Thermal power distribution and storage project comprehensive evaluation method and system

Abstract

The invention discloses a thermal power distribution and storage engineering comprehensive evaluation method and system, and relates to the technical field of power distribution network engineering, and the method comprises the steps: formulating a power distribution network engineering comprehensive evaluation index, and selecting a first-level index according to a clustering algorithm; and performing secondary index selection based on an analytic hierarchy process. And analyzing the evaluation indexes to obtain a power distribution network process comprehensive evaluation scheme. The average information contribution rate of each evaluation index in the power distribution network engineering evaluation index system constructed by the method to the number of sea selection indexes can reach 90% or above. The obtained comprehensive evaluation system can provide feasibility guidance for links such as power distribution network planning and transformation, so that the power utilization experience and the economic benefits of power enterprises are improved, and the method has wide application potential and economic value in actual design of power distribution network engineering.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network engineering, and particularly relates to a comprehensive evaluation method and system for thermal power distribution and energy storage projects. Background Art

[0002] In the process of evaluating conventional distribution network projects, there are phenomena such as incomplete unification of work standards and complex standards. The existing relevant standards and specifications for distribution network projects mainly focus on generality, with relatively low operability, and it is difficult to effectively reveal the comprehensive benefits and technical levels of distribution network project items. Against the background of a substantial increase in distribution network project items, a targeted research and analysis on the comprehensive evaluation system of actual distribution network project items is required.

[0003] Currently, relevant scholars have carried out corresponding research on the comprehensive evaluation index system of distribution network project items. The literature proposes an evaluation method based on an improved Elman neural network (Elman neural network, Elman), a feedback-type dynamic neural network. A self-feedback connection gain coefficient is introduced into the Elman neural network to optimize the relevant parameters of the Elman neural network, and to evaluate the influence degree of the historical information of the distribution network on the future state. The literature proposes an evaluation method based on the optimal combination weighting method. This model only needs to use four evaluation indicators, namely reliability, economy, electrical characteristics, and adaptability, to achieve the optimal quantitative evaluation of the planning scheme. The above methods have a certain degree of feasibility, but in the actual application process, due to problems such as insufficient depth of feasibility study of the distribution network and inadequate implementation of technical principles, the corresponding index evaluation effects deviate from the expectations.

[0004] The purpose of data mining is to extract the effective information required for the comprehensive evaluation indicators of engineering projects from a large amount of sample data, remove the influence of defective data, noise and random factors, and provide necessary data references for various decision-making instructions. Distribution network projects have characteristics such as complex internal structures and extreme external environments, and there are multi-level relationships among the control elements, which makes the identification and analysis of key control elements during the construction process of distribution network projects extremely complex, and the demand for creating corresponding comprehensive evaluation systems is increasing. Data mining technology can extract potentially useful information from seemingly unrelated data in the sample database, and it is the mainstream method when constructing the comprehensive evaluation system for distribution network projects, with very broad application prospects. According to the strategy of the State Grid Corporation of China, to build an energy Internet enterprise, it is necessary to adapt to the development situation of business digitalization and data valueization, and fully explore the value of data. Therefore, by using the existing professional business platforms and relying on the company's three platforms (data middle platform, business middle platform, cloud platform) to build an integrated platform for digital review of distribution networks, to solve problems such as the complex current situation of distribution network projects, the lack of effective information management in design review, and the single review mode and management level of distribution network projects. Thus, researching a comprehensive evaluation method for the design and review of distribution network projects based on data mining has important research significance and practical value. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the technical problem solved by the present invention is that although the existing technology has a certain degree of feasibility, in the actual application process, due to problems such as insufficient depth of feasibility study of the distribution network and inadequate implementation of technical principles, the corresponding index evaluation effect deviates from the expectation.

[0007] To solve the above technical problem, the present invention provides the following technical solution: A comprehensive evaluation method for thermal power with energy storage projects, which includes the following steps: formulating comprehensive evaluation indicators for distribution network projects, and selecting primary indicators according to the clustering algorithm; using the R clustering method to cluster the standardized indicators, dividing each indicator into multiple evaluation indicator categories, the R clustering method divides the categories based on the sum of squared deviations, and uses the non-parametric statistical test method to test the significance level of the clustering result to determine the primary indicators of the distribution network project; selecting secondary indicators based on the analytic hierarchy process; analyzing the evaluation indicators to obtain a comprehensive evaluation plan for the distribution network process; the analysis of the evaluation indicators includes obtaining the sample data of the distribution network project, on the basis of the determined evaluation indicators and corresponding weights, using the multiple regression analysis algorithm to construct a regression equation, and estimating the regression coefficients through independent tests, calculating the comprehensive score according to the regression equation, and ranking each evaluation object.

