Power transformer health condition determination method and system

Abstract

The application discloses a kind of power transformer health state determination methods, including obtaining the attribute data of power transformer;Attribute data is normalized and corresponding grey correlation coefficient is calculated;The weight value of attribute data information is calculated using analytic hierarchy process, and grey correlation degree value is calculated;The attribute of power transformer is divided into health grade and the priority of each health grade is determined by comparison;Health grade is associated with grey correlation degree value to construct fuzzy logic control rule;The health state determination of actual power transformer is carried out using fuzzy logic control rule.The application also discloses a kind of system for realizing the power transformer health state determination method.The application obtains and processes the data of power transformer, and predicts the state of power transformer based on analytic hierarchy process and grey correlation analysis scheme, so that the application can not only realize the health state determination of power transformer, but also has high reliability and good accuracy.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of electrical automation, and particularly relates to a power transformer health state determination method and system. BACKGROUND

[0002] With the development of economy and technology and the improvement of people's living standards, electric energy has become an essential secondary energy in people's production and life, bringing endless convenience to people's production and life. Therefore, ensuring the stable and reliable supply of electric energy has become one of the most important tasks of the power system.

[0003] The power transformer is an important part of the power system, and plays a crucial role in the safe and stable operation of the power transmission network. During the operation of the power transformer, the insulation system will be affected by heat, mechanical, electrical and environmental stress, and the harsh working environment of the power transformer itself will also have a negative impact on its working performance. Therefore, it is particularly important to monitor and determine the health state of the power transformer.

[0004] At present, the commonly used health state determination scheme of the power transformer includes various health evaluation technologies based on artificial intelligence, including neural network models and multi-criteria analysis models. However, the neural network model scheme often requires a large amount of training data for data support, and there are not enough related data of the power transformer in reality, which makes the reliability of the neural network model scheme poor. The multi-criteria analysis model lacks consideration of the priority of diagnostic attributes, so the accuracy of this type of scheme is also not high. SUMMARY

[0005] One of the purposes of the present application is to provide a power transformer health state determination method with high reliability and good accuracy.

[0006] The second purpose of the present application is to provide a system for implementing the power transformer health state determination method.

[0007] The power transformer health state determination method provided by the present application comprises the following steps:

[0008] S1. Obtain attribute data information of the power transformer;

[0009] S2. Normalize the attribute data information obtained in step S1, and calculate the corresponding grey correlation coefficient;

[0010] S3. Calculate the weight value of the attribute data information obtained in step S1 using the analytic hierarchy process, and calculate the grey correlation degree value in combination with the grey correlation coefficient obtained in step S2;

[0011] S4. According to the grey correlation degree value obtained in step S3, the health level of the property of the power transformer is divided, and the priority of each health level is determined by comparison;

[0012] S5. Based on the fuzzy control scheme, the health level determined in step S4 is associated with the grey correlation degree value, and a fuzzy logic control rule is constructed;

[0013] S6. The fuzzy logic control rule obtained in step S5 is used to determine the health state of the actual power transformer.

[0014] The attribute data information of the power transformer obtained in step S1 includes the following steps:

[0015] The attribute data information of the power transformer is obtained, and the attribute data information includes H2 concentration, CH4 concentration, C2H2 concentration, C2H4 concentration, C2H6 concentration, acidity, breakdown voltage, CO concentration and CO2 concentration, etc.

[0016] The attribute data information obtained in step S1 is normalized and the corresponding grey correlation coefficient is calculated, which includes the following steps:

[0017] The obtained attribute data information is normalized by the following rules:

[0018] For attribute data information of positive attributes, the following formula is used for normalization:

[0019]

[0020] For attribute data information of negative attributes, the following formula is used for normalization:

[0021]

[0022] In the formula, x ij is the grey relationship unit of the jth attribute of the ith sample after normalization; y ij is the attribute value of the jth attribute of the ith sample before normalization; Min{y i} is the minimum value of the jth attribute of the ith sample; Max{y i} is the maximum value of the jth attribute of the ith sample;

[0023] Wherein, the definition of positive attribute is: the higher the value, the better the performance of the attribute; the definition of negative attribute is: the lower the value, the better the performance of the attribute;

[0024] The grey correlation coefficient of each attribute data is calculated by the following formula:

[0025]

[0026] wherein γ ij is the grey correlation coefficient; Δ ij is the deviation value, and Δ ij is the deviation value, and Δ ij is the deviation value, and Δ ij = |1-x ij |; ζ is a set coefficient; Δ min is the minimum value of the deviation value; Δ max is the maximum value of the deviation value.

