation system for multi-dimensional planning indicators of county distribution networks, including:

[0051] A partitioning module is used to analyze the operating characteristics of county-level distribution networks that integrate multiple flexible resources. It then divides the planning scenarios of county-level distribution networks that integrate spatiotemporal resources based on the energy endowment, flexible resource configuration, and digitalization characteristics of different county-level distribution networks.

[0052] The first construction module is used to construct multi-dimensional planning evaluation indicators for different scenarios based on the results of the different planning scenarios and according to the complete indicator system for county distribution network planning evaluation. The complete indicator system for county distribution network planning evaluation includes multiple primary indicators, secondary indicators, and tertiary indicators in the aspects of clean and low-carbon, safe and sufficient, economic and efficient, flexible and intelligent, and supply and demand coordination;

[0053] The second construction module is used to construct a data-driven multi-dimensional planning evaluation indicator based on the multi-dimensional planning evaluation indicators of the different scenarios, taking into account the importance and correlation between the planning schemes and the indicators, and comprehensively ranking the indicators using game theory;

[0054] An evaluation module is used to evaluate the planning scheme through an entropy weight method based on the data-driven multi-dimensional planning evaluation indicators.

[0055] In a third aspect, the present invention provides a computer device, comprising:

[0056] memory and processor;

[0057] 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 method for flexibly constructing and evaluating multi-dimensional planning indicators of a county distribution network are implemented.

[0058] In a fourth aspect, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of a method for flexibly constructing and evaluating multi-dimensional planning indicators of a county distribution network.

[0059] Compared with the prior art, the present application has the beneficial effects: the present application adopts a multi-stage and multi-dimensional comprehensive evaluation method to realize county power distribution network planning scheme evaluation, which not only considers the complexity and diversity of power distribution network operation, but also integrates advanced data analysis technology and machine learning algorithm, and on the basis of in-depth analysis of the operation characteristics of county power distribution network integrating various flexible resources, combined with the energy endowment, flexible resource allocation, digitalization degree and other characteristics of different county power distribution networks, time and space resource integration planning scene division is carried out; for different planning scenes, according to the expert experience, the planning evaluation indexes of different scenes are screened according to the power distribution system planning evaluation index system; on the basis of multi-dimensional planning evaluation index based on scene, data-driven method is adopted, considering the importance and correlation between planning scheme and index, game theory is used for comprehensive sorting; the multi-dimensional planning evaluation index system based on data driving is constructed, and the entropy weight method is used for evaluation, and finally the county power distribution network planning is realized. Scene, difference and flexibility evaluation. BRIEF DESCRIPTION OF DRAWINGS

[0060] In order to more clearly illustrate the technical scheme of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0061] Figure 1 The overall flowchart of a county power distribution network multi-dimensional planning index flexible construction and evaluation method according to an embodiment of the present application;

[0062] Figure 2 The object-scene-two-mRMR feature importance sorting diagram of a county power distribution network multi-dimensional planning index flexible construction and evaluation method provided by an embodiment of the present application;

[0063] Figure 3 The object-scene-two-random forest feature importance sorting diagram of a county power distribution network multi-dimensional planning index flexible construction and evaluation method provided by an embodiment of the present application;

[0064] Figure 4 The object-scene-two-game theory feature importance sorting diagram of a county power distribution network multi-dimensional planning index flexible construction and evaluation method provided by an embodiment of the present application;

[0065] Figure 5 The object-scene-two-entropy weight method overall score diagram of a county power distribution network multi-dimensional planning index flexible construction and evaluation method provided by an embodiment of the present application;

[0066] Figure 6An embodiment of the present invention provides a flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network; an mRMR screening feature importance ranking diagram for scenario 6;

[0067] Figure 7 An embodiment of the present invention provides a method for flexibly constructing and evaluating multi-dimensional planning indicators for a county-level distribution network; an important ranking diagram of random forest screening features in scenario 6;

[0068] Figure 8 An embodiment of the present invention provides a flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network; an object two scenario six game theory screening feature importance ranking diagram;

[0069] Figure 9 An overall score diagram of the object two scene six entropy weight method of a flexible construction and evaluation method of multi-dimensional planning indicators of a county distribution network provided by an embodiment of the present invention;

[0070] Figure 10 A complete indicator system for county distribution network planning and evaluation, providing a flexible construction and evaluation method for multi-dimensional planning indicators of county distribution networks, provided by an embodiment of the present invention Figure 1 ;

[0071] Figure 11 A complete indicator system for county distribution network planning and evaluation, providing a flexible construction and evaluation method for multi-dimensional planning indicators of county distribution networks, provided by an embodiment of the present invention Figure 2 . DETAILED DESCRIPTION

[0072] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0073] Example 1, with reference to Figure 1 、 Figures 10-11 Tables 1 to 4 are examples of an embodiment of the present invention, providing a flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network, including:

[0074] S101: Analyze the operational characteristics of county-level distribution networks that integrate multiple flexible resources. Combined with the energy endowment, flexible resource configuration, and digitalization characteristics of different county-level distribution networks, divide the planning scenarios of county-level distribution networks that integrate spatiotemporal resources.

[0075] S102: Based on the results of the different planning scenarios and the complete indicator system for county distribution network planning evaluation, a multi-dimensional planning evaluation indicator for different scenarios is constructed;

[0076] S103, a multi-dimensional planning evaluation indicator based on different scenarios, adopts a data-driven approach, takes into account the importance and correlation between planning schemes and indicators, and uses game theory to comprehensively rank indicators to construct a data-driven multi-dimensional planning evaluation indicator;

[0077] S104, based on data-driven multi-dimensional planning evaluation indicators, the planning scheme is evaluated using the entropy weight method.

