Evaluation Methods and Related Equipment for the Development Level of Power Grids in Coastal Cities

Abstract

This application discloses a method and related equipment for evaluating the development level of coastal city power grids, relating to the field of power system operation analysis technology. The method includes: acquiring operational indicator data of the coastal city power grid to be evaluated; determining the optimal combination weight of each operational indicator data; calculating a first evaluation score and a second evaluation score of the operational indicator data based on the optimal combination weight, using a first evaluation method and a second evaluation method respectively; dynamically allocating fusion weights according to a first degree of dispersion of the first evaluation score and a second degree of dispersion of the second evaluation score; and weightedly fusing the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid. This application improves the accuracy of the evaluation results of the development level of coastal city power grids.

Description

Technical Field

[0001] This application relates to the field of power system operation analysis technology, and in particular to methods and related equipment for evaluating the development level of power grids in coastal cities. Background Technology

[0002] The safe and stable operation of power grids in coastal cities is directly related to the normal conduct of urban economic and social activities. Given the frequent occurrence of extreme weather events such as typhoons, coastal city power grids must possess robust infrastructure to withstand natural disasters like strong winds and heavy rains, while also meeting the green transformation needs of large-scale renewable energy integration and consumption. Conducting a multi-dimensional comprehensive evaluation of the development level of coastal city power grids is a fundamental task in guiding the grid to achieve coordinated development of safe operation and green, low-carbon practices.

[0003] When conventional technologies evaluate the development level of power grids, the indicator system often starts from traditional dimensions such as power supply reliability and economy, without incorporating key dimensions such as disaster resilience and fault recovery to address the special damage mechanism of typhoons on coastal power grids. This makes it difficult to accurately reflect the defense needs under extreme conditions. The weighting process relies on a single objective weighting method or subjective experience, and the weight allocation is not stable enough under high-dimensional and heterogeneous data. The comprehensive evaluation process generally implies the assumption that decision-makers are completely rational, ignores nonlinear psychological perceptions such as loss avoidance, and has a single evaluation dimension, which is easily affected by differences in indicator scales, resulting in poor accuracy of the evaluation results of the power grid development level of coastal cities. Summary of the Invention

[0004] The main purpose of this application is to provide a method and related equipment for evaluating the development level of power grids in coastal cities, aiming to solve the technical problem of poor accuracy in the evaluation results of the development level of power grids in coastal cities.

[0005] To achieve the above objectives, this application proposes a method for evaluating the development level of power grids in coastal cities, the method comprising: Obtain the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data; Based on the optimal combination weights, the first evaluation score and the second evaluation score of the operational indicator data are calculated using the first evaluation method and the second evaluation method, respectively. The fusion weights are dynamically allocated based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score. The first evaluation score and the second evaluation score are then weighted and fused based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

[0006] In one embodiment, the step of obtaining the operational indicator data of the coastal city power grid to be evaluated and determining the optimal combination weight of each operational indicator data in the operational indicator data includes: An objective weighting method for obtaining a preset number of values, wherein the objective weighting method includes the backpropagation neural network weighting method, the weight determination method based on index correlation, and the entropy weighting method; Calculate the initial weights of each operational indicator data according to the objective weighting methods described above; The initial weights are combined according to the principle of minimizing deviation to generate the optimal combined weights.

[0007] In one embodiment, the step of calculating the initial weights of each operational indicator data according to each of the objective weighting methods includes: When the objective weighting method is the backpropagation neural network weighting method, the connection weights from the input layer to the hidden layer in the trained backpropagation neural network are obtained. Divide the sum of the absolute values ​​of the connection weights of each running indicator data to all hidden neurons by the sum of the absolute values ​​of the connection weights of all running indicator data to all hidden neurons, and use the quotient as the initial weights of the backpropagation neural network for the current running indicator data. When the objective weighting method is a weight determination method based on the correlation of indicators, the standard deviation of each operating indicator data and the correlation coefficient between each operating indicator data and other operating indicator data are calculated. Based on the standard deviation of each operational indicator data and its correlation with all other operational indicator data, a comprehensive measure of the current operational indicator data is determined; The proportion of the comprehensive measure of each operational indicator data to the sum of the comprehensive measures of all operational indicator data is used as the initial weight of the current operational indicator data based on the weight determination method of indicator correlation. When the objective weighting method is the entropy weighting method, the information entropy of each operational indicator data is calculated. The degree of dispersion of the current operational indicator data is determined based on the information entropy of each operational indicator data. The proportion of the dispersion of each operational indicator data to the sum of the dispersion of all operational indicator data is used as the initial weight of the current operational indicator data using the entropy weight method.

[0008] In one embodiment, the step of combining the initial weights according to the principle of minimizing deviation to generate the optimal combined weights includes: The initial weights of the backpropagation neural network for each of the aforementioned operational indicator data, the initial weights of the weight determination method based on indicator correlation, and the initial weights of the entropy weight method are combined into a weight matrix. Calculate the covariance matrix of the weight matrix, and solve for the combination coefficients that minimize the deviation of the weight vector based on the covariance matrix and the all-one vector, thereby generating the optimal combination weight.

[0009] In one embodiment, before the step of calculating the first evaluation score and the second evaluation score of the operational indicator data based on the optimal combination weights using the first evaluation method and the second evaluation method respectively, the method further includes: The operational indicator data are converted into psychologically perceived prospect values ​​based on the preset improved cumulative prospect theory. The operational indicator data is updated based on the psychologically perceived prospective value.

[0010] In one embodiment, the step of converting the operational indicator data into psychologically perceived prospect values ​​according to the improved cumulative prospect theory includes: Define the reference point, gain-sensitive parameter, loss-sensitive parameter, and loss-aversion parameter in the improved cumulative prospect theory; When the operational indicator data is greater than or equal to the reference point, the power of the profit-sensitive parameter of the difference between the operational indicator data and the reference point is used as the psychologically perceived prospect value. When the operational indicator data is less than the reference point, the loss sensitivity parameter of the difference between the reference point and the operational indicator data is raised to the power of the loss aversion parameter, and the negative value is taken as the psychologically perceived prospect value.

[0011] In one embodiment, the first evaluation method is grey relational analysis, and the second evaluation method is an approximation of the ideal solution ranking method based on the similarity of the included angle. The step of calculating the first evaluation score and the second evaluation score of the operating indicator data using the first evaluation method and the second evaluation method respectively includes: The grey relational analysis evaluation score of the operational indicator data is calculated based on the grey relational analysis method and used as the first evaluation score. The approximation ideal solution ranking method evaluation score of the running index data is calculated based on the approximation ideal solution ranking method based on the angle similarity, and the second evaluation score is obtained.

[0012] In one embodiment, the step of calculating the approximation ideal solution ranking method evaluation score of the operating index data based on the approximation ideal solution ranking method based on the angle similarity, and obtaining the second evaluation score, includes: Obtain the predetermined positive and negative ideal solution vectors; The similarity between the vector corresponding to the operational index data and the positive ideal solution vector is calculated using cosine similarity to obtain the first similarity. The second similarity is obtained by calculating the similarity between the vector corresponding to the operational index data and the negative ideal solution vector using cosine similarity. Based on the first similarity and the second similarity, the evaluation score of the approximation ideal solution ranking method is calculated to obtain the second evaluation score.

[0013] In one embodiment, after the step of obtaining the operational indicator data of the coastal city power grid to be evaluated, and before the step of determining the optimal combination weight of each operational indicator data in the operational indicator data, the method further includes: The operational indicator data is dimensionless according to preset standardization rules to obtain standardized operational indicator data. The operational indicator data is updated based on the standardized operational indicator data.

[0014] In one embodiment, the preset standardization rules include a positive index extreme value standardization formula and a negative index extreme value standardization formula; The formula for standardizing the extreme values ​​of the positive index is as follows: ; The formula for standardizing the extreme values ​​of the negative index is: ; in, For standardized values, The original value, and The first The minimum and maximum values ​​of each indicator. This is a preset, small constant.

[0015] In one embodiment, the first degree of dispersion is the first variance of the first evaluation score; the second degree of dispersion is the second variance of the second evaluation score. The step of dynamically allocating fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and weightedly fusing the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid includes: Calculate the first variance of the first evaluation score for all coastal city power grids to be evaluated; Calculate the second variance of the second evaluation score for all coastal city power grids to be evaluated; The proportion of the second variance to the sum of the first variance and the second variance is used as the fusion weight of the first evaluation score; The proportion of the first variance to the sum of the first variance and the second variance is used as the fusion weight of the second evaluation score; Multiply the first evaluation score by the fusion weight of the first evaluation score to obtain the first weighted score; Multiply the second evaluation score by the fusion weight of the second evaluation score to obtain the second weighted score; The first weighted score and the second weighted score are added together, and the sum is used as the evaluation result of the development level of the power grid in the coastal city.

[0016] Furthermore, to achieve the above objectives, this application also proposes a device for evaluating the development level of power grids in coastal cities, the device comprising: The acquisition module is used to acquire the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data. The calculation module is used to calculate the first evaluation score and the second evaluation score of the operation indicator data based on the optimal combination weights, respectively, through the first evaluation method and the second evaluation method. The evaluation module is used to dynamically allocate fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and to perform weighted fusion of the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

[0017] In addition, to achieve the above objectives, this application also proposes a device for evaluating the development level of coastal city power grids, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the coastal city power grid development level evaluation method as described above.

[0018] In addition, to achieve the above objectives, this application also proposes a storage medium, which is a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, it implements the steps of the coastal city power grid development level evaluation method described above.

[0019] In addition, to achieve the above objectives, this application also provides a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the coastal city power grid development level evaluation method described above.