[0027] The weight value of the attribute data information obtained in step S1 is calculated by using the analytic hierarchy process, and the grey correlation degree value is calculated in combination with the grey correlation coefficient obtained in step S2, and specifically includes the following steps:

[0028] The weight value w j of the attribute data information obtained in step S1 is calculated by using the analytic hierarchy process.

[0029] The grey correlation degree value is calculated by using the following formula:

[0030]

[0031] wherein Γ z is the grey correlation degree value; k is the number of attributes of each health level; and n is the number of transformer attributes.

[0032] The attribute of the power transformer is divided into health levels according to the grey correlation degree value obtained in step S3, and the priority of each health level is determined, and specifically includes the following steps:

[0033] The attribute of the power transformer is divided into three health levels: dissolved gas level, oil health level and paper health level according to the grey correlation degree value obtained in step S3;

[0034] The priority of each health level is determined by using the analytic hierarchy process.

[0035] The health level determined in step S4 is associated with the grey correlation degree value based on the fuzzy control scheme, and the fuzzy logic control rule is constructed, and specifically includes the following steps:

[0036] The grey correlation degree value of the attribute is taken as the input;

[0037] Fuzzy stage:

[0038] The input data is calculated to correspond to the membership degree by using the trapezoidal function as the membership function:

[0039]

[0040] wherein μ A (t) is the membership of the trapezoidal membership function; t is the input data; a is the upper limit value of the membership function; b is the lower limit value of the membership function; c1 is the first center of the membership function; c2 is the second center of the membership function;

[0041] The obtained membership is divided into very good VG, medium M, poor P and very poor VP; wherein the very good VG corresponds to the range of 0.1634-0.2021, the medium M corresponds to the range of 0.1296-0.1734, the poor P corresponds to the range of 0.0958-0.1396, and the very poor VP corresponds to the range of 0.0669-0.1058;

[0042] The reasoning stage:

[0043] The fuzzy rules are constructed, and the fuzzy output value is obtained according to the input data; wherein the fuzzy output value corresponds to the health state judgment result of the power transformer;

[0044] The constructed fuzzy rules are as follows:

[0045] When the dissolved gas rating is VG:

[0046] When the oil health record is VG and the paper health rating is VG, the total health degree is the best; when the oil health record is VG and the paper health rating is M, the total health degree is good; when the oil health record is VG and the paper health rating is P, the total health degree is good; when the oil health record is VG and the paper health rating is VP, the total health degree is medium; when the oil health record is M and the paper health rating is VG, the total health degree is good; when the oil health record is M and the paper health rating is M, the total health degree is good; when the oil health record is M and the paper health rating is P, the total health degree is medium; when the oil health record is M and the paper health rating is VP, the total health degree is poor; when the oil health record is P and the paper health rating is VG, the total health degree is good; when the oil health record is P and the paper health rating is M, the total health degree is medium; when the oil health record is P and the paper health rating is P, the total health degree is poor; when the oil health record is P and the paper health rating is VP, the total health degree is poor; when the oil health record is VP and the paper health rating is VG, the total health degree is medium; when the oil health record is VP and the paper health rating is M, the total health degree is poor; when the oil health record is VP and the paper health rating is P, the total health degree is poor; and when the oil health record is VP and the paper health rating is VP, the total health degree is poor;

[0047] When the dissolved gas rating is M:

[0048] When the oil health rating is VG and the paper health rating is VG, the overall health is good; when the oil health rating is VG and the paper health rating is M, the overall health is good; when the oil health rating is VG and the paper health rating is P, the overall health is fair; when the oil health rating is VG and the paper health rating is VP, the overall health is poor; when the oil health rating is M and the paper health rating is VG, the overall health is good; when the oil health rating is M and the paper health rating is M, the overall health is fair; when the oil health rating is M and the paper health rating is P, the overall health is poor; when the oil health rating is M and the paper health rating is VP, the overall health is poor; when the oil health rating is P and the paper health rating is VG, the overall health is fair; when the oil health rating is P and the paper health rating is M, the overall health is poor; when the oil health rating is P and the paper health rating is P, the overall health is poor; when the oil health rating is P and the paper health rating is VP, the overall health is very poor; when the oil health rating is VP and the paper health rating is VG, the overall health is poor; when the oil health rating is VP and the paper health rating is M, the overall health is poor; when the oil health rating is VP and the paper health rating is P, the overall health is very poor; and when the oil health rating is VP and the paper health rating is VP, the overall health is very poor.