[0078] It should be noted that this embodiment uses data-driven analysis of the characteristics of county distribution networks, integrates multiple flexibility resources, and constructs a scenario-based, differentiated, and multi-dimensional planning indicator system to achieve a shift in planning evaluation from experience-driven to data-driven, assist in the green transformation of county distribution networks, and promote the construction of new distribution systems.

[0079] In a preferred embodiment, performing the scenario division of the county distribution network planning by integrating spatiotemporal resources includes:

[0080] The scenario division of county-level distribution network planning based on the integration of spatiotemporal resources includes the division of energy endowment, flexible resource allocation, and digitalization level of different power supply areas or power supply areas at different development stages;

[0081] Energy endowment divides energy resources into energy input type and energy output type. Energy input type means that the installed capacity of centralized and distributed energy in the county distribution network planning scheme is small, while energy output type means that the installed capacity of centralized and distributed energy in the county distribution network planning scheme is large.

[0082] Flexible resource configuration refers to the installed capacity of flexible resources in the county distribution network planning scheme;

[0083] The degree of digitalization indicates the degree of digitalization configuration in the county distribution network planning scheme.

[0084] Specifically, energy endowment divides energy resources into input and output types. Energy input refers to a county-level distribution network plan with small installed capacity of centralized and distributed energy resources, meaning the region's energy supply primarily relies on the upper-level grid. Energy output refers to a county-level distribution network plan with large installed capacity of centralized and distributed energy resources, meaning that the network not only meets the region's load but also has a certain degree of external transmission capability.

[0085] The energy density and load density in the county distribution network planning scheme are combined to determine the input and output types. The principles are as follows:

[0086] Energy density = total installed capacity / power supply area (MW / km 2)

[0087] Load density = maximum power supply load / power supply area (MW / km 2 )

[0088] Table 1: Input and output judgment table of county distribution network planning scheme

[0089] ,

[0090] Flexible resource allocation refers to the installed capacity of flexible resources within county-level distribution network planning schemes. Flexible resources include distributed power sources (DGs), energy storage, electric vehicles, and demand-side response resources. Distributed power sources include distributed photovoltaics, wind power, and micro-hydropower. Energy storage systems include electrochemical, physical, and hydrogen storage. Electric vehicles can serve as both mobile energy storage and provide regulation capabilities through charging networks. Demand-side response resources include adjustable load devices for industrial and commercial users, smart home appliances, and traditional power generation equipment with adjustable output.

[0091] The degree of digitalization refers to the degree of digital configuration in the county distribution network planning scheme. The differentiated configuration requirements for digital substations, lines, and distribution transformers are shown in Table 2.

[0092] Table 2: Differentiated configurations of digital substations, lines, and distribution transformers

[0093] ,

[0094] According to the requirements of digitalization level classification, the digital configuration plan for county distribution networks should comprehensively consider the design requirements of key links such as substations, lines, equipment and users according to the different levels of power supply areas, and formulate differentiated configuration strategies: For Class A+ areas, as priority development areas, advanced configurations should be fully adopted in core facilities such as digital substations, distribution lines, and medium-voltage equipment to ensure that the equipment has high automation, intelligence and remote control capabilities, and supports refined management and efficient operation; Class A areas need to maintain the same high-standard configuration as A+ areas, further strengthen the coverage of digital distribution stations and smart terminal equipment, and meet the needs of high-reliability power supply.

[0095] For Class B areas, intermediate configurations should be adopted in substations and lines, and standardized construction of distribution stations and medium-voltage equipment should be gradually realized, while taking into account the coverage of digital user terminals to meet the basic intelligent power supply needs in the region; Class C areas should mainly use standardized configurations, and digital substations and lines should ensure the stability of basic functions, while appropriately configuring medium-voltage equipment to achieve reliable power supply in the region; Class D areas focus on flexible configuration of core facilities, such as arranging digital substations and user terminal facilities according to actual conditions to ensure the digital capabilities of basic power supply services.

[0096] According to the above principles, the county distribution network planning scenarios can be divided into 8 typical scenarios as shown in Table 3.

[0097] Table 3: 8 typical scenarios

[0098] ,

[0099] In a preferred embodiment, constructing multi-dimensional planning evaluation indicators for different scenarios includes:

[0100] The comprehensive indicator system for county-level distribution network planning and evaluation includes multiple primary, secondary, and tertiary indicators in terms of clean and low-carbon, safe and sufficient, economical and efficient, flexible and intelligent, and supply-demand coordination.

[0101] Based on the complete indicator system for planning and evaluation of county distribution networks, combined with the characteristic data of different planning scenarios, differentiated planning evaluation indicators for different scenarios are constructed, and planning evaluation indicators for different scenarios are screened out from the indicator system.

[0102] Specifically, the complete indicator system for distribution network planning and evaluation is constructed from five aspects: clean and low-carbon, safe and sufficient, economical and efficient, flexible and intelligent, and supply and demand coordination. It has a total of 5 first-level indicators, 13 second-level indicators and 55 third-level indicators. According to the complete indicator system for distribution network planning and evaluation, based on different planning scenarios and expert experience, the planning evaluation indicators for different scenarios are screened, and the distribution system planning and evaluation indicator system is constructed, and the planning evaluation indicators for different scenarios are differentiated.

[0103] Among them, for example, scenario one is applicable to energy input-type scenarios with low degree of digitization and limited flexible resource allocation capabilities, and focuses on improving power supply reliability through grid improvement; scenario two is applicable to scenarios that use digital technology to improve power supply security and flexibility, but flexible resource allocation is still low; scenario three is applicable to scenarios with high flexible resource allocation and the need for efficient collaborative management of flexible resources, but low digital participation; scenario four is applicable to scenarios with high degree of digitization and flexible resource collaboration, focusing on intelligent management and supply and demand matching; scenario five is applicable to scenarios with low carbonization as the goal, with low flexible resource allocation and degree of digitization; scenario six is ​​applicable to scenarios that combine low carbonization and digitization, emphasizing intelligent low-carbon development; and scenario seven is applicable to scenarios with high flexible resource allocation but low degree of digitization; scenario eight is applicable to scenarios that fully combine high digitization, flexible resource allocation and low carbonization goals.