[0020] One or more technical solutions proposed in this application have at least the following technical effects: Compared with conventional techniques that use a single objective weighting method, resulting in insufficient weight stability and neglecting the psychological perception factors of decision-makers, leading to a disconnect between evaluation and actual decision-making logic, this application obtains the operational indicator data of the coastal city power grid to be evaluated, determines the optimal combination weight of each operational indicator data in the operational indicator data; based on the optimal combination weight, calculates the first evaluation score and the second evaluation score of the operational indicator data through a first evaluation method and a second evaluation method, respectively; dynamically allocates fusion weights according to the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and performs weighted fusion of the first evaluation score and the second evaluation score based on the fusion weight to obtain the evaluation result of the development level of the coastal city power grid. Understandably, this application employs a method of determining the optimal combination weights. When the weights of each operational indicator data are assigned, the overall weight allocation is optimized through the optimal combination weights, thereby obtaining a more reasonable indicator weight configuration. Simultaneously, it adopts a dual-evaluation method for parallel evaluation and dynamic fusion based on the degree of dispersion. After the dispersion of the first and second evaluation scores is obtained, the level of dispersion reflects the stability of the scores from each evaluation method, and the fusion weights are dynamically allocated accordingly. This allows evaluation methods with higher stability to receive greater fusion weights, thereby achieving automatic adjustment of the credibility of different evaluation methods. Therefore, based on the dual mechanism of optimal combination weights and adaptive fusion of dispersion, a multi-dimensional comprehensive judgment and adaptive optimization of the evaluation results of the development level of coastal city power grids can be achieved, leading to a more accurate comprehensive evaluation ranking and ultimately improving the accuracy of the evaluation results of the development level of coastal city power grids. Attached Figure Description

[0021] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application.

[0022] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 A flowchart illustrating an embodiment of the evaluation method for the development level of power grids in coastal cities in this application; Figure 2 A flowchart illustrating the method for evaluating the development level of power grids in coastal cities, as provided in this application. Figure 3 Weighting diagram of each indicator used in the evaluation method for the development level of power grids in coastal cities in this application; Figure 4 A comprehensive evaluation map of the development level of power grids in various cities provided for the evaluation method of power grid development level in coastal cities in this application; Figure 5 A schematic diagram illustrating the comprehensive ranking of cities under different decision-making psychologies, provided for the evaluation method of power grid development level in coastal cities in this application; Figure 6 This is a schematic diagram of the module structure of the coastal city power grid development level evaluation device according to an embodiment of this application; Figure 7 This is a schematic diagram of the equipment structure of the hardware operating environment involved in the evaluation method for the development level of the power grid in coastal cities in this application embodiment.

[0024] The purpose, features, and advantages of this application will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0025] It should be understood that the specific embodiments described herein are merely illustrative of the technical solutions of this application and are not intended to limit this application.

[0026] To better understand the technical solution of this application, a detailed description will be provided below in conjunction with the accompanying drawings and specific implementation methods.

[0027] The main solution in this application's embodiments is: Obtain the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data; Based on the optimal combination weights, the first evaluation score and the second evaluation score of the operational indicator data are calculated using the first evaluation method and the second evaluation method, respectively. The fusion weights are dynamically allocated based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score. The first evaluation score and the second evaluation score are then weighted and fused based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

[0028] In this embodiment, the application uses a coastal city power grid development level evaluation device as the execution subject. For ease of description, it will be referred to as "device" in the following detailed description.

[0029] When using conventional technologies to evaluate the development level of power grids in coastal cities, the indicator system lacks a specific characterization of typhoon disaster prevention and green and low-carbon transformation. The weighting method in the weighting stage is too simple and the weight stability is insufficient. The comprehensive evaluation stage ignores the nonlinear psychological perception of decision-makers and has a single evaluation dimension, resulting in poor accuracy of the evaluation results.

[0030] This application provides a solution aimed at improving the accuracy of evaluation results for the development level of power grids in coastal cities. The research and development background addressed in this embodiment is the need to assess the safe operation and renewable energy absorption capacity of power grids in typhoon-prone coastal cities under extreme weather conditions. The application scenarios addressed in this embodiment are differentiated planning, resilience enhancement, and investment strategy optimization for coastal city power grids.

[0031] In this embodiment, a multi-dimensional evaluation index system covering power supply reliability, grid intelligence, infrastructure level, disaster resilience, and green and low-carbon level is constructed to ensure that the evaluation dimensions accurately correspond to the disaster prevention needs and low-carbon transformation requirements of the coastal power grid. By integrating three objective weighting methods—backpropagation neural network weighting, index correlation-based weight determination, and entropy weighting—and generating the optimal combined weights based on the principle of minimizing deviation, the advantages of different weighting methods are effectively combined, avoiding weight distortion caused by model assumptions or data sensitivity in single weighting methods. By introducing an improved cumulative prospect theory, standardized index data is transformed into psychologically perceived prospect values ​​reflecting decision-makers' risk attitudes and loss aversion, allowing the evaluation input to break through the assumption of complete rationality and more closely align with actual decision-making logic. Evaluation scores are calculated using grey relational analysis and an approximation of ideal solutions based on angle similarity, and the fusion weights are dynamically allocated based on the dispersion of the score sequence, allowing evaluation methods with high score stability to obtain greater fusion weights, achieving adaptive weighted fusion, and eliminating the one-sidedness of single evaluation methods. Based on the above mechanisms, this application achieves an accurate evaluation of the development level of coastal urban power grids, solving the technical problem of poor accuracy in conventional technical evaluation results.

[0032] Based on this, the embodiments of this application provide a method for evaluating the development level of power grids in coastal cities, referring to... Figure 1 , Figure 1 This is a flowchart illustrating the first embodiment of the evaluation method for the development level of power grids in coastal cities according to this application.

[0033] In this embodiment, the evaluation method for the development level of the power grid in coastal cities includes steps S10 to S30: Step S10: Obtain the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data. It should be noted that the coastal city power grid to be evaluated refers to the urban power network system located in coastal areas that requires a development level assessment, including but not limited to the distribution and transmission networks under the jurisdiction of coastal administrative regions of prefecture-level cities or municipalities directly under the central government. Operational indicator data refers to the measured or statistical values ​​of various quantitative indicators used to measure the development level of the power grid, including but not limited to power supply reliability indicators, power grid intelligence indicators, infrastructure level indicators, disaster resilience indicators, and green and low-carbon level indicators. Optimal combined weights refer to the weight values ​​of each operational indicator data obtained by integrating multiple objective weighting methods and optimizing the combination of weights obtained from each method, minimizing the overall deviation between the combined weight vector and the weight vectors of each individual method.

[0034] Understandably, this step provides a comprehensive data foundation for subsequent evaluation by acquiring multi-dimensional operational indicator data of the coastal city power grid to be evaluated; by determining the optimal combination of weights, it overcomes the weight bias caused by model assumptions or data sensitivity in a single objective weighting method, making the weight allocation of each operational indicator data more scientific and reasonable, thereby improving the accuracy and robustness of the overall evaluation.

[0035] Step S20: Based on the optimal combination weights, calculate the first evaluation score and the second evaluation score of the operation indicator data using the first evaluation method and the second evaluation method, respectively. It should be noted that the first evaluation method refers to a comprehensive evaluation based on the geometric similarity of the indicator numerical sequence, and in this embodiment, grey relational analysis is preferred. The second evaluation method refers to a comprehensive evaluation based on the similarity of the angle between the evaluation object and the ideal solution, and in this embodiment, the approximation of the ideal solution ranking method based on the angle similarity is preferred. The first evaluation score refers to the numerical value calculated using the first evaluation method that reflects the quality of the development level of the power grid in the coastal city to be evaluated. The second evaluation score refers to the numerical value calculated using the second evaluation method that reflects the quality of the development level of the power grid in the coastal city to be evaluated.

[0036] Understandably, this step weights the operational indicator data using the optimal combination of weights and calculates scores using two different evaluation methods. This approach comprehensively characterizes the power grid development level from two dimensions: the geometric similarity of the indicator numerical sequence and the similarity of the angle between the ideal solution vectors. It avoids the one-sidedness caused by the limited evaluation perspective of a single evaluation method and provides multi-dimensional basic score data for subsequent integrated evaluation.

[0037] Step S30: Dynamically allocate fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and perform weighted fusion of the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

[0038] It should be noted that the first degree of dispersion refers to the degree of dispersion or fluctuation in the numerical distribution of the first evaluation scores of all coastal city power grids to be evaluated. In this embodiment, the variance of the first evaluation scores is preferably used as a measure. The second degree of dispersion refers to the degree of dispersion or fluctuation in the numerical distribution of the second evaluation scores of all coastal city power grids to be evaluated. In this embodiment, the variance of the second evaluation scores is preferably used as a measure. The fusion weight refers to the weight coefficient assigned to the first evaluation score and the second evaluation score when weighted and fused. This weight coefficient is dynamically determined according to the relationship between the magnitude of the first degree of dispersion and the second degree of dispersion. Weighted fusion refers to the calculation process of multiplying the first evaluation score and the second evaluation score by their respective fusion weights and then adding them together. The development level evaluation result refers to the final score or ranking obtained after a comprehensive quantitative assessment of the development level of the coastal city power grids to be evaluated. It is used to reflect the comprehensive development level of each coastal city power grid in terms of power supply capacity, intelligence level, infrastructure status, disaster resilience, and green and low-carbon development.

[0039] Understandably, this step, by calculating the dispersion of the first and second evaluation scores, can automatically identify the score stability of each evaluation method on different coastal city power grid data. By converting the dispersion into fusion weights, evaluation methods with high score stability receive larger fusion weights, while evaluation methods with low score stability receive smaller fusion weights. This achieves adaptive evaluation and dynamic weighting of the credibility of the two evaluation methods, thereby avoiding the one-sidedness of simple average fusion and further improving the accuracy and robustness of the comprehensive evaluation results.

[0040] This embodiment provides a method for evaluating the development level of power grids in coastal cities. This application employs a method of determining the optimal combination weights. When weights are assigned to each operational indicator data, the overall weight allocation is optimized through the optimal combination weights, thereby obtaining a more reasonable indicator weight configuration. Simultaneously, a dual-evaluation method is used for parallel evaluation and dynamic fusion based on the degree of dispersion. After the dispersion of the first and second evaluation scores is obtained, the level of dispersion reflects the stability of the scores from each evaluation method, and the fusion weight is dynamically allocated accordingly. This allows evaluation methods with higher stability to receive greater fusion weights, thereby achieving automatic adjustment of the credibility of different evaluation methods. Therefore, based on the dual mechanism of optimal combination weights and adaptive fusion of dispersion, a multi-dimensional comprehensive judgment and adaptive optimization of the evaluation results of the development level of power grids in coastal cities can be achieved, thereby determining a more accurate comprehensive evaluation ranking and ultimately improving the accuracy of the evaluation results of the development level of power grids in coastal cities.

