A two-layer index system optimization method for wind, solar and storage multi-system

Abstract

The present invention discloses a double-layer indicator system optimization method for a wind, solar and storage multi-system, which belongs to the field of new energy technology. The method constructs a hierarchical indicator architecture for a wind, solar and storage multi-system and generates a double-layer indicator system for the wind, solar and storage multi-system; simplifies the preliminary indicators in the double-layer indicator system, calculates the correlation between the preliminary indicators, and optimizes the preliminary indicators; generates subjective weights of secondary indicators according to the correlation and sequence of primary indicators, generates objective weights based on the standard information entropy of the secondary indicators, combines the subjective weights and the objective weights to generate comprehensive weights of the secondary indicators, and generates comprehensive weights of the corresponding primary indicators based on the comprehensive weights of the secondary indicators. The weights are flexibly configured and adjusted, thereby improving the pertinence and effectiveness of the evaluation; constructs a single scoring structure for each indicator, and integrates the single scoring structures to generate a double-layer indicator system for the wind, solar and storage multi-system.

Description

Technical Field

[0001] The present invention belongs to the field of new energy technology and relates to a double-layer index system optimization method for a wind, solar and storage multi-system. Background Art

[0002] With the transformation of the global energy structure and the rapid development of renewable energy, the importance of wind, solar and storage multi-systems as the key link connecting renewable energy power generation and electricity consumption markets is becoming increasingly prominent. Such systems effectively improve energy utilization and grid stability by integrating multiple renewable energy sources such as wind and solar energy and using energy storage technology to store and regulate energy.

[0003] However, in the practical application of wind, solar and storage multi-systems, how to scientifically and accurately evaluate their performance and optimize and improve the system accordingly has become a technical problem that needs to be solved urgently. Traditional performance evaluation methods often focus on the examination of a single indicator or several key indicators, lacking comprehensiveness and systematicness. At the same time, since wind, solar and storage multi-systems involve multiple disciplines, such as energy, electricity, control, etc., their performance evaluation indicators are numerous and complex, and there may be redundancy or high correlation between the indicators, which not only increases the complexity of the evaluation process, but also affects the accuracy and reliability of the evaluation results.

[0004] The existing Chinese patent application with publication number CN103761677A discloses an evaluation index system and method that integrates wind power, photovoltaic, energy storage and transmission projects. According to the different evaluation contents and applicable objects, a multi-link evaluation index system is established, including a project overall effect evaluation index system and a project link evaluation index system; secondly, the relevant standard content in the relevant national or industry regulations is referred to, and the actual situation and specific conditions of wind power, photovoltaic, energy storage and transmission projects are combined to form an index evaluation standard; thirdly, a "multi-dimensional evaluation method system" is proposed that combines the comprehensive evaluation method of the "macro level" with the basic evaluation method of the "micro level" to obtain objective and scientific evaluation results.

[0005] Although the existing technology has formed an evaluation standard with reference and guiding significance, filling the gap in the field of post-evaluation of new energy combined power generation in my country, it has not considered the specific optimization strategy of the index system, especially the dependence between various indicators and the flexibility of the index system. For example, increasing the capacity of energy storage equipment may increase the economic cost of the system, but at the same time it can also improve the reliability and stability of the system. Therefore, this application provides a two-layer index system optimization method and method for a wind, solar and energy storage multi-system, which calculates the correlation between various indicators, eliminates highly correlated indicators, and assigns weight coefficients to the corresponding indicators. Summary of the Invention

[0006] In response to the shortcomings of the existing technology, the purpose of the present invention is to provide a two-layer indicator system optimization method and method for a wind, solar and energy storage multi-system. The indicators are judged and screened through the correlation between the various preliminary indicators, which reduces the complexity of the evaluation. The comprehensive weight coefficient of the indicator is calculated by combining the subjective weight and the objective weight, thereby enhancing the flexibility and adaptability of the indicator system.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A dual-layer index system optimization method for a wind, solar, and energy storage multi-system includes:

[0009] Step S1: constructing a hierarchical indicator framework for the wind, solar, and storage multi-system, and generating a two-layer indicator system for the wind, solar, and storage multi-system;

[0010] Step S2: simplifying the preliminary selected indicators in the two-tier indicator system, calculating the correlation between the preliminary selected indicators, and constructing a correlation sequence for each indicator. The correlation sequences in the two-tier indicator system are judged and screened in turn, and the preliminary selected indicators are optimized.