[0049] When the dissolved gas rating is P:

[0050] When the oil health rating is VG and the paper health rating is VG, the overall health is good; when the oil health rating is VG and the paper health rating is M, the overall health is fair; when the oil health rating is VG and the paper health rating is P, the overall health is poor; when the oil health rating is VG and the paper health rating is VP, the overall health is poor; when the oil health rating is M and the paper health rating is VG, the overall health is fair; when the oil health rating is M and the paper health rating is M, the overall health is poor; when the oil health rating is M and the paper health rating is P, the overall health is poor; when the oil health rating is M and the paper health rating is VP, the overall health is poor; when the oil health rating is P and the paper health rating is VG, the overall health is fair; when the oil health rating is P and the paper health rating is M, the overall health is poor; when the oil health rating is P and the paper health rating is P, the overall health is poor; when the oil health rating is P and the paper health rating is VP, the overall health is very poor; when the oil health rating is VP and the paper health rating is VG, the overall health is poor; when the oil health rating is VP and the paper health rating is M, the overall health is poor; when the oil health rating is VP and the paper health rating is P, the overall health is very poor; and when the oil health rating is VP and the paper health rating is VP, the overall health is very poor.

[0051] When the dissolved gas rating is VP:

[0052] When the oil health record is VG and the paper health grade is VG, the total health degree is medium; when the oil health record is VG and the paper health grade is M, the total health degree is poor; when the oil health record is VG and the paper health grade is P, the total health degree is poor; when the oil health record is VG and the paper health grade is VP, the total health degree is poor; when the oil health record is M and the paper health grade is VG, the total health degree is poor; when the oil health record is M and the paper health grade is M, the total health degree is poor; when the oil health record is M and the paper health grade is P, the total health degree is poor; when the oil health record is M and the paper health grade is VP, the total health degree is very poor; when the oil health record is P and the paper health grade is VG, the total health degree is poor; when the oil health record is P and the paper health grade is M, the total health degree is poor; when the oil health record is P and the paper health grade is P, the total health degree is very poor; when the oil health record is P and the paper health grade is VP, the total health degree is very poor; when the oil health record is VP and the paper health grade is VG, the total health degree is poor; when the oil health record is VP and the paper health grade is M, the total health degree is very poor; when the oil health record is VP and the paper health grade is P, the total health degree is very poor; and when the oil health record is VP and the paper health grade is VP, the total health degree is worst.

[0053] De-fuzzification:

[0054] The truncated fuzzy set of the output is calculated, and the certainty index is determined; and the de-fuzzification is performed by using the following formula:

[0055]

[0056] In the formula, Z0 is the de-fuzzified output; z is the input variable; and μ(z) is the membership degree of the truncated output membership function.

[0057] The application further provides a system for realizing the power transformer health state determination method, comprising a data acquisition module, a data processing module, an association analysis module, a priority determination module, a fuzzy control module and a state determination module; the data acquisition module, the data processing module, the association analysis module, the priority determination module, the fuzzy control module and the state determination module are sequentially connected; the data acquisition module is used for acquiring attribute data information of the power transformer and uploading the data information to the data processing module; the data processing module is used for normalizing the acquired attribute data information according to the received data information, calculating corresponding grey correlation coefficients and uploading the data information to the association analysis module; the association analysis module is used for calculating the weight value of the acquired attribute data information by using the analytic hierarchy process according to the received data information, calculating the grey correlation degree value in combination with the obtained grey correlation coefficients and uploading the data information to the priority determination module; the priority determination module is used for dividing the attributes of the power transformer into health grades according to the obtained grey correlation degree value according to the received data information, comparing and determining the priority of each health grade and uploading the data information to the fuzzy control module; the fuzzy control module is used for associating the determined health grade with the grey correlation degree value, constructing a fuzzy logic control rule based on a fuzzy control scheme according to the received data information and uploading the data information to the state determination module; and the state determination module is used for determining the health state of the actual power transformer by using the obtained fuzzy logic control rule according to the received data information.

[0058] The power transformer health state determination method and system provided by the application can realize the health state determination of the power transformer, have higher reliability and better accuracy by acquiring and processing the data of the power transformer and predicting the state of the power transformer based on the analytic hierarchy process and the grey correlation analysis scheme. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 The application further provides a system for realizing the power transformer health state determination method, comprising a data acquisition module, a data processing module, an association analysis module, a priority determination module, a fuzzy control module and a state determination module; the data acquisition module, the data processing module, the association analysis module, the priority determination module, the fuzzy control module and the state determination module are sequentially connected; the data acquisition module is used for acquiring attribute data information of the power transformer and uploading the data information to the data processing module; the data processing module is used for normalizing the acquired attribute data information according to the received data information, calculating corresponding grey correlation coefficients and uploading the data information to the association analysis module; the association analysis module is used for calculating the weight value of the acquired attribute data information by using the analytic hierarchy process according to the received data information, calculating the grey correlation degree value in combination with the obtained grey correlation coefficients and uploading the data information to the priority determination module; the priority determination module is used for dividing the attributes of the power transformer into health grades according to the obtained grey correlation degree value according to the received data information, comparing and determining the priority of each health grade and uploading the data information to the fuzzy control module; the fuzzy control module is used for associating the determined health grade with the grey correlation degree value, constructing a fuzzy logic control rule based on a fuzzy control scheme according to the received data information and uploading the data information to the state determination module; and the state determination module is used for determining the health state of the actual power transformer by using the obtained fuzzy logic control rule according to the received data information.