[0008] As a preferred solution of the comprehensive evaluation method of thermal power distribution and storage engineering described in the present invention, the step of formulating the comprehensive evaluation index of the distribution network engineering includes determining the evaluation index, designing the index weight and evaluating the evaluation method; standardizing the comprehensive evaluation index of the distribution network engineering, including standardizing the positive index and the negative index; assuming that The first evaluation object The standardized value of the evaluation index is , the positive indicators are standardized as follows: , If the value of the indicator is small, it means that the comprehensive performance of the distribution network engineering design and review is good. In this case, such evaluation indicators are defined as negative indicators, and the standardization process is expressed as: , in, Indicates The first evaluation object Evaluation index value, for In 1 to Take between Minimum value, for In 1 to Take between Maximum value, As the evaluation index, For the evaluation object, Indicates the number of selected indicators after standardization.

[0009] As a preferred solution of the comprehensive evaluation method of thermal power distribution and storage engineering described in the present invention, wherein: the selection of the primary index according to the clustering algorithm includes, after the comprehensive evaluation of the distribution network engineering is standardized, clustering the comprehensive evaluation of the distribution network engineering using the R clustering algorithm, The standardized indicators are divided into categories, using and Respectively represent The vector of the sum of squared deviations and sample mean values of the class evaluation index is The sum of squares of the class evaluation index is expressed as: , in, Indicates The number of evaluation indicators of the class evaluation indicators, Indicates The first evaluation object The sample value vector after the evaluation index is standardized; The total within-class sum of squares of the evaluation index categories is expressed as: , where is the total within-class sum of squares of the evaluation index categories, is the th evaluation index category; the process of the within-class sum of squares clustering method includes, first, dividing the number of pre-screening indicators after standardization into categories; randomly selecting two evaluation indicators from the pre-screening indicators and grouping them into one category, while keeping the remaining indicators unchanged, to obtain a total of unified results; finally, determining the total within-class sum of squares corresponding to different indicator classification results based on the within-class sum of squares of the th category of evaluation indicators, and repeating the above process until the number of classifications is ; through the R clustering algorithm, the comprehensive evaluation indicators for the design and review of distribution network projects are divided into 6 categories, including the distribution network design result B1, the coordinated development degree of the distribution network B2, the intelligent level of the distribution network B3, the economic benefits of the distribution network B4, the operation energy efficiency of the distribution network B5, and the application of the distribution network B6; they are set as the first-level indicators in the comprehensive evaluation index system for the design and review of distribution network projects.

[0010] As a preferred solution of the comprehensive evaluation method for thermal power with energy storage projects described in the present invention, wherein: the selection of secondary indicators based on the analytic hierarchy process includes decomposing the evaluation indicators into a linear combination of a few common factors, and the factor analysis model is expressed as: , where represents the th evaluation indicator, represents the common factor of the th evaluation object; m represents the number of indicators, and k represents the number of common factors of the evaluation objects; represents the factor loading,

[0011] As a preferred solution of the comprehensive evaluation method for thermal power distribution and storage engineering described in the present invention, wherein: the secondary indicator layer includes evaluation indicators of different numbers in the secondary indicator layer, and an analysis network structure is formed between different levels; when establishing a comprehensive evaluation system for distribution network engineering, a hierarchical structural model is established for the evaluation target to form a corresponding evaluation system; a corresponding discriminant matrix is constructed, and a judgment matrix between two indicators is established layer by layer; the discriminant matrix It is expressed as: , in, express Relative to The importance index, is the order of the discriminant matrix; solve the attribute weights of the discriminant matrix of each layer, and calculate the geometric mean of the elements of each row of the discriminant matrix: , in, is the discriminant matrix The geometric mean of the row elements, For Equal to 1 to of Find the product; Perform normalization to obtain the corresponding indicator weights , It is expressed as: , in, is the discriminant matrix The geometric mean of the row elements, For Equal to 1 to of Sum; perform consistency test on the established matrix to scalarize the judgment, and the consistency test formula is: , in, is the consistency ratio, is the random consistency indicator, It represents the consistency index; represents the characteristic root, satisfy ; According to the working characteristics of the hierarchical analysis method, when When the discriminant matrix is established, it meets the consistency test requirements; the comprehensive evaluation value of the distribution network project is calculated and sorted, and the final score is calculated according to the weights of each indicator and the scoring results. , expressed as: , Among them, represents the score of a certain evaluation index within the th layer in the hierarchical structure, represents the number of sub-evaluation indicators at the level of any evaluation index, represents the score of the th sub-evaluation indicator of a certain evaluation index at the level; respectively represent the weights of the th sub-evaluation indicator of a certain evaluation index; the correlation between evaluation indexes and common factors is described by the absolute value of the factor loading The value reflects the influence of evaluation index i on the evaluation result. If the value of the evaluation index is large, it should be retained; If the value is small, it should be deleted.

