Weight combination optimization transformer quality assurance capability assessment method

Abstract

The invention discloses a transformer quality assurance capability evaluation method based on weight combination optimization. The method comprises the following steps: 1, obtaining indexes for evaluating the quality assurance capability of a transformer; 2, determining a relative importance weight of a transformer quality assurance capability evaluation index based on a superior sequence diagram method; 3, acquiring sample data corresponding to the transformer quality assurance capability evaluation index, and determining an information amount weight of the index based on a variable coefficient method; 4, establishing an optimization model fusing two transformer quality assurance capability evaluation index weights; 5, solving the weight optimization model; 6, calculating evaluation values of all the transformer samples; and 7, selecting a supplier of the transformer sample product corresponding to the highest evaluation value. Based on the principle of minimum deviation of evaluation values under subjective and objective empowerment, the defects of a single empowerment method in use are reduced, and the index weight obtained by the method is obviously higher than that of an original detailed rule in the discrimination degree of an evaluation result, so that a reference is provided for more accurately evaluating the quality of the transformer.

Description

Technical Field

[0001] The invention belongs to the technical field of power transmission and transformation equipment quality supervision, and in particular relates to a transformer quality assurance capability evaluation method with weight combination optimization. Background Art

[0002] As the core equipment for voltage conversion in the power system, the quality of transformers directly affects the safe and stable operation of large power grids and the power supply reliability of regional power grids. The transformer quality assurance capability assessment evaluates the consistency of the product design of transformer equipment with the relevant technical requirements in the current national standards, industry standards, enterprise standards, and accident prevention measures, and evaluates the transformer product design in the electric field design, winding impedance short-circuit capacity design, magnetic field design, temperature field design, assembly design, seismic design, mechanical strength design and other processes. It is an important research topic to improve the quality and safety level of transformer equipment access to the grid, and is widely used in the national grid system. At present, the transformer quality assurance capability assessment adopts a fixed weighting method in the weight allocation of evaluation items such as electric field design, winding impedance short-circuit capacity design, magnetic field design, temperature field design, assembly design, seismic design, and mechanical strength design. It mainly relies on historical evaluation experience and manual subjective opinions, lacks systematicity and scientificity, and is difficult to meet the needs of new power systems for transformer quality assessment and selection. In order to more truly and comprehensively reflect the actual situation of transformers and evaluate the quality assurance capability of transformers, many standardization organizations have stipulated relevant current standards for power transformers, but at the same time they are also faced with the problem of how to reasonably allocate the weights of different evaluation indicators. Unreasonable weight allocation may lead to distortion of evaluation results and affect the accuracy of transformer quality assurance capability evaluation. Therefore, it is particularly important to study a scientific optimization method for the weights of transformer quality assurance capability evaluation indicators.

[0003] There are several problems in the existing transformer quality assurance capability evaluation indicators and their weight distribution. First, the indicator weight distribution is highly subjective. The existing subjective weighting methods mostly rely on expert experience or historical data, with large subjective factors. Different evaluators may come up with different weight distribution schemes. The weights given by the existing objective weighting methods are absolutely objective, which may conflict with the economic (or technical) value of the existing indicators. At the same time, once the sample changes, the corresponding weights will also change accordingly. All of these lead to poor consistency and repeatability of the evaluation results. Second, the existing methods lack sufficient theoretical support and data verification when determining the weights, making the weight setting not scientific and reasonable. Third, the results of the evaluation work carried out according to the existing evaluation rules using the existing indicator weight distribution scheme show that the final evaluation values ​​of the transformers of various suppliers are not highly differentiated, and cannot provide a more valuable reference for the selection of transformer suppliers. The existence of these problems limits the efficiency and results of the transformer quality assurance capability evaluation. Summary of the invention

[0004] The purpose of the present invention is to solve the above-mentioned shortcomings and defects, and proposes a transformer quality assurance capability evaluation method with weight combination optimization, aiming to determine the most appropriate evaluation index weights, thereby improving the accuracy and reliability of the quality assurance capability evaluation results, so as to ensure that a transformer with reliable and stable operation can be selected.

