Distribution network comprehensive evaluation model optimization method based on distance correlation coefficient

Abstract

The invention discloses a power distribution network comprehensive evaluation model optimization method based on a distance correlation coefficient, and belongs to the technical field of power distribution management systems, and the method comprises the steps: S1, building a power distribution network comprehensive evaluation model based on an analytic hierarchy process, and obtaining the original weight of a comprehensive evaluation index in the power distribution network comprehensive evaluation model; s2, obtaining a first distance correlation coefficient between the comprehensive evaluation indexes based on historical data of the power distribution network and the original weight; and S3, obtaining a second distance correlation coefficient between the comprehensive evaluation indexes based on the real-time data of the power distribution network, obtaining a final weight of the comprehensive evaluation indexes based on the first distance correlation coefficient and the second distance correlation coefficient, and optimizing the comprehensive evaluation model of the power distribution network based on the final weight. The problem that the comprehensive evaluation accuracy is low due to the fact that the corresponding weight is difficult to adjust according to the change of the index data is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution management systems, and specifically to an optimization method for a comprehensive evaluation model of a distribution network based on distance correlation coefficients. Background Art

[0002] With the continuous expansion of the power system, in order to ensure the stable operation of the power grid and improve the power supply quality, it is extremely important to obtain the state of the distribution network. The existing technology mainly reflects the state of the distribution network through comprehensive evaluation of the distribution network. For example, the comprehensive evaluation method of a distribution network considering low-carbon benefits with the patent number CN116683518A includes: establishing a corresponding carbon emission model, analyzing the impact of distributed photovoltaics, energy storage power stations, and electric vehicle loads on the carbon emissions of the distribution network, verifying the rationality and effectiveness of each influencing factor through simulation calculations, and establishing corresponding low-carbon benefit evaluation indicators according to the influencing factors; establishing a comprehensive evaluation index system for the distribution network considering low-carbon benefits; obtaining the subjective weights of the indicators; obtaining the objective weights of the indicators; adopting a combination method with strong interpretability for the subjective and objective weights; using an improved fuzzy comprehensive evaluation method to establish a fuzzy comprehensive judgment matrix, and multiplying the weight coefficient matrix by the fuzzy comprehensive judgment matrix to obtain the comprehensive score of the distribution network. The above solution considers the low-carbon benefit indicators, establishes a comprehensive and new distribution network-adapted comprehensive evaluation index system, and at the same time improves the fuzzy comprehensive evaluation method to effectively reflect the weak links in the distribution network, making the evaluation results more accurate. However, the subjective weights and objective weights obtained in the above solution are static. When the indicator data corresponding to the subjective weight or the objective weight exceeds a certain threshold or is lower than a certain threshold, the weight of the corresponding indicator will change accordingly. However, the above solution does not consider the weight change caused by the dynamic change of the indicator data. Therefore, the above solution has the problem that it is difficult to adjust the corresponding weight according to the change of the indicator data, resulting in inaccurate comprehensive evaluation. Summary of the Invention

[0003] Aiming at the problem that the existing technology is difficult to adjust the corresponding weight according to the change of the indicator data, resulting in low accuracy of the comprehensive evaluation, the present invention provides an optimization method for a comprehensive evaluation model of a distribution network based on distance correlation coefficients. By obtaining the first distance correlation coefficient between the comprehensive evaluation indicators through historical data and the original weights obtained by the analytic hierarchy process, the quantification of the relationship between the historical data corresponding to the comprehensive evaluation indicators is realized. The second distance correlation coefficient between the comprehensive evaluation indicators is obtained according to the real-time data, realizing the dynamic capture of the relationship between the real-time data corresponding to the comprehensive evaluation indicators. Taking the original weight as the benchmark, the final weight is obtained according to the difference between the second distance correlation coefficient and the first distance correlation coefficient, solving the problem that the comprehensive evaluation accuracy is low due to the difficulty of adjusting the corresponding weight according to the change of the indicator data. Significantly improves the flexibility and accuracy of the comprehensive evaluation of the distribution network.