[0012] As a preferred solution of the comprehensive evaluation method for a thermal power energy storage project described in the present invention, among them: the analysis of the evaluation indexes includes constructing an evaluation index system, determining the weights of the evaluation indexes, approximately rounding the weights of the evaluation indexes to ensure that the sum of the weights of the evaluation indexes at the same level is 1; normalizing the weights of the evaluation indexes, and the normalization process is expressed as: , where represents the normalized weight value of the th evaluation index, represents the initial weight value of the th evaluation index after normalization, represents the number of indexes at the level to which the th evaluation index belongs after adjustment, represents the th evaluation index.

[0013] As a preferred solution of the comprehensive evaluation method for a thermal power energy storage project described in the present invention, among them: the process of obtaining the comprehensive evaluation plan for the distribution network includes determining the non-linear relationship between each sample data index and the comprehensive score through a multiple regression analysis algorithm, which is used to calculate the design score of the distribution network project under different sample data, and determining the comprehensive evaluation result; the equation fitting process of the multiple regression analysis is expressed as: , where represents the dependent variable; represents the regression coefficient; represents the independent variable; represents the unmeasurable random error distribution term, which satisfies ; where Represents the theoretical regression equation; when obtaining the relevant evaluation indicators of distribution network engineering design and establishing a multivariate regression equation based on it, and calculating the comprehensive scores corresponding to each evaluation indicator and weight of the project, it is necessary to use Independent tests to determine sample data The estimated regression coefficients , expressed as: , After using the multivariate regression analysis algorithm to obtain the nonlinear relationship expression between each sample data indicator and the comprehensive score, the comprehensive evaluation score of the distribution network engineering project under different evaluation indicators and corresponding weights is determined.

[0014] Another object of the present invention is to provide a comprehensive evaluation system for thermal power distribution and storage projects.

[0015] To solve the above technical problems, the present invention provides the following technical solutions: a comprehensive evaluation system for thermal power distribution and storage projects, comprising: selecting a primary indicator module, formulating comprehensive evaluation indicators for distribution network projects, and selecting primary indicators based on a clustering algorithm; selecting a secondary indicator module, and selecting secondary indicators based on a hierarchical analysis method; and an analysis and calculation module, analyzing the evaluation indicators to obtain a comprehensive evaluation plan for the distribution network process.

[0016] The present invention provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and is characterized in that when the processor executes the computer program, the steps of the comprehensive evaluation method for a thermal power distribution and storage project are implemented.

[0017] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the steps of the comprehensive evaluation method for a thermal power distribution and storage project are implemented.

[0018] Beneficial effects of the present invention: The average information contribution rate of each evaluation index in the distribution network engineering evaluation index system constructed by the present invention to the number of selected indicators can reach more than 90%. The obtained comprehensive evaluation system can provide feasibility guidance for distribution network planning and transformation, thereby improving the electricity consumption experience and the economic benefits of power companies, and has broad application potential and economic value in the actual design of distribution network engineering. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 The overall flowchart of a comprehensive evaluation method for a thermal power with energy storage project provided by an embodiment of the present invention.

[0021] Figure 2 The calculation result diagram of the weight of the first-level evaluation indicators of a comprehensive evaluation method for a thermal power with energy storage project provided by an embodiment of the present invention.

[0022] Figure 3 The comprehensive evaluation result diagram of a comprehensive evaluation method for a thermal power with energy storage project provided by an embodiment of the present invention.

[0023] Figure 4 The comparison diagram of the results of different evaluation methods of a comprehensive evaluation method for a thermal power with energy storage project provided by an embodiment of the present invention.

[0024] Figure 5 The comparison result diagram of the rationality of indicators of a comprehensive evaluation method for a thermal power with energy storage project provided by an embodiment of the present invention. Specific Embodiments

[0025] To make the above objects, features, and advantages of the present invention more obvious and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the scope of protection of the present invention.

[0026] Example 1, referring to Figure 1 , which is an embodiment of the present invention, provides a comprehensive evaluation method for a thermal power with energy storage project, including: S1: Formulate comprehensive evaluation indicators for the distribution network project, and select the first-level indicators according to the clustering algorithm.

[0027] It should be noted that the comprehensive evaluation of the distribution network project refers to making a comprehensive and overall assessment of the distribution network project system based on the working characteristics and sample data of different aspects and levels. With the exploration and research of the comprehensive evaluation method, the accuracy and efficiency of grasping the overall situation of the distribution network project in the macro dimension are improved. The steps of formulating the comprehensive evaluation indicators for the distribution network project can be divided into: the determination of evaluation indicators, the design of indicator weights, and the evaluation method evaluation steps.

