Transformer operation state analysis method and device

Abstract

The invention discloses a transformer operation state analysis method and device, and the method comprises the steps: carrying out the clustering of the obtained data of a target transformer, obtaining an evaluation index for evaluating the operation state of the target transformer, and constructing a first sample matrix according to the evaluation index; calculating through the first sample matrix to obtain an information utility value of each evaluation index, and determining a first object score of the target transformer according to the information utility value; calculating according to the first sample matrix to obtain a grey correlation degree of each evaluation index, and determining a second object score of the target transformer according to the grey correlation degree of the evaluation indexes; and evaluating the operation state of the target transformer according to the first object score of the target transformer and the second object score of the target transformer. By implementing the application, the one-sidedness defect existing in single weight analysis is effectively solved, so that the operation state of the target transformer can be evaluated more accurately.

Description

Technical Field

[0001] The present invention relates to the field of transformer status analysis, and in particular to a transformer operating status analysis method and device. Background Art

[0002] Power transformers are extremely critical node-type equipment in the power system. Comprehensive evaluation of their operating status can effectively prevent transformer failures, reduce failure rates, and ensure long-term stable power supply to the power system. However, existing status assessment methods have defects in indicator selection, either too single or redundant. A single indicator is difficult to fully reflect the true operating status of the transformer, and may miss some important performance parameters or potential failure risks; redundant indicators will increase time and computing costs, and are not conducive to identifying key defects. In addition, a large number of studies use a single method to determine weights, which may lead to one-sidedness. For example, the subjective weighting method relies on expert experience and knowledge and is highly subjective; the objective weighting method may ignore the value of subjective experience. At the same time, a single weight analysis will excessively magnify or reduce the weight of a certain state quantity in the evaluation process, thereby affecting the accuracy of the evaluation. Summary of the invention

[0003] The present invention provides a transformer operating status analysis method and device, wherein the method can more accurately evaluate the operating status of a target transformer.

[0004] An embodiment of the present invention provides a transformer operating status analysis method, comprising:

[0005] Obtain the operation data set, maintenance data set and experimental data set of the target transformer;

[0006] According to the operation data set, the maintenance data set and the experimental data set, the target index of the target transformer is obtained by screening;

[0007] Clustering the target indicators of the target transformer to obtain evaluation indicators for evaluating the operating status of the target transformer;

[0008] The number of target transformers is used as the number of matrix rows, and the number of evaluation indicators is used as the number of matrix columns to construct a first sample matrix; wherein the first sample matrix includes specific values ​​of all target transformers under each evaluation indicator;

[0009] Calculating the information utility value of the evaluation indicator according to the first sample matrix, and determining the weight value of the evaluation indicator according to the information utility value of the evaluation indicator;

[0010] Calculate a first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix;

[0011] According to the first sample matrix, a grey relational degree of the evaluation index is calculated, and according to the first sample matrix and the grey relational degree of the evaluation index, a second object score of the target transformer is obtained;

[0012] An operating state of the target transformer is evaluated according to the first object score of the target transformer and the second object score of the target transformer.

[0013] Furthermore, the process of generating the experimental data set includes:

[0014] Acquire an initial experimental data set; wherein the initial experimental data set includes an insulating oil chromatogram index, insulating oil chromatogram data corresponding to the insulating oil chromatogram index, an insulating oil quality index, insulating oil quality experimental data corresponding to the insulating oil quality index, a furfural index, and a furfural volume fraction corresponding to the furfural index;

[0015] Determine the first level information of each gas according to the content data of each gas in the insulating oil chromatogram data;

[0016] Calculate the oil chromatogram health index based on the preset oil chromatogram weight table and the first level information;

[0017] Determine the second level information of each oil quality index according to the inspection data of each oil quality index in the insulating oil quality experimental data, and calculate the oil quality health index based on the preset oil quality weight table and the second level information;

[0018] The furfural health index is calculated according to a preset furfural health formula and the furfural volume fraction;

[0019] The experimental data set is generated according to the insulating oil chromatogram index, the insulating oil quality index, the furfural index, the oil chromatogram health index, the oil quality health index and the furfural health index.

[0020] Furthermore, the target indicators of the target transformer are obtained by screening according to the operation data set, the maintenance data set and the experimental data set, including:

[0021] The number of target transformers is used as the number of matrix rows, and the sum of the number of indicators of the operating data set, the number of indicators of the maintenance data set, and the number of indicators of the experimental data set is used as the number of matrix columns to construct a second sample matrix; wherein the second sample matrix includes specific values ​​of all target transformers under each indicator;

[0022] Standardizing the second sample matrix, and calculating a first covariance matrix according to the standardized second sample matrix;

[0023] Calculating the eigenvalues ​​and eigenvectors of the first covariance matrix;

[0024] The cumulative contribution rate corresponding to each eigenvalue is calculated, and the eigenvalue whose cumulative contribution rate exceeds the preset threshold is taken as the target eigenvalue, and the indicator corresponding to the target eigenvalue is taken as the target indicator.