[0060] Figure 2 The application further provides a system for realizing the power transformer health state determination method, comprising a data acquisition module, a data processing module, an association analysis module, a priority determination module, a fuzzy control module and a state determination module; the data acquisition module, the data processing module, the association analysis module, the priority determination module, the fuzzy control module and the state determination module are sequentially connected; the data acquisition module is used for acquiring attribute data information of the power transformer and uploading the data information to the data processing module; the data processing module is used for normalizing the acquired attribute data information according to the received data information, calculating corresponding grey correlation coefficients and uploading the data information to the association analysis module; the association analysis module is used for calculating the weight value of the acquired attribute data information by using the analytic hierarchy process according to the received data information, calculating the grey correlation degree value in combination with the obtained grey correlation coefficients and uploading the data information to the priority determination module; the priority determination module is used for dividing the attributes of the power transformer into health grades according to the obtained grey correlation degree value according to the received data information, comparing and determining the priority of each health grade and uploading the data information to the fuzzy control module; the fuzzy control module is used for associating the determined health grade with the grey correlation degree value, constructing a fuzzy logic control rule based on a fuzzy control scheme according to the received data information and uploading the data information to the state determination module; and the state determination module is used for determining the health state of the actual power transformer by using the obtained fuzzy logic control rule according to the received data information. DETAILED DESCRIPTION

[0061] As Figure 1 The application further provides a system for realizing the power transformer health state determination method, comprising a data acquisition module, a data processing module, an association analysis module, a priority determination module, a fuzzy control module and a state determination module; the data acquisition module, the data processing module, the association analysis module, the priority determination module, the fuzzy control module and the state determination module are sequentially connected; the data acquisition module is used for acquiring attribute data information of the power transformer and uploading the data information to the data processing module; the data processing module is used for normalizing the acquired attribute data information according to the received data information, calculating corresponding grey correlation coefficients and uploading the data information to the association analysis module; the association analysis module is used for calculating the weight value of the acquired attribute data information by using the analytic hierarchy process according to the received data information, calculating the grey correlation degree value in combination with the obtained grey correlation coefficients and uploading the data information to the priority determination module; the priority determination module is used for dividing the attributes of the power transformer into health grades according to the obtained grey correlation degree value according to the received data information, comparing and determining the priority of each health grade and uploading the data information to the fuzzy control module; the fuzzy control module is used for associating the determined health grade with the grey correlation degree value, constructing a fuzzy logic control rule based on a fuzzy control scheme according to the received data information and uploading the data information to the state determination module; and the state determination module is used for determining the health state of the actual power transformer by using the obtained fuzzy logic control rule according to the received data information.

[0062] S1. acquiring attribute data information of the power transformer; specifically comprising the following steps:

[0063] Attribute data information of the power transformer is acquired; the attribute data information includes H2 concentration, CH4 concentration, C2H2 concentration, C2H4 concentration, C2H6 concentration, acidity, breakdown voltage, CO concentration and CO2 concentration, etc.

[0064] S2. The attribute data information acquired in step S1 is normalized, and corresponding grey correlation coefficients are calculated; specifically including the following steps:

[0065] The acquired attribute data information is normalized by using the following rules:

[0066] For attribute data information of positive attributes, the following formula is used for normalization:

[0067]

[0068] For attribute data information of negative attributes, the following formula is used for normalization:

[0069]

[0070] In the formula, x ij is the grey relation unit of the jth attribute of the ith sample after normalization; y ij is the attribute value of the jth attribute of the ith sample before normalization; Min{y i} is the minimum value of the jth attribute of the ith sample; Max{y i} is the maximum value of the jth attribute of the ith sample;

[0071] Wherein, the definition of positive attribute is: the higher the value, the better the performance of the attribute; the definition of negative attribute is: the lower the value, the better the performance of the attribute;

[0072] The following formula is used to calculate the grey correlation coefficient of each attribute data:

[0073]

[0074] In the formula, γ ij is the grey correlation coefficient of y ij ; Δ ij is the deviation value of x ij from the ideal value, and Δ ij = |1-x ij |; ζ is a set coefficient, and the value range is [0, 1]; Δ min is the minimum value of the deviation value; Δ max is the maximum value of the deviation value;

[0075] S3. The weight value of the attribute data information obtained in step S1 is calculated by using the analytic hierarchy process, and the grey correlation degree value is calculated in combination with the grey correlation coefficient obtained in step S2; specifically comprising the following steps:

[0076] The weight value w of the attribute data information obtained in step S1 is calculated by using the analytic hierarchy process j ;

[0077] The grey correlation degree value is calculated by using the following formula:

[0078]

[0079] In the formula, Γ z is the grey correlation degree value; k is the number of attributes of each health level; n is the number of transformer attributes.