[0104] The purpose of setting different scenarios in distribution network planning is to adapt to the differentiated needs of different objects in the same period, as well as the development and changing needs of the same object in different periods. Different regions or users differ in resource endowments, load demands, technical conditions and policy environments. Planning goals and focus will also change dynamically over time. Therefore, by constructing differentiated scenarios through expert experience, matching evaluations can be carried out for different scenarios such as traditional distribution networks, smart distribution networks or new distribution networks.

[0105] Based on expert experience, planning evaluation indicators for different scenarios are screened as shown in Table 4.

[0106] Table 4: Planning evaluation index screening table for different scenarios

[0107] ,

[0108] ,

[0109] ,

[0110] ,

[0111] ,

[0112] In a preferred embodiment, constructing a data-driven multi-dimensional planning evaluation indicator includes:

[0113] A data-driven approach is used to preliminarily screen potential indicators for each scenario using the maximum correlation minimum redundancy algorithm to maintain the maximum correlation between the indicators and the target variable and minimize the redundant information between the indicators.

[0114] The importance of the selected indicators is evaluated through the random forest algorithm to deal with the nonlinear relationship between the indicators;

[0115] By combining relevance and importance through the game theory optimization model, planning indicators are sorted and screened, and indicator weights are allocated and dynamically adjusted.

[0116] In an optional implementation, the maximum relevance minimum redundancy (mRMR) algorithm, mutual information as an effective information measurement method, can not only measure the linear relationship between variables, but also well evaluate the nonlinear relationship, and convert two random variables into and The mutual information between is defined as:

[0117] ,

[0118] in, and are random variables and The probability density of is a random variable and The joint probability density of , the definitions of maximum correlation and minimum redundancy are expressed as:

[0119] ,

[0120] ,

[0121] in, is the feature set, is the number of features in the feature set, is the target variable, Features and the target variable The mutual information between Set the features and target variables in the feature set The mean mutual information between Features and features The mutual information between is the mutual information between the features in the feature set.

[0122] Combining the above formulas, we get the maximum correlation and minimum redundancy criterion, which is expressed as:

[0123] ,

[0124] In a preferred embodiment, the feature selection process based on mRMR uses the maximum correlation minimum redundancy algorithm to preliminarily screen the potential indicators of each scenario, maintain the maximum correlation between the indicators and the target variable, and minimize the redundant information between the indicators, including:

[0125] The data input stage first performs data preprocessing and inputs the indicator data set and target variable c, perform data standardization and missing value processing;

[0126] Calculate each feature With the target variable Mutual information between , expressed as:

[0127] ,

[0128] in, represents the joint probability distribution, and represents the marginal probability distribution;

[0129] Calculate mutual information between features , evaluate the redundancy between features, expressed as:

[0130] ,

[0131] Perform MIQ criterion to calculate the correlation between features and target variables , expressed as:

[0132] ,

[0133] Perform redundancy measurement calculation and calculate the average mutual information between the feature and the features in the selected feature set S , expressed as:

[0134] ,

[0135] in, Indicates the size of the selected feature set;

[0136] In each iteration, the feature with the maximum MIQ score is selected, expressed as:

[0137] ,

[0138] And add the selected features to the selected feature set, expressed as:

[0139] ,

[0140] Get mRMR importance score , expressed as:

[0141] ,

[0142] in, Represents the total number of indicators that are effectively ranked by mRMR, and is the indicator Repeat the above steps until the preset number of features is met in the mRMR sorting list. Or if the MIQ score is lower than the set threshold, the final optimal feature subset will be selected. The preset number of features and the set threshold can be set according to the actual application scenario requirements or expert experience.

[0143] In an optional implementation, the basic idea of ​​the random forest algorithm is to use multiple decision trees to train and predict sample sequences. It can be divided into two parts: decision tree generation and forest generation. In the regression problem of this embodiment, the optimal split point is mainly determined by the mean square error (MSE), and the importance of each feature is evaluated based on the reduction in node impurity.

[0144] Assume Node Include The mean square error of a sample can be represented as:

[0145] ,

[0146] wherein, is the actual value of the sample, is the average value of all samples in the node ;

[0147] When a node is split using a feature , the split gain can be represented as:

[0148] ,

[0149] wherein, and are the number of samples in the left and right child nodes, and are the mean square errors of the corresponding child nodes;

[0150] The importance of a feature in a single decision tree in a random forest can be represented as:

[0151] ,

[0152] wherein, is the sample weight of the node , and is the impurity reduction of the node.

[0153] In a preferred embodiment, the importance of the screened indicators is evaluated by a random forest algorithm, and the non-linear relationship between the indicators is processed, including:

[0154] Bootstrap sampling is performed on the training set to randomly generate k sample subsets Si, and corresponding decision trees Ti are generated;

[0155] When a node is split, m try features are randomly selected from m features as a candidate feature set;

[0156] The optimal split point of the candidate features is calculated for each node, the mean square error of the parent node is calculated, different split thresholds are used, wherein the split threshold can be set according to the actual application scenario requirement or expert experience, and the mean square errors of the left and right child nodes are calculated, and the split gain is calculated, which is represented as:

[0157] ,

[0158] Select the feature with the largest split gain and the corresponding split point. Select the feature with the largest mean square error reduction for splitting, and then calculate the importance of each feature in a single tree, which is expressed as:

[0159] ,

[0160] Calculate the average importance of a feature in the entire forest, expressed as:

[0161] ,

[0162] The importance of feature j is obtained through normalization, which is expressed as:

[0163] ,

[0164] Where T is the total number of decision trees, is the sum of all feature importances;

[0165] Finally, the feature importance ranking is output.