[0028] The evaluation index of distribution network engineering is a sample parameter or abstract concept that comprehensively reflects its different functional characteristics. After all the evaluation indexes are weighted, a clear structure and reasonable stratification evaluation system are formed. When formulating the corresponding evaluation indexes, the principles of pertinence, comprehensiveness, independence, feasibility and practicality should be followed to determine the most accurate and effective comprehensive evaluation system for distribution network engineering, so as to achieve the purpose of fully grasping the operation status of distribution network engineering. According to the data accessibility standard, some indicators that cannot collect data should be deleted when formulating the evaluation index of distribution network engineering. Standardization of the comprehensive evaluation index of distribution network engineering is carried out, including standardization of positive indicators and negative indicators.

[0029] In the present invention Represents the evaluation index, Indicates the evaluation object. Indicates the number of selected indicators after standardization.

[0030] Assume The first evaluation object The standardized value of the evaluation index is , the positive indicators are standardized as follows: , If the value of the indicator is smaller, it means that the comprehensive performance of the distribution network engineering design and review is better. This type of evaluation indicator is defined as a negative indicator, and the standardization process is expressed as: , in, Indicates The first evaluation object Evaluation index value, for In 1 to Take between Minimum value, for In 1 to Take between Maximum value, As the evaluation index, For the evaluation object, Indicates the number of selected indicators after standardization.

[0031] After the comprehensive evaluation of distribution network projects is standardized, the R clustering algorithm is used to cluster the comprehensive evaluation of distribution network projects. The standardized indicators are divided into categories, using and Respectively represent The vector of the sum of squared deviations and sample mean values of the class evaluation index is The sum of squared deviations of the class evaluation indicators is expressed as: , where represents the number of evaluation indicators of the th class of evaluation indicators, represents the sample value vector after standardization of the th evaluation object for the th evaluation indicator.

[0032] The overall sum of squared deviations of the th class of evaluation indicators is expressed as: , where is the overall sum of squared deviations of the th class of evaluation indicators, is the rd class of evaluation indicators.

[0033] The process of the sum of squared deviations clustering method includes, first, dividing the number of sea - selected indicators after standardization into classes.

[0034] Randomly select two evaluation indicators from the sea - selected indicators and group them into one class, keeping the remaining indicators unchanged, obtaining a total of kinds of unified results.

[0035] Finally, determine the overall sum of squared deviations corresponding to different indicator classification results according to the sum of squared deviations of the th class of evaluation indicators, and use the R clustering algorithm to cluster the comprehensive evaluation of the distribution network project in a loop until the number of classifications is .

[0036] It should be noted that the number of classifications is usually set based on human experience. To avoid the subjective influence in the process of setting the number of classifications and improve the scientificity of the clustering results, the present invention uses non - parametric K - W test to test the results after clustering. If the significance level value of each class of evaluation indicators is greater than 0.05, it is considered that there is no obvious difference between the evaluation indicators in the same class, that is, the value is set reasonably. If the significance level value of a certain class of evaluation indicators is less than or equal to 0.05, it is considered that there is an obvious difference between the evaluation indicators in the same class, that is, the value is set unreasonably.

[0037] The comprehensive evaluation indicators for the design and review of distribution network projects are divided into 6 categories through the R clustering algorithm, including the distribution network design result B1, the degree of coordinated development of the distribution network B2, the intelligent level of the distribution network B3, the economic benefits of the distribution network B4, the operation energy efficiency of the distribution network B5, and the application of the distribution network B6. They are set as the first-level indicators in the comprehensive evaluation index system for the design and review of distribution network projects.

[0038] S2: Select the secondary indicators based on the analytic hierarchy process.

[0039] It should be noted that the analytic hierarchy process decomposes the problem into mutually independent constituent factors, combines them according to different levels based on the correlation effects and internal coupling relationships between different factors, and finally establishes an analysis model with a multi-level structure.

[0040] The evaluation indicators are decomposed into a linear combination of a few common factors, and the factor analysis model is expressed as: , Among them, represents the i-th evaluation indicator, represents the common factor of the j-th evaluation object. m represents the number of evaluation indicators, and k represents the number of common factors of the evaluation objects. represents the factor loading, represents the special factor that only affects the indicator .

[0041] According to the internal cycle of the ladder hierarchical structure and the correlation between hierarchical structures, different evaluation objectives included in the comprehensive evaluation process of distribution network project design review are classified and decision-making is carried out according to the target layer, the first-level indicator layer, and the second-level indicator layer.