[0011] Step S3: Generate subjective weights of secondary indicators based on the correlation and sequence of primary indicators, generate objective weights based on the standard information entropy of secondary indicators, combine the subjective weights and objective weights to generate comprehensive weights of the secondary indicators, and generate comprehensive weights of the corresponding primary indicators based on the comprehensive weights of the secondary indicators;

[0012] Step S4: Using a trapezoidal fuzzy membership function for fitting, constructing a single scoring structure for each indicator, and fusing the single scoring structures to generate an optimized two-layer indicator system for the wind-solar-storage multi-system;

[0013] Wherein, the step S3 includes:

[0014] S3.1: The first-level indicator of the two-tier indicator system is {D1, D2, ..., D k}, first-level indicator D x The sub-indicators are Among them, D x is the xth indicator in the first-level indicator, s x First-level indicator D x The number of categories for the next sub-indicator;

[0015] S3.2: Obtain the first-level indicator D x The judgment matrix C of the next sub-indicator x , and calculate the secondary index D xz Judgment value Among them, D xz The first-level indicator D xThe zth sub-index under x ;

[0016] S3.3: Obtain the first-level indicator D x Correlation matrix after correlation screening For the correlation matrix M x The first-level index D is obtained by summing up the relevant sequences within x Correlation and sequence

[0017] S3.4: Using the correlation and sequence sum(M x ) and the judgment value Calculate secondary indicator D xz The approximate weight of The expression is as follows:

[0018]

[0019] S3.5: Approximate weights Perform standardization to obtain the secondary index D xz The subjective weight of

[0020] Specifically, step S2 includes:

[0021] S2.1: Set the first-level indicators of the preliminary indicator system as {H1, H2, ..., H k}, first-level indicator H x The sub-indicators are Where k is the number of categories of the first-level indicators, x=0,1,…,k, H x is the xth indicator in the first-level indicator, t x H is the first-level indicator x The number of categories for the next sub-indicator;

[0022] S2.2: Collect operational data for each indicator in the preliminary indicator system;

[0023] S2.3: Clean, organize, and standardize the collected data;

[0024] S2.4: Calculate the correlation coefficients between the secondary indicators under the same primary indicator;

[0025] S2.5: Sub-index H xy Construct a correlation sequence R xy ;

[0026] S2.6: Set the correlation threshold r th, compare and judge the relevant sequences of each secondary indicator, screen and eliminate highly correlated indicators, and generate a two-layer indicator system.

[0027] Specifically, the S2.6 further includes:

[0028] S2.61: Get the first level indicator H x The related sequence of each secondary index generates the first-level index H x Correlation matrix

[0029] S2.62: If |r| ≥ r th , the correlation coefficient is defined as the correlation index coefficient; if |r| <r th , the correlation coefficient is defined as the irrelevant index coefficient; where r is the correlation matrix R x The correlation coefficient in ;

[0030] S2.63: Read related sequence R x1 The category of the correlation coefficient; when R x1 If there is no relevant index coefficient in the , then the secondary index H is retained x1 ; When the related sequence R x1 If there is a related index coefficient in the , then the secondary index H is eliminated. x1 , and update the correlation matrix R x ;

[0031] S2.64: Update the correlation matrix R x Other related sequences in the sparse matrix are retained or eliminated until the correlation matrix R is traversed. x All related sequences in .