[0041] For example, refer to Figure 2 This application proposes a comprehensive evaluation method for the development level of power grids in coastal cities, including: Step S101: Construct an evaluation index system for the development level of power grids in coastal cities, and standardize the collected index data.

[0042] Specifically, an evaluation index system was first constructed, encompassing five dimensions: power supply reliability, grid intelligence, infrastructure level, disaster resilience, and green and low-carbon level, covering a total of 20 secondary indicators. After collecting raw operational data of the coastal city power grid under each of the above indicators, the raw operational data was standardized to eliminate scale differences between indicators of different dimensions, resulting in dimensionless standardized index data. The standardized data values ​​are located within the range [0,1], serving as the input basis for subsequent steps.

[0043] Further, in step S102, an objective combination weighting model based on the principle of minimizing deviation is constructed, and the optimal combination weight of each indicator is calculated.

[0044] Specifically, by integrating three objective weighting methods—backpropagation neural network, weight determination method based on indicator correlation, and entropy weight method—an objective combined weighting model based on the principle of deviation minimization is constructed to calculate the optimal combined weights of each operational indicator data.

[0045] Specifically, using the standardized index data output in step S101 as input, the initial weights of each operational index data are calculated using the backpropagation neural network weighting method, the weight determination method based on index correlation, and the entropy weighting method, respectively. Then, based on the principle of minimizing deviation, the optimal combined weights of each operational index data are determined by solving the covariance matrix to minimize the deviation between the weight vectors obtained by each individual weighting method. This optimal combined weight integrates the advantages of the three objective weighting methods and avoids weight bias caused by model assumptions or data sensitivity of individual weighting methods.

[0046] Further, in step S103, the ICPT-GRA and ICPT-TOPSIS evaluation models are constructed to calculate the correlation and similarity scores.

[0047] Specifically, based on the cumulative prospect theory, the perceived prospect value is calculated, and combined with grey relational analysis and the approximation ideal solution ranking method, ICPT-GRA and ICPT-TOPSIS evaluation models are constructed to calculate the correlation and similarity scores.

[0048] Specifically, based on the standardized index data output in step S101, the Improved Cumulative Prospect Theory (ICPT) is introduced. Reference points, gain-sensitive parameters, loss-sensitive parameters, and loss-averse parameters are set to convert the standardized index data into psychologically perceived prospect values ​​for decision-makers. Then, the ICPT-GRA evaluation model and the ICPT-TOPSIS evaluation model are constructed using Grey Relational Analysis (GRA) and Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS), respectively. Based on the optimal combination weights determined in step S102, the grey relational degree and cosine similarity between the prospect value sequence of each evaluated object and the positive and negative ideal solutions are calculated, thereby obtaining the evaluation scores of each evaluated object under the two evaluation models.

[0049] Further, in step S104, the evaluation scores are weighted and fused to obtain the final comprehensive evaluation result.

[0050] Specifically, dynamic fusion weights are calculated based on the concept of variance, and the evaluation scores of ICPT-GRA and ICPT-TOPSIS are weighted and fused to obtain the final comprehensive evaluation result.

[0051] Specifically, using the two sets of evaluation scores for each evaluated object output in step S103 under the ICPT-GRA and ICPT-TOPSIS evaluation models as input, the variance of the two sets of evaluation scores is calculated separately. Then, the fusion weights are dynamically allocated according to the magnitude of the variance: the evaluation model score with a larger variance is assigned a smaller weight, and the evaluation model score with a smaller variance is assigned a larger weight. The two sets of evaluation scores for each evaluated object are weighted and fused according to the fusion weights, and the result is the final comprehensive evaluation score of the power grid development level of each coastal city. The fusion weights are adaptively determined based on the set of evaluated objects, without the need for manual pre-setting, ensuring the objectivity and robustness of the fusion result.

[0052] In one feasible implementation, the step of obtaining the operational indicator data of the coastal city power grid to be evaluated and determining the optimal combination weight of each operational indicator data includes: An objective weighting method for obtaining a preset number of values, wherein the objective weighting method includes the backpropagation neural network weighting method, the weight determination method based on index correlation, and the entropy weighting method; Calculate the initial weights of each operational indicator data according to the objective weighting methods described above; The initial weights are combined according to the principle of minimizing deviation to generate the optimal combined weights.

[0053] It should be noted that the preset number of objective weighting methods refers to a collection of pre-selected methods that calculate indicator weights based on the characteristics of the data itself without relying on expert subjective judgment. In this embodiment, three methods are preferred. The backpropagation neural network weighting method uses the connection weights from the input layer to the hidden layer in a trained backpropagation neural network to determine the indicator weights by calculating the mean of the absolute values ​​of the connection weights corresponding to each indicator. The weighting method based on indicator correlation considers both the degree of variation of the indicator itself and the conflict between indicators. The degree of variation is measured by the standard deviation, and the conflict is measured by the correlation coefficient. The entropy weighting method is based on the principle of information entropy, determining the weights by measuring the dispersion of indicator data. The greater the dispersion of indicator data, the smaller the information entropy, and the greater the weight assigned to that indicator. The initial weights refer to the uncombined independent weight values ​​calculated separately by each individual objective weighting method for each running indicator data. The deviation minimization principle refers to the optimization principle of solving the combination coefficients through mathematical programming to minimize the overall distance between the final combined weight vector and the weight vectors of each individual weighting method.

[0054] It is understandable that this implementation method integrates the backpropagation neural network weighting method's ability to fit nonlinear relationships, the weight determination method based on index correlation's emphasis on index conflict, and the entropy weighting method's consideration of data dispersion. These three methods complement each other in terms of weighting mechanism. By solving for the optimal combination weights through the principle of minimizing deviation, the final weights can comprehensively reflect the inherent laws of the data, the correlation structure between indicators, and the amount of information in the indicators. This effectively overcomes the weight distortion problem caused by model assumptions or data sensitivity in single objective weighting methods, improves the stability and reliability of weight allocation when dealing with high-dimensional and heterogeneous coastal power grid operation data, and provides a more scientific and reasonable weight basis for the subsequent calculation of evaluation scores.

[0055] In one feasible implementation, the step of calculating the initial weights of each operational indicator data according to each of the objective weighting methods includes: When the objective weighting method is the backpropagation neural network weighting method, the connection weights from the input layer to the hidden layer in the trained backpropagation neural network are obtained. Divide the sum of the absolute values ​​of the connection weights of each running indicator data to all hidden neurons by the sum of the absolute values ​​of the connection weights of all running indicator data to all hidden neurons, and use the quotient as the initial weights of the backpropagation neural network for the current running indicator data. When the objective weighting method is a weight determination method based on the correlation of indicators, the standard deviation of each operating indicator data and the correlation coefficient between each operating indicator data and other operating indicator data are calculated. Based on the standard deviation of each operational indicator data and its correlation with all other operational indicator data, a comprehensive measure of the current operational indicator data is determined; The proportion of the comprehensive measure of each operational indicator data to the sum of the comprehensive measures of all operational indicator data is used as the initial weight of the current operational indicator data based on the weight determination method of indicator correlation. When the objective weighting method is the entropy weighting method, the information entropy of each operational indicator data is calculated. The degree of dispersion of the current operational indicator data is determined based on the information entropy of each operational indicator data. The proportion of the dispersion of each operational indicator data to the sum of the dispersion of all operational indicator data is used as the initial weight of the current operational indicator data using the entropy weight method.

[0056] It should be noted that a trained backpropagation neural network refers to a backpropagation neural network model that has completed learning and fixed the connection weights between neurons in each layer of the network structure. This network includes, but is not limited to, an input layer, at least one hidden layer, and an output layer. The connection weights from the input layer to the hidden layer refer to the connection strength parameters between each neuron in the input layer and each neuron in the hidden layer of the backpropagation neural network. Their absolute values ​​reflect the contribution of the input metric to the feature extraction of the hidden layer. All hidden neurons refer to all neuron nodes contained in the hidden layer. The currently running metric data refers to the running metric data for which weights are being calculated. The initial weights of the backpropagation neural network refer to the uncombined weight values ​​obtained by the backpropagation neural network weighting method using a single weighting method.

[0057] Standard deviation measures the dispersion of operational indicator data across all coastal city power grids being evaluated; a larger standard deviation indicates more significant data variability. Correlation coefficient measures the linear correlation between two operational indicator data points, ranging from -1 to +1; a value closer to one indicates a stronger correlation. Correlation with all other operational indicator data reflects the combined correlation coefficients between the current operational indicator data and every other operational indicator data point; a higher correlation indicates lower conflict with other indicators. A comprehensive measure combines the indicator's own variability with inter-indicator conflict to reflect the amount of information contained in the indicator. Initial weights in the correlation-based weighting method refer to the uncombined weight values ​​obtained from the single weighting method calculated using the correlation-based weighting method.

[0058] Information entropy is a measure of the amount of information contained in operational indicator data. A lower information entropy indicates greater dispersion and a larger amount of information contained in the data. Dispersion refers to the uneven distribution of operational indicator data across all coastal city power grids being evaluated. Greater dispersion indicates a stronger ability of the indicator to distinguish between different evaluation objects. The initial weights in the entropy weighting method refer to the uncombined weight values ​​obtained from the single weighting method calculated using this method.

[0059] Understandably, this implementation describes the specific process of calculating initial weights using three objective weighting methods. The backpropagation neural network weighting method reflects the contribution of the indicator to the output by extracting network connection weights; the weight determination method based on indicator correlation considers both indicator variability and conflict by combining standard deviation and correlation coefficient; and the entropy weighting method measures the distinguishing ability of the indicator through information entropy. These three methods quantify the importance of the indicators from different perspectives, such as nonlinear fitting contribution, the strength of comparison between indicators, and the amount of information in the indicator data. This provides diversified sources of initial weights for subsequent deviation minimization combined weighting, enabling the combined weights to integrate the advantages of different weighting mechanisms.

