Electric power system digital excellent unit selection method and device

Abstract

The invention provides a digital excellent unit selection method and device for a power system, and belongs to the field of power industry evaluation. The method comprises the following steps: acquiring unit basic data of a power unit participating in evaluation; the unit basic data type at least comprises a digital evaluation index; constructing an original data matrix according to the unit basic data, standardizing the original data matrix, and calculating the weight of the digital evaluation index according to the standardized data matrix; weighting the standardized data matrix according to the weight of the digital evaluation index to form a weighted matrix; and constructing a benefit-type solution and a cost-type solution according to the weighting matrix, calculating the relative closeness of each power unit participating in evaluation according to the benefit-type solution, the cost-type solution and the weighting matrix, sorting the relative closeness from small to large, and determining an excellent unit according to a sorting result. According to the invention, the problem of scientifically evaluating the digital transformation effect of the power unit can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power industry evaluation, and particularly to a method and device for selecting excellent units for the digitalization of power systems. Background Art

[0002] With the transformation of the global energy structure and power market, the power industry is undergoing an unprecedented digital transformation. Digital technologies, such as big data, Internet of Things, and cloud computing, are changing the traditional operation and management methods of power systems, making power systems more intelligent, efficient, and reliable. However, the success of this transformation not only depends on the introduction of technology but also requires an effective evaluation mechanism to monitor and guide this process. Most traditional power unit evaluation methods rely on financial and operational performance indicators, which, although able to reflect the operating status of the unit, often fall short in evaluating the effects of digital transformation. These traditional methods usually lack consideration of specific digital factors, such as the integration degree of information systems, data processing capabilities, and their specific impacts on operational efficiency. In addition, existing evaluation methods often rely on the subjective judgments of experts, which is not only time-consuming and laborious but also may lead to inconsistent evaluation results due to individual preferences and experience differences. For example, expert evaluation largely relies on experience and intuitive judgment, which may overlook some important data-driven indicators. Facing these challenges, existing technical solutions do not provide a comprehensive and objective evaluation system, which limits the ability of power units to formulate strategies and evaluate effects on the path of digital transformation. Summary of the Invention

[0003] Embodiments of the present invention provide a method and device for selecting excellent units for the digitalization of power systems to solve the problem of scientifically evaluating the digital transformation effects of power units.

[0004] In a first aspect, embodiments of the present invention provide a method for selecting excellent units for the digitalization of power systems, including: obtaining the unit basic data of power units participating in the evaluation; the types of unit basic data at least include digital evaluation indicators;

[0005] Constructing an original data matrix based on the unit basic data, standardizing the original data matrix, and calculating the weights of digital evaluation indicators according to the standardized data matrix;

[0006] Weighting the standardized data matrix according to the weights of digital evaluation indicators to form a weighted matrix; constructing a benefit-type solution and a cost-type solution according to the weighted matrix, calculating the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, cost-type solution, and weighted matrix, sorting the relative closeness from small to large, and determining excellent units according to the sorting results.

[0007] In a second aspect, an embodiment of the present invention provides an apparatus for selecting excellent units in the digitalization of a power system, including: a data acquisition module, configured to acquire the basic unit data of the power units participating in the evaluation; the basic unit data type includes at least digital evaluation indicators;

[0008] a weight calculation module, configured to construct an original data matrix based on the basic unit data, standardize the original data matrix, and calculate the weights of the digital evaluation indicators according to the standardized data matrix;

[0009] an excellent unit determination module, configured to weight the standardized data matrix according to the weights of the digital evaluation indicators to form a weighted matrix; construct a benefit-type solution and a cost-type solution according to the weighted matrix, calculate the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, the cost-type solution and the weighted matrix, sort the relative closeness from small to large, and determine the excellent units according to the sorting result.

[0010] An embodiment of the present invention provides a method and apparatus for selecting excellent units in the digitalization of a power system. By determining the weights of digital indicators, the importance of each digital evaluation indicator is reflected, thereby improving the accuracy and fairness of the evaluation. Weight the standardized data matrix according to the weights of the digital evaluation indicators to form a weighted matrix; construct a benefit-type solution and a cost-type solution according to the weighted matrix, calculate the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, the cost-type solution and the weighted matrix, sort the relative closeness from small to large, and determine the excellent units according to the sorting result. The excellent units are selected by comparing the relative closeness of each power unit participating in the evaluation with the benefit-type solution (optimal situation) and the cost-type solution (worst situation), so as to objectively evaluate the performance of each power unit participating in the evaluation. By determining the weights of each digital evaluation indicator and multi-criteria decision analysis technology, a scientific, objective and comprehensive evaluation framework is provided. This method can automatically process a large amount of performance data. It can effectively identify and reward power units that perform well in the digital transformation process, and promote the digitalization process of the entire industry. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0012] Figure 1 is a flowchart of the method for selecting excellent units in the digitalization of a power system provided by an embodiment of the present invention;

[0013] Figure 2It is the structural diagram of the excellent unit selection device for power system digitization provided by the embodiments of the present invention. Detailed implementation manners

[0014] In the following description, specific details such as specific system structures and technologies are presented for the purpose of illustration rather than limitation, so as to thoroughly understand the embodiments of the present invention. However, those skilled in the art should clearly understand that the present invention can also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present invention.