[0166] In an optional embodiment, a game theory optimization model is constructed based on the mRMR algorithm and the random forest algorithm. Game theory uses the Shapley value cooperative game. Formally, the definition of a coalition game is: there is a set ( ) and a function , which maps the subset of players to real numbers: ,in, represents the empty set, function is called the characteristic function.

[0167] function The meaning of is as follows: if S is a player coalition, then v(S) is called a coalition The value of The total expected rewards that members can obtain through cooperation.

[0168] The Shapley value is a way of distributing the total payoff among the players, assuming they all cooperate. It is a "fair" distribution because it is the only distribution that has certain desirable properties: In the player The score gain is expressed as:

[0169] ,

[0170] in, It is from Medium Any subset of all players except is a subset The number of elements in, that is, except for the player The number of other players besides is a collection The total number of players in Indicates that in the collection Add players The benefits of the alliance formed later, Indicates that only the set The benefits brought by the alliance formed by the players in the game.

[0171] The specific process of Shapley value game theory is as follows:

[0172] Step 1: Data input and preprocessing

[0173] Loading feature datasets from mRMR and random forest calculation results , the missing values ​​in the dataset are filled with the mean, and the feature values ​​are scaled to the [0, 1] interval using Min-Max normalization.

[0174] Step 2: Correlation calculation of mRMR features

[0175] Feature selection, using the mRMR algorithm to sort features and select features according to the MIQ index;

[0176] Importance score, which is assigned according to the selection order of features, is expressed as:

[0177] ,

[0178] in, is the total number of features, Features The order of selection in the mRMR algorithm.

[0179] Step 3: Random Forest Feature Importance Calculation

[0180] Model training: Train the random forest regression model on the preprocessed data;

[0181] Importance Extraction: Extract the feature importance of the model, calculated based on the mean square error reduction, expressed as:

[0182] ,

[0183] SHAP value calculation: Use the trained random forest model to calculate the SHAP value of each feature:

[0184] ,

[0185] Importance evaluation: Take the average absolute value of the SHAP value as the importance of the feature:

[0186] ,

[0187] Step 4: Normalization

[0188] Min-Max normalization: Normalize the feature importance obtained by each method to ensure that the importance scores are on the same scale, expressed as:

[0189] ,

[0190] Here, k represents the mRMR, RF, or SHAP method.

[0191] Step 5: Feature Importance Fusion Based on Shapley Value

[0192] Fusion strategy: Using the average as the fusion method, the three normalized importance scores are combined, expressed as:

[0193] ,

[0194] It should be noted that the game theory perspective treats each feature evaluation method as a "player" in the game, and the combined importance score is equivalent to the average of the "Shapley values" of the features contributed by different methods. Combining the advantages of multiple feature evaluation methods results in a more robust feature importance assessment.

[0195] In a preferred embodiment, evaluating the planning scheme using the entropy weight method includes:

[0196] Set indicator types, including benefit indicators, cost indicators and intermediate indicators. Divide planning evaluation indicators according to the definitions of each indicator type, and perform data preprocessing on cost indicators and intermediate indicators.

[0197] The probability matrix is ​​introduced through the entropy weight method to calculate the information entropy value of the information contained in each indicator. Based on the information entropy value, the redundancy and indicator weight of indicators at all levels are calculated.

[0198] It should be noted that the entropy weighting method in this embodiment is used to objectively evaluate and determine the relative importance (i.e., weight) of each planning indicator. The evaluation process includes preprocessing the indicator data, calculating the information entropy of each indicator to measure the effectiveness of the information it provides, and determining the weight of each indicator based on this information. The direct output of the entropy weighting method is a set of objective weights that quantitatively reflect the contribution of each indicator in distinguishing different planning schemes. Subsequently, the standardized performance value of each planning scheme on each indicator is multiplied by its corresponding indicator weight and summed to obtain a final comprehensive score for each planning scheme. This comprehensive score is the final evaluation result of the planning scheme. The high or low score can clearly distinguish the overall advantages and disadvantages of different planning schemes. By further analyzing the contribution of each dimension to the score, planners can identify the strengths and weaknesses of specific schemes in specific scenarios. This comprehensive evaluation method based on objective weighting improves the scientific and targeted nature of the evaluation, providing strong support for accurate decision-making in distribution network planning.

[0199] Specifically, the planning evaluation indicators are classified and preprocessed, and the planning evaluation indicators are divided into benefit indicators, cost indicators and intermediate indicators, and the cost indicators and intermediate indicators are preprocessed;

[0200] Benefit-based indicators are those whose larger values ​​indicate better performance or benefits of the evaluation object. For example, a higher "power supply reliability" indicates better power supply service quality; a higher "renewable energy utilization rate" indicates greater utilization of renewable energy.

[0201] Cost-based indicators are those whose smaller values ​​indicate better performance or lower costs. For example, a lower "line loss rate" indicates higher energy transmission efficiency, and a lower "total investment cost" indicates a more economical solution.

[0202] Intermediate indicators (or moderate indicators) are those whose values ​​do not monotonically improve but rather have an optimal range or interval. Values ​​that are too high or too low are not ideal. For these indicators, the optimal value or range can be set based on specific engineering experience, industry recommendations, or specific planning objectives.

[0203] According to the definitions of the above indicator types, each planning evaluation indicator is matched and the division of planning evaluation indicators is completed.