[0060] For example, a back propagation neural network (BP neural network) is a multi-layer feedforward network trained using an error backpropagation algorithm. This application utilizes the connection weights from the input layer to the hidden layers in the trained BP neural network to calculate the objective weights of various performance metrics.

[0061] Specifically, by utilizing the connection weights from the input layer to the hidden layer in the trained network, the first... The weight of each indicator is obtained by the average absolute value of the weights connected to all hidden neurons: ; In the formula, The first one calculated by the BP neural network The objective weights of each evaluation indicator; For the first The input metrics to the first The connection weights of each hidden neuron; This represents the number of neurons in the hidden layer. This represents the total number of evaluation indicators.

[0062] In the formula, the numerator is the sum of the absolute values ​​of the connection weights connecting the j-th operational index data to all hidden neurons, and the denominator is the sum of the absolute values ​​of the connection weights connecting all operational index data to all hidden neurons. The resulting quotient is the initial weight of the backpropagation neural network for that operational index data.

[0063] As an alternative, the connection weights from the output layer to the hidden layer of the BP neural network can be used for calculation, or the sum of the absolute values ​​of the weights between the input layer and the hidden layer can be nonlinearly transformed (e.g., squared and normalized) and used as the weights. The choice can be made flexibly according to the actual network structure.

[0064] The CRITIC (Criteria Importance Through Intercriteria Correlation) method comprehensively measures the objective weight of indicators by considering their standard deviation and the degree of conflict between them. The weight of the j-th running indicator is calculated based on its standard deviation and the degree of conflict. ; in, The first number calculated by the CRITIC method The objective weights of each evaluation indicator; For the first Standard deviation of each indicator; For the first The first indicator and the first The correlation coefficients of the indicators.

[0065] In the formula, the molecule This is the comprehensive measure of the j-th operational indicator data, with the denominator being the sum of the comprehensive measures of all operational indicator data. The resulting ratio is the initial CRITIC weight of that operational indicator data.

[0066] As an alternative, Spearman's rank correlation coefficient or Kendall's rank correlation coefficient can be used instead of Pearson's correlation coefficient to adapt to non-normally distributed data scenarios.

[0067] Entropy weighting method uses information entropy to measure the dispersion of performance indicator data. For a dataset with a total of m evaluated objects and n total performance indicator data, the weight of the i-th evaluated object under the j-th performance indicator data is first calculated. : ; Then calculate the information entropy of the j-th performance indicator data. : ; Finally, the entropy weight objective weight of the j-th performance indicator data is calculated. : ; In the formula, For the first Information entropy of each evaluation indicator; The total number of objects being evaluated; For the first The evaluated object was in the first The proportion of each indicator; For the first The evaluated object was in the first Standardized values ​​for each indicator; The first one calculated by the entropy weight method Objective weights of each evaluation indicator.

[0068] In the formula, This reflects the dispersion of the j-th operational indicator data. The greater the dispersion, the smaller the information entropy, and the greater the weight. The resulting ratio is the initial weight of the operational indicator data using the entropy weight method.

[0069] As an alternative, it can be calculated Add very small positive numbers to avoid ln If the value becomes infinite, an improved entropy weighting method can be used, such as replacing the proportion calculation with standardized Euclidean distance, which can be adjusted appropriately according to the data quality.

[0070] In one feasible implementation, the step of combining the initial weights according to the principle of minimizing deviation to generate the optimal combined weights includes: The initial weights of the backpropagation neural network for each of the aforementioned operational indicator data, the initial weights of the weight determination method based on indicator correlation, and the initial weights of the entropy weight method are combined into a weight matrix. Calculate the covariance matrix of the weight matrix, and solve for the combination coefficients that minimize the deviation of the weight vector based on the covariance matrix and the all-one vector, thereby generating the optimal combination weight.

[0071] It should be noted that the weight matrix refers to the matrix formed by arranging the initial weights of each operational indicator data calculated by the three objective weighting methods column-wise. Each column corresponds to the weight vector obtained by one weighting method, and each row corresponds to the weight value of an operational indicator data under each of the three methods. The covariance matrix is ​​a matrix that reflects the pairwise correlation and variability between the weight vectors in each column of the weight matrix, and is used to measure the consistency and difference of the weight results obtained by different weighting methods. The all-one vector refers to a column vector in which all elements are one, and is used as a constraint condition to solve for the combination coefficient in this step. The combination coefficient refers to the weighting coefficient assigned to the weight vector of each individual weighting method, reflecting the contribution ratio of each weighting method in the final combined weight. The optimal combined weight refers to the final weight value of each operational indicator data that minimizes the overall deviation between the combined weight vector and the weight vectors of each individual weighting method, obtained by solving the deviation minimization principle.

[0072] Understandably, this implementation method uses the covariance matrix to characterize the intrinsic correlation and differences between weights obtained by different weighting methods, so that the solution of the combination coefficient can automatically consider the information overlap and complementarity of the weight results of each method; by solving the combination coefficient through the principle of minimizing deviation, the overall distance between the final combined weight and the weight vector of each individual weighting method is minimized, thereby maximizing the extraction of consensus of multiple weighting methods, avoiding the deviation caused by simple averaging or subjective allocation of combination coefficients, and enabling the optimal combined weight to robustly integrate the advantages of different weighting mechanisms, thereby improving the scientificity and credibility of weight allocation.

[0073] For example, refer to Figure 3 The weight distribution diagram of each operational indicator data obtained by the combined weighting method of this application shows that the combined weight integrates the information of three individual weights, avoiding the influence of extreme values. The evaluation index system for the development level of coastal city power grids constructed in this embodiment includes 5 primary indicators (power supply reliability, power grid intelligence, infrastructure level, disaster resilience, and green and low-carbon level) and 20 secondary indicators. Each secondary indicator is related to... Figure 3 The correspondence between the indicators C1-C20 is as follows: Power Supply Reliability Dimension: C1: Power Supply Reliability Rate (RS-3); C2: Average Power Outage Time per User; C3: Average Number of Power Outages per User; C4: Distribution Network Interconnection Rate. Power Grid Intelligence Dimension: C5: Distribution Automation Coverage Rate; C6: Smart Meter Coverage Rate; C7: Power Grid Dispatch Automation Level; C8: Information System Coverage Rate. Infrastructure Level Dimension: C9: Substation N-1 Completion Rate; C10: Line Insulation Rate; C11: Cable Coverage Rate; C12: Average Distribution Transformer Capacity. Disaster Resilience Dimension: C13: Proportion of Wind-Reinforced Lines; C14: Backup Power Configuration Rate for Important Users; C15: Power Grid Fault Self-Healing Rate; C16: Post-Disaster Power Restoration Time. Green and Low-Carbon Level Dimension: C17: New Energy Power Generation Consumption Rate; C18: Clean Energy Installed Capacity Ratio; C19: Power Grid Line Loss Rate; C20: Carbon Emission Intensity Index. Figure 3 The presentation shows the weight distributions of the aforementioned 20 indicators calculated using the BP neural network weighting method, the CRITIC method, the entropy weighting method, and the combined weighting method proposed in this application, respectively. Figure 3 It allows for a direct comparison of the differences in the importance of each indicator based on different weighting methods.

[0074] Specifically, this application constructs a weight matrix. The combined weights are solved using the covariance matrix: ; In the formula, Σ is the final combined weight vector; Σ is the covariance matrix of W; 1 is an all-1 vector. Let W be the inverse covariance matrix, and T be the transpose.

[0075] In one feasible implementation, before the step of calculating the first evaluation score and the second evaluation score of the operational indicator data based on the optimal combination weights using the first evaluation method and the second evaluation method respectively, the method further includes: The operational indicator data are converted into psychologically perceived prospect values ​​based on the preset improved cumulative prospect theory. The operational indicator data is updated based on the psychologically perceived prospective value.

[0076] It should be noted that the pre-defined improved cumulative prospect theory refers to a theoretical model that modifies the prospect value function by introducing decision-maker risk attitude parameters and loss aversion parameters on the basis of traditional cumulative prospect theory. This model is used to transform objective indicator data into a quantitative value reflecting the decision-maker's subjective psychological perception. The psychologically perceived prospect value refers to the value quantity reflecting the decision-maker's psychological perception of the indicator data after transformation by the improved cumulative prospect theory. Its value can be positive or negative; a positive value indicates that the decision-maker perceives the indicator data as a gain, and a negative value indicates a perceived loss. Updating the operational indicator data means replacing the original standardized operational indicator data with the psychologically perceived prospect value, so that subsequent evaluation calculations are based on the psychologically perceived value rather than the original indicator values.

[0077] Understandably, this implementation method, by introducing an improved cumulative prospect theory, transforms operational indicator data into psychologically perceptual data before evaluation calculation, thus overcoming the implicit assumption of perfectly rational decision-makers in conventional evaluation techniques. By setting reference points, gain-sensitive parameters, loss-sensitive parameters, and loss-averse parameters, it can simulate the differentiated perceptual responses of different types of decision-makers when faced with indicator values ​​that are better or worse than their psychological expectations, particularly the loss-avoidance psychology that is more sensitive to losses. After transforming objective data into psychologically perceptual prospect values, subsequent evaluation scores can dynamically reflect the profile of power grid development level from different decision-making psychological perspectives, making the evaluation results closer to actual decision-making logic and meeting diversified decision-making needs.

[0078] In one feasible implementation, the step of converting the operational indicator data into psychologically perceived prospect values ​​according to the improved cumulative prospect theory includes: Define the reference point, gain-sensitive parameter, loss-sensitive parameter, and loss-aversion parameter in the improved cumulative prospect theory; When the operational indicator data is greater than or equal to the reference point, the power of the profit-sensitive parameter of the difference between the operational indicator data and the reference point is used as the psychologically perceived prospect value. When the operational indicator data is less than the reference point, the loss sensitivity parameter of the difference between the reference point and the operational indicator data is raised to the power of the loss aversion parameter, and the negative value is taken as the psychologically perceived prospect value.