[0015] To make the objectives, technical solutions, and advantages of the present invention clearer, the following will be described through specific embodiments with reference to the accompanying drawings.

[0016] Figure 1 It is the flowchart of the excellent unit selection method for power system digitization provided by the embodiments of the present invention.

[0017] As Figure 1 shown, the excellent unit selection method for power system digitization includes the following steps:

[0018] Step 101: Obtain the unit basic data of the power units participating in the evaluation; the types of unit basic data include at least digital evaluation indicators;

[0019] Automatically capture the required unit basic data from the information systems of each power unit participating in the evaluation. The systems involved in unit basic data capture include the equipment asset lean management system, enterprise resource planning system, etc. Use scripts and APIs to capture unit basic data. Clean the captured data, and use data cleaning tools to remove missing values, outliers, and duplicate data to ensure data quality. Use interpolation or mean filling to process missing data, and use box plot method or Z-score method to identify and process outliers.

[0020] The unit basic data of power units includes various possible evaluation indicators related to digital transformation. Optionally, these evaluation indicators include "digital society index", "digital ability index", and "digital element index". In other embodiments, other evaluation indicators are also included.

[0021] Determine some evaluation indicators of the unit basic data as digital evaluation indicators through the Delphi method or principal component analysis method.

[0022] The digital evaluation indicators are used to measure the digital transformation effect of the power units participating in the evaluation.

[0023] Step 102: Construct an original data matrix based on the unit basic data, standardize the original data matrix, and calculate the weights of the digital evaluation indicators according to the standardized data matrix.

[0024] Since the dimensions and data ranges of the various digital evaluation indicators are different, it is necessary to standardize the original data matrix to eliminate the influence of dimensions, so that the digital evaluation indicators can be compared at the same level, thereby eliminating the deviation caused by inconsistent scales.

[0025] In a possible implementation, the unit basic data is constructed into an original data matrix according to the determined digital evaluation indicators. Each row represents a participating power unit, and each column corresponds to a digital evaluation indicator.

[0026] The digital evaluation indicators are divided into two categories, namely benefit-type indicators and cost-type indicators. The original data matrix of the corresponding type is standardized for different indicators to obtain a standardized data matrix, which includes the standardized benefit-type indicators and cost-type indicators.

[0027] Among them, the benefit-type indicators include one or more benefit-type sub-indicators, and the cost-type indicators include one or more cost-type sub-indicators.

[0028] Weight refers to the importance of the digital evaluation indicator relative to the digital effect. It is different from the general proportion. It reflects not only the percentage of the digital evaluation indicator, but emphasizes the relative importance of the digital evaluation indicator, tending to the contribution degree or importance. That is, in the embodiments of the present application, the contribution degrees of the benefit-type indicators and cost-type indicators to the relative importance of the digital evaluation results are different.

[0029] Step 103: Weight the standardized data matrix according to the weights of the digital evaluation indicators to form a weighted matrix; construct a benefit-type solution and a cost-type solution according to the weighted matrix, calculate the relative closeness of each participating power unit according to the benefit-type solution, cost-type solution and weighted matrix, sort the relative closeness from small to large, and determine the excellent units according to the sorting results.

[0030] The excellent units are selected by comparing the relative distances of each unit from the benefit-type solution (optimal situation) and the cost-type solution (worst situation). Considering the relative closeness of each participating power unit to the optimal and worst solutions, the performance of each power unit can be objectively evaluated.

[0031] The benefit-type solution refers to the solution that hopes to obtain the maximum value on multiple attributes in the optimal decision-making, that is, the solution that can reach the optimal level on all attributes.

[0032] A cost-based solution refers to a solution that aims to achieve the minimum value on multiple attributes in the optimal decision-making, that is, a solution that can reach the worst level on all attributes.

[0033] In a possible implementation manner, the standardized weighted Euclidean distances of the benefit-based solution and the cost-based solution are calculated respectively according to the weighted matrix;

[0034] The relative closeness degrees of the power units participating in the evaluation to the benefit-based solution and the cost-based solution are calculated according to the standardized weighted Euclidean distances.

[0035] In another possible implementation manner, the ratios of the benefit-based solution and the cost-based solution are calculated respectively according to the weighted matrix;

[0036] The relative closeness degrees of the power units participating in the evaluation to the benefit-based solution and the cost-based solution are calculated according to the ratios.