[0005] In order to achieve the above-mentioned purpose, the present invention adopts the following technical scheme:

[0006] The characteristic of the transformer quality assurance capability evaluation method with weight combination optimization of the present invention is that it comprises the following steps:

[0007] S1. Obtain n evaluation indicators for evaluating the quality assurance capability of transformers, where represents the jth transformer quality assurance capability evaluation index, n represents the total number of evaluation indicators, ;

[0008] S2. Determine the relative importance weight vector of transformer quality assurance capability evaluation indicators based on the priority diagram method ;in, express The relative importance weight of , T represents transposition;

[0009] S3. Obtain sample data corresponding to the transformer quality assurance capability evaluation index, and determine the information weight vector of the transformer quality assurance capability evaluation index based on the coefficient of variation method. ;in, express The corresponding information weight;

[0010] S4. Use equations (8) and (9) to establish a weight optimization model that integrates relative importance weight and information weight:

[0011] (8)

[0012] (9)

[0013] In formula (8), is the weight coefficient, express The corresponding weights; represents the objective function of the combined weight optimization model, represents the weight vector; express The corresponding standardized i-th transformer sample;

[0014] S5. Solve the weight optimization model to obtain the optimized weight vector ,in, express The corresponding optimized weights;

[0015] S6. Calculate the quality assurance capability evaluation value of the i-th transformer sample using formula (10): , thus obtaining the quality assurance capability evaluation value of m transformer samples;

[0016] (10)

[0017] S7. After sorting the quality assurance capability evaluation values ​​of the m transformer samples in descending order, select the supplier of the transformer samples corresponding to the first s quality assurance capability evaluation values ​​as the transformer supplier with the best quality.

[0018] The transformer quality assurance capability evaluation method with weight combination optimization described in the present invention is also characterized in that S2 includes:

[0019] S2.1. Initialization of transformer quality assurance capability evaluation indicators The average importance score , thus obtaining the average score of n indicators;

[0020] S2.2. Compare the average scores of n transformer quality assurance capability evaluation indicators to obtain the relative importance of each indicator, thereby constructing a relative importance judgment matrix with a dimension of n×n;

[0021] S2.3. After adding up all the column elements of each row of the relative importance judgment matrix, the relative importance vector of n transformer quality assurance capability evaluation indicators is obtained: and the sum of relative importance ,in, represents the j-th transformer quality assurance capability evaluation index The relative importance vector of ; and = ,in, is the element in the rth row and jth column of the relative importance judgment matrix, representing the rth transformer quality assurance capability evaluation index and the jth transformer quality assurance capability evaluation index The relative importance between them;

[0022] S2.4, using formula (1) to obtain The relative importance weight of , thus obtaining the relative importance weight vector of n transformer quality assurance capability evaluation indicators: ;

[0023] (1)

[0024] In formula (1), ,and , ; Represents the rth transformer quality assurance capability evaluation index The relative importance vector of .

[0025] Furthermore, S3 includes:

[0026] S3.1. Obtain the j-th transformer quality assurance capability evaluation index The corresponding m transformer samples, thus constructing a dimension of The transformer sample matrix ,in, represents the j-th transformer quality assurance capability evaluation index The corresponding i-th transformer sample;

[0027] S3.2, using formula (2) Perform forward processing to obtain the jth transformer quality assurance capability evaluation index The corresponding i-th transformer sample after forward processing , thus obtaining the transformer sample matrix after forward processing ';

[0028] (2)

[0029] In formula (2), k is the coefficient, express The corresponding m transformer samples, express The absolute value of

[0030] S3.3, using formula (3) After standardization, we get The corresponding standardized i-th transformer sample , thus obtaining the standardized transformer sample matrix R;

[0031] (3)

[0032] S3.4, Calculate using formula (4) The mean of the corresponding m transformer samples , calculated using formula (5) The standard deviation of the corresponding m transformer samples , and then use formula (6) to calculate The coefficient of variation ;

[0033] (4)

[0034] (5)

[0035] (6)

[0036] S3.5, Calculate using formula (7) The corresponding information weight , thus obtaining the information weight vector of n transformer quality assurance capability evaluation indicators: ;

[0037] (7)

[0038] In formula (7), ,and , .