[0079] It should be noted that the reference point refers to the psychologically neutral benchmark value used by decision-makers to judge whether operational indicator data represents a gain or a loss. Values ​​above this benchmark are perceived as gains, and values ​​below it are perceived as losses. The value of the reference point can be set according to the decision-making scenario, including but not limited to the mean, median, or expected value set by the decision-maker based on experience for each operational indicator data. The gain sensitivity parameter controls the decision-maker's sensitivity to perceived gains, typically ranging from zero to one. A smaller value indicates a faster decrease in the marginal perception of gains. The loss sensitivity parameter controls the decision-maker's sensitivity to perceived losses, typically ranging from zero to one. A smaller value indicates a faster decrease in the marginal perception of losses. The loss aversion parameter reflects the psychological characteristic that decision-makers are more sensitive to losses than gains, typically greater than one, indicating that the perceived intensity of a loss of the same magnitude is greater than that of a gain. The gain sensitivity parameter exponentiation refers to the power of the difference between the operational indicator data and the reference point, with the gain sensitivity parameter as the exponent. The loss sensitivity parameter exponentiation refers to raising the difference between the reference point and the operating index data to the power of the loss sensitivity parameter. Taking a negative value means taking the opposite of the product result to indicate that the perceived foreground value is a loss attribute.

[0080] For example, the traditional utility function is modified by introducing decision-making psychological parameters such as risk attitude and loss aversion: when hour: ; when hour: ; In the formula, After modification by the improved cumulative prospect theory, the first The evaluated object was in the first The perceived value of each indicator; This serves as a reference point, set as the psychologically neutral benchmark for decision-makers. Values ​​above this benchmark are considered "gains," while values ​​below it are considered "losses." A parameter sensitive to returns; For loss-sensitive parameters; This is the loss aversion parameter.

[0081] Understandably, this implementation method divides operational indicator data into two psychological regions—gain and loss—by setting a reference point. Positive psychological perception is generated when the indicator data reaches or exceeds expectations, while negative psychological perception is generated when the indicator data falls below expectations. The method controls the diminishing marginal perception patterns in the gain and loss regions using gain-sensitive and loss-sensitive parameters, respectively, making the psychological perception more consistent with actual decision-making behavior. A loss-aversion parameter amplifies the intensity of psychological perception in the loss region, simulating the loss-avoidance tendency commonly observed in decision-makers. This implementation method can flexibly adapt to the risk preferences of different types of decision-makers. By adjusting parameter combinations, it can generate psychological perception prospect values ​​from different decision-making psychological perspectives, such as aggressive, balanced, and cautious, enabling the evaluation results to dynamically reflect diverse decision-making needs.

[0082] In one feasible implementation, the first evaluation method is grey relational analysis, and the second evaluation method is an approximation of the ideal solution ranking method based on the similarity of the included angle. The steps of calculating the first evaluation score and the second evaluation score of the operating indicator data using the first evaluation method and the second evaluation method respectively include: The grey relational analysis evaluation score of the operational indicator data is calculated based on the grey relational analysis method and used as the first evaluation score. The approximation ideal solution ranking method evaluation score of the running index data is calculated based on the approximation ideal solution ranking method based on the angle similarity, and the second evaluation score is obtained.

[0083] It should be noted that grey relational analysis refers to a method of comprehensive evaluation that compares the geometric similarity of the indicator data sequences between the object to be evaluated and the ideal solution. The more similar the shapes, the higher the correlation and the better the evaluation result. The grey relational analysis evaluation score refers to the comprehensive score calculated based on grey relational analysis, reflecting the overall correlation between the coastal city power grid to be evaluated and the ideal solution. This score ranges from zero to one, with values ​​closer to one indicating that the power grid development level is closer to the ideal state. The approximation of the ideal solution ranking method based on angle similarity refers to a comprehensive evaluation method that measures the similarity between the vector of the object to be evaluated and the vector of the ideal solution by calculating the cosine value of the angle between them, using cosine similarity instead of the traditional Euclidean distance as the measure of closeness. The approximation of the ideal solution ranking method evaluation score refers to the comprehensive score calculated based on angle similarity, reflecting the relative closeness between the coastal city power grid to be evaluated and the positive ideal solution. This score ranges from zero to one, with values ​​closer to one indicating a better power grid development level.

[0084] For example, this application uses cosine similarity to calculate the similarity between the foreground value vector and the ideal solution vector: ; ; ; In the formula, and For the first Cosine similarity between the foreground value vector of an evaluated object and the positive and negative ideal solution vectors; For the first A vector of foreground values ​​for each evaluated object; and These are the positive ideal solution vector and the negative ideal solution vector, where each element represents the maximum and minimum values ​​of all prospective values ​​for the corresponding index. For the first The closer the ICPT-TOPSIS evaluation score of the evaluated object is to 1, the better the solution is.

[0085] Understandably, this implementation method uses grey relational analysis and an approximation-to-ideal-solution ranking method based on angle similarity to evaluate the development level of the power grid in coastal cities from different dimensions. Grey relational analysis focuses on characterizing the similarity between the index data sequence and the ideal solution in terms of geometric shape change trends, effectively capturing the overall coordination level of power grid development. The approximation-to-ideal-solution ranking method based on angle similarity uses cosine similarity instead of traditional Euclidean distance, focusing on measuring the similarity between the evaluation object and the ideal solution in the vector direction, eliminating the interference of differences in index dimensional scales on the evaluation results. The two evaluation methods complement each other in mechanism, providing a multi-perspective evaluation scoring basis for subsequent dynamic fusion based on the degree of discreteness.

[0086] In one feasible implementation, the step of calculating the approximate ideal solution ranking evaluation score of the operating index data based on the approximate ideal solution ranking method based on the angle similarity, and obtaining the second evaluation score, includes: Obtain the predetermined positive and negative ideal solution vectors; The similarity between the vector corresponding to the operational index data and the positive ideal solution vector is calculated using cosine similarity to obtain the first similarity. The second similarity is obtained by calculating the similarity between the vector corresponding to the operational index data and the negative ideal solution vector using cosine similarity. Based on the first similarity and the second similarity, the evaluation score of the approximation ideal solution ranking method is calculated to obtain the second evaluation score.

[0087] It should be noted that the positive ideal solution vector refers to a reference vector composed of the optimal prospect values ​​of each operational indicator data across all coastal city power grids to be evaluated, with each element being the maximum value of the perceived prospect value of the corresponding operational indicator data across all coastal city power grids to be evaluated. The negative ideal solution vector refers to a reference vector composed of the worst prospect values ​​of each operational indicator data across all coastal city power grids to be evaluated, with each element being the minimum value of the perceived prospect value of the corresponding operational indicator data across all coastal city power grids to be evaluated. The vector corresponding to the operational indicator data refers to a multi-dimensional vector formed by arranging the values ​​of the coastal city power grid to be evaluated in each operational indicator data in order. Cosine similarity is a metric that measures the degree of similarity between two vectors by calculating the cosine of the angle between them, with a value between -1 and +1. The closer the value is to +1, the more consistent the vector directions are. The first similarity refers to the cosine similarity between the vector of the coastal city power grid to be evaluated and the positive ideal solution vector, reflecting its proximity to the optimal state in the vector direction. The second similarity refers to the cosine similarity between the vector of the coastal city power grid to be evaluated and the negative ideal solution vector, reflecting its proximity to the worst state in the vector direction.

[0088] For example, the grey relational degree of positive and negative ideal solutions is calculated based on the prospect values ​​of ICPT (Improved Cumulative Prospect Theory): ; ; In the formula, and For the first The evaluated object was in the first Grey relational coefficients between each index and the positive and negative ideal solutions; and For the first The evaluated object was in the first The absolute difference between the prospective value and the positive and negative ideal solutions for each indicator; and These are the maximum and minimum values ​​among all absolute differences; The resolution coefficient has a range of (0,1) and is usually set to 0.5. It is a very small positive number, used for numerical stability.

[0089] Calculate the ICPT-GRA score comprehensively: ; In the formula, For the first The closer the ICPT-GRA score of the evaluated object is to 1, the better the solution is. For the first The combined weights of each indicator.

[0090] Understandably, this implementation method uses cosine similarity instead of traditional Euclidean distance to calculate the proximity of the ideal solution, effectively eliminating the interference of differences in the dimensional scale of various operational index data on the evaluation results. This allows the evaluation to focus on the consistency of vector direction rather than absolute numerical distance. By calculating the similarity between the evaluated object and both the positive and negative ideal solutions and synthesizing the results to arrive at the final score, the method can comprehensively measure its relative position to the optimal and worst states, making the evaluation results more objective and accurate. It also avoids the ranking bias caused by the sensitivity of traditional Euclidean distance to index scale.

[0091] In one feasible implementation, after the step of acquiring the operational indicator data of the coastal city power grid to be evaluated, and before the step of determining the optimal combination weight of each operational indicator data in the operational indicator data, the method further includes: The operational indicator data is dimensionless according to preset standardization rules to obtain standardized operational indicator data. The operational indicator data is updated based on the standardized operational indicator data.

[0092] It should be noted that the preset standardization rules refer to pre-defined data processing specifications used to eliminate differences in dimensions and orders of magnitude between different operational indicator data, including but not limited to extreme value standardization formulas used for positive and negative indicators respectively. Dimensionless processing refers to the process of converting raw operational indicator data with different units of measurement into dimensionless values, so that the values ​​of each indicator data are within the same order of magnitude range. Standardized operational indicator data refers to the operational indicator values ​​obtained after dimensionless processing, which have eliminated the influence of dimensions, and their value range is usually mapped to the interval between zero and one. Updating the operational indicator data refers to replacing the originally collected operational indicator data with standardized dimensionless values ​​as input data for subsequent weighting and evaluation calculations.

[0093] Understandably, this implementation method eliminates interference from different units of measurement and vastly different numerical ranges of operational indicator data in subsequent calculations by standardizing the operational indicator data before determining the optimal combination weights. If the original indicator data is used directly for weighting and evaluation, the larger number of indicator data will dominate the calculation, leading to distorted evaluation results. After standardization, all indicator data are on the same scale, ensuring that subsequent weight calculations accurately reflect the information content and importance of each indicator, thus guaranteeing the fairness and scientific rigor of the comprehensive evaluation.

[0094] In one feasible implementation, the preset standardization rules include a standardization formula for positive index extreme values ​​and a standardization formula for negative index extreme values. The formula for standardizing the extreme values ​​of the positive index is as follows: ; The formula for standardizing the extreme values ​​of the negative index is: ; in, For standardized values, The original value, and The first The minimum and maximum values ​​of each indicator. This is a preset, small constant.