[0004] To solve the above technical problems, the present invention provides an optimization method for a comprehensive evaluation model of a distribution network based on distance correlation coefficients, comprising the following steps: S1: Construct a comprehensive evaluation model of the distribution network based on the analytic hierarchy process, and obtain the original weights of the comprehensive evaluation indicators in the comprehensive evaluation model of the distribution network; S2: Obtain the first distance correlation coefficient between the comprehensive evaluation indicators based on the historical data of the distribution network and the original weights; S3: Obtain the second distance correlation coefficient between the comprehensive evaluation indicators based on the real-time data of the distribution network, obtain the final weights of the comprehensive evaluation indicators based on the first distance correlation coefficient and the second distance correlation coefficient, and optimize the comprehensive evaluation model of the distribution network based on the final weights.

[0005] After adopting the above technical solution, the present invention has the following advantages: Since the weights of the comprehensive evaluation indicators are relatively set, and the correlation between the comprehensive evaluation indicators is difficult to quantify, it is difficult to adjust the weights according to the quantization results for the data changes corresponding to the comprehensive evaluation indicators. Considering that the distance correlation coefficient can capture the linear and non-linear relationships between data, the relationship between the historical data corresponding to the comprehensive evaluation indicators is quantified by obtaining the distance correlation coefficient between the comprehensive evaluation indicators. On this basis, the first distance correlation coefficient is obtained from the original weights obtained by the analytic hierarchy process and the historical data, providing an adjustment benchmark for the weights of the comprehensive evaluation indicators. The second distance correlation coefficient is obtained from the real-time data, realizing the dynamic capture of the relationship between the dynamic data corresponding to the comprehensive evaluation indicators. Furthermore, the original weights are adjusted according to the difference between the second distance correlation coefficient and the first distance correlation coefficient, significantly improving the accuracy and flexibility of the comprehensive evaluation, and at the same time improving the comprehensive evaluation efficiency; It solves the problem of low accuracy of comprehensive evaluation caused by the difficulty of adjusting the corresponding weights according to the changes in the index data.

[0006] Preferably, the S1 includes: S11: Decompose the distribution network evaluation indicators into different levels according to the distribution network evaluation objectives, obtain the comprehensive evaluation model of the distribution network according to the hierarchical relationship between different levels, and perform pairwise comparison of the comprehensive evaluation indicators in the comprehensive evaluation model of the distribution network to obtain a judgment matrix; S12: Perform a consistency test on the judgment matrix based on the characteristic roots and orders of the judgment matrix. If the consistency test passes, obtain the original weights based on the judgment matrix. If the consistency test fails, correct the judgment matrix according to the weight relationship preset by the expert opinions in the analytic hierarchy process, and execute S12.

[0007] Preferably, in S11, the judgment matrix satisfies where aij represents the element in the \(i\)-th row and \(j\)-th column of the judgment matrix, \(a\) ji represents the element in the \(j\)-th row and \(i\)-th column of the judgment matrix; The consistency test of the judgment matrix based on the eigenvalue and order of the judgment matrix includes: Obtain the first index characterizing consistency where \(\lambda\) is the eigenvalue and \(n\) is the order; Obtain the second index characterizing consistency where \(RI\) is the random consistency index obtained based on the order; If the second index is less than the preset value, it indicates that the consistency test passes; if the second index is greater than or equal to the preset value, it indicates that the consistency test fails.

[0008] Preferably, the S2 includes: S21: Obtain the theoretical comprehensive evaluation of the distribution network based on the historical data of the distribution network and the original weight, and compare the theoretical comprehensive evaluation with the actual comprehensive evaluation of the distribution network corresponding to the historical data of the distribution network; S22: Obtain the historical data of the distribution network with successful comparison, and obtain the first distance correlation coefficient between the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison.