[0042] The target layer is the comprehensive evaluation result of the distribution network project design review. The first-level indicator layer is the level where the core evaluation indicator categories obtained through data mining clustering analysis methods are located. The second-level indicator layer is the level where the terminal demand evaluation indicators are located.

[0043] There are an unequal number of evaluation indicators in the second-level indicator layer, and an analysis network structure is formed between different levels. When establishing the comprehensive review system for distribution network projects, the analytic hierarchy process includes establishing a hierarchical structure model with clear levels for the evaluation objectives to form a corresponding evaluation system. Constructing a corresponding discriminant matrix and establishing a pairwise judgment matrix layer by layer. The discriminant matrix is expressed as: , Among them, represents 's importance index relative to , is the order of the discriminant matrix.

[0044] Solve the attribute weights of each layer's discrimination matrix by calculating the geometric mean of the elements in each row of the discrimination matrix: , where, is the geometric mean of the elements in the th row of the discrimination matrix, is the product of from 1 to . Product.

[0045] Normalize the calculated to obtain the corresponding index weight , expressed as: , where, is the geometric mean of the elements in the th row of the discrimination matrix, is the sum of from 1 to . Sum.

[0046] Conduct a consistency test on the established matrix to scalarize the judgment. The consistency test formula is: , where, is the consistency ratio, is the random consistency index, represents the consistency index; represents the eigenvalue, satisfies .

[0047] According to the working characteristics of the analytic hierarchy process, when , the established discrimination matrix meets the requirements of the consistency test.

[0048] Calculate the comprehensive evaluation value of the distribution network project and rank it. Calculate the final score according to the weights of each index and the scoring results , expressed as: , where, represents the score of a certain evaluation index within the th layer of the hierarchical structure, represents the number of hierarchical sub-evaluation indexes of any evaluation index, represents the score of the hierarchical sub-evaluation index i of a certain evaluation index, respectively represent a certain evaluation index The weight of the hierarchical sub - evaluation index i.

[0049] Evaluation index The correlation between common factors is described by the absolute value of the factor loading The value size reflects the influence of evaluation index i on the evaluation result. The larger the value of the evaluation index, the more it should be retained. The smaller the value, the more it should be deleted.

[0050] Furthermore, through the factor analysis process, it can be ensured that the initial information is described by a small number of common factors, thereby obtaining the secondary indicators of the comprehensive evaluation index system for the distribution network engineering design and review. Combining the six primary indicators, a comprehensive evaluation index system for the distribution network engineering design and review is constructed.

[0051] S3: Analyze the evaluation indicators to obtain a comprehensive evaluation plan for the distribution network process.

[0052] It should be noted that how to set the index weights is the key to the research of the comprehensive evaluation system, which can be used to reflect the relative importance among the indicators at the same level and has a great impact on the comprehensive evaluation result and the balance of the system. The present invention constructs an evaluation index system, determines the evaluation index weights, and considering factors such as operability, approximately rounds the determined evaluation index weights to ensure that the sum of the evaluation index weight values within the same level is 1.

[0053] Considering that the data collected for each evaluation index is affected by different factors, there are phenomena such as data missing and pollution. In the actual evaluation process, some evaluation indicators need to be abandoned, and the evaluation index weight values are normalized. The normalization process is expressed as: , where represents the normalized weight value of the th evaluation index, represents the initial weight value of the th evaluation index after normalization, represents the number of indicators at the level to which the th evaluation index belongs after adjustment, represents the th evaluation index.

[0054] The non - linear relationship between each sample data index and the comprehensive score is determined through the multiple regression analysis algorithm, which is used to calculate the distribution network engineering design scores under different sample data and determine the comprehensive evaluation result.

[0055] Furthermore, the multiple regression analysis algorithm (MRA) is a mathematical analysis method that processes the statistical relationships between different variables by fitting the observed data equation and is applicable to solving mathematical problems where there is no definite functional relationship between the independent variable and the dependent variable. The multiple regression analysis algorithm can determine the mathematical expression of the relationship between different variables, describe the coupling relationship between the independent variable and the dependent variable using the corresponding regression equation, and thus achieve relatively accurate result prediction.

[0056] The equation fitting process of multiple regression analysis is expressed as: , where, represents the dependent variable. represents the regression coefficient. represents the independent variable. represents the unmeasurable random error distribution term, satisfying . Where represents the theoretical regression equation.

[0057] When obtaining the relevant evaluation indicators of the distribution network engineering design and establishing a multiple regression equation therefrom to calculate the comprehensive scores corresponding to each evaluation indicator and weight of the project, it is necessary to determine the estimated regression coefficient of the sample data through independent tests, which is expressed as: , After obtaining the non-linear relationship expression between each sample data index and the comprehensive score using the multiple regression analysis algorithm, determine the comprehensive evaluation score of the distribution network engineering project under different evaluation indicators and corresponding weights.