[0095] It should be noted that the positive indicator extreme value standardization formula refers to a mathematical expression for standardizing operational indicator data where larger values ​​indicate a better level of development. This formula linearly maps the original data to the interval between zero and one by dividing the difference between the original value and the minimum value by the sum of the range and a small constant. The closer the standardized value is to one, the better the performance of the indicator. The negative indicator extreme value standardization formula refers to a mathematical expression for standardizing operational indicator data where smaller values ​​indicate a better level of development. This formula divides the difference between the maximum value and the original value by the sum of the range and a small constant, so that the smaller the original value, the closer the standardized value is to one. The original value refers to the actual value of the operational indicator data directly collected or statistically obtained from the coastal city power grid to be evaluated without any mathematical transformation. The minimum value of the j-th indicator refers to the minimum original value of all coastal city power grids to be evaluated for the j-th operational indicator data. The maximum value of the j-th indicator refers to the maximum original value of all coastal city power grids to be evaluated for the j-th operational indicator data. The preset small constant refers to an extremely small positive number pre-set to prevent the denominator from being zero, including but not limited to values ​​on the order of ten to the power of negative six.

[0096] It is understandable that this implementation method uses corresponding extreme value standardization formulas to process positive and negative indicators respectively. This not only eliminates the incommensurability problem caused by differences in dimensions and orders of magnitude between different operating indicator data, but also maintains the relative difference information between the evaluation objects in the original data. At the same time, by introducing a small constant, it avoids the calculation anomaly of zero denominator in extreme cases, so that the standardized data are all within the same scale range, providing a standardized input data basis for subsequent weighting and evaluation calculations.

[0097] This implementation method aims to systematically evaluate the development level of power grids in typhoon-prone coastal cities. It constructs a comprehensive evaluation system that includes 5 primary indicators and 20 secondary indicators, covering power supply reliability, power grid intelligence, infrastructure level, disaster resilience, and green and low-carbon level, eliminating the dimensional influence of indicators with different dimensions.

[0098] In one feasible implementation, the first degree of dispersion is the first variance of the first evaluation score; the second degree of dispersion is the second variance of the second evaluation score. The step of dynamically allocating fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and weightedly fusing the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid includes: Calculate the first variance of the first evaluation score for all coastal city power grids to be evaluated; Calculate the second variance of the second evaluation score for all coastal city power grids to be evaluated; The proportion of the second variance to the sum of the first variance and the second variance is used as the fusion weight of the first evaluation score; The proportion of the first variance to the sum of the first variance and the second variance is used as the fusion weight of the second evaluation score; Multiply the first evaluation score by the fusion weight of the first evaluation score to obtain the first weighted score; Multiply the second evaluation score by the fusion weight of the second evaluation score to obtain the second weighted score; The first weighted score and the second weighted score are added together, and the sum is used as the evaluation result of the development level of the power grid in the coastal city.

[0099] It should be noted that the first variance refers to the variance of the data sequence consisting of the first evaluation scores of all coastal city power grids to be evaluated. It is used to measure the dispersion of the grey relational analysis evaluation scores among different evaluation objects. The smaller the variance, the more concentrated the score distribution and the higher the stability of the evaluation method. The second variance refers to the variance of the data sequence consisting of the second evaluation scores of all coastal city power grids to be evaluated. It is used to measure the dispersion of the evaluation scores of the approximation ideal solution ranking method based on angle similarity among different evaluation objects. The sum of the first and second variances is the arithmetic sum of the first and second variances, used as the normalized denominator for calculating the fusion weight. The proportion refers to the weight of a part relative to the whole. In this embodiment, the proportion of the second variance to the sum of the first and second variances is the fusion weight of the first evaluation score, and the proportion of the first variance to the sum of the first and second variances is the fusion weight of the second evaluation score. The first weighted score is the result obtained by multiplying the first evaluation score by the corresponding fusion weight. The second weighted score is the result obtained by multiplying the second evaluation score by the corresponding fusion weight. The sum of the first weighted score and the second weighted score is used as the final comprehensive development level evaluation score of the power grid in coastal cities.

[0100] For example, this application introduces the concept of variance to achieve dynamic weight allocation and integrates the scores of ICPT-GRA and ICPT-TOPSIS: ; ; ; In the formula, The dynamic fusion weights assigned to the ICPT-GRA method have values ​​that are similar to... The weight of the GRA method is directly proportional to the variance of the TOPSIS method; that is, when the TOPSIS method is unstable (large variance), the weight of the GRA method will increase. Dynamic fusion weights assigned to the ICPT-TOPSIS method; The variance of the ICPT-GRA scores for all evaluated subjects; The variance of the ICPT-TOPSIS scores for all evaluated objects; For the first The final comprehensive evaluation score of each evaluated object.

[0101] Understandably, this implementation method quantifies the score stability by calculating the variance of the score sequences of each evaluation method, and uses the variance ratio as the basis for allocating the fusion weights. When the variance of a certain evaluation method's score is small, it indicates that the method's differentiation results on different objects being evaluated are more stable and reliable, and its opposing evaluation method will receive a larger fusion weight, and vice versa. This method of adaptively adjusting weights based on the degree of dispersion allows the fusion weights to dynamically reflect the reliability differences between the two evaluation methods in practical applications, avoiding evaluation bias caused by fixed weights or simple average fusion that ignores differences in method stability. This implementation method, through a variance-driven dynamic weighting mechanism, achieves intelligent fusion of two evaluation methods with different mechanisms, further improving the accuracy and robustness of the final comprehensive evaluation results.

[0102] For example, to help understand the implementation process of the coastal city power grid development level evaluation method obtained by combining this embodiment with the above embodiment one, refer to Figure 4 , Figure 4A comprehensive evaluation chart of the power grid development level of various cities is provided. Specifically, the horizontal axis represents the power grid of the coastal cities to be evaluated (e.g., city A, city B, city C, etc.), and the vertical axis represents the comprehensive evaluation score of each city's power grid. This score is based on the method described in this invention. It is calculated by determining the optimal combination weights of each operational indicator data, using Grey Relational Analysis (GRA) and the Technique for Order Preference by Similarity to an Ideal Solution (TOPSIS) to obtain the first and second evaluation scores respectively. These scores are then dynamically allocated and weighted according to variance to obtain the final evaluation score. The chart displays the final evaluation score of each city in bar chart form; a higher score indicates a better level of power grid development in that city. As an example, Figure 4 City A received the highest evaluation score, while city H received the lowest. Figure 4 This paper presents the final comprehensive evaluation scores and ranking results of 12 city power grids under the methodology described in this application. Based on the evaluation scores, the 12 cities can be divided into three tiers: Tier 1 (comprehensive score greater than 0.8): including City A (0.931), City J (0.908), City B (0.897), City G (0.878), and City D (0.876). These cities demonstrate excellent performance in comprehensive dimensions such as power supply reliability, disaster resilience, and green and low-carbon development, indicating a leading level of development. Tier 2 (comprehensive score between 0.4 and 0.7): only City L (0.602), representing a moderate level of development. Tier 3 (comprehensive score less than 0.4): including City F (0.390), City C (0.386), City K (0.364), City E (0.344), City I (0.321), and City H (0.221). These cities have shortcomings in areas such as power grid infrastructure, smart grids, or disaster resilience, and still have significant room for improvement.

[0103] For example, refer to Figure 5 , Figure 5 This invention provides a comprehensive ranking diagram for cities under different decision-making psychology. Specifically, the horizontal axis represents the power grid of the coastal city to be evaluated, and the vertical axis represents the comprehensive ranking order (the smaller the value, the higher the ranking). This invention, by improving the reference point, gain-sensitive parameter α, loss-sensitive parameter β, and loss-averse parameter λ in the Improved Cumulative Prospect Theory (ICPT), can simulate the evaluation results under different decision-making psychology (such as aggressive, balanced, cautious, loss-averse, and gain-sensitive). Figure 5This indicates the ranking changes of various cities' power grid development levels under a preset combination of decision-making psychological parameters. As an example, when decision-makers are biased towards loss aversion, some cities with stronger disaster resilience (such as city L) will see a significant improvement in their ranking; when decision-makers are biased towards profit pursuit, cities with higher levels of green and low-carbon development (such as city E) will rank higher. Figure 5 It intuitively reflects the dynamic impact of different decision-making psychology on evaluation ranking, providing power grid planners with differentiated decision-making references. Figure 5 A heatmap showing the ranking changes of 12 coastal city power grids (codes A to L) under six different combinations of decision-making psychological parameters (aggressive-loss sensitive, balanced-loss sensitive, cautious-loss sensitive, aggressive-gain sensitive, balanced-gain sensitive, cautious-gain sensitive) is presented. Figure 5 It is evident that due to differences in their own indicator structures, cities exhibit varying degrees of sensitivity to decision-making psychology, specifically as follows: City A: Its ranking is extremely stable and absolutely leading, consistently ranking 1st or 2nd under all six decision-making psychological conditions, indicating a highly balanced development across all indicators and the strongest overall competitiveness; City B: Its ranking is stable between 3rd and 4th, demonstrating strong and balanced comprehensive strength; City C: Its ranking fluctuates the most dramatically, jumping to 1st place under the "balance-loss sensitivity" psychological condition, but falling to 10th place under the "cautious-gain sensitivity" psychological condition, indicating a significant advantage under specific evaluation expectations; City D: Its ranking shows no significant fluctuation, consistently remaining between 5th and 6th, belonging to the stable, upper-middle-tier city category; City E: Its overall ranking is relatively low, mainly in the 9th to 12th range, ranking at the bottom at 12th under the "cautious-gain sensitivity" psychological condition, indicating a comprehensive disadvantage in gain-related indicators; City F: Its ranking is extremely sensitive to decision-making psychology, consistently ranking 1st or 2nd under the "balance-gain sensitivity" psychological condition. Under the "sensitive" psychological condition, the ranking jumped to 1st place, but under the other 5 psychological conditions, it fell back to 8th to 10th place, showing an extreme coexistence of "weaknesses" and "strengths"; City G: The ranking is also highly stable, always between 3rd and 5th, a solid member of the first tier; City H: Overall performance is in a very poor position, ranking consistently between 10th and 12th, with a significantly insufficient comprehensive development level; City I: The ranking is mainly between 7th and 8th, only dropping to 12th under the "balance-gain sensitivity" psychological condition, overall at a lower-middle level; City J: The ranking is highly stable and excellent, always maintaining between 2nd and 3rd, demonstrating an outstanding comprehensive development level; City K: Under most psychological conditions, the ranking is in the lower tier of 10th to 12th, only under the "cautious-gain sensitivity" psychological condition can the ranking improve to 6th; City L: The overall ranking fluctuates between 5th and 9th, with a slightly better ranking under the "loss sensitivity" psychological condition than under the "gain sensitivity" psychological condition, belonging to a city with a medium level of development. In summary, Figure 5 It clearly reflects the differentiated performance of the power grid rankings of various coastal cities in the AL under different decision-making psychological orientations, and can provide power grid planners with multi-perspective decision-making references.