[0032] Specifically, the specific steps of step S3 also include:

[0033] S3.6: Construct the first-level indicator D x The sample data matrix G x , and calculate the secondary index D xz Information entropy E xz ;

[0034] S3.7: Based on the information entropy E xz , calculate the secondary index D xz The standard information entropy The expression is as follows:

[0035]

[0036] Where, Var(E x ) is the first-level indicator D x The variance of the information entropy;

[0037] S3.8: Information entropy according to the standard Calculate the secondary index D xz The objective weight ζ xz , the expression is as follows:

[0038]

[0039] S3.9: Based on the sub-indicator D xz The subjective weight and objective weight are used to calculate the secondary index D xz The comprehensive weight w xz .

[0040] Specifically, the specific steps of step S3 also include:

[0041] S3.10: The first level indicator D x The comprehensive weights of all secondary indicators are summarized and the arithmetic average is used to calculate the primary indicator D x The weight ε x ;

[0042] S3.11: Normalize the weight of the first-level index to generate the first-level index D x The comprehensive weight w x , the expression is as follows:

[0043]

[0044] Where w x D is the first-level index after normalization x The comprehensive weight of .

[0045] Specifically, step S4 includes:

[0046] S4.1: For each indicator, design a trapezoidal fuzzy membership function, where the first-level indicator D x The membership function is denoted as μ x (α), secondary indicator D xz The membership function is denoted as μ xz (α);

[0047] S4.2: Obtain the secondary indicator D xz The actual value of α xz , and calculate the secondary index D xz The membership value μ xz (α xz );

[0048] S4.3: Set the secondary indicator D xz The rating range is [S xz-min ,S xz-max], and calculate the secondary index D xz Rating S xz , the expression is as follows:

[0049] S xz =S xz-min +(S xz-max -S xz-min )·μ xz (α xz )

[0050] Where S xz It is the single scoring structure of the zth secondary indicator under the xth first-level indicator.

[0051] Specifically, the step S4 further includes:

[0052] S4.4: Calculate the primary indicator D based on the comprehensive weight of the secondary indicators x Calculation score of x ;

[0053] S4.5: Calculate the score s x As the membership function μ x (α) input, calculate the first-level index D x The membership value μ x (s x );

[0054] S4.6: Based on Level 1 Indicator D x Rating range [S x-min ,S x-max ], use linear interpolation to calculate the first-level index D x Rating S x ;

[0055] S4.7: Based on the comprehensive weights of the first-level indicators in S3.11, calculate the comprehensive score S of the wind, solar and storage multi-system and generate an optimized two-tier indicator system for the wind, solar and storage multi-system.

[0056] Beneficial effects of the present invention:

[0057] 1. Construct a hierarchical indicator framework for the wind, solar, and storage multi-system to ensure the comprehensiveness and hierarchy of the indicator system, allowing evaluators to clearly see the performance of each level of the system; and strictly screen and optimize the preliminary indicators. By calculating the correlation between indicators, highly redundant and repeatedly calculated indicators are removed to ensure the simplicity and effectiveness of the indicator system. This not only reduces the complexity of the evaluation, but also improves the accuracy and reliability of the evaluation results.

[0058] 2. The judgment values ​​of secondary indicators are generated through hierarchical analysis, and the influence of each secondary indicator in the corresponding primary indicator is quantified according to the correlation and sequence of the primary indicators, and the subjective weights of the secondary indicators are generated. At the same time, the correlation and relative importance between the secondary indicators are taken into account, avoiding information redundancy and duplication of weight configuration; and the information entropy of the secondary indicators is calculated using the entropy weight method, and the standard information entropy is calculated using the variance of the information entropy. The discrete degree of the information entropy of all secondary indicators is taken into account to evaluate the importance of each secondary indicator in the overall evaluation. At the same time, the flexibility and adaptability of the indicator system are enhanced, and the objective weights of the secondary indicators are generated so that smaller objective weights are given when the data fluctuates greatly, and finally the comprehensive weights of the secondary indicators are generated. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 This is a flow chart of a two-tier index system optimization method for a wind, solar and storage multi-system;