[0204] In an optional implementation, data preprocessing is performed on the cost-type indicators and the intermediate-type indicators to facilitate effective and reasonable comparison and analysis of the indicators;

[0205] For benefit-type indicators, take their original values, and for cost-type indicators, take their reciprocals, expressed as:

[0206] ,

[0207] For interval type indicators , it satisfies the following:

[0208] ,

[0209] wherein, is the minimum value or the allowable lower limit of the indicator , and is the maximum value or the allowable upper limit of the indicator , while different indicators have different dimensions, which will have a significant impact on the analysis of indicator data, and the indicator data needs to be standardized and dimensionless to obtain normalized data, and the normalization formula is:

[0210] ,

[0211] It should be noted that the entropy weight method is used, and a probability matrix is introduced, which can reflect the distribution of each indicator data value, and the information entropy value calculated therefrom can describe the amount of information contained in each indicator. The smaller the information entropy value, the more concentrated the distribution of the indicator data, the greater the degree of influence on the overall goal, and therefore the greater the weight of the indicator. Based on the obtained information entropy value, the redundancy and indicator weight of each level can be further calculated.

[0212] The evaluation process mainly includes:

[0213] Calculate the information entropy, based on the standardized decision matrix, calculate the proportion of each planning scheme performance value under each indicator, and further calculate the information entropy value of each indicator.

[0214] Information entropy calculation determines the weight, according to the information entropy value, the information utility value (or difference degree) of each indicator is calculated, the greater the information utility value, the greater the contribution of the indicator to the differentiation of different planning schemes, and finally, the information utility value of each indicator is normalized to obtain the objective weight coefficient W j of each indicator in this specific scenario.

[0215] Using the above indicator weight W j determined by the entropy weight method, the final comprehensive performance evaluation of each planning scheme is carried out, specifically, the standardized performance value Z ij of each planning scheme i on each (preprocessed and dimensionless) indicator j is multiplied by the corresponding indicator weight W j , and then the weighted performance values of all indicators are summed to obtain the final comprehensive score S i of each planning scheme i.

[0216] The comprehensive score S i That is, the final evaluation result of the planning scheme, which can quantitatively reflect the overall advantages and disadvantages of each planning scheme in a specific scenario, thereby realizing effective sorting and optimization of different schemes.

[0217] The present application adopts a multi-stage and multi-dimensional comprehensive evaluation method to realize the county distribution network planning scheme evaluation. This method not only considers the complexity and diversity of the distribution network operation, but also integrates advanced data analysis technology and machine learning algorithm. Based on in-depth analysis of the operation characteristics of the county distribution network integrating various flexible resources, combined with the energy endowment, flexible resource allocation, and digitalization degree of different county distribution networks, the time and space resource integration planning scenario is divided. According to the distribution system planning evaluation index system, the planning evaluation indexes of different scenarios are selected according to expert experience. Based on the multi-dimensional planning evaluation index based on the scenario, the importance and correlation between the planning scheme and the index are considered by using data-driven method, and the comprehensive sorting is carried out by using game theory. A multi-dimensional planning evaluation index system based on data-driven is constructed, and the evaluation is carried out by using entropy weight method, and finally the county distribution network planning scenario, difference and flexibility evaluation is realized.

[0218] The above is a schematic scheme of the county distribution network multi-dimensional planning index flexible construction and evaluation method of the present embodiment. It should be noted that the technical scheme of the county distribution network multi-dimensional planning index flexible construction and evaluation system belongs to the same concept as the technical scheme of the county distribution network multi-dimensional planning index flexible construction and evaluation method described above. The technical scheme of the county distribution network multi-dimensional planning index flexible construction and evaluation system in the present embodiment is not described in detail. The details can be referred to the description of the technical scheme of the county distribution network multi-dimensional planning index flexible construction and evaluation method.

[0219] The county distribution network multi-dimensional planning index flexible construction and evaluation system in the present embodiment comprises:

[0220] The division module is used to analyze the operation characteristics of the county distribution network integrating various flexible resources, and combine the energy endowment, flexible resource allocation, and digitalization degree characteristics of different county distribution networks to divide the time and space resource integration county distribution network planning scenario.

[0221] The first construction module is used to construct multi-dimensional planning evaluation indexes of different scenarios based on different planning scenario division results according to the county distribution network planning evaluation complete index system.

[0222] The second building block is used to construct multi-dimensional planning evaluation indicators based on different scenarios. It adopts a data-driven approach, takes into account the importance and correlation between planning schemes and indicators, and uses game theory to comprehensively rank the indicators to construct a data-driven multi-dimensional planning evaluation indicator.

[0223] The evaluation module is used to evaluate planning schemes based on data-driven multi-dimensional planning evaluation indicators and the entropy weight method.

[0224] This embodiment further provides a computer device suitable for flexibly constructing and evaluating multi-dimensional planning indicators for county-level distribution networks, including:

[0225] Memory and processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the flexible construction and evaluation method of multi-dimensional planning indicators of county distribution networks proposed in the above embodiment.

[0226] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network as proposed in the above embodiment.

[0227] The storage medium proposed in this embodiment and the method for realizing flexible construction and evaluation of multi-dimensional planning indicators of county distribution networks proposed in the above embodiment belong to the same inventive concept. For technical details not described in detail in this embodiment, please refer to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.

[0228] Through the above description of the embodiments, those skilled in the art will clearly understand that the present invention can be implemented using software and necessary general-purpose hardware. Of course, it can also be implemented using hardware, but in many cases the former is a more preferred embodiment. Based on this understanding, the technical solution of the present invention, or the portion 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 computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk, or optical disk, and includes a number of instructions for causing a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.

[0229] Example 2, reference Figures 2-11 Tables 5 to 11 are an embodiment of the present invention, which provides a flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network. In order to verify its beneficial effects, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0230] To verify the effectiveness and practicability of the proposed power distribution network performance index screening method, the actual power distribution network data of Object One is selected for case analysis in this embodiment. This region has typical urban-rural mixed characteristics, including highly urbanized areas and rural areas, and the new energy penetration rate differs significantly, which is an ideal scenario for verifying the research method. The following is the basis for selecting Object One.