[0009] In this solution, considering that the original weight obtained by the analytic hierarchy process needs to be used as a benchmark to obtain the final weight, and obtaining the final weight requires obtaining the historical data of the comprehensive evaluation indexes corresponding to the original weight. To ensure the accuracy and stability of the historical data of the comprehensive evaluation indexes corresponding to the original weight, the first distance correlation coefficient is obtained by obtaining the historical data of the distribution network with successful comparison, thereby indirectly improving the accuracy of the first distance correlation coefficient. At the same time, it also provides an accurate reference standard for the second distance correlation coefficient, thereby improving the accuracy of the final weight.

[0010] Preferably, in S22, the obtaining the first distance correlation coefficient between the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison includes: Obtain the first distance covariance between the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison, and obtain the first distance standard deviation of the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison; Obtain the first distance correlation coefficient based on the first distance covariance and the first distance standard deviation. By obtaining the distance correlation coefficient through the distance covariance and the distance standard deviation, the quantification of the correlation degree between the historical data corresponding to the comprehensive evaluation indexes is realized, and at the same time, the unification of the correlation scale between different comprehensive evaluation indexes is realized, thereby improving the convenience of obtaining the final weight through the distance standard deviation.

[0011] Preferably, S3 includes: S31: Obtain the second distance covariance between the comprehensive evaluation indicators based on the real-time data of the distribution network, and obtain the second distance standard deviation of the comprehensive evaluation indicators based on the real-time data of the distribution network; S32: Obtain the second distance correlation coefficient based on the second distance covariance and the second distance standard deviation, and obtain the final weight of the comprehensive evaluation indicators based on the first distance correlation coefficient and the second distance correlation coefficient. By obtaining the distance correlation coefficient through the distance covariance and the distance standard deviation, the quantification of the correlation degree between the real-time data corresponding to the comprehensive evaluation indicators is realized, and at the same time, the unification of the correlation scale between different comprehensive evaluation indicators is realized, thereby further improving the convenience of obtaining the final weight through the distance standard deviation.

[0012] Preferably, in S32, the obtaining the final weight of the comprehensive evaluation indicators based on the first distance correlation coefficient and the second distance correlation coefficient includes: If the absolute difference between the second distance correlation coefficient and the first distance correlation coefficient is greater than the preset difference, then based on the magnitude relationship of the original weights, use the absolute difference to obtain the final weight. If the absolute difference is less than or equal to the preset difference, then the original weight is the final weight.

[0013] Preferably, in S1, the comprehensive evaluation indicators include the distributed power source consumption capacity, power supply capacity, power supply reliability, and extreme disaster scenario recovery ability.

[0014] Preferably, S3 further includes: The optimized distribution network comprehensive evaluation model T = k 1 C + k 2 (α 1 η + α 2 ΔV + α 3 M + α 4 R), where k 1 , k 2 are both proportionality coefficients, C represents the distribution network carbon emission rights trading cost considering low-carbon constraints, α 1 is the final weight of the distributed power source consumption capacity, η represents the distributed power source consumption capacity, α 2 is the final weight of the power supply capacity, ΔV represents the power supply capacity, α 3 is the final weight of the power supply reliability, M represents the power supply reliability, α 4 is the final weight of the extreme disaster scenario recovery ability, and R represents the extreme disaster scenario recovery ability.