[0058] Example 2, referring to Figures 2 to 5 , is the second embodiment of the present invention, which provides a comprehensive evaluation method for thermal power with energy storage projects. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments.

[0059] To verify the actual application performance of the proposed comprehensive evaluation method for distribution network engineering design and review based on data mining, the distribution network engineering of Hebei Southern Network is taken as the experimental target for testing. The selected distribution network engineering projects reach more than ten thousand each year, and there are problems such as few relevant practitioners, a large number of engineering projects, and inability to guarantee the quality of project review.

[0060] Randomly select 10 distribution network engineering projects in 2022 (named Target A - J respectively), conduct experimental tests on the designed evaluation system for distribution network engineering projects, and conduct comprehensive evaluations on them using the method of the present invention. Taking Project A of the 35kV distribution network new construction project in the southern region of Hebei as an example, use the method proposed in the present invention to determine the weight values corresponding to each first - level index. The distribution of the weight coefficients of each index is as follows Figure 2 as shown.

[0061] According to Figure 2 it can be seen that after calculating the weights of each first - level evaluation index using the method proposed in the present invention, the weight of the distribution network design result index (B1) is the highest, reaching 0.24, and the weight of the distribution network economic benefit index (B4) is the lowest, being 0.11. The triangular marks are the weights corresponding to the indexes.

[0062] Based on Figure 2 the calculated results of the evaluation index weights shown above, clean, integrate, and fuse the source data collected during the design process of this distribution network project through processing means such as cleaning, integration, and fusion, and conduct a comprehensive evaluation on it in combination with the current situation of the target area. Set the full score of the evaluation index system to 100 points, and use the multiple regression analysis algorithm to calculate the specific scores. The results are as follows Figure 3 as shown.

[0063] According to Figure 3 it can be seen that the scores of each first - level evaluation index in Project A of the distribution network new construction project are 19.7, 15.36, 10.11, 8.10, 14.80, and 14.93 points respectively. The indexes with the highest score loss are the distribution network intelligent level (B3), distribution network economic benefit (B4), distribution network operation energy efficiency (B5), and distribution network design result evaluation (B1), and the score loss rates are all higher than 20%. This shows that there are certain defects in Project A of the distribution network new construction project in the above four aspects. The reason is that the construction speed of the distribution communication facilities in the project is relatively slow, affecting its intelligent level. At the same time, there are certain limitations in the live working rate and fault recovery time during the construction process, affecting its operation energy efficiency.

[0064] In order to verify the feasibility and effectiveness of the distribution network comprehensive design and evaluation method designed by the present invention, conduct a comparative analysis on the improved Elman feedback - type dynamic neural network algorithm proposed for the reliability assessment of the distribution network based on the improved Elman feedback - type dynamic neural network and the optimal combination weighting method evaluation model proposed for the application of the optimal combination weighting method in the distribution network transformation - taking the distribution network transformation in large cities of less - developed countries as an example, and analyze the effects of different evaluation methods applied to the comprehensive evaluation of the distribution network. Use different evaluation methods to evaluate the 10 selected distribution network engineering design projects, and the comparison values are shown in Table 1, and the comparison results are as follows Figure 4 as shown.

[0065] Table 1 Specific values of different evaluation methods

[0066] according to Figure 4 As shown in Table 1, compared with the improved Elman feedback dynamic neural network algorithm and the optimal combination weighting method, the R clustering, hierarchical analysis method and multivariate regression algorithm proposed in the present invention have a higher consistency with the expected score for the comprehensive design of the distribution network, while the evaluation results obtained by the other two algorithms are relatively far from the expected value. Compared with the other two comparison methods, the evaluation results of the method proposed in the present invention are more in line with the actual situation and have practical value.

[0067] Taking the average information contribution rate of all evaluation indicators in the constructed distribution network engineering comprehensive evaluation index system to the number of selected indicators as the standard, the prior art improved Elman feedback dynamic neural network method and the prior art optimal combination weighting method method and the method proposed by the present invention are compared in rationality. Figure 5 shown.

[0068] according to Figure 5 It can be seen that the evaluation index system constructed based on R clustering, hierarchical analysis method and multiple regression algorithm has an average information contribution rate of more than 90% for the number of selected indicators, which is about 7% higher than that of the other two algorithms. The comprehensive evaluation index system of distribution network designed by the present invention is more reasonable and has greater application potential.

[0069] Embodiment 3 is the third embodiment of the present invention, which is different from the first two embodiments in that: If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0070] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a definable sequence list of executable instructions for implementing logical functions, which can be specifically implemented in any computer-readable medium for use by an instruction execution system, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device), or used in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by or in combination with an instruction execution system, apparatus, or device.