[0080] Step S4: According to the grey correlation degree value obtained in step S3, the attributes of the power transformer are divided into health levels, and the priority of each health level is determined, specifically comprising the following steps:

[0081] According to the grey correlation degree value obtained in step S3, the attributes of the power transformer are divided into three health levels: dissolved gas level (DGG), oil health level (OHG) and paper health level (PHG); the weight of the jth attribute w j is calculated by using the formula in step S3, γ ij is the grey correlation coefficient corresponding to the attribute, Γ1, Γ2 and Γ3 represent DGG (k=5), OHG (k=4) and PHG (k=3) respectively.

[0082] The priority of each health level is determined by using the analytic hierarchy process;

[0083] S5. Based on the fuzzy control scheme, the health level determined in step S4 is associated with the grey correlation degree value to construct a fuzzy logic control rule; specifically comprising the following steps:

[0084] The grey correlation degree value of the attribute is taken as the input; the health state determination value of the power transformer is taken as the output;

[0085] Fuzzy stage:

[0086] The input data is calculated to obtain the corresponding membership degree by using the trapezoidal function as the membership function:

[0087]

[0088] In the formula, μ A(t) is a membership of trapezoidal membership function; t is input data; a is an upper limit value of membership function; b is a lower limit value of membership function; c1 is a first center of membership function; c2 is a second center of membership function;

[0089] The obtained membership is divided into very good VG, medium M, poor P and very poor VP; wherein, very good VG corresponds to a range of 0.1634-0.2021, medium M corresponds to a range of 0.1296-0.1734, poor P corresponds to a range of 0.0958-0.1396, and very poor VP corresponds to a range of 0.0669-0.1058;

[0090] The reasoning stage is:

[0091] The fuzzy rule is constructed, and a fuzzy output value is obtained according to the input data; wherein the fuzzy output value corresponds to the health state judgment result of the power transformer;

[0092] The constructed fuzzy rule is:

[0093] When the dissolved gas level is VG:

[0094] When the oil health record is VG and the paper health level is VG, the total health degree is the best; when the oil health record is VG and the paper health level is M, the total health degree is good; when the oil health record is VG and the paper health level is P, the total health degree is good; when the oil health record is VG and the paper health level is VP, the total health degree is medium; when the oil health record is M and the paper health level is VG, the total health degree is good; when the oil health record is M and the paper health level is M, the total health degree is good; when the oil health record is M and the paper health level is P, the total health degree is medium; when the oil health record is M and the paper health level is VP, the total health degree is poor; when the oil health record is P and the paper health level is VG, the total health degree is good; when the oil health record is P and the paper health level is M, the total health degree is medium; when the oil health record is P and the paper health level is P, the total health degree is poor; when the oil health record is P and the paper health level is VP, the total health degree is poor; when the oil health record is VP and the paper health level is VG, the total health degree is medium; when the oil health record is VP and the paper health level is M, the total health degree is poor; when the oil health record is VP and the paper health level is P, the total health degree is poor; and when the oil health record is VP and the paper health level is VP, the total health degree is poor;

[0095] When the dissolved gas level is M:

[0096] When the oil health rating is VG and the paper health rating is VG, the overall health is good; when the oil health rating is VG and the paper health rating is M, the overall health is good; when the oil health rating is VG and the paper health rating is P, the overall health is fair; when the oil health rating is VG and the paper health rating is VP, the overall health is poor; when the oil health rating is M and the paper health rating is VG, the overall health is good; when the oil health rating is M and the paper health rating is M, the overall health is fair; when the oil health rating is M and the paper health rating is P, the overall health is poor; when the oil health rating is M and the paper health rating is VP, the overall health is poor; when the oil health rating is P and the paper health rating is VG, the overall health is fair; when the oil health rating is P and the paper health rating is M, the overall health is poor; when the oil health rating is P and the paper health rating is P, the overall health is poor; when the oil health rating is P and the paper health rating is VP, the overall health is very poor; when the oil health rating is VP and the paper health rating is VG, the overall health is poor; when the oil health rating is VP and the paper health rating is M, the overall health is poor; when the oil health rating is VP and the paper health rating is P, the overall health is very poor; and when the oil health rating is VP and the paper health rating is VP, the overall health is very poor.