[0039] Further, the step S5 is solved according to the following steps to obtain :

[0040] S5.1. Calculation The sum of the square values ​​of the corresponding m transformer samples , thus obtaining the sum of the square values ​​of the n evaluation indicators, and then using formula (11) to obtain the diagonal matrix , :

[0041] (11)

[0042] S5.2. Calculate according to formula (8) Deviation of the estimated value , thereby calculating the deviation of n evaluation indicators, and then using formula (12) to construct the vector :

[0043] (12)

[0044] S5.3, using formula (13) to obtain :

[0045] (13)

[0046] In formula (13), Represents a column vector of dimension n where all elements are 1.

[0047] An electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the transformer quality assurance capability assessment method, and the processor is configured to execute the program stored in the memory.

[0048] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program executes the steps of the transformer quality assurance capability evaluation method when the computer program is executed by a processor.

[0049] Compared with the prior art, the present invention has the following beneficial effects:

[0050] 1. The present invention takes into account the defects of the weighting methods of various indicators in the existing transformer quality assurance capability evaluation process. In the process of optimizing the transformer quality assurance capability evaluation indicator weight system, a subjective and objective combined weighting method is adopted to balance the indicator value and the specific data value, reducing the deviation caused by a single method. This method proposes improvement suggestions and scientific references for the optimization of the transformer quality assurance capability evaluation indicator system, which will help transformer users to more accurately evaluate the design quality of the transformer and discover potential problems in a timely manner.

[0051] 2. The method used in the present invention promotes the flexibility of the transformer quality assurance capability assessment process. The proposed method can be flexibly adjusted according to the actual conditions of transformers of different models and suppliers. It can not only quickly adapt to the application of new technologies in products, but also perform personalized weighting for different types of transformers, and has a wide range of applicable scenarios. At the same time, for the evaluator, this method reduces computing resources and time consumption, and enhances the efficiency and convenience of transformer quality assurance capability assessment.

[0052] 3. Based on the existing transformer quality assurance capability evaluation scheme, the present invention takes into account improving the differentiation of evaluation results for different evaluation objects. By combining the historical cognition of the relative importance of transformer quality assurance capability evaluation indicators within the enterprise, collecting and utilizing the statistical characteristics of the data, the derived indicator weights can more finely distinguish the differences in quality assurance capabilities of different transformers, and strengthen the capture of specific characteristics between transformer product designs from different suppliers. This ensures that the transformer quality assurance capability evaluation indicators are both scientific and close to actual needs in the weighting process, and at the same time can also improve the differentiation of evaluation results for transformers from different suppliers, which is more conducive to the selection of the evaluator. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A flow chart of a transformer quality assurance capability evaluation method for weight combination optimization of the present invention;

[0054] Figure 2 It is a flow chart of determining the weights of transformer quality assurance capability evaluation indicators based on the priority diagram method of the present invention;

[0055] Figure 3 The present invention provides a flow chart for determining transformer quality assurance capability evaluation index weights based on the coefficient of variation method. DETAILED DESCRIPTION

[0056] In this embodiment, a transformer quality assurance capability evaluation method with weight combination optimization is proposed, such as Figure 1 As shown, first, according to the existing "Implementation Rules for Technical Conformity Assessment of 220 kV Transformers of State Grid Corporation of China" and "Implementation Rules for Technical Conformity Assessment of 330~750 kV Transformers of State Grid Corporation of China", etc., appropriate indicators are determined; secondly, according to the determined indicators, the relative importance weights of the indicators are determined in combination with the superiority diagram method; then, the sample data corresponding to the transformer quality assurance capability assessment indicators are obtained, and the information weight of the indicators is determined in combination with the coefficient of variation method; then, on the basis of using the superiority diagram method and the coefficient of variation method to obtain the respective weight optimization results, based on the calculation in the sense of least squares, the principle of minimizing the deviation of the evaluation value under the fusion method and the subjective and objective weighting is used to calculate the optimized indicator weight for each evaluation indicator under the combined evaluation method; then, according to the calculated optimization weight corresponding to each indicator, the evaluation value of each transformer sample to be evaluated is calculated; finally, the appropriate transformer supplier is selected according to the calculated evaluation value. Specifically, the steps of the transformer quality assurance capability assessment method of weight combination optimization are as follows:

[0057] S1. Obtain n indicators for evaluating the quality assurance capability of transformers, where: represents the jth indicator used to evaluate the quality assurance capability of the transformer, n represents the total number of indicators, In this embodiment, the quality assurance capability indicator set for evaluating n transformers is determined based on the existing "Implementation Rules for Technical Conformity Assessment of 220 kV Transformers of State Grid Corporation of China" and "Implementation Rules for Technical Conformity Assessment of 330-750 kV Transformers of State Grid Corporation of China".

[0058] In this embodiment, indicators are selected based on four sources: quality management theory, quality problem traceability, supplier best practices, and company quality control requirements, and 13 indicators for evaluating the quality assurance capability of the transformer are obtained, including: poor design selection, electric field design, winding impedance short-circuit capacity design, magnetic field design, temperature field design, assembly design, seismic design, mechanical strength design, DC bias magnetic tolerance design, overload capacity design, non-electrical protection design, design process management, and R&D design tools.

[0059] S2, such as Figure 2 As shown, the relative importance weight vector of transformer quality assurance capability evaluation indicators is determined based on the priority diagram method;

[0060] S2.1. According to the use scenarios and requirements of the priority diagram method, it is necessary to obtain the scoring data including the relative importance of the evaluation party's understanding of the quality assurance capability evaluation indicators of n transformers and initialize the indicators The average score , thus obtaining the average score of n indicators;

[0061] S2.2. Compare the average scores of n transformer quality assurance capability evaluation indicators to obtain the relative importance of each indicator. The specific method is to establish a judgment matrix for the comparison of different indicators in pairs. The first row and the first column of the matrix are all indicators listed in sequence. The numbers in the squares where the rows and columns intersect represent the results of the comparison of the indicators in pairs. Among them, "0" and "1" are used to represent the importance of the indicators: "1" means that the indicator is relatively more important in the comparison of the two indicators, and "0" means that the indicator is relatively less important. If the comparison between the two indicators is equal, the score can be 0.5, that is, for a certain two indicators and ,like ,but The corresponding row and The score of the cell where the corresponding columns intersect should be 1. The corresponding columns and The score of the cell crossed by the corresponding rows should be 0; if ,but The corresponding row and The cells where the corresponding columns intersect, The corresponding columns and The scores of cells crossed by corresponding rows are all 0.5, thereby constructing a relative importance judgment matrix with a dimension of n×n.

[0062] S2.3. Add up all the column elements of each row of the judgment matrix to get the total superiority number of n indicators. and the total superiority number ,in, represents the j-th transformer quality assurance capability evaluation index The relative importance vector of ; and = ,in, is the element in the rth row and jth column of the relative importance judgment matrix, representing the rth transformer quality assurance capability evaluation index and the jth transformer quality assurance capability evaluation index The relative importance between them.

[0063] S2.4, using formula (1) to obtain The relative importance weight of , thus obtaining the relative importance weight vector of n transformer quality assurance capability evaluation indicators: ;

[0064] (1)

[0065] In formula (1), ,and , , Represents the rth transformer quality assurance capability evaluation index The relative importance vector of .

[0066] S3, such as Figure 3 As shown, sample data corresponding to the transformer quality assurance capability evaluation index is obtained, and the information weight vector of the transformer quality assurance capability evaluation index is determined based on the coefficient of variation method;

[0067] S3.1. According to the scenarios and requirements of the coefficient of variation method, it is necessary to obtain the jth transformer quality assurance capability evaluation index from the original data of each transformer quality assurance capability evaluation index. The corresponding m transformer samples.