[0231] (1) The resource endowment characteristics are prominent. The total installed capacity of the 110 kV below photovoltaic power generation project of Object One is only 0.038 MW, the power supply area is 148 square kilometers, and the power supply load reaches 132.3 kilowatts. It is a typical area where electricity demand is greater than local supply capacity. As a new urban area, it mainly relies on the power supply of the upper power grid, and there are few local power generation facilities. Although the region is rich in wind energy resources (annual utilization hours reach more than 2200 hours), the current new energy development scale is still limited.

[0232] Table 5: Object One Data Table

[0233] ,

[0234] (2) There are fewer flexible resources, and the load is mainly composed of commercial loads and smart home, and the construction of electric vehicle charging facilities has not yet formed a scale, and energy storage resources have not yet been put on the agenda.

[0235] (3) The digitalization development is rapid. Through digitalization construction, it realizes comprehensive real-time monitoring of distribution network, full coverage of self-healing, and greatly improves the operation reliability of distribution network.

[0236] Based on the above analysis, the planning scheme of Object One is energy input type, flexible configuration is low, and digitalization degree is high. The planning scenario corresponds to Scenario Two.

[0237] The planning scenario of Object One is Scenario Two. According to expert experience, a total of 41 indicators are selected, covering clean and low carbon, economic and efficient, flexible and intelligent, safe and abundant, and supply and demand coordination. These indicators comprehensively reflect the characteristics of Scenario Two.

[0238] Table 6: Scenario Two Based Evaluation Index Data Table of Object One

[0239] ,

[0240] ,

[0241] ,

[0242] Table 7: Scenario Two Target Variables

[0243] ,

[0244] Based on the indicator construction based on scenario 2 above, a data-driven approach is used to reduce the indicator dimension.

[0245] Among them, the feature evaluation system is constructed based on the mRMR algorithm theory:

[0246] For indicator set and target variable c (initial score of the solution), calculate each feature With the target variable Mutual information between , expressed as:

[0247] ,

[0248] in, represents the joint probability distribution, and represents the marginal probability distribution;

[0249] Calculation indicators Mutual information between , evaluate the redundancy between features, expressed as:

[0250] ,

[0251] Perform MIQ criterion to calculate the correlation between features and target variables , expressed as:

[0252] ,

[0253] Perform redundancy measurement calculation and calculate the average mutual information between the feature and the features in the selected feature set S , expressed as:

[0254] ,

[0255] in, Indicates the size of the selected feature set;

[0256] In each iteration, the feature with the maximum MIQ score is selected, expressed as:

[0257] ,

[0258] And add the selected features to the selected feature set, expressed as:

[0259] ,

[0260] Finally, we get the mRMR importance score :

[0261] ,

[0262] Represents the total number of indicators that are effectively ranked by mRMR, and is the indicator Repeat the above steps until the preset number of features is met in the mRMR sorting list. Or if the MIQ score is lower than the set threshold, the final optimal feature subset will be selected. The preset number of features and the set threshold can be set according to the actual application scenario requirements or expert experience. The top 20 features are displayed, and a horizontal bar chart is used to show the feature importance scores.

[0263] Rating by importance Arrange in descending order, and get Figure 2 Object 1 of the flexible construction method of multi-dimensional planning indicators of county distribution networks based on digital means. Importance ranking diagram of mRMR screening features based on scenario 2.

[0264] Based on the random forest algorithm, we can get:

[0265] Construct a random forest regression model and optimize the training process; feature importance analysis Importance calculation: Calculate feature importance based on the split gain of the decision tree, use an ensemble method to combine the evaluation results of multiple trees, standardize feature importance scores, and facilitate comparison of results visualization: Select the top 20 features for key display and add specific numerical labels to provide accurate importance information, as shown below. Figure 3 Object 1 is based on the random forest feature importance ranking diagram of scenario 2.

[0266] The calculation based on game theory can be obtained:

[0267] mRMR feature selection; random forest model evaluation; Shapley value calculation; feature fusion and normalization, normalize the scores of the three algorithms respectively, build a unified feature score data frame to ensure the comparability of the scores; multi-algorithm fusion strategy; build a comprehensive scoring system, select the top 20 indicators to generate the final feature importance ranking results, such as Figure 4 Object 1 is based on the game theory screening feature importance ranking diagram of scenario 2.

[0268] Based on the entropy weight method, we get Figure 5 Scores for Subject 1's planning scheme based on Scenario 2:

[0269] Among them, the planning scheme 1-1 in scenario 2 scored 0.8275, the planning scheme 1-2 scored 0.9079, the planning scheme 1-3 scored 0.9504, and the planning scheme 1-4 scored 0.8946.

[0270] In order to verify the effectiveness and practicality of the proposed distribution network performance index screening method, this embodiment selected the actual distribution network data of Object 2 for case analysis;

[0271] For object 2 selected in this example, the following is the basis for selecting the planning scenario:

[0272] (1) The resource endowment advantage is significant. Object 2 fully utilizes its unique wind and light resources and strives to build an energy structure with multiple energy complementarities of "wind, light, water, fire, and hydrogen energy". In terms of energy supply, relying on the local abundant coal resources, a number of large thermal power plants have been built with sufficient installed capacity to ensure the basic load supply. At the same time, a number of photovoltaic power generation bases have been built, making full use of the resources of barren mountains and slopes, and using the local high-quality wind resources to layout and build centralized wind farms. A number of small hydropower stations have been built based on river systems to provide regulation capabilities for regional power supply. In recent years, it has actively explored the development of the hydrogen energy industry, laid out hydrogen production, hydrogen storage and hydrogen energy application facilities, and further enriched the form of energy supply. The total installed capacity is 2855MW, the power supply area is 4056.7 square kilometers, the maximum power supply load is 661,200 kilowatts, and the energy density reaches 0.7037MW / km 2 , the load density reaches 0.162 MW / km 2 It is a typical output-oriented region. It has currently formed a diversified energy supply system based on coal-fired power and the rapid development of new energy. It not only ensures the safety and reliability of power supply, but also promotes the clean and low-carbon transformation of the energy structure.