[0104] It should be noted that the above examples are only for understanding this application and do not constitute a limitation on the evaluation method for the development level of coastal urban power grids in this application. Any simple modifications based on this technical concept are within the scope of protection of this application.

[0105] This application also provides a device for evaluating the development level of power grids in coastal cities. Please refer to [reference needed]. Figure 6 The coastal city power grid development level evaluation device includes: The acquisition module 10 is used to acquire the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data. The calculation module 20 is used to calculate the first evaluation score and the second evaluation score of the operation indicator data based on the optimal combination weights, respectively through the first evaluation method and the second evaluation method. Evaluation module 30 is used to dynamically allocate fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and to perform weighted fusion of the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

[0106] And / or, the coastal city power grid development level evaluation device includes: The first acquisition module is used to acquire a preset number of objective weighting methods, wherein the objective weighting methods include the backpropagation neural network weighting method, the weight determination method based on index correlation, and the entropy weighting method. The first calculation module is used to calculate the initial weights of each operational indicator data according to each of the objective weighting methods described above. The first generation module is used to combine the initial weights according to the principle of minimizing deviation to generate the optimal combined weights.

[0107] And / or, the coastal city power grid development level evaluation device includes: The second acquisition module is used to acquire the connection weights from the input layer to the hidden layer in the trained backpropagation neural network when the objective weighting method is the backpropagation neural network weighting method. The second calculation module is used to divide the sum of the absolute values ​​of the connection weights of each running indicator data to all hidden neurons by the sum of the absolute values ​​of the connection weights of all running indicator data to all hidden neurons, and the quotient is used as the initial weight of the backpropagation neural network of the current running indicator data. The third calculation module is used to calculate the standard deviation of each operating indicator data and the correlation coefficient between each operating indicator data and other operating indicator data when the objective weighting method is a weight determination method based on indicator correlation. The first determining module is used to determine a comprehensive measure of the current operating indicator data based on the standard deviation of each operating indicator data and its correlation with all other operating indicator data. The fourth calculation module is used to take the proportion of the comprehensive measurement of each operational indicator data to the sum of the comprehensive measurements of all operational indicator data as the initial weight of the current operational indicator data based on the weight determination method of indicator correlation. The fifth calculation module is used to calculate the information entropy of each operational indicator data when the objective weighting method is the entropy weighting method; The second determining module is used to determine the degree of dispersion of the current operating indicator data based on the information entropy of each operating indicator data. The sixth calculation module is used to take the proportion of the dispersion of each operating indicator data to the total dispersion of all operating indicator data as the initial weight of the current operating indicator data using the entropy weight method.

[0108] And / or, the coastal city power grid development level evaluation device includes: The first component module is used to construct a weight matrix by combining the initial weights of the backpropagation neural network, the initial weights of the weight determination method based on indicator correlation, and the initial weights of the entropy weight method for each of the aforementioned operational indicator data. The seventh calculation module is used to calculate the covariance matrix of the weight matrix, and to solve for the combination coefficients that minimize the deviation of the weight vector based on the covariance matrix and the all-one vector, thereby generating the optimal combination weight.

[0109] And / or, the coastal city power grid development level evaluation device includes: The first conversion module is used to convert the operational index data into psychologically perceived prospect values ​​according to a preset improved cumulative prospect theory. The first update module is used to update the operational indicator data based on the psychologically perceived prospective value.

[0110] And / or, the coastal city power grid development level evaluation device includes: The first setting module is used to set the reference point, the gain-sensitive parameter, the loss-sensitive parameter, and the loss-averse parameter in the improved cumulative prospect theory. The eighth calculation module is used to use the power of the profit-sensitive parameter of the difference between the operating indicator data and the reference point as the psychologically perceived prospect value when the operating indicator data is greater than or equal to the reference point. The ninth calculation module is used to, when the operating index data is less than the reference point, multiply the loss sensitivity parameter of the difference between the reference point and the operating index data by the loss aversion parameter and take the negative value as the psychologically perceived prospect value.

[0111] And / or, the coastal city power grid development level evaluation device includes: The tenth calculation module is used to calculate the grey relational analysis evaluation score of the operation index data as the first evaluation score based on the grey relational analysis method. The eleventh calculation module is used to calculate the approximate ideal solution ranking method evaluation score of the running index data according to the approximate ideal solution ranking method based on the angle similarity, and obtain the second evaluation score.

[0112] And / or, the coastal city power grid development level evaluation device includes: The third acquisition module is used to acquire the predetermined positive ideal solution vector and negative ideal solution vector; The twelfth calculation module is used to calculate the similarity between the vector corresponding to the operational index data and the positive ideal solution vector using cosine similarity, and obtain the first similarity. The thirteenth calculation module is used to calculate the similarity between the vector corresponding to the running index data and the negative ideal solution vector using cosine similarity, and obtain the second similarity. The fourteenth calculation module is used to calculate the evaluation score of the approximation ideal solution ranking method based on the first similarity and the second similarity, and obtain the second evaluation score.

[0113] And / or, the coastal city power grid development level evaluation device includes: The first processing module is used to perform dimensionless processing on the operation index data according to preset standardization rules to obtain standardized operation index data. The second update module is used to update the operational indicator data based on the standardized operational indicator data.

[0114] And / or, the coastal city power grid development level evaluation device includes: The fifteenth calculation module is used to calculate the first variance of the first evaluation score for all coastal city power grids to be evaluated; The sixteenth calculation module is used to calculate the second variance of the second evaluation score for all coastal city power grids to be evaluated; The seventeenth calculation module is used to take the proportion of the second variance to the sum of the first variance and the second variance as the fusion weight of the first evaluation score; The eighteenth calculation module is used to take the proportion of the first variance to the sum of the first variance and the second variance as the fusion weight of the second evaluation score; The nineteenth calculation module is used to multiply the first evaluation score by the fusion weight of the first evaluation score to obtain the first weighted score; The twentieth calculation module is used to multiply the second evaluation score by the fusion weight of the second evaluation score to obtain the second weighted score; The twenty-first calculation module is used to add the first weighted score and the second weighted score, and the sum is used as the evaluation result of the development level of the coastal city power grid.

[0115] The coastal city power grid development level evaluation device provided in this application adopts the coastal city power grid development level evaluation method in the above embodiments, which can solve the technical problem of poor accuracy of the evaluation results of the coastal city power grid development level. Compared with the prior art, the beneficial effects of the coastal city power grid development level evaluation device provided in this application are the same as the beneficial effects of the coastal city power grid development level evaluation method provided in the above embodiments, and other technical features in the coastal city power grid development level evaluation device are the same as the features disclosed in the methods of the above embodiments, and will not be repeated here.

[0116] This application provides a device for evaluating the development level of power grids in coastal cities. The device includes: at least one processor; and a memory communicatively connected to the at least one processor. The memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the coastal city power grid development level evaluation method in Embodiment 1 above.

[0117] The following is for reference. Figure 7 The diagram illustrates a structural schematic suitable for implementing the coastal city power grid development level evaluation device in the embodiments of this application. The coastal city power grid development level evaluation device in the embodiments of this application may include, but is not limited to, mobile terminals such as mobile phones, tablets, laptops, digital broadcast receivers, PDAs (Personal Digital Assistants), PMPs (Portable Media Players), vehicle terminals (e.g., vehicle navigation terminals), and fixed terminals such as digital televisions and desktop computers. Figure 7 The coastal city power grid development level evaluation device shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of this application.

[0118] like Figure 7As shown, the coastal city power grid development level assessment device may include a processing unit 1001, which can perform various appropriate actions and processes according to a program stored in ROM 1002 or a program loaded from storage device 1003 into RAM 1004. RAM 1004 also stores various programs and data required for the operation of the coastal city power grid development level assessment device. The processing unit 1001, ROM 1002, and RAM 1004 are interconnected via bus 1005. I / O interface 1006 is also connected to the bus. Typically, the following systems can be connected to I / O interface 1006: input devices 1007 including, for example, touch screens, touchpads, keyboards, mice, image sensors, microphones, accelerometers, gyroscopes, etc.; output devices 1008 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1003 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1009. Communication device 1009 allows the coastal city power grid development level assessment device to exchange data with other devices wirelessly or via wired communication. Although the diagram shows equipment for evaluating the development level of coastal city power grids with various systems, it should be understood that the implementation or presence of all the systems shown is not required. More or fewer systems may be implemented alternatively.

[0119] Specifically, according to the embodiments disclosed in this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments disclosed in this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device, or installed from storage device 1003, or installed from ROM 1002. When the computer program is executed by processing device 1001, it performs the functions defined in the methods of the embodiments disclosed in this application.

[0120] The coastal city power grid development level evaluation equipment provided in this application adopts the coastal city power grid development level evaluation method in the above embodiments, which can solve the technical problem of poor accuracy of the evaluation results of the coastal city power grid development level. Compared with the prior art, the beneficial effects of the coastal city power grid development level evaluation equipment provided in this application are the same as the beneficial effects of the coastal city power grid development level evaluation method provided in the above embodiments, and other technical features in the coastal city power grid development level evaluation equipment are the same as the features disclosed in the previous embodiment method, and will not be repeated here.