[0068] S3.1.1. For the 10 evaluation indicators obtained for evaluating the quality assurance capability of 220kV-750kV transformer products, including: poor design selection, winding impedance short-circuit capability design, temperature field design, assembly design, seismic design, mechanical strength design, DC bias magnetic tolerance capability design, overload capability design, non-electrical protection design, and design process management indicators, use the document inspection method to obtain specific sample data; among them, the document inspection method: review and item j of the transformer quality assurance capability evaluation indicator Relevant design data are obtained to obtain the corresponding data under each evaluation index; further, within an evaluation cycle, two contracts are drawn for transformers of different voltage levels, and all contracts that are less than two are drawn, and the transformer reports corresponding to the two contracts are checked to obtain specific data. For example, for the poor design selection index, the transformer assembly design report can be checked to determine whether there is a poor design selection and obtain the sample data corresponding to the index. The same is true for other indicators.

[0069] S3.1.2. For the three evaluation indicators for evaluating the quality assurance capability of 220kV-750kV transformer products, including electric field design, magnetic field design, and R&D design tool indicators, specific sample data are obtained by using document inspection and on-site verification methods; among which, the on-site verification method is: on-site review and the jth transformer quality assurance capability evaluation indicator The actual design performance and key parts of the relevant equipment to obtain the corresponding sample data under each evaluation indicator;

[0070] S3.1.3, based on the acquired data, the original sample data matrix of the evaluation index is composed of poor design selection, electric field design, winding impedance short-circuit capacity design, magnetic field design, temperature field design, assembly design, seismic design, mechanical strength design, DC bias magnetic tolerance design, overload capacity design, non-electrical protection design, design process management, and R&D design tools, thus constructing a dimension of The original sample data matrix of the evaluation index of the transformer sample ,in, Represents the j-th transformer quality assurance capability evaluation index The data corresponding to the i-th transformer sample.

[0071] S3.2. In this embodiment, in order to convert all indicators into positive indicators, all indicators need to be positively converted. Then, formula (2) is used to convert Perform positive processing to obtain the jth evaluation index after processing The corresponding i-th transformer sample , thus obtaining the data matrix of the evaluation index of the processed transformer samples ';

[0072] (2)

[0073] In formula (2), k is a coefficient, and its value is generally 0.1 or 0.2; Represents the jth evaluation indicator The corresponding m transformer samples, express The absolute value of .

[0074] S3.3. In this embodiment, for different indicators, the units of the corresponding data may be different, so it is necessary to eliminate the influence of different units through data standardization so that the evaluation values ​​of all sample data can be compared. After standardization, the jth evaluation index is obtained The specific data of the corresponding i-th transformer sample , thus obtaining the standardized transformer sample data matrix R;

[0075] (3)

[0076] S3.4. Based on the standardized data, use formula (4) to calculate the jth evaluation index: The mean of the specific data of the corresponding m transformer samples to be evaluated , use formula (5) to calculate the jth evaluation index The standard deviation of the specific data of the corresponding m transformer samples to be evaluated , and then use formula (6) to calculate the jth evaluation index The coefficient of variation ;

[0077] (4)

[0078] (5)

[0079] (6).

[0080] S3.5. Based on the coefficient of variation corresponding to each indicator, use formula (7) to calculate the jth evaluation indicator: The corresponding information weight , so the information weight vector is , where T represents transposition, and the calculated information weight can highlight the relative change range of each indicator, that is, the degree of variation;

[0081] (7)

[0082] In formula (7), ,and , .

[0083] S4. Establish and solve the combined weight optimization model integrating relative importance weight and information weight in the sense of least squares to obtain the weight vector of the optimized transformer quality assurance capability evaluation index. ,The specific steps of building a weight optimization model are as follows:

[0084] S4.1. According to the scenarios and requirements of the least squares method, the relative importance weight vector reflecting the value of the indicator is obtained. and the information weight vector reflecting the information content of the indicator Then, using equations (8) and (9), we can get the deviation between the fusion weighting method and the superiority graph method and the coefficient of variation method:

[0085] (8)

[0086] (9)

[0087] S4.2. In order to minimize the difference between the combined weight and the weight obtained by using the two methods separately, the objective function of the global model for optimizing the weight of the transformer quality assurance capability evaluation index is obtained using formula (10): :

[0088] (10)