[0071] More specific examples (non-exhaustive list) of computer-readable media include the following: an electrical connection part (electronic device) having one or more wirings, a portable computer disk cartridge (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0072] It should be understood that the various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application-specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0073] Example 4, the fourth embodiment of the present invention, this embodiment provides a comprehensive evaluation system for thermal power distribution and energy storage projects, including: A first-level index selection module, formulating comprehensive evaluation indexes for distribution network projects, and selecting first-level indexes according to a clustering algorithm.

[0074] A second-level index selection module, selecting second-level indexes based on the analytic hierarchy process.

[0075] An analysis and calculation module, analyzing the evaluation indexes to obtain a comprehensive evaluation plan for the distribution network process.

[0076] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. An integrated evaluation method for thermal power with energy storage project, characterized in that: including, formulating comprehensive evaluation indicators for distribution network projects and selecting first-level indicators according to the clustering algorithm; using the R clustering method to cluster the standardized indicators, dividing each indicator into multiple evaluation indicator categories. The R clustering method divides categories based on the sum of squared deviations and uses non-parametric statistical test methods to test the significance level of the clustering results to determine the first-level indicators of the distribution network project; selecting second-level indicators based on the analytic hierarchy process; analyzing the evaluation indicators to obtain a comprehensive evaluation plan for the distribution network process; the analysis of the evaluation indicators includes obtaining sample data of the distribution network project, constructing a regression equation using the multiple regression analysis algorithm based on the determined evaluation indicators and corresponding weights, estimating the regression coefficients through independent testing, calculating the comprehensive score according to the regression equation, and ranking each evaluation object.

2. The comprehensive evaluation method for a thermal power energy storage project according to claim 1, wherein: the steps of formulating comprehensive evaluation indicators for distribution network projects include the determination of evaluation indicators, the design of indicator weights, and the evaluation of evaluation methods; standardizing the comprehensive evaluation indicators for distribution network projects, including the standardization of positive indicators and negative indicators; Suppose the -th evaluation object's -th standardized value of the evaluation index is , and the implementation of the standardization process for positive indicators is expressed as ， if the value of the indicator is small, it indicates good comprehensive performance in the design and review of the distribution network project, and such evaluation indicators are defined as negative indicators. The standardization process is expressed as ， Among them, represents the th evaluation index value of the th evaluation object, is to take the minimum value between 1 and ; is to take the maximum value between 1 and ; is the evaluation index, is the evaluation object, represents the number of pre-screening indicators after standardization.

3. The comprehensive evaluation method for a thermal power energy storage project according to claim 2, wherein: The selection of the first-level indicators according to the clustering algorithm includes, after the standardized processing of the comprehensive evaluation of the distribution network project, using the R clustering algorithm to cluster the comprehensive evaluation of the distribution network project, and standardized indicators are divided into categories, and and respectively represent the sample mean vectors of the sum of squared deviations of the th category of evaluation indicators. Then the sum of squared deviations of the th category of evaluation indicators is expressed as ， Among them, represents the number of evaluation indicators for the category of evaluation indicators, represents the vector of sample values after standardization of the th evaluation indicator of the th evaluation object; The total within-class sum of squares of the evaluation index categories is expressed as ， Among them, is the total sum of squared deviations of all evaluation index categories, is the th evaluation index category; The process of the sum of squared deviations clustering method includes, first, dividing the number of pre-screening indicators after standardization into categories; From Randomly select two evaluation indicators from the selections of the audition indicators and classify them into one category, while keeping the remaining indicators unchanged, resulting in a total of kinds of unified results; Finally, according to the sum of squared deviations of the -type evaluation indicators, determine the overall sum of squared deviations corresponding to the classification results of different indicators. Loop the above process until the number of classifications is ; classifying the comprehensive evaluation indicators for the design and review of distribution network projects into 6 categories through the R clustering algorithm, including the distribution network design result B1, the coordinated development degree of the distribution network B2, the intelligent level of the distribution network B3, the economic benefits of the distribution network B4, the operation energy efficiency of the distribution network B5, and the application of the distribution network B6; these are set as the first-level indicators in the comprehensive evaluation index system for the design and review of distribution network projects.

4. The comprehensive evaluation method for a thermal power energy storage project according to claim 3, wherein: the selection of second-level indicators based on the analytic hierarchy process includes decomposing the evaluation indicators into a linear combination of a few common factors. The factor analysis model is expressed as ， Among them, represents the i-th evaluation index, represents the common factor of the j-th evaluation object; m represents the number of indicators, and k represents the number of common factors of the evaluation objects; represents the factor loading, represents the special factor that only affects the indicator ; according to the internal cycle of the ladder hierarchy structure and the correlation between hierarchy structures, the different evaluation objectives included in the comprehensive evaluation process of the distribution network project design review are hierarchically decision-making according to the target layer, the first-level indicator layer, and the second-level indicator layer; the target layer is the comprehensive evaluation result of the distribution network project design review, the first-level indicator layer is the level where the core evaluation indicator categories obtained by the data mining clustering analysis method are located, and the second-level indicator layer is the level where the terminal demand evaluation indicators are located.