[0015] Advantages of this solution: Since the weights of the comprehensive evaluation indicators are relatively set, and it is difficult to quantify the correlation between the comprehensive evaluation indicators, it is difficult to adjust the weights according to the data changes corresponding to the comprehensive evaluation indicators based on the quantification results. Considering that the distance correlation coefficient can capture the linear and non-linear relationships between data, the relationship between the historical data corresponding to the comprehensive evaluation indicators is quantified by obtaining the distance correlation coefficient between the comprehensive evaluation indicators. On this basis, the original weight obtained by the analytic hierarchy process and the historical data are used to obtain the first distance correlation coefficient, which provides an adjustment benchmark for the weights of the comprehensive evaluation indicators. The second distance correlation coefficient is obtained through real-time data, realizing the dynamic capture of the relationship between the dynamic data corresponding to the comprehensive evaluation indicators. Furthermore, the original weight is adjusted according to the difference between the second distance correlation coefficient and the first distance correlation coefficient, significantly improving the accuracy, flexibility, and efficiency of the comprehensive evaluation; The distance correlation coefficient is obtained through distance covariance and distance standard deviation, realizing the quantification of the correlation degree between the data corresponding to the comprehensive evaluation indicators, and at the same time realizing the unification of the correlation scales between different comprehensive evaluation indicators, thereby improving the convenience of obtaining the final weight through the distance standard deviation; It solves the problem of low accuracy of comprehensive evaluation caused by the difficulty of adjusting the corresponding weights according to the changes in index data.

[0016] The present invention also provides a storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, the steps of the optimization method for the distribution network comprehensive evaluation model based on the distance correlation coefficient are realized. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, objects, and advantages of the present invention will become more apparent. The drawings are only for the purpose of showing the preferred embodiments and are not considered as limiting the present invention. Moreover, throughout the drawings, the same reference numerals are used to represent the same components.

[0018] Figure 1 It is a flowchart of the optimization method for the distribution network comprehensive evaluation model based on the distance correlation coefficient of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0019] In order to make the purpose, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only the best embodiments of the present invention and are only used to explain the present invention, without limiting the protection scope of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts fall within the protection scope of the present invention.

[0020] Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts depict the operations (or steps) as sequential processes, many of the operations (or steps) can be implemented in parallel, concurrently, or simultaneously. In addition, the order of the operations can be rearranged. The process can be terminated when its operations are completed, but it can also have additional steps not included in the drawings; the process can correspond to a method, function, procedure, subroutine, subprogram, and so on.

[0021] Embodiment 1: As Figure 1 shown, an optimization method for a comprehensive evaluation model of a distribution network based on a distance correlation coefficient includes the following steps: S1: Construct a comprehensive evaluation model of the distribution network based on the analytic hierarchy process, and obtain the original weights of the comprehensive evaluation indicators in the comprehensive evaluation model of the distribution network.

[0023] The S1 includes: S11: Decompose the distribution network evaluation indicators to different levels according to the distribution network evaluation objectives, and obtain the comprehensive evaluation model of the distribution network according to the hierarchical relationship between different levels. Compare the comprehensive evaluation indicators in the comprehensive evaluation model of the distribution network pairwise to obtain a judgment matrix; S12: Perform a consistency test on the judgment matrix based on the characteristic root and order of the judgment matrix. If the consistency test passes, obtain the original weights based on the judgment matrix. If the consistency test fails, correct the judgment matrix according to the weight relationship preset based on the expert opinions in the analytic hierarchy process, and execute S12.

[0024] In S11, the judgment matrix satisfies where a ij represents the element in the i-th row and j-th column of the judgment matrix, and a ji represents the element in the j-th row and i-th column of the judgment matrix; The performing a consistency test on the judgment matrix based on the characteristic root and order of the judgment matrix includes: Obtain a first index representing consistency where λ is the characteristic root and n is the order; Obtain a second index representing consistency where RI is a random consistency index obtained based on the order; If the second index is less than a preset value, it indicates that the consistency test passes. If the second index is greater than or equal to the preset value, it indicates that the consistency test fails.

[0025] In S1, the comprehensive evaluation index includes the distributed power source consumption capacity, power supply capacity, power supply reliability, and extreme disaster scenario recovery capacity.