[0097] When the dissolved gas rating is P:

[0098] When the oil health rating is VG and the paper health rating is VG, the overall health is good; when the oil health rating is VG and the paper health rating is M, the overall health is fair; when the oil health rating is VG and the paper health rating is P, the overall health is poor; when the oil health rating is VG and the paper health rating is VP, the overall health is poor; when the oil health rating is M and the paper health rating is VG, the overall health is fair; when the oil health rating is M and the paper health rating is M, the overall health is poor; when the oil health rating is M and the paper health rating is P, the overall health is poor; when the oil health rating is M and the paper health rating is VP, the overall health is poor; when the oil health rating is P and the paper health rating is VG, the overall health is fair; when the oil health rating is P and the paper health rating is M, the overall health is poor; when the oil health rating is P and the paper health rating is P, the overall health is poor; when the oil health rating is P and the paper health rating is VP, the overall health is very poor; when the oil health rating is VP and the paper health rating is VG, the overall health is poor; when the oil health rating is VP and the paper health rating is M, the overall health is poor; when the oil health rating is VP and the paper health rating is P, the overall health is very poor; and when the oil health rating is VP and the paper health rating is VP, the overall health is very poor.

[0099] When the dissolved gas rating is VP:

[0100] When the oil health is registered as VG and the paper health is registered as VG, the total health is medium; when the oil health is registered as VG and the paper health is registered as M, the total health is poor; when the oil health is registered as VG and the paper health is registered as P, the total health is poor; when the oil health is registered as VG and the paper health is registered as VP, the total health is poor; when the oil health is registered as M and the paper health is registered as VG, the total health is poor; when the oil health is registered as M and the paper health is registered as M, the total health is poor; when the oil health is registered as M and the paper health is registered as P, the total health is poor; when the oil health is registered as M and the paper health is registered as VP, the total health is very poor; when the oil health is registered as P and the paper health is registered as VG, the total health is poor; when the oil health is registered as P and the paper health is registered as M, the total health is poor; when the oil health is registered as P and the paper health is registered as P, the total health is very poor; when the oil health is registered as P and the paper health is registered as VP, the total health is very poor; when the oil health is registered as VP and the paper health is registered as VG, the total health is poor; when the oil health is registered as VP and the paper health is registered as M, the total health is very poor; when the oil health is registered as VP and the paper health is registered as P, the total health is very poor; when the oil health is registered as VP and the paper health is registered as VP, the total health is worst.

[0101] De-fuzzification:

[0102] The truncated fuzzy set of the output is calculated, and the certainty index is determined; the de-fuzzification is processed by using the following formula:

[0103]

[0104] In the formula, Z0 is the de-fuzzified output; z is the input variable; and μ(z) is the membership degree of the truncated output membership function.

[0105] S6. The health state of the actual power transformer is determined by using the fuzzy logic control rule obtained in step S5.

[0106] As Figure 2The system for realizing the power transformer health state determination method comprises a data acquisition module, a data processing module, an association analysis module, a priority determination module, a fuzzy control module and a state determination module; the data acquisition module, the data processing module, the association analysis module, the priority determination module, the fuzzy control module and the state determination module are sequentially connected; the data acquisition module is used for acquiring attribute data information of the power transformer and uploading the data information to the data processing module; the data processing module is used for normalizing the acquired attribute data information according to the received data information, calculating corresponding grey correlation coefficients and uploading the data information to the association analysis module; the association analysis module is used for calculating weight values of the acquired attribute data information according to the received data information by using an analytic hierarchy process, calculating grey correlation degree values in combination with the obtained grey correlation coefficients and uploading the data information to the priority determination module; the priority determination module is used for dividing the attributes of the power transformer into health grades according to the received data information and the obtained grey correlation degree values, comparing and determining the priority of each health grade and uploading the data information to the fuzzy control module; the fuzzy control module is used for associating the determined health grades with the grey correlation degree values, constructing fuzzy logic control rules based on a fuzzy control scheme and uploading the data information to the state determination module; and the state determination module is used for determining the actual health state of the power transformer by using the obtained fuzzy logic control rules.