[0060] Figure 2 This is an indicator optimization flow chart for a dual-layer indicator system optimization method for a wind, solar, and storage multi-system;

[0061] Figure 3 A weight configuration flow chart for a two-tier indicator system optimization method for a wind, solar, and storage multi-system;

[0062] Figure 4 Generate a two-tier indicator system flow chart for a two-tier indicator system optimization method for a wind, solar and energy storage multi-system. DETAILED DESCRIPTION

[0063] The technical solution of the present invention is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present invention and the specific features in the embodiments are detailed descriptions of the technical solution of the present invention, rather than limitations on the technical solution of the present invention. In the absence of conflict, the embodiments of the present invention and the technical features in the embodiments can be combined with each other.

[0064] Example 1

[0065] refer to Figures 1 to 3 As shown, this embodiment introduces a two-layer index system optimization method for a wind, solar and storage multi-system, including the following steps:

[0066] Step S1: Construct a hierarchical indicator framework for the wind, solar, and storage multi-system, and generate a two-tier indicator system for the wind, solar, and storage multi-system. The preliminary indicators of the two-tier indicator system are shown in Table 1. The two-tier indicator system adopts a two-tier design, with the first tier being the primary indicators and the second tier being the secondary indicators. There is a clear hierarchical relationship between the primary indicators and the secondary indicators, which together form a rigorous and well-organized two-tier indicator system, which helps to ensure the comprehensiveness and systematicity of the evaluation and also facilitates subsequent data collection and analysis.

[0067] Table 1 Preliminary indicators of the two-tier indicator system

[0068]

[0069]

[0070] Step S2: simplify the preliminary selected indicators in the two-layer indicator system, calculate the correlation between the preliminary selected indicators, and construct a correlation sequence for each indicator. Then, judge and screen the correlation sequences in the two-layer indicator system in turn, and optimize the preliminary selected indicators.

[0071] Step S3: Configure the weights of the indicators. Generate the subjective weights of the secondary indicators based on the correlation and sequence of the primary indicators. Generate the objective weights based on the standard information entropy of the secondary indicators. Combine the subjective weights and the objective weights to generate the comprehensive weights of the secondary indicators. Calculate the comprehensive weights of the primary indicators based on the comprehensive weights of the secondary indicators. This implements the weight configuration of the indicators. The weights can be flexibly configured and adjusted according to different situations and needs, so that the indicator system can better adapt to different evaluation scenarios and objects, thereby improving the pertinence and effectiveness of the evaluation.

[0072] Step S4: Use the trapezoidal fuzzy membership function for fitting, obtain the corresponding indicator membership function according to the properties of each indicator, and determine the single-item scoring structure; and organically integrate multiple single-item scoring structures to generate an optimized two-layer indicator system for the wind, solar and storage multi-system.

[0073] Specifically, the specific steps of step S2 include:

[0074] S2.1: The first-level indicators of the two-tier indicator system are divided into k categories, represented by {H1, H2, …, H k}, first-level indicator H x The sub-indicators under are divided into t x Class, represented by Among them, H x is the xth index in the first-level index, x=0,1,…,k, H xy H is the first-level indicator x The y-th sub-index in , y=0,1,…,t x ;

[0075] S2.2: Collect operational data for each indicator in the preliminary indicator system from each component of the wind, solar and energy storage multi-system (e.g., wind power generation system, photovoltaic power generation system, energy storage system);

[0076] S2.3: Clean and organize the collected data, remove outliers, missing values, and other data that do not meet the requirements, and standardize the data to ensure comparability between different indicators;