[0037] In this embodiment, by providing a method and device for selecting excellent units in the digitalization of power systems, the importance of each digital evaluation index is reflected by determining the weights of the digitalization indexes, thereby improving the accuracy and fairness of the evaluation. The standardized data matrix is weighted according to the weights of the digital evaluation indexes to form a weighted matrix; a benefit-based solution and a cost-based solution are constructed according to the weighted matrix, and the relative closeness degrees of each power unit participating in the evaluation are calculated according to the benefit-based solution, the cost-based solution and the weighted matrix, and the relative closeness degrees are sorted from small to large, and the excellent units are determined according to the sorting results. By comparing the relative closeness degrees of each power unit participating in the evaluation to the benefit-based solution (optimal situation) and the cost-based solution (worst situation) to select excellent units, the performance of each power unit participating in the evaluation can be objectively evaluated. By determining the weights of each digital evaluation index and multi-criteria decision analysis technology, a scientific, objective and comprehensive evaluation framework is provided. It can effectively identify and reward power units that perform well in the digital transformation process and promote the digitalization process of the entire industry.

[0038] The preferred implementation manners of the steps of this embodiment are described in detail below.

[0039] In a possible implementation manner, the digital evaluation indexes are determined based on the Delphi method and expert scoring, and the steps include:

[0040] According to the historical unit basic data of the power units participating in the evaluation obtained, a digital evaluation index pool is established and stored; the digital evaluation index pool contains alternative digital evaluation indexes;

[0041] The alternative digital evaluation indexes are adjusted according to the initial expert survey results obtained, and the initial digital evaluation indexes are determined according to the adjusted alternative digital evaluation indexes; among them, the expert survey results are obtained through expert surveys based on the Delphi method.

[0042] Adjust the initial digital evaluation indicators according to the scoring and ranking results obtained by experts based on the initial digital evaluation indicators, and obtain the digital evaluation indicators to be determined according to the adjusted initial digital evaluation indicators;

[0043] Determine the digital evaluation indicators based on the coverage of the digital evaluation indicators to be determined and the consistency of expert opinions.

[0044] Among them, the anonymous survey conducted by the Delphi method includes at least three rounds to ensure the scientificity and representativeness of the digital evaluation indicators. Specifically, form an expert group and select at least 10 experts in the power industry with extensive experience and high recognition to ensure the diversity of the expert group, including power system operation and maintenance experts, digital transformation experts, data analysis experts, etc. Send questionnaires to experts via email or online survey tools to collect experts' initial opinions on the digital evaluation indicators of the power system. The questionnaire design includes open-ended questions, inviting experts to list the digital evaluation indicators they think are important.

[0045] Summarize and feedback the initial survey results to the experts, and request the experts to score and prioritize the determined initial digital evaluation indicators. The scoring uses a Likert scale (e.g., 1-5 points), inviting experts to evaluate the importance of each digital evaluation indicator, and further supplement and adjust the digital evaluation indicators.

[0046] Finally, based on the results of multiple rounds of surveys, comprehensively consider the coverage of the digital evaluation indicators and the consistency of expert opinions to determine the digital evaluation indicators. Use the Kendall concordance coefficient to test the consistency of expert opinions to ensure the scientificity and reliability of indicator selection.

[0047] As a structured communication technology, the Delphi method mainly collects experts' opinions and predictions on specific topics through multiple rounds of anonymous surveys to determine the key indicators for evaluating digital excellent units. This method helps to reach a consensus among experts and reduce the influence of single expert bias.

[0048] The digital evaluation indicators include but are not limited to the automation coverage rate of distribution network lines, the accuracy rate of distribution network data, the application rate of mobile operations, the heavy load rate of distribution network lines, the defect rate of distribution network equipment, and the comprehensive line loss rate.

[0049] According to the opinions of the expert group, construct a digital effectiveness evaluation index system including 3 first-level indicators, 15 second-level indicators, and 85 third-level indicators around the three aspects of "digital society index", "digital ability index", and "digital element index".

[0050] Set the first-level indicator of "Digital Society Index". Since digital transformation strategy, process-oriented organization, and digitalization of leadership process are the core contents of enterprise digital construction, which can achieve more efficient operation and management of enterprises, and the digital ecosystem facilitates the establishment of a good digital business environment for enterprises, the setting of this first-level indicator aims to measure the strategic planning, organizational management, leadership construction, and the degree and ability of ecosystem construction in the process of enterprise digital transformation. The secondary indicators include digital transformation strategy, process-oriented organization, digitalization of leadership process, and digital ecosystem. The specific indicators are shown in Table 1.

[0051] Table 1 Digital Society Index

[0052]

[0053] Set the first-level indicator of "Digital Capability Index". Since digital transformation plan, digitalization of research and development, digitalization of production process, digitalization of marketing management, digitalization of supply management, digital human resource management, and digital infrastructure are the key areas for enterprises to enhance digital capabilities, which can achieve comprehensive digital transformation and improvement of enterprises, the setting of this first-level indicator aims to measure the degree and ability of plan formulation, R & D innovation, production management, marketing management, supply chain management, human resource management, and infrastructure construction in the process of enterprise digital capability construction. The secondary indicators include digital transformation plan, digitalization of research and development, digitalization of production process, digitalization of marketing management, digitalization of supply management, digital human resource management, and digital infrastructure. The specific indicators are shown in Table 2.