[0025] Furthermore, clustering the target indicators of the target transformer to obtain evaluation indicators for evaluating the operating status of the target transformer includes:

[0026] For each of the target indicators, a component value of the target indicator is calculated according to the standardized second sample matrix and the target eigenvalue corresponding to the target indicator;

[0027] The target indicators are clustered according to the component values ​​of the target indicators to obtain evaluation indicators for evaluating the operating status of the target transformer.

[0028] Further, the information utility value of the evaluation indicator is calculated according to the first sample matrix, and the weight value of the evaluation indicator is determined according to the information utility value of the evaluation indicator, including:

[0029] Performing a forward processing on the first sample matrix to obtain a forward-processed first sample matrix;

[0030] Performing standardization processing on the first sample matrix after the forward transformation to obtain a standardized first sample matrix;

[0031] For each evaluation index in the standardized first sample matrix, the specific values ​​of all target transformers under the evaluation index are added to obtain a total value; according to the specific values ​​of each target transformer under the evaluation index and the total value, the information utility value of the evaluation index is calculated;

[0032] According to the information utility value of each evaluation indicator, the weight value of each evaluation indicator is calculated.

[0033] Furthermore, the calculating a first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix includes:

[0034] Screening out the minimum element value and the maximum element value of each column vector in the standardized first sample matrix;

[0035] According to all the minimum element values ​​obtained by screening, a minimum value vector is formed;

[0036] According to all the maximum element values ​​obtained by screening, a maximum value vector is constructed;

[0037] For each target transformer, a first distance value is calculated according to the maximum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a second distance value is calculated according to the minimum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a first object score of the target transformer is calculated according to the first distance value and the second distance value;

[0038] A first object score of each of the target transformers is obtained.

[0039] Furthermore, the information utility value of the evaluation index is calculated based on the specific value of each target transformer under the evaluation index and the total value, including:

[0040] For each evaluation index, the probability value of each target transformer under the evaluation index is calculated according to the specific value of each target transformer under the evaluation index and the total value; the information entropy of the evaluation index is calculated according to the probability value of each target transformer under the evaluation index; and the information utility value of the evaluation index is calculated according to the information entropy.

[0041] Further, the grey correlation degree of the evaluation index is calculated according to the first sample matrix, and the second object score of the target transformer is obtained according to the first sample matrix and the grey correlation degree of the evaluation index, including:

[0042] Normalizing the first sample matrix after the forward transformation to obtain a normalized first sample matrix;

[0043] Comparing the elements in each row vector in the normalized first sample matrix to obtain the maximum value of each row vector, and constructing a virtual mother sequence according to the maximum value of each row vector;

[0044] Calculating the grey relational degree of each evaluation index according to the normalized first sample matrix and the virtual mother sequence;

[0045] According to the normalized first sample matrix and the grey relational degree of each evaluation index, a second object score of each target transformer is obtained.

[0046] Furthermore, the step of normalizing the first sample matrix after the forward transformation to obtain a normalized first sample matrix includes:

[0047] According to the first sample matrix after forward transformation, the mean value of each evaluation index is obtained by solving;

[0048] For each evaluation index, according to the specific value of each target transformer under the evaluation index and the mean value of the evaluation index, determine the normalized element value of each target transformer under the evaluation index;

[0049] According to the normalized element values ​​of each target transformer under each evaluation index, a normalized first sample matrix is ​​obtained.

[0050] An embodiment of the present invention further provides a transformer operation status analysis device, comprising: a data set acquisition module, a target index determination module, an evaluation index determination module, a first sample matrix construction module, a weight value determination module, a first object score determination module, a second object score determination module and an evaluation module;

[0051] The data set acquisition module is used to acquire the operation data set, maintenance data set and experimental data set of the target transformer;

[0052] The target indicator determination module is used to screen and obtain the target indicator of the target transformer according to the operation data set, the maintenance data set and the experimental data set;

[0053] The evaluation index determination module is used to cluster the target indicators of the target transformer to obtain the evaluation indicators for evaluating the operating status of the target transformer;

[0054] The first sample matrix construction module is used to construct a first sample matrix with the number of target transformers as the number of matrix rows and the number of evaluation indicators as the number of matrix columns; wherein the first sample matrix includes specific values ​​of all target transformers under each evaluation indicator;

[0055] The weight value determination module is used to calculate the information utility value of the evaluation indicator according to the first sample matrix, and determine the weight value of the evaluation indicator according to the information utility value of the evaluation indicator;

[0056] The first object score determination module is used to calculate the first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix;

[0057] The second object score determination module is used to calculate the grey correlation degree of the evaluation index according to the first sample matrix, and obtain the second object score of the target transformer according to the first sample matrix and the grey correlation degree of the evaluation index;

[0058] The evaluation module is used to evaluate the operating status of the target transformer according to the first object score of the target transformer and the second object score of the target transformer.