[0089] Formula (10) indicates that for all indicators of all evaluation objects, the deviation between the evaluation values ​​under the fusion method and the two weighting methods should be as small as possible. is the weight coefficient of the main and objective weighting methods, indicating the proportion of the weighting of the priority diagram method and the coefficient of variation method in the combination weight. , indicating that the weight obtained by the coefficient of variation method accounts for a larger proportion, and the result is more affected by objective factors. When , the weight obtained by the priority diagram method accounts for a large proportion, and by controlling The value of can allow users to flexibly adjust the weight optimization ratio according to the specific conditions of indicators and data, and obtain weight optimization results that are more in line with the actual situation;

[0090] S4.3. Using equations (11) and (12), the global model for optimizing the weights of transformer quality assurance capability evaluation indicators is obtained:

[0091] (11)

[0092] (12)

[0093] Combining the Lagrangian function and the matrix equation solution method, the model given by equation (11) and equation (12) can be solved;

[0094] S4.4. The specific steps for solving the global model are as follows:

[0095] S4.4.1. Based on the given objective function, use formula (13) to make the Lagrangian function:

[0096] (13)

[0097] S4.4.2. Use equations (14) and (15) to and Find the partial derivatives separately and set them both equal to 0:

[0098] (14)

[0099] (15)

[0100] S4.4.3. Calculate the jth evaluation index The sum of the square values ​​of the corresponding m transformer samples to be evaluated , thus obtaining the sum of the square values ​​of the data of the m transformer samples to be evaluated under the n evaluation indicators, and then using formula (11) to obtain the diagonal matrix , ;

[0101] (16)

[0102] S4.4.4. According to formula (11), for the jth evaluation index , calculate the deviation of its evaluation value , and then calculate the deviation of n evaluation indicators, and then use formula (17) to construct the vector :

[0103] (17)

[0104] S4.3. Use equation (18) to transform equation (14) and equation (15) into the form of matrix equations, and use equation (19) to obtain ,in, express The corresponding optimized weights;

[0105] (18)

[0106] (19)

[0107] In formula (19), Represents a column vector of dimension n where all elements are 1.

[0108] S5. Calculate the evaluation value of the i-th transformer sample to be evaluated using formula (20): , thereby obtaining the evaluation values ​​of m transformer samples to be evaluated. The calculated evaluation values ​​make transformer products from different suppliers comparable, and the higher the evaluation value, the better the quality of the product;

[0109] (20)

[0110] S6. After sorting the evaluation values ​​of the m transformer samples to be evaluated in descending order, the suppliers of the transformer sample products corresponding to the first s evaluation values ​​are selected as the transformer suppliers with the best quality according to the actual needs within the organization.

[0111] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0112] In this embodiment, a computer-readable storage medium stores a computer program on the computer-readable storage medium, and the computer program executes the steps of the above method when executed by a processor.

[0113] In summary, based on the existing evaluation rules, the present invention takes into account that the determination of existing indicator weights is highly subjective and has no clear scientific basis, which leads to the phenomenon that the final evaluation results have low discrimination in actual evaluation work, and the problem that the traditional subjective and objective weighting methods have defects when used alone. A combined optimization weighting method is adopted to optimize the weights of transformer quality assurance capability evaluation indicators. According to the optimized indicator weights, the appropriate transformer supplier can be selected by comparing the evaluation values ​​of each transformer sample to be evaluated. This method applied to transformer quality assurance capability evaluation is more scientific, improves the discrimination of evaluation results, and provides a scientific basis for decision makers to select high-quality transformer suppliers.