Claims

1. A power transformer health state determination method, comprising the following steps: S1. Obtaining attribute data information of the power transformer; specifically comprising the following steps: Obtaining attribute data information of the power transformer; the attribute data information comprises H2 concentration, CH4 concentration, C2H2 concentration, C2H4 concentration, C2H6 concentration, acidity, breakdown voltage, CO concentration and CO2 concentration; S2. Normalizing the attribute data information obtained in step S1 and calculating the corresponding grey correlation coefficient; S3. Calculating the weight value of the attribute data information obtained in step S1 by using the analytic hierarchy process, and calculating the grey correlation degree value in combination with the grey correlation coefficient obtained in step S2; specifically comprising the following steps: The weight value w of the attribute data information obtained in step S1 is calculated by using the analytic hierarchy process j ; The grey correlation degree value is calculated by using the following formula: where Γ z is the grey correlation degree value; k is the number of attributes of each health level; n is the number of transformer attributes; S4. Dividing the attributes of the power transformer into health levels according to the grey correlation degree value obtained in step S3, and comparing and determining the priority of each health level; S5. Associating the health level determined in step S4 with the grey correlation degree value based on a fuzzy control scheme to construct a fuzzy logic control rule; S6. Using the fuzzy logic control rule obtained in step S5 to determine the health state of the actual power transformer.

2. The power transformer health condition determination method according to claim 1, characterized in that The normalization of the attribute data information obtained in step S1 and the calculation of the corresponding grey correlation coefficient in step S2 specifically comprise the following steps: The obtained attribute data information is normalized by using the following rules: For attribute data information of positive attributes, the following formula is used for normalization: For attribute data information of negative attributes, the following formula is used for normalization: where x ij is the grey relation unit of the jth attribute of the ith sample after normalization; y ij is the attribute value of the jth attribute of the ith sample before normalization; Min{y i} is the minimum value of the jth attribute of the ith sample; Max{y i} is the maximum value of the jth attribute of the ith sample. Wherein, the definition of positive attributes is: the higher the value, the better the performance; the definition of negative attributes is: the lower the value, the better the performance; The grey correlation coefficient of each attribute data is calculated by using the following formula: In the formula γ ij For y ij The grey relational coefficient; Δ ij For x ij The deviation from the ideal value, and Δ ij =|1-x ij |;ζ is the set common coefficient; Δ min The minimum value of the deviation; Δ max This represents the maximum value of the deviation.

3. The power transformer health condition determination method according to claim 2, characterized in that The division of the attributes of the power transformer into health levels according to the grey correlation degree value obtained in step S3 and the comparison and determination of the priority of each health level in step S4 specifically comprise the following steps: According to the grey correlation degree value obtained in step S3, the attributes of the power transformer are divided into three health levels: dissolved gas level, oil health level and paper health level; The priority of each health level is compared and determined by using the analytic hierarchy process.