[0077] S2.4: Use the Pearson correlation coefficient formula to calculate the correlation coefficient between each secondary indicator under the same primary indicator. The expression is as follows:

[0078]

[0079] In the formula, τ and σ are sample sequences of two comparison indicators, r τσ is the correlation coefficient between the indicators corresponding to the two comparison indicators, L is the number of samples in the sample sequence, τ l and σ l are the lth sample values ​​in the corresponding sample sequence, and is the average value of the sample series of the two comparison indicators, The size of the correlation coefficient reflects the degree of correlation between indicators. The closer the correlation coefficient is to 1 or -1, the stronger the correlation between indicators. The closer the correlation coefficient is to 0, the weaker the correlation between indicators.

[0080] S2.5: Sub-index H xy Construct a correlation sequence in, Secondary indicator H xy and secondary indicators The correlation coefficient between 11 The related sequence is index The related sequence is

[0081] S2.6: Set the correlation threshold r th , compare and judge the correlation sequence of each secondary indicator, screen and eliminate highly correlated indicators, and generate a two-layer indicator system after correlation screening.

[0082] Specifically, the specific steps of S2.6 also include:

[0083] S2.61: Get the first level indicator H x The related sequence of each secondary index is used to generate the first-level index H x Correlation matrix

[0084] S2.62: The correlation matrix R x The correlation coefficient r and the correlation threshold r in thCompare and determine the specific categories of each correlation coefficient, where the categories include relevant indicator coefficients and irrelevant indicator coefficients; if |r| ≥ r th , the correlation coefficient is defined as the correlation index coefficient, if |r| <r th , the correlation coefficient is defined as the coefficient of irrelevant indicators;

[0085] S2.63: Read related sequence R x1 The category of the correlation coefficient; when the correlation sequence R x1 If there is no relevant index coefficient in the , then retain the secondary index H x1 ; When the related sequence R x1 If there is a related index coefficient in the x1 , and the correlation matrix R x Middle and secondary indicators H x1 The relevant correlation coefficients are deleted to update the correlation matrix R x ;

[0086] S2.64: Update the correlation matrix R x Other related sequences in the sparse matrix are retained or eliminated until the correlation matrix R is traversed. x All relevant sequences in the are used to generate a two-layer indicator system after correlation screening.

[0087] Specifically, the specific steps of step S3 include:

[0088] S3.1: After optimization, the first-level indicators of the two-tier indicator system are still divided into k categories, represented by {D1, D2, …, D k}, first-level indicator D x The sub-indicators under x Class, represented by Among them, D xz First-level indicator D x The zth sub-index under x ;

[0089] S3.2: Level 1 indicator D x By s x Secondary indicators are determined to obtain the primary indicator D obtained through expert experience x Judgment matrix To quantify the relative importance of each sub-indicator under the first-level indicator and calculate the sub-indicator D xz Judgment value It reflects the relative importance between the secondary indicators. The expression is as follows:

[0090]

[0091] Where c ab First-level indicator D x Next level indicator D xa For secondary indicator D xb The importance of a=1,2,…,s x ,b=1,2,…,s x , and when a=b, c ab =1;

[0092] S3.3: Obtain the first-level indicator D after relevant screening x Correlation matrix Sum each correlation sequence in the correlation matrix to obtain the first-level index D x Correlation and sequence The influence of each secondary indicator on the corresponding primary indicator is quantified; among them, M xz is the optimized secondary indicator D xz , sum(·) is the summation formula;

[0093] S3.4: Use Level 1 Indicator D x The correlation and sequence sum(M x ) and the judgment value Calculate secondary indicator D xz The approximate weight of The correlation and relative importance between the secondary indicators are taken into account at the same time, avoiding information redundancy and duplication of weight configuration. The expression is as follows:

[0094]

[0095] S3.5: For sub-indicator D xz The approximate weight of the normalized process is used to obtain the secondary index D xz The subjective weight of The expression is as follows:

[0096]