[0026] In this embodiment, the evaluation objective of the distribution network is to evaluate the comprehensive capacity of the distribution network. The comprehensive capacity of the distribution network includes the bearing capacity, interactivity, self-healing ability, and efficiency of the distribution network. The comprehensive evaluation index is used to characterize the comprehensive capacity of the distribution network. At the same time, the distributed power source consumption capacity in the comprehensive evaluation index is characterized by the consumption rate, and the consumption rate where P DG is the power generation of the distributed power source, and P tol is the load demand of the distribution network; the power supply capacity in the comprehensive evaluation index is characterized by the voltage deviation, and the voltage deviation where V actual is the actual voltage of the distribution network node, and V c is the rated voltage of the distribution network; the power supply reliability in the comprehensive evaluation index is characterized by the reliability margin index, and the reliability margin index where P b is the spare capacity of the distribution network, and P cl is the critical load demand; the extreme disaster scenario recovery capacity in the comprehensive evaluation index is characterized by the recovery rate, and the recovery rate where L res (t) represents the load restored within time t, and L total represents the total load of the affected area before the disaster. By continuously quantifying the comprehensive capacity of the distribution network at a lower level, it is convenient to evaluate the comprehensive capacity of the distribution network according to actual data, thereby improving the accuracy and objectivity of the evaluation. At the same time, the distribution network is evaluated from multiple dimensions such as the distributed power source consumption capacity, power supply capacity, power supply reliability, and extreme disaster scenario recovery capacity, improving the comprehensiveness of the evaluation. Through comprehensive evaluation, the bottlenecks and weak links in the distribution network can be identified, and then targeted measures can be taken for improvement, which helps to enhance the system performance of the entire distribution network and ensure the reliability and stability of power supply.

[0027] As an implementation method, the evaluation objective of the distribution network can also be adjusted correspondingly according to the distribution network evaluation requirements. At the same time, when the evaluation objective of the distribution network is adjusted, the corresponding comprehensive evaluation index of the distribution network evaluation objective also needs to be adjusted accordingly. If the evaluation objective of the distribution network is to obtain the reliability of the distribution network, the corresponding comprehensive evaluation index is the index characterizing the reliability of the distribution network. By making corresponding adjustments to the distribution network evaluation objective, the flexibility of the distribution network comprehensive evaluation model is improved.

[0028] In this embodiment, the comprehensive evaluation model of the distribution network consists of an objective layer, an index layer, and weights pointing from the index layer to the objective layer. According to the hierarchical relationship between different levels, it can be known that the comprehensive evaluation indexes are located in the index layer, and the distribution network evaluation objectives are located in the objective layer. The comprehensive evaluation indexes are compared pairwise to obtain a judgment matrix. The order of the judgment matrix is the number of comprehensive evaluation indexes, and the preset value is 0.1. When the consistency test fails, the judgment matrix is adjusted according to the magnitude relationship of the original weights between the original weights obtained based on expert opinions. The specific adjustment steps are as follows: If the weight of comprehensive evaluation index A in the original weights is greater than the weight of comprehensive evaluation index B, and the weight of comprehensive evaluation index B is greater than the weight of comprehensive evaluation index C, but in the judgment matrix, the weight of comprehensive evaluation index C is greater than the weight of comprehensive evaluation index A, then the weight of comprehensive evaluation index C in the judgment matrix is adjusted so that the weight of comprehensive evaluation index C is less than that of comprehensive evaluation indexes A and B, thereby ensuring the rationality and scientificity of the weights between the evaluation indexes.

[0029] S2: Obtain the first distance correlation coefficient between the comprehensive evaluation indexes based on the historical data of the distribution network and the original weights.

[0030] The S2 includes: S21: Obtain the theoretical comprehensive evaluation of the distribution network based on the historical data of the distribution network and the original weights, and compare the theoretical comprehensive evaluation with the actual comprehensive evaluation of the distribution network corresponding to the historical data of the distribution network. S22: Obtain the historical data of the distribution network with successful comparison, and obtain the first distance correlation coefficient between the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison.

[0031] In S22, the obtaining the first distance correlation coefficient between the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison includes: Obtain the first distance covariance between the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison, and obtain the first distance standard deviation of the comprehensive evaluation indexes based on the historical data of the distribution network with successful comparison. Obtain the first distance correlation coefficient based on the first distance covariance and the first distance standard deviation.