[0054] Table 2 Digital Capability Index

[0055]

[0056]

[0057] Set the first-level indicator of "Digital Element Index". Since data utilization, digital technologies, digitalization of innovation, and digital integration provide important resources and technical support for enterprises in the process of digital transformation, which can achieve the efficient operation and innovative development of enterprises, the setting of this first-level indicator aims to measure the data utilization level, technology application degree, innovation ability, and integration ability of enterprises in the process of digital transformation. The secondary indicators include data utilization, digital technologies, digitalization of innovation, and digital integration. The specific indicators are shown in Table 3.

[0058] Table 3 Digital Element Index

[0059]

[0060]

[0061] Correspondingly, the steps for calculating the weights of digital evaluation indicators based on the standardized data matrix include:

[0062] Calculate the contribution degree of each participating power unit under the digital evaluation indicator according to the standardized first data matrix, calculate the information entropy of the digital evaluation indicator according to the contribution degree, calculate the difference coefficient of the digital evaluation indicator according to the information entropy, and calculate the first weight of the digital evaluation indicator according to the difference coefficient.

[0063] Correspondingly, the types of digital evaluation indicators include first benefit type indicators and first cost type indicators;

[0064] Construct the original data matrix with the cleaned data according to the determined digital evaluation indicators. Each row represents a participating power unit, and each column corresponds to a digital evaluation indicator. Suppose there are M participating power units, and each M corresponds to N digital evaluation indicators.

[0065] The original data matrix is:

[0066] A=(a ij ) m×n

[0067] The standardized first data matrix is:

[0068] B=(b ij ) m×n

[0069] The calculation formula for the first benefit type indicator is:

[0070]

[0071] The calculation formula for the first cost type indicator is:

[0072]

[0073] The calculation formula for the contribution degree is:

[0074]

[0075] The calculation formula for the information entropy is:

[0076]

[0077] The calculation formula for the difference coefficient is:

[0078] g j =1-e j

[0079] The calculation formula for the first weight is:

[0080]

[0081] wherein, a ij is the value of the j-th digital evaluation index of the i-th power unit participating in the evaluation; b ij is the value of the j-th digital evaluation index of the i-th power unit participating in the evaluation after standardization; a j min is the minimum value of the j-th digital evaluation index of the power units participating in the evaluation, is the maximum value of the j-th digital evaluation index of the power units participating in the evaluation, g j is the difference coefficient, and m is the total number of all digital evaluation indexes.

[0082] The entropy weight method is used to calculate the weights of each digital evaluation index. The entropy weight method is a method for determining weights based on the principle of information entropy, which can objectively reflect the variation degree and importance of each digital evaluation index.

[0083] The determination of the weight depends on the dispersion degree of the digital evaluation index data. The greater the dispersion degree of the digital evaluation index, the higher its weight, indicating that the index has stronger discrimination ability in the overall evaluation.

[0084] In a possible implementation manner, the digital evaluation index is determined based on principal component analysis, and the steps include:

[0085] Obtain the historical unit basic data of the power units participating in the evaluation, construct an original data matrix according to the unit basic data, standardize the original data matrix, and calculate the covariance matrix according to the standardized second data matrix; perform eigenvalue decomposition on the principal components of the covariance matrix to obtain an eigenvalue matrix and an eigenvector matrix; calculate the cumulative contribution rate of the principal components according to the eigenvalue matrix, and select the principal components with a cumulative contribution rate higher than 85%; determine the digital evaluation index according to the original digital evaluation indexes corresponding to the principal components with a cumulative contribution rate higher than 85%.

[0086] Correspondingly, the steps for calculating the weights of the digital evaluation indexes according to the standardized data matrix include:

[0087] Calculate the covariance matrix according to the standardized second data matrix, perform eigenvalue decomposition on the principal components of the covariance matrix to obtain an eigenvalue matrix and an eigenvector matrix;

[0088] Calculate the principal component loading matrix according to the eigenvalue matrix and the eigenvector matrix;

[0089] Determine the second weight of the digital evaluation index according to the principal component loading matrix.

[0090] The standardized data matrix has the same dimension, making subsequent analysis more accurate. Calculate the covariance matrix based on the standardized data matrix to evaluate the correlation between various digital evaluation indicators.

[0091] The eigenvalue represents the variance magnitude of the principal component, and the eigenvector represents the direction of the principal component.

[0092] In a possible implementation, use the loadings on the principal components as row vectors; construct the principal component loading matrix with digital evaluation indicators as column vectors.

[0093] Correspondingly, the types of digital evaluation indicators include second benefit type indicators and second cost type indicators:

[0094] The original data matrix is:

[0095] A = (a ij ) m×n

[0096] The standardized second data matrix is:

[0097] X′ = (x′) m×n

[0098] The calculation formula for the second benefit type indicator is:

[0099]

[0100] The calculation formula for the second cost type indicator is:

[0101]

[0102] The calculation formula for the covariance matrix is:

[0103]

[0104] The calculation formula for eigenvalue decomposition is:

[0105] C = P∧P T

[0106] The calculation formula for the cumulative contribution rate is:

[0107]

[0108] The calculation formula for the principal component loading matrix is:

[0109] A = V∧ 1 / 2

[0110] The calculation formula for the second weight is:

[0111]

[0112] where x ′ is the value of the digital evaluation index after standardization, x is the value of the digital evaluation index of the power unit participating in the evaluation, min(x) is the minimum value of the digital evaluation index of the power unit participating in the evaluation, max(x) is the maximum value of the digital evaluation index of the power unit participating in the evaluation, X′ is the data matrix after standardization, is the mean vector of the standardized data matrix, n is the number of samples; Λ is the diagonal matrix of the eigenvalue matrix, the diagonal elements of which are eigenvalues, P is the eigenvector matrix; λ i is the i-th eigenvalue, k is the number of principal components selected, m is the total number of all eigenvalues; V is the eigenvector matrix, p ij is the element in the principal component loading matrix, representing the loading of the j-th digital index on the i-th principal component, k is the number of principal components selected, m is the total number of all digital evaluation indexes.

[0113] In a possible implementation, assume that the principal component loading matrix A is:

[0114]

[0115] Then the weight ω i of the i-th digital evaluation index is calculated as:

[0116]

[0117] where p i1 is the loading of the 1st digital evaluation index on the i-th principal component, p i2 is the loading of the 2nd digital evaluation index on the i-th principal component.

[0118] Through the above steps, the digital evaluation indexes and their weights of the power system can be scientifically determined using the principal component analysis method.

[0119] In a possible implementation, the steps of calculating the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, cost-type solution and weighted matrix include:

[0120] Calculate the distances of each power unit participating in the evaluation to the benefit-type solution and cost-type solution, and calculate the relative closeness of each power unit participating in the evaluation according to the distances.

[0121] Correspondingly, the calculation formula of the weighted matrix is:

[0122] c ij =b ij ×ω j

[0123] The weighted matrix is:

[0124]

[0125] The beneficial ideal solution is:

[0126]

[0127] The cost ideal solution is:

[0128]

[0129] Among them, the calculation formula for the distance of each participating power unit to the beneficial ideal solution is:

[0130]

[0131] The calculation formula for the distance of each participating power unit to the cost ideal solution is:

[0132]

[0133] The calculation formula for the relative closeness of each participating power unit is:

[0134]

[0135] In the formula, b ij is the value of the jth digital index of the ith power unit; ω j is the weight of the jth digital evaluation index, maxc ij is the maximum attribute of the digital evaluation index in the weighted matrix, and minc ij is the minimum attribute of the digital evaluation index in the weighted matrix.

[0136] Through the present invention, the power grid system will obtain a scientific, comprehensive and dynamic digital transformation effectiveness evaluation tool. This tool can help the power system accurately identify which power units perform better during the digital transformation process, providing an important basis for future strategic decision-making. Implementing this evaluation method and device will promote the more efficient and intelligent allocation of resources in the power grid system, improve service quality, and enhance innovation capabilities. It can effectively identify and reward power units that perform well during the digital transformation process, thereby promoting the sustainable development of enterprises and further promoting the digital process of the entire industry.

[0137] It should be understood that the magnitudes of the sequence numbers of the steps in the above embodiments do not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0138] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiments above.

[0139] Figure 2 This is the structural diagram of the device for evaluating excellent units in the digitalization of the power system provided by the embodiments of the present invention. For the sake of convenience of description, only the parts related to the embodiments of the present application are shown and are described in detail as follows:

[0140] As Figure 2 shown, the device 2 for evaluating excellent units in the digitalization of the power system includes:

[0141] A data acquisition module 21, configured to acquire the unit basic data of the power units participating in the evaluation; the types of the unit basic data include at least digital evaluation indicators;

[0142] A weight calculation module 22, configured to construct an original data matrix according to the unit basic data, standardize the original data matrix, and calculate the weights of the digital evaluation indicators according to the standardized data matrix;

[0143] An excellent unit determination module 23, configured to weight the standardized data matrix according to the weights of the digital evaluation indicators to form a weighted matrix; construct a benefit-type solution and a cost-type solution according to the weighted matrix, calculate the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, the cost-type solution and the weighted matrix, sort the relative closeness from small to large, and determine the excellent units according to the sorting result.

[0144] The embodiments of the present invention provide a method and a device for evaluating excellent units in the digitalization of the power system. By determining the weights of the digital indicators, the importance of each digital evaluation indicator is reflected, thereby improving the accuracy and fairness of the evaluation. Weight the standardized data matrix according to the weights of the digital evaluation indicators to form a weighted matrix; construct a benefit-type solution and a cost-type solution according to the weighted matrix, calculate the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, the cost-type solution and the weighted matrix, sort the relative closeness from small to large, and determine the excellent units according to the sorting result. By comparing the relative closeness of each power unit participating in the evaluation with the benefit-type solution (optimal situation) and the cost-type solution (worst situation) to select excellent units, the performance of each power unit participating in the evaluation can be objectively evaluated. By determining the weights of each digital evaluation indicator and the multi-criteria decision analysis technology, a scientific, objective and comprehensive evaluation framework is provided. It can effectively identify and reward the power units that perform well in the digital transformation process and promote the digitalization process of the entire industry.