[0097] S3.6: Level 1 indicator D x The sample data is standardized and the sample data matrix is ​​generated. And calculate the secondary index D xz Information entropy E xz , the expression is as follows:

[0098]

[0099] Where p zf Secondary indicator D xz The proportion of sample f, and

[0100] S3.7: When the sample data matrix G x When the data fluctuates greatly, the quantitative results of the corresponding secondary indicators are of poor quality and should be given a smaller objective weight. xz , calculate the secondary index G xz The standard information entropy While considering the information entropy of a single secondary indicator, the discrete degree of the information entropy of all secondary indicators is also considered to evaluate the importance of each secondary indicator in the overall evaluation. The expression is as follows:

[0101]

[0102] Where, Var(E x ) is the first-level indicator D x The variance of all information entropy under is the variance adjustment factor;

[0103] S3.8: Calculate the secondary index D based on the size of the standard information entropy xz The objective weight ζ xz , the expression is as follows:

[0104]

[0105] S3.9: Based on Sub-Indicator D xz The subjective weight and objective weight of the secondary index D are calculated. xz The comprehensive weight w xz , the expression is as follows:

[0106]

[0107] S3.10: Since each first-level indicator is composed of multiple sub-indicators, the first-level indicator D x The comprehensive weights of all secondary indicators are summarized and the arithmetic average is used to calculate the primary indicator D x The weight ε x , the expression is as follows:

[0108]

[0109] S3.11: In order to make the sum of the weights of the first-level indicators equal to 1, the weights of the first-level indicators are normalized to generate the first-level indicator D x The comprehensive weight w x , the expression is as follows:

[0110]

[0111] Where wx D is the first-level index after normalization x The comprehensive weight of .

[0112] Specifically, the specific steps of step S4 include:

[0113] S4.1: For each indicator, including primary indicators and secondary indicators, design a trapezoidal fuzzy membership function. The primary indicator D x The membership function is denoted as μ x (α), secondary indicator D xz The membership function is denoted as μ xz (α); the expression is as follows:

[0114]

[0115] Wherein, μ(α) is the expression of the trapezoidal fuzzy membership function, (e1, e2, e3, e4) are the key parameters of the function μ(α), e1 is the lowest value point of the membership, e2 is the highest value point of the membership, e3 is the transition point where the membership increases from 0 to 1, and e4 is the transition point where the membership decreases from 1 to 0, and the parameters of each membership function are different, and the specific parameter values ​​are defined by those skilled in the art;

[0116] S4.2: Obtain secondary indicator D xz The actual value of α xz , substitute the actual value into the corresponding membership function and calculate the secondary index D xz The membership value μ xz (α xz );

[0117] S4.3: Set sub-indicator D xz The rating range is [S xz-min ,S xz-max ], use linear interpolation to calculate the secondary index D xz Rating S xz , the expression is as follows:

[0118] S xz =S xz-min +(S xz-max -S xz-min )·μ xz (α xz )

[0119] Where S xz It is the single scoring structure of the zth secondary indicator under the xth first-level indicator;

[0120] S4.4: Calculate the primary indicator D based on the comprehensive weight of the secondary indicators in the weight configuration. x Calculation score of x, the expression is as follows:

[0121]

[0122] S4.5: The score s will be calculated x As the membership function μ x (α) input, calculate the first-level index D x The membership value μ x (s x );

[0123] S4.6: Based on Level 1 Indicator D x Rating range [S x-min ,S x-max ], use linear interpolation to calculate the first-level index D x Rating S x , the expression is as follows:

[0124] S x =S x-min +(S x-max -S x-min )·μ x (s x )

[0125] S4.7: Based on the comprehensive weights of the first-level indicators in S3.11, calculate the comprehensive score S of the wind, solar and storage multi-system. The expression is as follows:

[0126]

[0127] In the formula, the comprehensive score S reflects the overall performance and expression of the wind-solar-storage multi-system. By organically integrating multiple individual scoring structures and combining the comprehensive weight coefficients of each indicator, an optimized two-layer indicator system for the wind-solar-storage multi-system is generated.