[0032] In this embodiment, the comprehensive evaluation results of the distribution network are quantified and divided, and each interval represents different evaluation situations of the distribution network. Then, by calculating the product of the historical data of the distribution network and the original weight, the obtained product is compared with the actual comprehensive evaluation corresponding to the historical data of the distribution network, so as to identify the historical data of the distribution network with successful comparison, ensuring the accuracy and stability of the historical data of the comprehensive evaluation index corresponding to the original weight. At the same time, the first distance correlation coefficient is obtained according to the historical data of the distribution network with successful comparison, thereby indirectly improving the accuracy of the first distance correlation coefficient. At the same time, an accurate reference standard is provided for the second distance correlation coefficient, thereby improving the accuracy of the final weight.

[0033] In this embodiment, by calculating the first distance covariance between two comprehensive evaluation indexes and the first distance standard deviation of each comprehensive evaluation index, the first distance correlation coefficient between every two comprehensive evaluation indexes is obtained. For example, if the comprehensive evaluation indexes are A, B, and C, the first distance correlation coefficients between A and B, B and C, and A and C need to be calculated. The calculation formula of the first distance correlation coefficient is: In the formula, dCor(X, Y) represents the first distance correlation coefficient between the comprehensive evaluation indexes X and Y, dCov(X1, Y1) represents the first distance covariance between the comprehensive evaluation indexes X and Y, dVar(X1) represents the first distance standard deviation of the comprehensive evaluation index X, dVar(Y1) represents the first distance standard deviation of the comprehensive evaluation index Y. Among them, X1 represents the historical data of the distribution network corresponding to the comprehensive evaluation index X, and Y1 represents the historical data of the distribution network corresponding to the comprehensive evaluation index Y. By calculating the first distance correlation coefficient between every two comprehensive evaluation indexes through the first distance covariance and the first distance standard deviation, the quantification of the correlation degree between the historical data corresponding to the comprehensive evaluation indexes is realized, and at the same time, the unification of the correlation scale between different comprehensive evaluation indexes is realized, thereby improving the convenience of obtaining the final weight through the distance standard deviation.

[0034] S3: Obtain the second distance correlation coefficient between the comprehensive evaluation indexes based on the real-time data of the distribution network, obtain the final weight of the comprehensive evaluation index based on the first distance correlation coefficient and the second distance correlation coefficient, and optimize the comprehensive evaluation model of the distribution network based on the final weight.

[0035] The S3 includes: S31: Obtain the second distance covariance between the comprehensive evaluation indexes based on the real-time data of the distribution network, and obtain the second distance standard deviation of the comprehensive evaluation indexes based on the real-time data of the distribution network; S32: Obtain the second distance correlation coefficient based on the second distance covariance and the second distance standard deviation, and obtain the final weight of the comprehensive evaluation index based on the first distance correlation coefficient and the second distance correlation coefficient.

[0036] In S32, the obtaining of the final weight of the comprehensive evaluation index based on the first distance correlation coefficient and the second distance correlation coefficient includes: If the absolute difference between the second distance correlation coefficient and the first distance correlation coefficient is greater than a preset difference, then based on the magnitude relationship of the original weights, the absolute difference is used to obtain the final weight; if the absolute difference is less than or equal to the preset difference, then the original weight is the final weight.

[0037] S3 further includes: The optimized comprehensive evaluation model of the distribution network T = k 1 C + k 2 (α 1 η + α 2 ΔV + α 3 M + α 4 R), where k 1 , k 2 are both proportionality coefficients, C represents the distribution network carbon emission rights trading cost considering low-carbon constraints, α 1 is the final weight of the distributed power source consumption capacity, η represents the distributed power source consumption capacity, α 2 is the final weight of the power supply capacity, ΔV represents the power supply capacity, α 3 is the final weight of the power supply reliability, M represents the power supply reliability, α 4 is the final weight of the extreme disaster scenario recovery ability, and R represents the extreme disaster scenario recovery ability.