[0059] The following beneficial effects are achieved by implementing the present invention:

[0060] The present invention provides a transformer operating status analysis method and device. The method first obtains three types of information, namely, status information of a target transformer during operation, transformer maintenance data, and various test data, so as to use the three types of information as the basis for determining subsequent evaluation indicators, and can comprehensively, scientifically, and multi-angle reflect the status of the target transformer; then, through a clustering method, the evaluation indicators are screened out from a large number of indicators, so as to streamline the data and improve the evaluation efficiency; finally, the first object score and the second object score are solved for the evaluation indicators obtained by clustering, so as to effectively solve the defect of one-sidedness of single weight analysis, and thus the operating status of the target transformer can be more accurately evaluated. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the implementation manner will be briefly introduced below. Obviously, the drawings described below are only some implementation manners of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0062] Figure 1 It is a flowchart of a transformer operating status analysis method provided by a certain embodiment of the present application;

[0063] Figure 2 It is a structural schematic diagram of a transformer operating status analysis device provided in a certain embodiment of the present application. DETAILED DESCRIPTION

[0064] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in the embodiments of this application. Obviously, the described embodiments are part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0065] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by technicians in the technical field to which this application belongs; the terms used herein are only for the purpose of describing specific embodiments and are not intended to limit this application; the terms "including" and "having" in the specification and claims of this application and the above-mentioned figure descriptions and any variations thereof are intended to cover non-exclusive inclusions.

[0066] In the description of the embodiments of the present application, the technical terms "first", "second", etc. are only used to distinguish different objects, and cannot be understood as indicating or implying relative importance or implicitly indicating the number, specific order or primary and secondary relationship of the indicated technical features. In the description of the embodiments of the present application, the meaning of "multiple" is more than two, unless otherwise clearly and specifically defined.

[0067] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0068] In the description of the embodiments of the present application, the term "and / or" is only a description of the association relationship of the associated objects, indicating that there may be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0069] In the description of the embodiments of the present application, the term "multiple" refers to more than two (including two). Similarly, "multiple groups" refers to more than two groups (including two groups), and "multiple pieces" refers to more than two pieces (including two pieces).

[0070] In the description of the embodiments of the present application, unless otherwise clearly specified and limited, technical terms such as "installed", "connected", "connected", "fixed" and the like should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, and it can be the internal connection of two elements or the interaction relationship between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present application can be understood according to the specific circumstances.

[0071] See also Figure 1 , is a flow chart of a transformer operation status analysis method provided by an embodiment of the present invention, comprising:

[0072] S1. Obtain the operation data set, maintenance data set and experimental data set of the target transformer;

[0073] Schematically, the target transformer is in contact with the outside world during operation, which will lead to status information indicators such as load conditions, ambient temperature, heat dissipation conditions, cooling effect, number of overvoltage shocks, and design life;

[0074] Specifically, the indicators of the operating data set include load conditions, ambient temperature, heat dissipation conditions, cooling effect, number of overvoltage shocks and design life;

[0075] Indicatively, according to the professional work summary of primary substation equipment of power grid companies in each year, the parts of transformers with emergency (major) defects are tap changers, bushings, main bodies, coolers, gas relays, oil pillows, etc., and the major defect types are leakage, abnormal oil / gas test data, and heating of joints;

[0076] Specifically, therefore, based on the above summary, the indicators in the maintenance data set include maintenance records, bushings, body, cooler, gas relay and oil pillow;

[0077] Schematically, the transformer test can obtain the changing trend of the data under the same test conditions, and relate this changing trend to the transformer status, so as to make a judgment on the transformer status;

[0078] Due to the wide variety of tests, this application refers to the transformer test content and regulations in the "China Southern Power Grid Electric Power Preventive Test Procedure" and the "Electrical Equipment Handover Test Standard for Electrical Test Equipment Engineering", and conducts a comprehensive analysis of various test items, functions, and significance. Considering the comprehensiveness, effectiveness, and applicability of parameters in reflecting the equipment status, the application finally chooses to analyze the insulating oil chromatography data, insulating oil quality experimental data, and furfural volume fraction. At the same time, the health index is introduced to quantify the stable operation capability of the transformer and reduce the dimension of the test data.

[0079] In a preferred embodiment, the process of generating the experimental data set includes:

[0080] Acquire an initial experimental data set; wherein the initial experimental data set includes an insulating oil chromatogram index, insulating oil chromatogram data corresponding to the insulating oil chromatogram index, an insulating oil quality index, insulating oil quality experimental data corresponding to the insulating oil quality index, a furfural index, and a furfural volume fraction corresponding to the furfural index;

[0081] Determine the first level information of each gas according to the content data of each gas in the insulating oil chromatogram data;

[0082] Calculate the oil chromatogram health index based on the preset oil chromatogram weight table and the first level information;

[0083] Determine the second level information of each oil quality index according to the inspection data of each oil quality index in the insulating oil quality experimental data, and calculate the oil quality health index based on the preset oil quality weight table and the second level information;

[0084] The furfural health index is calculated according to a preset furfural health formula and the furfural volume fraction;

[0085] generating the experimental data set according to the insulating oil chromatogram index, the insulating oil quality index, the furfural index, the oil chromatogram health index, the oil quality health index and the furfural health index;