[0273] Table 8: Object 2 data table

[0274] ,

[0275] (2) Flexible resources are relatively scarce, and the load is mainly based on traditional industrial electricity consumption. The construction of electric vehicle charging facilities has not yet reached a large scale, and energy storage resources have not yet been put on the agenda.

[0276] (3) The digital and intelligent construction has been steadily advancing, achieving comprehensive real-time monitoring of the distribution network, and deploying differentiated distribution automation technology routes, effectively improving the operational reliability and power supply quality of the distribution network.

[0277] Based on the above analysis, the planning scheme for Object 2 is energy output type, with low flexible configuration and high degree of digitalization, and the planning scenario corresponds to Scenario 6.

[0278] The planning scenario for Object 2 is Scenario 6. Based on expert experience, a total of 25 indicators were selected, covering five aspects: clean and low-carbon, economic and efficient, flexible and intelligent, safe and sufficient, and supply and demand coordination. These indicators fully reflect the characteristics of Scenario 6.

[0279] Table 9: Evaluation index data table of object 2 based on scenario 6

[0280] ,

[0281] ,

[0282] Table 10: Target variables for scenario six

[0283] ,

[0284] Based on the above indicator construction and data tables, use data-driven methods to reduce the indicator dimension;

[0285] Among them, the feature evaluation system is constructed based on the mRMR algorithm theory:

[0286] Rating by importance Arrange in descending order, select the top 20 features in terms of importance for display, and use a horizontal bar chart to display the feature importance score. Figure 6 Importance ranking diagram of mRMR screening features for object 2 based on scenario 6.

[0287] Based on the random forest algorithm, we can get:

[0288] Construct a random forest regression model and optimize the training process; feature importance analysis Importance calculation: Calculate feature importance based on the split gain of the decision tree, use an ensemble method to combine the evaluation results of multiple trees, standardize feature importance scores to facilitate comparison Result visualization: Select the top 20 features for key display and add specific numerical labels to provide accurate importance information, as shown below Figure 7 Object 2 is based on the random forest feature importance ranking diagram of scenario 6.

[0289] The calculation based on game theory can be obtained:

[0290] mRMR feature selection; random forest model evaluation; Shapley value calculation; feature fusion and normalization, normalize the scores of the three algorithms respectively, build a unified feature score data frame to ensure the comparability of the scores; multi-algorithm fusion strategy; build a comprehensive scoring system, select the top 20 indicators to generate the final feature importance ranking results, such as Figure 8 Object 2 is based on the game theory screening feature importance ranking diagram of scenario 6.

[0291] Based on the entropy weight method, we can get Figure 9 The scores of Object 2 calculated using the entropy weight method for Scenario 6 are: Planning Scheme 2-1 scored 0.8628, Planning Scheme 2-2 scored 0.893, Planning Scheme 2-3 scored 0.8887, and Planning Scheme 2-4 scored 0.9295.

[0292] After algorithm screening, the optimal planning scheme for subject one based on scenario two is 2-3, with a score of 0.9504; the optimal planning scheme for subject two based on scenario six is ​​6-4, with a score of 0.9295. The final constructed indicators are shown in Table 9.

[0293] Table 11: Final indicator construction table

[0294] ,

[0295] It should be noted that an in-depth analysis of the distribution network planning of two different counties, Object 1 and Object 2, shows that the method of the present invention can realize the flexible construction and evaluation of scenario-based and differentiated indicators for different county distribution network planning schemes.

[0296] In summary, the present invention proposes a flexible construction of multi-dimensional planning indicators for county distribution networks using digital means. Based on an in-depth analysis of the characteristics of county distribution networks that integrate a variety of flexible resources, the spatiotemporal resource fusion planning scenarios are divided in combination with the energy endowment, flexible resource configuration, degree of digitization and other characteristics of different county distribution networks. For different planning scenarios, the planning evaluation indicators for different scenarios are screened according to the distribution system planning evaluation indicator system and expert experience. On the basis of the scenario-based multi-dimensional planning evaluation indicators, a data-driven approach is adopted, taking into account the importance and correlation between planning schemes and indicators, and comprehensive ranking is performed using game theory. A data-driven multi-dimensional planning evaluation indicator system is constructed, and the entropy weight method is used for evaluation, ultimately realizing the scenario-based, differentiated, and flexible planning analysis of county distribution networks.

[0297] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A flexible construction and evaluation method for multi-dimensional planning indicators of county distribution networks, characterized by: include: Analyze the operating characteristics of county-level distribution networks that integrate multiple flexible resources, and divide planning scenarios for county-level distribution networks that integrate spatiotemporal resources based on the energy endowment, flexible resource configuration, and degree of digitization of different county-level distribution networks. Based on the results of different planning scenarios and the complete indicator system for county distribution network planning evaluation, multi-dimensional planning evaluation indicators for different scenarios are constructed; Based on the multidimensional planning evaluation indicators of different scenarios, a data-driven approach is adopted, taking into account the importance and correlation between planning schemes and indicators, and using game theory to comprehensively rank the indicators to construct a data-driven multidimensional planning evaluation indicator; Based on the data-driven multi-dimensional planning evaluation indicators, the planning scheme is evaluated by the entropy weight method; The multi-dimensional planning evaluation indicators for different scenarios include: The comprehensive indicator system for county-level distribution network planning and evaluation includes multiple first-level, second-level, and third-level indicators in terms of clean and low-carbon, safe and sufficient, economical and efficient, flexible and intelligent, and supply-demand coordination. Based on the distribution system planning evaluation index system and combined with the characteristic data of different planning scenarios, differentiated planning evaluation indicators for different scenarios are constructed, and planning evaluation indicators for different scenarios are screened from the planning evaluation index system; Constructing data-driven multi-dimensional planning evaluation indicators includes: A data-driven approach is used to preliminarily screen potential indicators for each scenario using the maximum correlation minimum redundancy algorithm to maintain the maximum correlation between the indicators and the target variable and minimize the redundant information between the indicators. The importance of the selected indicators is evaluated through the random forest algorithm to deal with the nonlinear relationship between the indicators; By combining relevance and importance through the game theory optimization model, planning indicators are sorted and screened, and indicator weights are allocated and dynamically adjusted.