[0145] In a possible implementation manner, the data acquisition module may be used for:

[0146] The digital evaluation indicators are determined based on the Delphi method and expert scoring, and the steps thereof include:

[0147] Based on the historical unit basic data of the power units participating in the evaluation obtained, establish and store a digital evaluation index pool; the digital evaluation index pool contains alternative digital evaluation indexes.

[0148] Adjust the alternative digital evaluation indexes according to the initial expert survey results obtained, and determine the initial digital evaluation indexes according to the adjusted alternative digital evaluation indexes; among them, the expert survey results are obtained through expert surveys based on the Delphi method.

[0149] Adjust the initial digital evaluation indexes according to the results of the experts' scoring and ranking based on the initial digital evaluation indexes, and obtain the digital evaluation indexes to be determined according to the adjusted initial digital evaluation indexes.

[0150] Determine the digital evaluation indexes according to the coverage of the digital evaluation indexes to be determined and the consistency of the experts' opinions.

[0151] In a possible implementation manner, the weight calculation module can be used for:

[0152] The steps of calculating the weights of the digital evaluation indexes according to the standardized data matrix include:

[0153] Calculate the contribution degrees of each power unit participating in the evaluation under the digital evaluation indexes according to the standardized first data matrix, calculate the information entropy of the digital evaluation indexes according to the contribution degrees, calculate the difference coefficient of the digital evaluation indexes according to the information entropy, and calculate the first weights of the digital evaluation indexes according to the difference coefficient.

[0154] In a possible implementation manner, the weight calculation module can be used for:

[0155] The types of digital evaluation indexes include first benefit type indexes and first cost type indexes.

[0156] The original data matrix is:

[0157] A=(a ij ) m×n

[0158] The standardized first data matrix is:

[0159] B=(b ij ) m×n

[0160] The calculation formula for the first benefit type index is:

[0161]

[0162] The calculation formula for the first cost type index is:

[0163]

[0164] The calculation formula for the contribution degree is as follows:

[0165]

[0166] The calculation formula for the information entropy is as follows:

[0167]

[0168] The calculation formula for the coefficient of difference is as follows:

[0169] g j = 1 - e j

[0170] The calculation formula for the first weight is as follows:

[0171]

[0172] In the formula, a ij is the value of the j-th digital evaluation index of the i-th power unit participating in the evaluation; b ij is the value of the j-th digital evaluation index of the i-th power unit participating in the evaluation after standardization; a j min is the minimum value of the j-th digital evaluation index of the power units participating in the evaluation, is the maximum value of the j-th digital evaluation index of the power units participating in the evaluation, g j is the coefficient of difference, and m is the total number of all digital evaluation indexes.

[0173] In a possible implementation manner, the data acquisition module can be used for:

[0174] The digital evaluation indexes are determined based on principal component analysis, and the steps include:

[0175] Obtain the historical unit basic data of the power units participating in the evaluation, construct the original data matrix according to the unit basic data, standardize the original data matrix, and calculate the covariance matrix according to the standardized second data matrix; perform eigenvalue decomposition on the principal components of the covariance matrix to obtain the eigenvalue matrix and the eigenvector matrix; calculate the cumulative contribution rate of the principal components according to the eigenvalue matrix, and select the principal components with a cumulative contribution rate higher than 85%; determine the digital evaluation indexes according to the original digital evaluation indexes corresponding to the principal components with a cumulative contribution rate higher than 85%.

[0176] In a possible implementation manner, the weight calculation module can be used for:

[0177] The steps for calculating the weights of the digital evaluation indexes according to the standardized data matrix include:

[0178] Calculate the covariance matrix based on the standardized second data matrix; perform eigenvalue decomposition on the principal components of the covariance matrix to obtain the eigenvalue matrix and the eigenvector matrix;

[0179] Calculate the principal component loading matrix based on the eigenvalue matrix and the eigenvector matrix;

[0180] Determine the second weight of the digital evaluation index according to the principal component loading matrix.

[0181] In a possible implementation, the weight calculation module can be used for:

[0182] The types of digital evaluation indexes include benefit type indexes and cost type indexes:

[0183] The original data matrix is:

[0184] A = (a ij ) m×n

[0185] The standardized second data matrix is:

[0186] X ′ = (x ′ )

[0187] m×n

[0188] The calculation formula for the second benefit type index is:

[0189]

[0190] The calculation formula for the second cost type index is:

[0191]

[0192] The calculation formula for the covariance matrix is:

[0193]

[0194] The calculation formula for eigenvalue decomposition is:

[0195] C = P∧P T

[0196] The calculation formula for the cumulative contribution rate is:

[0197]

[0198] The calculation formula for the principal component loading matrix is:

[0199] A = V∧ 1 / 2

[0200] The calculation formula for the second weight is:

[0201]

[0202] Wherein, x' is the value of the digital evaluation index after standardization, x is the value of the digital evaluation index of the power unit participating in the evaluation, min(x) is the minimum value of the digital evaluation index of the power units participating in the evaluation, max(x) is the maximum value of the digital evaluation index of the power units participating in the evaluation, X' is the data matrix after standardization, is the mean vector of the standardized data matrix, n is the number of samples; ∧ is the diagonal matrix of the eigenvalue matrix, the diagonal elements of which are eigenvalues, and P is the eigenvector matrix; λ i is the i-th eigenvalue, k is the number of selected principal components, and m is the total number of all eigenvalues; V is the eigenvector matrix, p ij is an element in the principal component loading matrix, indicating the loading of the j-th digital index on the i-th principal component, k is the number of selected principal components, and m is the total number of all digital evaluation indexes.