[0086] Schematically, the gas composition in the insulating oil of the target transformer reflects the internal operating state of the transformer. By judging the composition, the internal defects of the transformer can be found. When the transformer operates normally, CH4 and C2H6 gases will be generated in its insulating oil, and the generation of H2, C2H4, and C2H2 is obviously related to equipment failure. Therefore, this embodiment calculates the oil chromatogram health index HI1 according to the content of these five gases;

[0087] Specifically, Table 1 is a content level evaluation table of each gas in the insulating oil chromatogram data; therefore, according to the content data of each gas in the insulating oil chromatogram data, the level evaluation is performed in Table 1 to determine the first level information L of each gas;

[0088]

[0089]

[0090] Table 1

[0091] Specifically, after obtaining the first level information of each gas, the weight value C of each gas is determined based on a preset oil chromatogram weight table, and the oil chromatogram health index is calculated according to the weight value C of each gas and the first level information L of each gas; wherein the oil chromatogram weight table is shown in Table 2;

[0092]

[0093] Table 2

[0094] Specifically, the specific calculation expression of the oil chromatogram health index HI1 is as follows:

[0095]

[0096] In schematic form, the oil quality test mainly tests the relevant parameters of the insulating oil, reflecting the insulation condition of the insulating oil itself; the test data can be used to determine whether the insulating effect of the insulating oil meets the requirements and to discover defects such as moisture and aging; the calculation method and principle of the oil quality health index HI2 are the same as those of the oil chromatogram health index;

[0097] Specifically, according to the inspection data of each oil quality index in the insulating oil quality experimental data, the second level information of each oil quality index is determined, wherein the oil quality index in the insulating oil quality experimental data includes breakdown voltage, dielectric loss, acid value and trace water content; Table 3 is a content grade evaluation table of each oil quality index in the insulating oil quality experimental data; thus, according to the inspection data of each oil quality index in the insulating oil quality experimental data, the grade is evaluated in Table 3 to determine the second level information L of each oil quality index;

[0098]

[0099]

[0100] Table 3

[0101] Specifically, after obtaining the second level information L of each oil quality index, the weight value C of each oil quality index is determined based on a preset oil quality weight table, and the oil quality health index HI2 is calculated according to the weight value C of each oil quality index and the second level information L of each oil quality index, wherein the oil quality weight table is shown in Table 4;

[0102]

[0103] Table 4

[0104] Specifically, the level of furfural volume fraction FFA reflects the aging condition of the insulation material. New and high-quality insulation paper has a higher degree of polymerization DP. However, during operation, affected by various factors such as acidic substances and heat, the degree of polymerization will gradually decrease, and the corresponding furfural volume fraction FFA will also gradually decrease; therefore, according to the preset furfural health formula and the furfural volume fraction of the target transformer, the furfural health index HI3 is calculated. The specific calculation formula is as follows:

[0105] HI3=2.33×FFA 0.68 ;

[0106] Specifically, the experimental data set is generated according to the insulating oil chromatogram index, the insulating oil quality index, the furfural index, the oil chromatogram health index HI1, the oil quality health index HI2 and the furfural health index HI3.

[0107] S2. Screening and obtaining target indicators of the target transformer according to the operation data set, the maintenance data set and the experimental data set;

[0108] In a preferred embodiment, the target indicators of the target transformer are obtained by screening according to the operation data set, the maintenance data set and the experimental data set, including:

[0109] The number of target transformers is used as the number of matrix rows, and the sum of the number of indicators of the operating data set, the number of indicators of the maintenance data set, and the number of indicators of the experimental data set is used as the number of matrix columns to construct a second sample matrix; wherein the second sample matrix includes specific values ​​of all target transformers under each indicator;

[0110] Standardizing the second sample matrix, and calculating a first covariance matrix according to the standardized second sample matrix;

[0111] Calculating the eigenvalues ​​and eigenvectors of the first covariance matrix;

[0112] Calculate the cumulative contribution rate corresponding to each eigenvalue, and take the eigenvalue whose cumulative contribution rate exceeds the preset threshold as the target eigenvalue, and take the indicator corresponding to the target eigenvalue as the target indicator;

[0113] Indicatively, based on step S1, the total number of indicators in the operating data set, the maintenance data set, and the experimental data set is 15, and the number of indicators is still large, requiring a dimensionality reduction operation; therefore, in this embodiment, a principal component analysis is first performed on the initial indicators and data to obtain the target indicators, and then a clustering operation is performed on the screened target indicators to obtain an evaluation indicator for evaluating the operating status of the target transformer;

[0114] Specifically, assuming that there are n target transformers and p = 15 indicators, a second sample matrix x of size n×p is constructed. The specific expression is as follows:

[0115]

[0116] Schematically, the second sample matrix is ​​standardized, and a first covariance matrix is ​​calculated based on the standardized second sample matrix;

[0117] Specifically, calculate the mean of x by column and standard deviation Further standardized data After the second sample matrix is ​​normalized, a standardized second sample matrix is ​​obtained, which is specifically shown as follows:

[0118]

[0119] Then use the following formula to obtain the first covariance matrix R:

[0120]