2. A flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network according to claim 1, characterized in that: The scenario division of county distribution network planning for spatiotemporal resource integration includes: The scenario division of the county-level distribution network planning for the integration of spatiotemporal resources includes the division of energy endowments, flexible resource allocation, and digital intelligence levels in different power supply areas or at different development stages of power supply areas; The energy endowment divides energy resources into energy input type and energy output type. The energy input type indicates that the installed capacity of centralized and distributed energy resources in the county distribution network planning scheme is small, while the energy output type indicates that the installed capacity of centralized and distributed energy resources in the county distribution network planning scheme is large. The flexible resource configuration represents the installed capacity of the flexible resources in the county distribution network planning scheme; The degree of digitalization represents the degree of digitalization configuration in the county distribution network planning scheme.

3. A flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network according to claim 1, characterized in that: The potential indicators of each scenario are preliminarily screened using the maximum correlation and minimum redundancy algorithm to maintain the maximum correlation between the indicators and the target variable and minimize the redundant information between the indicators, including: For the indicator input data set X={x1,x2,...,x n } and target variable c, perform data standardization and missing value processing; Calculate the mutual information I(xi;c) between each feature xi and the target variable c, expressed as: I(xi;c)=∑∑p(xi,c)log(p(xi,c) / (p(xi)p(c))) Among them, p(xi,c) represents the joint probability distribution, p(xi) and p(c) represent the marginal probability distribution; Calculate the mutual information I(xi;xj) between features and evaluate the redundancy between features, which is expressed as: I(xi;xj)=∑∑p(xi,xj)log(p(xi,xj) / (p(xi)p(xj))) The MIQ criterion is used to calculate the correlation D between the feature and the target variable, which is expressed as: D=I(xi;c) Perform redundancy measurement calculation and calculate the average mutual information R between the feature and the features in the selected feature set S, which is expressed as: R=1 / |S|∑I(xi;xj),xj∈S Among them, |S| represents the size of the selected feature set; In each iteration, the feature with the maximum MIQ score is selected and added to the set of selected features to obtain the mRMR importance score. Expressed as: Among them, N v 、r j They represent the total number of indicators that are effectively sorted by mRMR, and are indicators x j The above steps are repeated until the preset number of features is reached or the MIQ score is lower than the set threshold.

4. A flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network according to claim 3, characterized in that: The importance of the selected indicators is evaluated through the random forest algorithm, and the nonlinear relationship between the indicators is processed, including: Bootstrap sampling is performed on the training set to randomly generate k sample subsets Si and the corresponding decision tree Ti; When a node splits, m features are randomly selected from m features. try features as candidate feature sets; Calculate the optimal splitting point of the candidate feature for each node, calculate the mean square error of the parent node using different splitting thresholds, and calculate the mean square error of the left and right child nodes at the same time. The calculation split gain is expressed as: Gain=MSE 父节点 -(NL / N×MSE 左节点 +NR / N×MSE 右节点 ) Select the feature with the largest split gain and the corresponding split point. Select the feature with the largest mean square error reduction for splitting, and then calculate the importance of each feature in a single tree, which is expressed as: Impj,tree = ∑(MSE reduction when node uses feature j) Calculate the average importance of a feature in the entire forest, expressed as: Impj=(1 / T)×∑(Impj,tree) The importance of features is obtained through normalization, and the feature importance ranking is output.

5. A flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network according to claim 4, characterized in that: The evaluation of planning schemes by entropy weight method includes: Setting indicator types, including benefit-based indicators, cost-based indicators, and intermediate indicators, dividing planning evaluation indicators according to the definitions of each indicator type, and performing data preprocessing on the cost-based indicators and intermediate indicators; The probability matrix is ​​introduced through the entropy weight method to calculate the information entropy value of the information contained in each indicator. Based on the information entropy value, the redundancy and indicator weight of each level of indicators are calculated.

6. A flexible construction and evaluation system for multi-dimensional planning indicators of a county distribution network, applying a flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network as described in any one of claims 1 to 5, characterized in that: include: A partitioning module is used to analyze the operating characteristics of county-level distribution networks that integrate multiple flexible resources. It then divides the planning scenarios of county-level distribution networks that integrate spatiotemporal resources based on the energy endowment, flexible resource configuration, and digitalization characteristics of different county-level distribution networks. The first construction module is used to construct multi-dimensional planning evaluation indicators for different scenarios based on the results of different planning scenarios and the complete indicator system for county distribution network planning evaluation; The second construction module is used to construct a data-driven multi-dimensional planning evaluation indicator based on the multi-dimensional planning evaluation indicators of the different scenarios, taking into account the importance and correlation between the planning schemes and the indicators, and comprehensively ranking the indicators using game theory; An evaluation module is used to evaluate the planning scheme through an entropy weight method based on the data-driven multi-dimensional planning evaluation indicators.

7. A computer device, characterized in that: include: memory and 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 method for flexible construction and evaluation of multi-dimensional planning indicators of a county distribution network as described in any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium, characterized in that It stores computer-executable instructions, which, when executed by a processor, implement the steps of a flexible construction and evaluation method for multi-dimensional planning indicators of a county distribution network as described in any one of claims 1 to 5.

Images