[0128] In summary, the present invention constructs a preliminary indicator system, judges and screens the indicators in the preliminary indicator system based on the correlation between the indicators, combines hierarchical analysis with the correlation and sequence of the primary indicators, generates the subjective weights of the secondary indicators, and uses the entropy weight method to combine the variance of the information entropy to calculate the standard information entropy, generates the objective weights of the secondary indicators, generates the comprehensive weights of the secondary indicators, summarizes the secondary indicators, generates the comprehensive weights of the primary indicators, uses the configured weights to generate the single scoring structure of each indicator, and integrates the single scoring structures to generate the optimized two-layer indicator system of the wind, solar and storage multi-system.

[0129] The above description is merely a preferred embodiment of the present invention. The scope of protection of the present invention is not limited to the above embodiment. All technical solutions based on the concept of the present invention are within the scope of protection of the present invention. It should be noted that for those skilled in the art, various improvements and modifications that do not depart from the principles of the present invention should also be considered within the scope of protection of the present invention.

Claims

1. A dual-layer index system optimization method for a wind, solar, and energy storage multi-system, characterized in that: include: Step S1: constructing a hierarchical indicator framework for the wind, solar, and storage multi-system, and generating a two-layer indicator system for the wind, solar, and storage multi-system; Step S2: simplifying the preliminary selected indicators in the two-tier indicator system, calculating the correlation between the preliminary selected indicators, and constructing a correlation sequence for each indicator. The correlation sequences in the two-tier indicator system are judged and screened in turn, and the preliminary selected indicators are optimized. Step S3: Generate subjective weights of secondary indicators based on the correlation and sequence of primary indicators, generate objective weights based on the standard information entropy of secondary indicators, combine the subjective weights and objective weights to generate comprehensive weights of the secondary indicators, and generate comprehensive weights of the corresponding primary indicators based on the comprehensive weights of the secondary indicators; Step S4: Using a trapezoidal fuzzy membership function for fitting, constructing a single scoring structure for each indicator, and fusing the single scoring structures to generate an optimized two-layer indicator system for the wind-solar-storage multi-system; Wherein, the step S3 includes: S3.1: The first-level indicator of the two-tier indicator system is {D1, D2, ..., D k }, first-level indicator D x The sub-indicators are Among them, D x is the xth indicator in the first-level indicator, s x First-level indicator D x The number of categories for the next sub-indicator; S3.2: Obtain the first-level indicator D x The judgment matrix C of the next sub-indicator x , and calculate the secondary index D xz Judgment value Among them, D xz The first-level indicator D x The zth sub-index under x ; S3.3: Obtain the first-level indicator D x Correlation matrix after correlation screening For the correlation matrix M x The first-level index D is obtained by summing up the relevant sequences within x Correlation and sequence S3.4: Using the correlation and sequence sum(M x ) and the judgment value Calculate secondary indicator D xz The approximate weight of The expression is as follows: S3.5: Approximate weights Perform standardization to obtain the secondary index D xz The subjective weight of 2. The method for optimizing a double-layer index system for a wind-solar-storage multi-system according to claim 1, characterized in that: The step S2 comprises: S2.1: Set the first-level indicators of the preliminary indicator system as {H1, H2, ..., H k }, first-level indicator H x The sub-indicators are Where k is the number of categories of the first-level indicators, x=0,1,…,k, H x is the xth indicator in the first-level indicator, t x H is the first-level indicator x The number of categories for the next sub-indicator; S2.2: Collect operational data for each indicator in the preliminary indicator system; S2.3: Clean, organize, and standardize the collected data; S2.4: Calculate the correlation coefficients between the secondary indicators under the same primary indicator; S2.5: Sub-index H xy Construct a correlation sequence R xy ; S2.6: Set the correlation threshold r th , compare and judge the relevant sequences of each secondary indicator, screen and eliminate highly correlated indicators, and generate a two-layer indicator system.