[0121] In the formula,

[0122] Schematically, calculating the eigenvalues ​​and eigenvectors of the first covariance matrix;

[0123] Specifically, because R is a semi-positive definite matrix, we can use the formula Calculate the eigenvalues ​​and satisfy: λ1≥λ2≥…λ p ≥0; and calculate the eigenvector:

[0124]

[0125] In schematic form, the contribution rate and cumulative contribution rate corresponding to each eigenvalue are calculated, and the eigenvalue whose cumulative contribution rate exceeds a preset threshold is taken as the target eigenvalue, and the indicator corresponding to the target eigenvalue is taken as the target indicator;

[0126] Specifically, the calculation formula for the contribution rate of each eigenvalue is as follows:

[0127]

[0128] Specifically, the calculation formula for the cumulative contribution rate of each eigenvalue is as follows:

[0129]

[0130] Therefore, m eigenvalues ​​with cumulative contribution rates exceeding 80% are taken as target eigenvalues, and the indicators corresponding to the target eigenvalues ​​are taken as target indicators, thereby obtaining m target indicators.

[0131] S3, clustering the target indicators of the target transformer to obtain evaluation indicators for evaluating the operating status of the target transformer;

[0132] In a preferred embodiment, clustering the target indicators of the target transformer to obtain evaluation indicators for evaluating the operating status of the target transformer includes:

[0133] For each of the target indicators, a component value of the target indicator is calculated according to the standardized second sample matrix and the target eigenvalue corresponding to the target indicator;

[0134] Clustering the target indicators according to the component values ​​of the target indicators to obtain evaluation indicators for evaluating the operating status of the target transformer;

[0135] Specifically, for each target indicator, the component value of each target indicator is calculated, and the specific calculation formula is as follows:

[0136] F i =a 1i X1+a 2i X2+…+a pi X p(i=1,2,…,m);

[0137] Then, according to the component value of each target indicator, it is input into the Spss software to complete the system clustering, the number of clusters is determined according to the pedigree diagram or the aggregation coefficient line chart, and the final clustering effect can be visualized;

[0138] Finally, the number of evaluation indicators used to evaluate the operating status of the target transformer is determined based on the clustering effect. Multiple indicators that are clustered into one category and have a large correlation are selected to replace them, and finally q evaluation indicators are obtained.

[0139] S4, taking the number of target transformers as the number of matrix rows and the number of evaluation indicators as the number of matrix columns, constructing a first sample matrix; wherein the first sample matrix includes specific values ​​of all target transformers under each evaluation indicator;

[0140] S5. Calculate the information utility value of the evaluation indicator according to the first sample matrix, and determine the weight value of the evaluation indicator according to the information utility value of the evaluation indicator;

[0141] In a preferred embodiment, the information utility value of the evaluation indicator is calculated according to the first sample matrix, and the weight value of the evaluation indicator is determined according to the information utility value of the evaluation indicator, including:

[0142] Performing a forward processing on the first sample matrix to obtain a forward-processed first sample matrix;

[0143] Performing standardization processing on the first sample matrix after the forward transformation to obtain a standardized first sample matrix;

[0144] For each evaluation index in the standardized first sample matrix, the specific values ​​of all target transformers under the evaluation index are added to obtain a total value; according to the specific values ​​of each target transformer under the evaluation index and the total value, the information utility value of the evaluation index is calculated;

[0145] According to the information utility value of each evaluation indicator, the weight value of each evaluation indicator is calculated;

[0146] In a preferred embodiment, the information utility value of the evaluation index is calculated based on the specific value of each target transformer under the evaluation index and the total value, including:

[0147] For each evaluation index, according to the specific value of each target transformer under the evaluation index and the total value, the probability value of each target transformer under the evaluation index is calculated; according to the probability value of each target transformer under the evaluation index, the information entropy of the evaluation index is calculated; according to the information entropy, the information utility value of the evaluation index is calculated;

[0148] Schematically, performing a forward processing on the first sample matrix to obtain a forward-processed first sample matrix;

[0149] Specifically, since the evaluation indicators include extremely large indicators, extremely small indicators, intermediate indicators and interval indicators, it is necessary to firstly perform a positive processing on the first sample matrix to uniformly convert the indicator types of all evaluation indicators into extremely large indicators.

[0150] The positive formulas for the three types of indicators are as follows:

[0151] Very small indicators:

[0152] Interval index: M = max{a-min{x i},max{x i}-b},

[0153] Intermediate index: M = max{|x i -x best |}

[0154] Specifically, after the forward processing is completed, the first sample matrix after the forward processing is obtained is as follows:

[0155]

[0156] In order to eliminate the influence of different indicator dimensions, it is necessary to perform a normalization operation on the first sample matrix X after forward transformation:

[0157] Specifically, it is known that the number of target transformers is n, and the number of evaluation indicators is q. Since the subsequent steps need to calculate the probability, it is necessary to ensure that each element of the first sample matrix X after the forward transformation is non-negative. The specific process is shown in the following formula:

[0158]

[0159] Thus, the standardized first sample matrix Z is obtained, and its specific expression is as follows:

[0160]