Claims

1. A method for evaluating transformer quality assurance capability with weight combination optimization, characterized in that: The steps include: S1. Obtain n evaluation indicators for evaluating the quality assurance capability of transformers, where represents the jth transformer quality assurance capability evaluation index, n represents the total number of evaluation indicators, ; S2. Determine the relative importance weight vector of transformer quality assurance capability evaluation indicators based on the priority diagram method ;in, express The relative importance weight of , T represents transposition; S3. Obtain sample data corresponding to the transformer quality assurance capability evaluation index, and determine the information weight vector of the transformer quality assurance capability evaluation index based on the coefficient of variation method. ;in, express The corresponding information weight; S4. Use equations (8) and (9) to establish a weight optimization model that integrates relative importance weight and information weight: (8) (9) In formula (8), is the weight coefficient, express The corresponding weights; represents the objective function of the combined weight optimization model, represents the weight vector; express The corresponding standardized i-th transformer sample; S5. Solve the weight optimization model to obtain the optimized weight vector ,in, express The corresponding optimized weights; S6. Calculate the quality assurance capability evaluation value of the i-th transformer sample using formula (10): , thus obtaining the quality assurance capability evaluation value of m transformer samples; (10) S7. After sorting the quality assurance capability evaluation values ​​of the m transformer samples in descending order, select the supplier of the transformer samples corresponding to the first s quality assurance capability evaluation values ​​as the transformer supplier with the best quality.

2. The transformer quality assurance capability evaluation method based on weight combination optimization according to claim 1 is characterized in that S2 include: S2.

1. Initialization of transformer quality assurance capability evaluation indicators The average importance score , thus obtaining the average score of n indicators; S2.

2. Compare the average scores of n transformer quality assurance capability evaluation indicators to obtain the relative importance of each indicator, thereby constructing a relative importance judgment matrix with a dimension of n×n; S2.

3. After adding up all the column elements of each row of the relative importance judgment matrix, the relative importance vector of n transformer quality assurance capability evaluation indicators is obtained. and the sum of relative importance ,in, represents the j-th transformer quality assurance capability evaluation index The relative importance vector of ; and = ,in, is the element in the rth row and jth column of the relative importance judgment matrix, representing the rth transformer quality assurance capability evaluation index and the j-th transformer quality assurance capability evaluation index The relative importance between them; S2.4, using formula (1) to obtain The relative importance weight of , thus the relative importance weight vector of n transformer quality assurance capability evaluation indicators is obtained as ; (1) In formula (1), ,and , ; Represents the rth transformer quality assurance capability evaluation index The relative importance vector of .

3. The transformer quality assurance capability evaluation method based on weight combination optimization according to claim 2 is characterized in that S3 include: S3.

1. Obtain the j-th transformer quality assurance capability evaluation index The corresponding m transformer samples, thus constructing a dimension of The transformer sample matrix ,in, represents the j-th transformer quality assurance capability evaluation index The corresponding i-th transformer sample; S3.2, using formula (2) Perform forward processing to obtain the jth transformer quality assurance capability evaluation index The corresponding i-th transformer sample after forward processing , thus obtaining the transformer sample matrix after forward processing '; (2) In formula (2), k is the coefficient, express The corresponding m transformer samples, express The absolute value of S3.3, using formula (3) After standardization, we get The corresponding standardized i-th transformer sample , thus obtaining the standardized transformer sample matrix R; (3) S3.4, Calculate using formula (4) The mean of the corresponding m transformer samples , calculated using formula (5) The standard deviation of the corresponding m transformer samples , and then use formula (6) to calculate The coefficient of variation ; (4) (5) (6) S3.5, Calculate using formula (7) The corresponding information weight , thus obtaining the information weight vector of n transformer quality assurance capability evaluation indicators: ; (7) In formula (7), ,and , .

4. The transformer quality assurance capability evaluation method based on weight combination optimization according to claim 3 is characterized in that: The step S5 is solved according to the following steps to obtain : S5.

1. Calculation The sum of the square values ​​of the corresponding m transformer samples , thus obtaining the sum of the square values ​​of the n evaluation indicators, and then using formula (11) to obtain the diagonal matrix , : (11) S5.

2. Calculate according to formula (8) Deviation of the estimated value , thereby calculating the deviation of n evaluation indicators, and then using formula (12) to construct the vector : (12) S5.3, using formula (13) to obtain : (13) In formula (13), Represents a column vector of dimension n where all elements are 1.

5. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the transformer quality assurance capability evaluation method according to any one of claims 1 to 4, and the processor is configured to execute the program stored in the memory.

6. 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 transformer quality assurance capability evaluation method according to any one of claims 1 to 4 are executed.

Images