[0038] In this embodiment, the second distance covariance between the comprehensive evaluation indexes is obtained based on the real-time data of the distribution network, and the second distance standard deviation of the comprehensive evaluation indexes is specifically obtained based on the real-time data of the distribution network as follows: If the real-time data of the distribution network corresponding to the comprehensive evaluation indexes A and B are A1 and B1, then In the formula, dCor(A, B) represents the first distance correlation coefficient between the comprehensive evaluation indexes A and B, dCov(A1, B1) represents the first distance covariance between the comprehensive evaluation indexes A and B, dVar(A1) represents the first distance standard deviation of the comprehensive evaluation index A, dVar(B1) represents the first distance standard deviation of the comprehensive evaluation index B. Calculating the second distance correlation coefficient between every two comprehensive evaluation indexes through the second distance covariance and the second distance standard deviation realizes the quantification of the correlation degree between the real-time data corresponding to the comprehensive evaluation indexes, and at the same time realizes the unification of the correlation scales between different comprehensive evaluation indexes, thereby improving the convenience of obtaining the final weight through the distance standard deviation.

[0039] In this embodiment, the preset difference is set according to the comprehensive evaluation requirements. If it is necessary to improve the comprehensive evaluation efficiency, the preset difference can be set relatively large. If it is necessary to improve the accuracy of the comprehensive evaluation, the preset difference can be set relatively small. Based on the magnitude relationship of the original weights, the final weights are obtained using the absolute difference. Specifically, when using the absolute difference to obtain the final weights, the following conditions need to be met: If the original weight of the comprehensive evaluation index A is greater than the original weight of the comprehensive evaluation index B, the final weight of A in the obtained final weights also needs to be greater than the final weight of B. To meet the above conditions, if the final weights do not meet the above conditions after reducing or increasing the absolute difference, each absolute difference is reduced by a multiple, and then the weights are adjusted based on the original weights, and double-checking is performed, thereby improving the accuracy of the obtained final weights.

[0040] As an implementation manner, when the absolute difference between two comprehensive evaluation indexes is too large, the final weights of the two comprehensive evaluation indexes are not limited by the magnitude relationship of the original weights, thereby further improving the accuracy and flexibility of the comprehensive evaluation.

[0041] Embodiment 2: This embodiment also provides a storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are loaded and executed by a processor, the steps of the optimization method for the comprehensive evaluation model of the distribution network based on the distance correlation coefficient are implemented.

[0042] The above specific implementation manners are the preferred implementation manners of the optimization method for the comprehensive evaluation model of the distribution network based on the distance correlation coefficient of the present invention, and do not limit the specific implementation scope of the present invention. The scope of the present invention includes but is not limited to this specific implementation manner. All equivalent changes made according to the shape and structure of the present invention are within the protection scope of the present invention.

Claims

1. A distribution network comprehensive evaluation model optimization method based on distance correlation coefficient, characterized in that: The following steps are involved: S1: Based on the analytic hierarchy process, a comprehensive evaluation model of the distribution network is constructed, and the original weights of the comprehensive evaluation indicators in the comprehensive evaluation model of the distribution network are obtained; S2: Acquire a first distance correlation coefficient between the comprehensive evaluation indicators based on the distribution network historical data and the original weight; S3: Obtain a second distance correlation coefficient between the comprehensive evaluation indicators based on the real-time data of the distribution network, obtain a final weight of the comprehensive evaluation indicator based on the first distance correlation coefficient and the second distance correlation coefficient, and optimize the comprehensive evaluation model of the distribution network based on the final weight.

2. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 1 is characterized in that: The S1 includes: S11: Decomposing the distribution network evaluation indicators into different levels according to the distribution network evaluation target, and obtaining the distribution network comprehensive evaluation model according to the hierarchical relationship between the different levels, and comparing the comprehensive evaluation indicators in the distribution network comprehensive evaluation model pairwise to obtain a judgment matrix; S12: Perform a consistency check on the judgment matrix based on the characteristic roots and the order of the judgment matrix. If the consistency check passes, obtain the original weights based on the judgment matrix. If the consistency check fails, modify the judgment matrix according to the weight relationship preset based on the expert opinions in the hierarchical analysis method, and execute S12.

3. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 2 is characterized in that: In S11, the judgment matrix satisfies Among them, a ij Represents the element in row i and column j in the judgment matrix, a ji Represents the element in the jth row and ith column of the judgment matrix; The consistency check of the judgment matrix based on the characteristic root and order of the judgment matrix includes: Get the first indicator of consistency Wherein, λ is the characteristic root, and n is the order; Obtaining a second indicator of consistency Wherein, RI is a random consistency index obtained based on the order; If the second indicator is less than the preset value, it indicates that the consistency check has passed; if the second indicator is greater than or equal to the preset value, it indicates that the consistency check has failed.

4. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 1, characterized in that: The S2 includes: S21: obtaining a theoretical comprehensive evaluation of the distribution network based on the historical data of the distribution network and the original weights, and comparing the theoretical comprehensive evaluation with an actual comprehensive evaluation of the distribution network corresponding to the historical data of the distribution network; S22: Acquire the historical data of the distribution network that has been successfully compared, and acquire the first distance correlation coefficient between the comprehensive evaluation indicators based on the historical data of the distribution network that has been successfully compared.

5. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 4 is characterized in that: In S22, the first distance correlation coefficient between the comprehensive evaluation indicators is obtained based on the successfully compared distribution network historical data, including: Based on the successfully compared historical data of the distribution network, a first distance covariance between the comprehensive evaluation indicators is obtained; based on the successfully compared historical data of the distribution network, a first distance standard deviation of the comprehensive evaluation indicators is obtained; The first distance correlation coefficient is acquired based on the first distance covariance and the first distance standard deviation.

6. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 1, characterized in that: The S3 includes: S31: acquiring a second distance covariance between the comprehensive evaluation indicators based on the real-time data of the distribution network, and acquiring a second distance standard deviation of the comprehensive evaluation indicators based on the real-time data of the distribution network; S32: Obtain the second distance correlation coefficient based on the second distance covariance and the second distance standard deviation, and obtain the final weight of the comprehensive evaluation index based on the first distance correlation coefficient and the second distance correlation coefficient.

7. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 6, characterized in that: In S32, the obtaining the final weight of the comprehensive evaluation index based on the first distance correlation coefficient and the second distance correlation coefficient includes: If the absolute difference between the second distance correlation coefficient and the first distance correlation coefficient is greater than the preset difference, then based on the size relationship of the original weights, the absolute difference is used to obtain the final weight; if the absolute difference is less than or equal to the preset difference, the original weight is the final weight.

8. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 1, characterized in that: In S1, the comprehensive evaluation indicators include distributed power supply absorption capacity, power supply capacity, power supply reliability and extreme disaster scenario recovery capability.

9. The method for optimizing the comprehensive evaluation model of distribution network based on distance correlation coefficient according to claim 8, characterized in that: The S3 further includes: The optimized distribution network comprehensive evaluation model T=k1C+k2(α1η+α2ΔV+α3M+α4R), where k1 and k2 are both proportional coefficients, C represents the carbon emission rights trading cost of the distribution network considering low-carbon constraints, α1 is the final weight of the distributed power supply absorption capacity, η represents the distributed power supply absorption capacity, α2 is the final weight of the power supply capacity, ΔV represents the power supply capacity, α3 is the final weight of the power supply reliability, M represents the power supply reliability, α4 is the final weight of the extreme disaster scenario recovery capability, and R represents the extreme disaster scenario recovery capability.

10. A storage medium, characterized in that: The storage medium stores computer executable instructions, and when the computer executable instructions are loaded and executed by the processor, the steps of the method for optimizing the comprehensive evaluation model of the distribution network based on the distance correlation coefficient as described in any one of claims 1 to 9 are implemented.

Images