[0203] In a possible implementation manner, the excellent unit determination module can be used for:

[0204] The steps of calculating the relative closeness of each power unit participating in the evaluation according to the benefit-type solution, cost-type solution and weighted matrix include:

[0205] Calculate the distances of each power unit participating in the evaluation to the benefit-type solution and cost-type solution, and calculate the relative closeness of each power unit participating in the evaluation according to the distances.

[0206] In a possible implementation manner, the excellent unit determination module can be used for:

[0207] The calculation formula of the weighted matrix is:

[0208] c ij = b ij × ω j

[0209] The weighted matrix is:

[0210]

[0211] The benefit-type ideal solution is:

[0212]

[0213] The cost-type ideal solution is:

[0214]

[0215] Among them, the calculation formula for the distance of each power unit participating in the evaluation to the benefit-type ideal solution is:

[0216]

[0217] The calculation formula for the distance of each participating power unit to the cost-type ideal solution is as follows:

[0218]

[0219] The calculation formula for the relative closeness degree of each participating power unit is as follows:

[0220]

[0221] In the formula, b ij is the value of the jth digital index of the ith power unit; ω j is the weight of the jth digital evaluation index, maxc ij is the maximum attribute of the digital evaluation index in the weighted matrix, and minc ij is the minimum attribute of the digital evaluation index in the weighted matrix.

[0222] In the above embodiments, the descriptions of the various embodiments have their own focuses. For the parts not detailed or recorded in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

[0223] Those of ordinary skill in the art can realize that, in combination with the templates, units, and algorithm steps of the examples described in this disclosure, they can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0224] If the above-mentioned module / unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such understanding, to implement all or part of the processes in the above-mentioned method embodiments of the present invention, it can also be completed by instructing relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of the above-mentioned method embodiments for selecting excellent units in digitalization of each power system can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in the form of source code, object code, executable file or some intermediate form, etc. The computer-readable medium can include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory, random access memory, electrical carrier signal, telecommunication signal, and software distribution medium, etc.

[0225] The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for selecting excellent units for digitalization of power systems, characterized in that: include: Obtain basic unit data of the power units participating in the evaluation; The unit basic data type at least includes digital evaluation indicators; Constructing an original data matrix according to the unit basic data, standardizing the original data matrix, and calculating the weight of the digital evaluation index according to the standardized data matrix; The standardized data matrix is ​​weighted according to the weights of the digital evaluation indicators to form a weighted matrix; a benefit solution and a cost solution are constructed according to the weighted matrix, and the relative closeness of each power unit participating in the evaluation is calculated according to the benefit solution, the cost solution and the weighted matrix, and the relative closeness is sorted from small to large, and the excellent unit is determined according to the sorting result.

2. A method for selecting excellent digital units in a power system according to claim 1, characterized in that: The digital evaluation index is determined based on the Delphi method and expert scoring, and the steps include: Establishing and storing a digital evaluation index pool based on the acquired historical unit basic data of the power units participating in the evaluation; wherein the digital evaluation index pool includes alternative digital evaluation indicators; Adjusting the candidate digital evaluation index according to the obtained initial expert survey results, and determining the initial digital evaluation index according to the adjusted candidate digital evaluation index; wherein the expert survey results are obtained by conducting an expert survey based on the Delphi method; According to the obtained expert scoring and ranking results based on the initial digital evaluation index, the initial digital evaluation index is adjusted, and the digital evaluation index to be determined is obtained according to the adjusted initial digital evaluation index; The digital evaluation index is determined according to the coverage of the digital evaluation index to be determined and the consistency of expert opinions.

3. The method for selecting excellent digital units of power system according to claim 2, characterized in that: The step of calculating the weight of the digital evaluation index according to the standardized data matrix includes: The contribution of each electric power unit participating in the evaluation under the digital evaluation index is calculated according to the standardized first data matrix, the information entropy of the digital evaluation index is calculated according to the contribution, the difference coefficient of the digital evaluation index is calculated according to the information entropy, and the first weight of the digital evaluation index is calculated according to the difference coefficient.