[0121] It should be understood that the various parts disclosed in this application can be implemented using hardware, software, firmware, or a combination thereof. In the description of the above embodiments, specific features, structures, materials, or characteristics can be combined in any suitable manner in one or more embodiments or examples.

[0122] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0123] This application provides a computer-readable storage medium having computer-readable program instructions (i.e., a computer program) stored thereon, which are used to execute the coastal city power grid development level evaluation method in the above embodiments.

[0124] The computer-readable storage medium provided in this application may be, for example, a USB flash drive, but is not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, system, or device. The program code contained on the computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wires, optical cables, RF (Radio Frequency), etc., or any suitable combination thereof.

[0125] The aforementioned computer-readable storage medium may be included in the coastal city power grid development level evaluation equipment; or it may exist independently and not be assembled into the coastal city power grid development level evaluation equipment.

[0126] The aforementioned computer-readable storage medium carries one or more programs, which, when executed by the coastal city power grid development level assessment device, cause the coastal city power grid development level assessment device to: Obtain the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data; Based on the optimal combination weights, the first evaluation score and the second evaluation score of the operational indicator data are calculated using the first evaluation method and the second evaluation method, respectively. The fusion weights are dynamically allocated based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score. The first evaluation score and the second evaluation score are then weighted and fused based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

[0127] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a Local Area Network (LAN) or a Wide Area Network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0128] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0129] The modules described in the embodiments of this application can be implemented in software or hardware. The names of the modules do not necessarily limit the functionality of the unit itself.

[0130] The readable storage medium provided in this application is a computer-readable storage medium that stores computer-readable program instructions (i.e., a computer program) for executing the above-described coastal city power grid development level evaluation method, thereby solving the technical problem of poor accuracy in the evaluation results of coastal city power grid development level. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in this application are the same as those of the coastal city power grid development level evaluation method provided in the above embodiments, and will not be repeated here.

[0131] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the coastal city power grid development level evaluation method described above.

[0132] The computer program product provided in this application can solve the technical problem of poor accuracy in the evaluation results of the development level of power grids in coastal cities. Compared with the prior art, the beneficial effects of the computer program product provided in this application are the same as the beneficial effects of the coastal city power grid development level evaluation method provided in the above embodiments, and will not be repeated here.

[0133] All acquisition of signals, information, or actions in this application are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization of the owner of the relevant device.

[0134] The above description is only a part of the embodiments of this application and does not limit the scope of protection of this application. All equivalent structural transformations made under the technical concept of this application and using the content of this application specification and drawings, or direct / indirect applications in other related technical fields, are included in the scope of protection of this application.

Claims

1. A method for evaluating the development level of power grids in coastal cities, characterized in that, The method includes: Obtain the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data; Based on the optimal combination weights, the first evaluation score and the second evaluation score of the operational indicator data are calculated using the first evaluation method and the second evaluation method, respectively. The fusion weights are dynamically allocated based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score. The first evaluation score and the second evaluation score are then weighted and fused based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

2. The method as described in claim 1, characterized in that, The steps of obtaining the operation index data of the coastal city power grid to be evaluated and determining the optimal combination weight of each operation index data include: An objective weighting method for obtaining a preset number of values, wherein the objective weighting method includes the backpropagation neural network weighting method, the weight determination method based on index correlation, and the entropy weighting method; Calculate the initial weights of each operational indicator data according to the objective weighting methods described above; The initial weights are combined according to the principle of minimizing deviation to generate the optimal combined weights.

3. The method as described in claim 2, characterized in that, The step of calculating the initial weights of each operational indicator data according to each of the objective weighting methods includes: When the objective weighting method is the backpropagation neural network weighting method, the connection weights from the input layer to the hidden layer in the trained backpropagation neural network are obtained. Divide the sum of the absolute values ​​of the connection weights of each running indicator data to all hidden neurons by the sum of the absolute values ​​of the connection weights of all running indicator data to all hidden neurons, and use the quotient as the initial weights of the backpropagation neural network for the current running indicator data. When the objective weighting method is a weight determination method based on the correlation of indicators, the standard deviation of each operating indicator data and the correlation coefficient between each operating indicator data and other operating indicator data are calculated. Based on the standard deviation of each operational indicator data and its correlation with all other operational indicator data, a comprehensive measure of the current operational indicator data is determined; The proportion of the comprehensive measure of each operational indicator data to the sum of the comprehensive measures of all operational indicator data is used as the initial weight of the current operational indicator data based on the weight determination method of indicator correlation. When the objective weighting method is the entropy weighting method, the information entropy of each operational indicator data is calculated. The degree of dispersion of the current operational indicator data is determined based on the information entropy of each operational indicator data. The proportion of the dispersion of each operational indicator data to the sum of the dispersion of all operational indicator data is used as the initial weight of the current operational indicator data using the entropy weight method.

4. The method as described in claim 3, characterized in that, The step of combining the initial weights according to the principle of minimizing deviation to generate the optimal combined weights includes: The initial weights of the backpropagation neural network for each of the aforementioned operational indicator data, the initial weights of the weight determination method based on indicator correlation, and the initial weights of the entropy weight method are combined into a weight matrix. Calculate the covariance matrix of the weight matrix, and solve for the combination coefficients that minimize the deviation of the weight vector based on the covariance matrix and the all-one vector, thereby generating the optimal combination weight.

5. The method as described in claim 1, characterized in that, Before the step of calculating the first evaluation score and the second evaluation score of the operational indicator data based on the optimal combination weights using the first evaluation method and the second evaluation method respectively, the method further includes: The operational indicator data are converted into psychologically perceived prospect values ​​based on the preset improved cumulative prospect theory. The operational indicator data is updated based on the psychologically perceived prospective value.

6. The method as described in claim 5, characterized in that, The step of converting the operational indicator data into psychologically perceived prospect values ​​based on the improved cumulative prospect theory includes: Define the reference point, gain-sensitive parameter, loss-sensitive parameter, and loss-aversion parameter in the improved cumulative prospect theory; When the operational indicator data is greater than or equal to the reference point, the power of the profit-sensitive parameter of the difference between the operational indicator data and the reference point is used as the psychologically perceived prospect value. When the operational indicator data is less than the reference point, the loss sensitivity parameter of the difference between the reference point and the operational indicator data is raised to the power of the loss aversion parameter, and the negative value is taken as the psychologically perceived prospect value.

7. The method as described in claim 1, characterized in that, The first evaluation method is grey relational analysis, and the second evaluation method is an approximation of the ideal solution ranking method based on the similarity of the included angle. The steps of calculating the first evaluation score and the second evaluation score of the operating indicator data using the first evaluation method and the second evaluation method respectively include: The grey relational analysis evaluation score of the operational indicator data is calculated based on the grey relational analysis method and used as the first evaluation score. The approximation ideal solution ranking method evaluation score of the running index data is calculated based on the approximation ideal solution ranking method based on the angle similarity, and the second evaluation score is obtained.

8. The method as described in claim 7, characterized in that, The step of calculating the approximate ideal solution ranking evaluation score of the operating index data based on the approximate ideal solution ranking method based on the angle similarity, and obtaining the second evaluation score, includes: Obtain the predetermined positive and negative ideal solution vectors; The similarity between the vector corresponding to the operational index data and the positive ideal solution vector is calculated using cosine similarity to obtain the first similarity. The second similarity is obtained by calculating the similarity between the vector corresponding to the operational index data and the negative ideal solution vector using cosine similarity. Based on the first similarity and the second similarity, the evaluation score of the approximation ideal solution ranking method is calculated to obtain the second evaluation score.

9. The method as described in claim 1, characterized in that, After the step of obtaining the operation index data of the coastal city power grid to be evaluated, and before the step of determining the optimal combination weight of each operation index data in the operation index data, the method further includes: The operational indicator data is dimensionless according to preset standardization rules to obtain standardized operational indicator data. The operational indicator data is updated based on the standardized operational indicator data.

10. The method as described in claim 9, characterized in that, The preset standardization rules include the standardization formula for positive index extreme values ​​and the standardization formula for negative index extreme values; The formula for standardizing the extreme values ​​of the positive index is as follows: ； The formula for standardizing the extreme values ​​of the negative index is: ； in, For standardized values, The original value, and The first The minimum and maximum values ​​of each indicator. This is a preset, small constant.

11. The method as described in claim 1, characterized in that, The first degree of dispersion is the first variance of the first evaluation score; the second degree of dispersion is the second variance of the second evaluation score. The step of dynamically allocating fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and weighting and fusing the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid includes: Calculate the first variance of the first evaluation score for all coastal city power grids to be evaluated; Calculate the second variance of the second evaluation score for all coastal city power grids to be evaluated; The proportion of the second variance to the sum of the first variance and the second variance is used as the fusion weight of the first evaluation score; The proportion of the first variance to the sum of the first variance and the second variance is used as the fusion weight of the second evaluation score; Multiply the first evaluation score by the fusion weight of the first evaluation score to obtain the first weighted score; Multiply the second evaluation score by the fusion weight of the second evaluation score to obtain the second weighted score; The first weighted score is added to the second weighted score, and the sum is used as the evaluation result of the development level of the power grid in the coastal city.

12. A device for evaluating the development level of power grids in coastal cities, characterized in that, The device includes: The acquisition module is used to acquire the operation index data of the coastal city power grid to be evaluated, and determine the optimal combination weight of each operation index data in the operation index data. The calculation module is used to calculate the first evaluation score and the second evaluation score of the operation indicator data based on the optimal combination weights, respectively, through the first evaluation method and the second evaluation method. The evaluation module is used to dynamically allocate fusion weights based on the first degree of dispersion of the first evaluation score and the second degree of dispersion of the second evaluation score, and to perform weighted fusion of the first evaluation score and the second evaluation score based on the fusion weights to obtain the evaluation result of the development level of the coastal city power grid.

13. A device for evaluating the development level of power grids in coastal cities, characterized in that, The device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the coastal city power grid development level evaluation method as described in any one of claims 1 to 11.

14. A storage medium, characterized in that, The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, it implements the steps of the coastal city power grid development level evaluation method as described in any one of claims 1 to 11.

15. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the coastal city power grid development level evaluation method as described in any one of claims 1 to 11.

Images