[0161] Specifically, since the standardized first sample matrix Z has been obtained, it is necessary to calculate the probability value of each element in the standardized first sample matrix Z and regard it as the probability used in the relative entropy calculation, where the probability value p of each element is ij The calculation formula is:

[0162]

[0163] Then, for each evaluation indicator in the standardized first sample matrix, the information entropy of each evaluation indicator is calculated, and the specific calculation formula is as follows:

[0164]

[0165] In the formula, -ln(p ij ) indicates the amount of information;

[0166] It should be noted that the purpose of dividing by lnn is to make the range of information entropy within the range of 0-1. Information entropy represents the expected amount of information, that is, the larger the information entropy, the less information the jth evaluation indicator has;

[0167] Define the information utility value d j =1-e j , the larger its value is, the greater the amount of information is. Finally, the information utility value is normalized to obtain the weight corresponding to each evaluation indicator:

[0168]

[0169] S6. Calculate a first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix;

[0170] In a preferred embodiment, the step of calculating the first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix includes:

[0171] Screening out the minimum element value and the maximum element value of each column vector in the standardized first sample matrix;

[0172] According to all the minimum element values ​​obtained by screening, a minimum value vector is formed;

[0173] According to all the maximum element values ​​obtained by screening, a maximum value vector is constructed;

[0174] For each target transformer, a first distance value is calculated according to the maximum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a second distance value is calculated according to the minimum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a first object score of the target transformer is calculated according to the first distance value and the second distance value;

[0175] Obtaining a first object score for each of the target transformers;

[0176] Specifically, it is necessary to screen out the minimum element value and the maximum element value of each column vector in the standardized first sample matrix;

[0177] Specifically, define the maximum value vector

[0179] Similarly, the minimum vector

[0181] Thus, a minimum value vector is formed according to all the minimum element values ​​obtained by screening; and a maximum value vector is formed according to all the maximum element values ​​obtained by screening;

[0182] Schematically, for each target transformer, a first distance value is calculated according to the maximum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a second distance value is calculated according to the minimum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a first object score of the target transformer is calculated according to the first distance value and the second distance value;

[0183] Specifically, the first distance value of the maximum value vector of the i-th (i=1, 2, ..., n) target transformer is defined as:

[0184]

[0185] Specifically, the second distance value of the minimum value vector of the i-th (i=1, 2, ..., n) target transformer is defined as:

[0186]

[0187] Therefore, the first object score of the i-th target transformer is calculated by the preset TOPSIS method, and its expression is as follows:

[0188]

[0189] Since the score is not normalized, it needs to be normalized:

[0190] And meet

[0191] S7. According to the first sample matrix, the grey relational degree of the evaluation index is calculated, and according to the first sample matrix and the grey relational degree of the evaluation index, a second object score of the target transformer is obtained;

[0192] In a preferred embodiment, the grey correlation degree of the evaluation index is calculated according to the first sample matrix, and the second object score of the target transformer is obtained according to the first sample matrix and the grey correlation degree of the evaluation index, including:

[0193] Normalizing the first sample matrix after the forward transformation to obtain a normalized first sample matrix;

[0194] Comparing the elements in each row vector in the normalized first sample matrix to obtain the maximum value of each row vector, and constructing a virtual mother sequence according to the maximum value of each row vector;

[0195] Calculating the grey relational degree of each evaluation index according to the normalized first sample matrix and the virtual mother sequence;

[0196] Obtaining a second object score for each target transformer according to the normalized first sample matrix and the grey correlation degree of each evaluation index;

[0197] In a preferred embodiment, the step of normalizing the first sample matrix after the forward transformation to obtain the normalized first sample matrix includes:

[0198] According to the first sample matrix after forward transformation, the mean value of each evaluation index is obtained by solving;

[0199] For each evaluation index, according to the specific value of each target transformer under the evaluation index and the mean value of the evaluation index, determine the normalized element value of each target transformer under the evaluation index;

[0200] According to the normalized element values ​​of each target transformer under each evaluation index, a normalized first sample matrix is ​​obtained;

[0201] Illustratively, the first sample matrix after the forward transformation needs to be normalized;

[0202] Specifically, for each evaluation index, it is necessary to first obtain the mean of the evaluation index, and then divide the specific values ​​of each target transformer under the evaluation index by the mean of the evaluation index, so as to achieve normalization, remove the dimension and reduce the range of variables, thereby simplifying the calculation. The specific calculation formula is as follows:

[0203]

[0204] Specifically, the elements in each row vector in the normalized first sample matrix are compared to obtain the maximum value of each row vector, and a virtual mother sequence X0 is constructed according to the maximum value of each row vector. The specific calculation formula is as follows:

[0205]

[0206] Then the sequence formed by each evaluation index in the normalized first sample matrix is ​​recorded as:

[0207] X1=x(1),x1(2),…,x1(n)) T …X q =(x q (1),x q (2),…,x q (n)) T ;

[0208] Define each evaluation indicator X j The grey correlation degree of (j=1,2,…,q) and the virtual mother sequence X0, is as follows:

[0209]

[0210] Therefore, the normalized weights of each evaluation index are calculated. The specific calculation formula is as follows:

[0211]

[0212] Thus, the second object score of each target transformer is calculated, and the specific calculation formula is as follows:

[0213]

[0214] And normalize it to get

[0215] S8. Evaluate the operating status of the target transformer according to the first object score of the target transformer and the second object score of the target transformer;

[0216] Specifically, a first preset weight θ1 of the first object score is obtained, and a second preset weight θ2 of the second object score is obtained; for the i-th target transformer, according to the first object score of the target transformer i and the second object score of the target transformer i Determine the operating status score of the target transformer

[0217] The operating status of the target transformer is determined according to the operating status score and a preset operating status comparison table.