5. The comprehensive evaluation method for a thermal power with energy storage project according to claim 4, characterized in that: the second-level indicator layer includes an unequal number of evaluation indicators within the second-level indicator layer, forming an analysis network structure between different levels; when establishing the comprehensive review system for distribution network projects, a hierarchical and clear structure model is established for the evaluation objectives to form a corresponding evaluation system; Construct a corresponding discrimination matrix and establish a judgment matrix between two indicators layer by layer; the discrimination matrix is expressed as ， Among them, represents the importance index relative to ; is the order of the discrimination matrix. solving the attribute weights of each layer's discriminant matrix and calculating the geometric mean of the elements in each row of the discriminant matrix; ， Among them, is the geometric mean of the elements in the th row of the discrimination matrix, is the product of from 1 to equal to ; The calculated is normalized to obtain the corresponding index weights , which is expressed as ， Among them, is the geometric mean of the elements in the th row of the discrimination matrix, is for equal to the sum from 1 to of ; conducting a consistency test on the established matrix to scalarize the judgment. The consistency test formula is ， Among them, is the consistency ratio, is the random consistency index, represents the consistency index; represents the eigenvalue, satisfies ; According to the working characteristics of the analytic hierarchy process, when the established discriminant matrix meets the requirements of consistency test. Calculate the comprehensive evaluation value of the distribution network project and sort it, and calculate the final score according to the weights of each index and the scoring results , expressed as ， Among them, represents the score of a certain evaluation index within the layer in the hierarchical structure, represents the number of sub-evaluation indexes at the hierarchical level of any evaluation index, represents the score of the sub-evaluation index i at the hierarchical level of a certain evaluation index, respectively represent the weights of the sub-evaluation index i at the hierarchical level of a certain evaluation index; Evaluation Index The correlation between common factors is described by the absolute value of the factor loading The value reflects the impact of evaluation index i on the evaluation result. For an evaluation index with a large value, it should be retained; with a small value, it should be deleted.

6. The comprehensive evaluation method for a thermal power energy storage project according to claim 4, characterized in that: the analysis of the evaluation indicators includes constructing an evaluation indicator system, determining the weights of the evaluation indicators, approximately rounding the weights of the evaluation indicators to ensure that the sum of the weights of the evaluation indicators within the same level is 1; performing normalization processing on the weights of the evaluation indicators. The normalization process is expressed as ， wherein represents the normalized weight value of the th evaluation index, represents the initial normalized weight value of the th evaluation index, represents the number of indexes at the level to which the th evaluation index belongs after adjustment, represents the th evaluation index.

7. The comprehensive evaluation method for a thermal power energy storage project according to claim 4, wherein: The obtained comprehensive evaluation scheme for the distribution network process includes determining the non-linear relationship between each sample data index and the comprehensive score through the multiple regression analysis algorithm, which is used to calculate the design score of the distribution network project under different sample data and determine the comprehensive evaluation result. The equation fitting process of the multiple regression analysis is expressed as ， Among them, represents the dependent variable; represents the regression coefficient; represents the independent variable; represents the unmeasurable random error distribution term, satisfying ; where represents the theoretical regression equation; When obtaining the relevant evaluation indicators of the distribution network engineering design and establishing a multiple regression equation based on this, and calculating the comprehensive scores corresponding to each evaluation indicator and weight of the project, it is necessary to determine the sample data through independent detections of the estimated regression coefficients , expressed as ， After obtaining the non-linear relationship expression between each sample data index and the comprehensive score by using the multiple regression analysis algorithm, the comprehensive evaluation score of the distribution network project under different evaluation indexes and corresponding weights is determined.

8. A comprehensive evaluation system for thermal power with energy storage project, which applies a comprehensive evaluation method for thermal power with energy storage project as described in any one of claims 1 to 7, characterized in that, It includes: Select the first-level index module, formulate the comprehensive evaluation index of the distribution network project, and select the first-level index according to the clustering algorithm; Select the second-level index module and select the second-level index based on the analytic hierarchy process; The analysis and calculation module analyzes the evaluation indexes to obtain the comprehensive evaluation scheme for the distribution network process.

9. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of a comprehensive evaluation method for a thermal power energy storage project according to any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of a comprehensive evaluation method for a thermal power energy storage project according to any one of claims 1 to 7.

Images