3. The method for optimizing a double-layer index system for a wind-solar-storage multi-system according to claim 2, characterized in that: Said S2.6 also includes: S2.61: Get the first level indicator H x The related sequence of each secondary index generates the first-level index H x Correlation matrix S2.62: If |r| ≥ r th , the correlation coefficient is defined as the correlation index coefficient; if |r| <r th , the correlation coefficient is defined as the irrelevant index coefficient; where r is the correlation matrix R x The correlation coefficient in ; S2.63: Read related sequence R x1 The category of the correlation coefficient; when R x1 If there is no relevant index coefficient in the , then the secondary index H is retained x1 ; When the related sequence R x1 If there is a related index coefficient in the , then the secondary index H is eliminated. x1 , and update the correlation matrix R x ; S2.64: Update the correlation matrix R x Other related sequences in the sparse matrix are retained or eliminated until the correlation matrix R is traversed. x All related sequences in .

4. The method for optimizing a double-layer index system for a wind, solar, and storage multi-system according to claim 3, characterized in that: The specific steps of step S3 also include: S3.6: Construct the first-level indicator D x The sample data matrix G x , and calculate the secondary index D xz Information entropy E xz ; S3.7: Based on the information entropy E xz , calculate the secondary index D xz The standard information entropy The expression is as follows: Where, Bar(E x ) is the first-level indicator D x The variance of the information entropy; S3.8: Information entropy according to the standard Calculate the secondary index D xz The objective weight ζ xz , the expression is as follows: S3.9: Based on the sub-indicator D xz The subjective weight and objective weight are used to calculate the secondary index D xz The comprehensive weight w xz .

5. The method for optimizing a double-layer index system for a wind, solar, and storage multi-system according to claim 4, characterized in that: The specific steps of step S3 also include: S3.10: The first level indicator D x The comprehensive weights of all secondary indicators are summarized and the arithmetic average is used to calculate the primary indicator D x The weight ε x ; S3.11: Normalize the weight of the first-level index to generate the first-level index D x The comprehensive weight w x , the expression is as follows: Where w x D is the first-level index after normalization x The comprehensive weight of .

6. The method for optimizing a double-layer index system for a wind, solar, and storage multi-system according to claim 5, characterized in that: The step S4 comprises: S4.1: For each indicator, design a trapezoidal fuzzy membership function, where the first-level indicator D x The membership function is denoted as μ x (α), secondary indicator D xz The membership function is denoted as μ xz (α); S4.2: Obtain the secondary indicator D xz The actual value of α xz , and calculate the secondary index D xz The membership value μ xz (α xz ); S4.3: Set the secondary indicator D xz The rating range is [S xz-min ,S xz-max ], and calculate the secondary index D xz Rating S xz , the expression is as follows: S xz =S xz-min +(S xz-max -S xz-min )·m xz (a xz ) Where S xz It is the single scoring structure of the zth secondary indicator under the xth first-level indicator.

7. The method for optimizing a double-layer index system for a wind, solar, and storage multi-system according to claim 6, characterized in that: The step S4 further includes: S4.4: Calculate the primary indicator D based on the comprehensive weight of the secondary indicators x Calculation score of x ; S4.5: Calculate the score s x As the membership function μ x (α) input, calculate the first-level index D x The membership value μ x (s x ); S4.6: Based on Level 1 Indicator D x Rating range [S x-min ,S x-max ], use linear interpolation to calculate the first-level index D x Rating S x ; S4.7: Based on the comprehensive weights of the first-level indicators in S3.11, calculate the comprehensive score S of the wind, solar and storage multi-system and generate an optimized two-tier indicator system for the wind, solar and storage multi-system.

Images