[0218] See also Figure 2 , is a transformer operation status analysis device provided by an embodiment of the present invention, comprising: a data set acquisition module, a target indicator determination module, an evaluation indicator determination module, a first sample matrix construction module, a weight value determination module, a first object score determination module, a second object score determination module and an evaluation module;

[0219] The data set acquisition module is used to acquire the operation data set, maintenance data set and experimental data set of the target transformer;

[0220] The target indicator determination module is used to screen and obtain the target indicator of the target transformer according to the operation data set, the maintenance data set and the experimental data set;

[0221] The evaluation index determination module is used to cluster the target indicators of the target transformer to obtain the evaluation indicators for evaluating the operating status of the target transformer;

[0222] The first sample matrix construction module is used to construct a first sample matrix with the number of target transformers as the number of matrix rows and the number of evaluation indicators as the number of matrix columns; wherein the first sample matrix includes specific values ​​of all target transformers under each evaluation indicator;

[0223] The weight value determination module is used to calculate the information utility value of the evaluation indicator according to the first sample matrix, and determine the weight value of the evaluation indicator according to the information utility value of the evaluation indicator;

[0224] The first object score determination module is used to calculate the first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix;

[0225] The second object score determination module is used to calculate the grey correlation degree of the evaluation index according to the first sample matrix, and obtain the second object score of the target transformer according to the first sample matrix and the grey correlation degree of the evaluation index;

[0226] The evaluation module is used to evaluate the operating status of the target transformer according to the first object score of the target transformer and the second object score of the target transformer.

[0227] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A transformer operating status analysis method, characterized in that: include: Obtain the operation data set, maintenance data set and experimental data set of the target transformer; According to the operation data set, the maintenance data set and the experimental data set, the target indicator of the target transformer is obtained by screening; Clustering the target indicators of the target transformer to obtain evaluation indicators for evaluating the operating status of the target transformer; The number of target transformers is used as the number of matrix rows, and the number of evaluation indicators is used as the number of matrix columns to construct a first sample matrix; wherein the first sample matrix includes specific values ​​of all target transformers under each evaluation indicator; Calculating the information utility value of the evaluation indicator according to the first sample matrix, and determining the weight value of the evaluation indicator according to the information utility value of the evaluation indicator; Calculate a first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix; According to the first sample matrix, a grey relational degree of the evaluation index is calculated, and according to the first sample matrix and the grey relational degree of the evaluation index, a second object score of the target transformer is obtained; An operating state of the target transformer is evaluated according to the first object score of the target transformer and the second object score of the target transformer.

2. The transformer operating status analysis method according to claim 1, characterized in that: The generation process of the experimental data set includes: Acquire an initial experimental data set; wherein the initial experimental data set includes an insulating oil chromatogram index, insulating oil chromatogram data corresponding to the insulating oil chromatogram index, an insulating oil quality index, insulating oil quality experimental data corresponding to the insulating oil quality index, a furfural index, and a furfural volume fraction corresponding to the furfural index; Determine the first level information of each gas according to the content data of each gas in the insulating oil chromatogram data; Calculate the oil chromatogram health index based on the preset oil chromatogram weight table and the first level information; Determine the second level information of each oil quality index according to the inspection data of each oil quality index in the insulating oil quality experimental data, and calculate the oil quality health index based on the preset oil quality weight table and the second level information; The furfural health index is calculated according to a preset furfural health formula and the furfural volume fraction; The experimental data set is generated according to the insulating oil chromatogram index, the insulating oil quality index, the furfural index, the oil chromatogram health index, the oil quality health index and the furfural health index.

3. The transformer operating status analysis method according to claim 1, characterized in that: The step of screening and obtaining the target indicator of the target transformer according to the operation data set, the maintenance data set and the experimental data set includes: The number of target transformers is used as the number of matrix rows, and the sum of the number of indicators of the operating data set, the number of indicators of the maintenance data set, and the number of indicators of the experimental data set is used as the number of matrix columns to construct a second sample matrix; wherein the second sample matrix includes specific values ​​of all target transformers under each indicator; Standardizing the second sample matrix, and calculating a first covariance matrix according to the standardized second sample matrix; Calculating the eigenvalues ​​and eigenvectors of the first covariance matrix; The cumulative contribution rate corresponding to each eigenvalue is calculated, and the eigenvalue whose cumulative contribution rate exceeds the preset threshold is taken as the target eigenvalue, and the indicator corresponding to the target eigenvalue is taken as the target indicator.