4. A method for selecting excellent digital units of power system according to claim 3, characterized in that: The types of digital evaluation indicators include first benefit-type indicators and first cost-type indicators; The original data matrix is: A=(a ij ) m×n The first data matrix after standardization is: B=(b ij ) m×n The calculation formula of the first benefit-based indicator is: The calculation formula of the first cost-based indicator is: The calculation formula of the contribution is: The calculation formula of the information entropy is: The calculation formula of the difference coefficient is: g j =1-e j The calculation formula of the first weight is: In the formula, a ij is the value of the jth digital evaluation index of the i-th power unit participating in the evaluation; b ij is the value of the jth digital evaluation index of the ith power unit participating in the evaluation after standardization; a j min is the minimum value of the jth digital evaluation index of the power units participating in the evaluation, is the maximum value of the jth digital evaluation index of the power unit participating in the evaluation, g j is the coefficient of variation, and m is the total number of all digital evaluation indicators.

5. The method for selecting excellent digital units of power system according to claim 1, characterized in that: The digital evaluation index is determined based on principal component analysis, and the steps include: Obtain historical unit basic data of the power units participating in the evaluation, construct an original data matrix based on the unit basic data, standardize the original data matrix, and calculate the covariance matrix based on the standardized second data matrix; perform eigenvalue decomposition on the principal components of the covariance matrix to obtain an eigenvalue matrix and an eigenvector matrix; calculate the cumulative contribution rate of the principal components based on the eigenvalue matrix, and select the principal components with a cumulative contribution rate higher than 85%; determine the digital evaluation index based on the original digital evaluation index corresponding to the principal components with a cumulative contribution rate higher than 85%.

6. A method for selecting excellent digital units in power systems according to claim 5, characterized in that: The step of calculating the weight of the digital evaluation index according to the standardized data matrix includes: Calculate a covariance matrix based on the standardized second data matrix; perform eigenvalue decomposition on the principal components of the covariance matrix to obtain an eigenvalue matrix and an eigenvector matrix; Calculate the principal component loading matrix according to the eigenvalue matrix and the eigenvector matrix; The second weight of the digital evaluation index is calculated according to the principal component loading matrix.

7. A method for selecting excellent digital units of power system according to claim 6, characterized in that: The types of digital evaluation indicators include second benefit indicators and second cost indicators: The original data matrix is: A=(a ij ) m×n The standardized second data matrix is: X ′ =(x ′ ) m×n The calculation formula of the second benefit-type indicator is: The calculation formula of the second cost-based indicator is: The calculation formula of the covariance matrix is: The calculation formula of the eigenvalue decomposition is: C=P∧P T The calculation formula of the cumulative contribution rate is: The calculation formula of the principal component loading matrix is: A=V∧ 1 / 2 The calculation formula of the second weight is: In the formula, x ′ is the value of the digital evaluation index after standardization, x is the value of the digital evaluation index of the power unit participating in the evaluation, min(x) is the minimum value of the digital evaluation index of the power unit participating in the evaluation, max(x) is the maximum value of the digital evaluation index of the power unit participating in the evaluation, X′ is the standardized data matrix, is the mean vector of the standardized data matrix, n is the number of samples; ∧ is the diagonal matrix of the eigenvalue matrix, whose diagonal elements are the eigenvalues, and P is the eigenvector matrix; λ i is the i-th eigenvalue, k is the number of principal components selected, m is the total number of all eigenvalues; V is the eigenvector matrix, p ij It is an element in the principal component loading matrix, which represents the loading of the jth digital index on the i-th principal component, k is the number of principal components selected, and m is the total number of all digital evaluation indicators.

8. The method for selecting excellent digital units of power system according to claim 1, characterized in that: The step of calculating the relative closeness of each power unit participating in the evaluation according to the benefit-based solution, the cost-based solution and the weighted matrix comprises: The distances from each power unit participating in the evaluation to the benefit-based solution and the cost-based solution are calculated, and the relative proximity of each power unit participating in the evaluation is calculated based on the distances.

9. A method for selecting excellent digital units in power systems according to claim 8, characterized in that: The calculation formula of the weighted matrix is: c ij =b ij ×ω j The weighting matrix is: The benefit-based ideal solution is: The cost-based ideal solution is: Among them, the calculation formula of the distance from each power unit participating in the evaluation to the benefit-based ideal solution is: The calculation formula of the distance from each power unit participating in the evaluation to the cost-based ideal solution is: The calculation formula for the relative proximity of each power unit participating in the evaluation is: Where b ij is the value of the jth digital indicator of the ith power unit; ω j is the weight of the jth digital evaluation index, maxc ij It is the maximum attribute of the digital evaluation index in the weighted matrix, minc ij It is the minimum attribute of the digital evaluation index in the weighted matrix.

10. A device for selecting excellent digital units in a power system, characterized in that: include: A data acquisition module, used to acquire basic unit data of the power units participating in the evaluation; The unit basic data type at least includes digital evaluation indicators; A weight calculation module, used to construct an original data matrix according to the unit basic data, standardize the original data matrix, and calculate the weight of the digital evaluation index according to the standardized data matrix; The excellent unit determination module is used to weight the standardized data matrix according to the weights of the digital evaluation indicators to form a weighted matrix; construct a benefit solution and a cost solution according to the weighted matrix, calculate the relative proximity of each power unit participating in the evaluation according to the benefit solution, the cost solution and the weighted matrix, sort the relative proximity from small to large, and determine the excellent unit according to the sorting result.

Images