4. The transformer operating status analysis method according to claim 3, characterized in that: The clustering of the target indicators of the target transformer to obtain an evaluation indicator for evaluating the operating status of the target transformer includes: For each of the target indicators, a component value of the target indicator is calculated according to the standardized second sample matrix and the target eigenvalue corresponding to the target indicator; The target indicators are clustered according to the component values ​​of the target indicators to obtain evaluation indicators for evaluating the operating status of the target transformer.

5. The transformer operating status analysis method according to claim 1, characterized in that: The step of calculating the information utility value of the evaluation indicator according to the first sample matrix, and determining the weight value of the evaluation indicator according to the information utility value of the evaluation indicator includes: Performing a forward processing on the first sample matrix to obtain a forward-processed first sample matrix; Performing standardization processing on the first sample matrix after the forward transformation to obtain a standardized first sample matrix; For each evaluation index in the standardized first sample matrix, the specific values ​​of all target transformers under the evaluation index are added to obtain a total value; according to the specific values ​​of each target transformer under the evaluation index and the total value, the information utility value of the evaluation index is calculated; According to the information utility value of each evaluation indicator, the weight value of each evaluation indicator is calculated.

6. The transformer operating status analysis method according to claim 5, characterized in that: The step of calculating a first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix includes: Screening out the minimum element value and the maximum element value of each column vector in the standardized first sample matrix; According to all the minimum element values ​​obtained by screening, a minimum value vector is formed; According to all the maximum element values ​​obtained by screening, a maximum value vector is constructed; For each target transformer, a first distance value is calculated according to the maximum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a second distance value is calculated according to the minimum value vector, the weight values ​​of the evaluation indicators, and the specific values ​​of the target transformer under each evaluation indicator; a first object score of the target transformer is calculated according to the first distance value and the second distance value; A first object score of each of the target transformers is obtained.

7. The transformer operating status analysis method according to claim 5, characterized in that: The information utility value of the evaluation index is calculated based on the specific value of each target transformer under the evaluation index and the total value, including: For each evaluation index, the probability value of each target transformer under the evaluation index is calculated according to the specific value of each target transformer under the evaluation index and the total value; the information entropy of the evaluation index is calculated according to the probability value of each target transformer under the evaluation index; and the information utility value of the evaluation index is calculated according to the information entropy.

8. The transformer operating status analysis method according to claim 5, characterized in that: The step of calculating the grey correlation degree of the evaluation index according to the first sample matrix, and obtaining the second object score of the target transformer according to the first sample matrix and the grey correlation degree of the evaluation index includes: Normalizing the first sample matrix after the forward transformation to obtain a normalized first sample matrix; Comparing the elements in each row vector in the normalized first sample matrix to obtain the maximum value of each row vector, and constructing a virtual mother sequence according to the maximum value of each row vector; Calculating the grey relational degree of each evaluation index according to the normalized first sample matrix and the virtual mother sequence; According to the normalized first sample matrix and the grey relational degree of each evaluation index, a second object score of each target transformer is obtained.

9. The transformer operating status analysis method according to claim 8, characterized in that: The step of normalizing the first sample matrix after the forward transformation to obtain the normalized first sample matrix includes: According to the first sample matrix after forward transformation, the mean value of each evaluation index is obtained by solving; For each evaluation index, according to the specific value of each target transformer under the evaluation index and the mean value of the evaluation index, determine the normalized element value of each target transformer under the evaluation index; According to the normalized element values ​​of each target transformer under each evaluation index, a normalized first sample matrix is ​​obtained.

10. A transformer operating status analysis device, characterized in that: include: A data set acquisition module, a target indicator determination module, an evaluation indicator determination module, a first sample matrix construction module, a weight value determination module, a first object score determination module, a second object score determination module, and an evaluation module; The data set acquisition module is used to acquire the operation data set, maintenance data set and experimental data set of the target transformer; The target indicator determination module is used to screen and obtain the target indicator of the target transformer according to the operation data set, the maintenance data set and the experimental data set; The evaluation index determination module is used to cluster the target indicators of the target transformer to obtain the evaluation indicators for evaluating the operating status of the target transformer; The first sample matrix construction module is used to construct a first sample matrix with the number of target transformers as the number of matrix rows and the number of evaluation indicators as the number of matrix columns; wherein the first sample matrix includes specific values ​​of all target transformers under each evaluation indicator; The weight value determination module is used to calculate the information utility value of the evaluation indicator according to the first sample matrix, and determine the weight value of the evaluation indicator according to the information utility value of the evaluation indicator; The first object score determination module is used to calculate the first object score of the target transformer according to the weight value of the evaluation index and the first sample matrix; The second object score determination module is used to calculate the grey correlation degree of the evaluation index according to the first sample matrix, and obtain the second object score of the target transformer according to the first sample matrix and the grey correlation degree of the evaluation index; The evaluation module is used to evaluate the operating status of the target transformer according to the first object score of the target transformer and the second object score of the target transformer.

Images