4. The power transformer health condition determination method according to claim 3, characterized in that The association of the health level determined in step S4 with the grey correlation degree value based on the fuzzy control scheme in step S5 to construct the fuzzy logic control rule specifically comprises the following steps: The grey correlation degree value of the attribute is taken as the input; Fuzzy stage: The input data is calculated to obtain the corresponding membership degree by using the trapezoidal function as the membership function: where μ A (t) is the membership of the trapezoidal membership function; t is the input data; a is the upper limit value of the membership function; b is the lower limit value of the membership function; c1 is the first center of the membership function; c2 is the second center of the membership function; The obtained membership degree is divided into very good VG, medium M, poor P and very poor VP; wherein, the corresponding range of very good VG is 0.1634-0.2021, the corresponding range of medium M is 0.1296-0.1734, the corresponding range of poor P is 0.0958-0.1396, and the corresponding range of very poor VP is 0.0669-0.1058; Inference stage: The fuzzy rule is constructed to obtain a fuzzy output value according to the input data, wherein the fuzzy output value corresponds to the health state determination result of the power transformer; The fuzzy rule constructed is as follows: When the dissolved gas rating is VG: When the oil health record is VG and the paper health rating is VG, the total health degree is best; when the oil health record is VG and the paper health rating is M, the total health degree is good; when the oil health record is VG and the paper health rating is P, the total health degree is good; when the oil health record is VG and the paper health rating is VP, the total health degree is medium; when the oil health record is M and the paper health rating is VG, the total health degree is good; when the oil health record is M and the paper health rating is M, the total health degree is good; when the oil health record is M and the paper health rating is P, the total health degree is medium; when the oil health record is M and the paper health rating is VP, the total health degree is poor; when the oil health record is P and the paper health rating is VG, the total health degree is good; when the oil health record is P and the paper health rating is M, the total health degree is medium; when the oil health record is P and the paper health rating is P, the total health degree is poor; when the oil health record is P and the paper health rating is VP, the total health degree is poor; when the oil health record is VP and the paper health rating is VG, the total health degree is medium; when the oil health record is VP and the paper health rating is M, the total health degree is poor; when the oil health record is VP and the paper health rating is P, the total health degree is poor; when the oil health record is VP and the paper health rating is VP, the total health degree is very poor; When the dissolved gas rating is M: When the oil health record is VG and the paper health rating is VG, the total health degree is good; when the oil health record is VG and the paper health rating is M, the total health degree is good; when the oil health record is VG and the paper health rating is P, the total health degree is medium; when the oil health record is VG and the paper health rating is VP, the total health degree is poor; when the oil health record is M and the paper health rating is VG, the total health degree is good; when the oil health record is M and the paper health rating is M, the total health degree is medium; when the oil health record is M and the paper health rating is P, the total health degree is poor; when the oil health record is M and the paper health rating is VP, the total health degree is poor; when the oil health record is P and the paper health rating is VG, the total health degree is medium; when the oil health record is P and the paper health rating is M, the total health degree is poor; when the oil health record is P and the paper health rating is P, the total health degree is poor; when the oil health record is P and the paper health rating is VP, the total health degree is poor; when the oil health record is VP and the paper health rating is VG, the total health degree is medium; when the oil health record is VP and the paper health rating is M, the total health degree is poor; when the oil health record is VP and the paper health rating is P, the total health degree is poor; when the oil health record is VP and the paper health rating is VP, the total health degree is very poor; When the dissolved gas rating is P: When the oil health record is VG and the paper health grade is VG, the total health degree is good; when the oil health record is VG and the paper health grade is M, the total health degree is medium; when the oil health record is VG and the paper health grade is P, the total health degree is poor; when the oil health record is VG and the paper health grade is VP, the total health degree is poor; when the oil health record is M and the paper health grade is VG, the total health degree is medium; when the oil health record is M and the paper health grade is M, the total health degree is poor; when the oil health record is M and the paper health grade is P, the total health degree is poor; when the oil health record is M and the paper health grade is VP, the total health degree is poor; when the oil health record is P and the paper health grade is VG, the total health degree is medium; when the oil health record is P and the paper health grade is M, the total health degree is poor; when the oil health record is P and the paper health grade is P, the total health degree is poor; when the oil health record is P and the paper health grade is VP, the total health degree is very poor; when the oil health record is VP and the paper health grade is VG, the total health degree is poor; when the oil health record is VP and the paper health grade is M, the total health degree is poor; when the oil health record is VP and the paper health grade is P, the total health degree is very poor; and when the oil health record is VP and the paper health grade is VP, the total health degree is worst. When the dissolved gas grade is VP: When the oil health record is VG and the paper health grade is VG, the total health degree is medium; when the oil health record is VG and the paper health grade is M, the total health degree is poor; when the oil health record is VG and the paper health grade is P, the total health degree is poor; when the oil health record is VG and the paper health grade is VP, the total health degree is poor; when the oil health record is M and the paper health grade is VG, the total health degree is poor; when the oil health record is M and the paper health grade is M, the total health degree is poor; when the oil health record is M and the paper health grade is P, the total health degree is poor; when the oil health record is M and the paper health grade is VP, the total health degree is very poor; when the oil health record is P and the paper health grade is VG, the total health degree is poor; when the oil health record is P and the paper health grade is M, the total health degree is poor; when the oil health record is P and the paper health grade is P, the total health degree is very poor; when the oil health record is P and the paper health grade is VP, the total health degree is very poor; when the oil health record is VP and the paper health grade is VG, the total health degree is poor; when the oil health record is VP and the paper health grade is M, the total health degree is very poor; when the oil health record is VP and the paper health grade is P, the total health degree is very poor; and when the oil health record is VP and the paper health grade is VP, the total health degree is worst. Defuzzification: The truncated fuzzy set of the output is calculated to determine the certainty index; the defuzzification is processed by using the following formula: In the formula, Z0 is the defuzzified output; z is the input variable; and μ(z) is the membership degree of the truncated output membership function.

5. A system for implementing the power transformer health condition determination method according to one of claims 1 to 4, characterized in that The method comprises a data acquisition module, a data processing module, a correlation analysis module, a priority determination module, a fuzzy control module and a state determination module; the data acquisition module, the data processing module, the correlation analysis module, the priority determination module, the fuzzy control module and the state determination module are sequentially connected; the data acquisition module is used for acquiring attribute data information of the power transformer and uploading the data information to the data processing module; The data processing module is used for normalizing the acquired attribute data information according to the received data information, calculating a corresponding grey correlation coefficient, and uploading the data information to the correlation analysis module; The correlation analysis module is used for calculating a weight value of the acquired attribute data information according to the received data information by using an analytic hierarchy process, calculating a grey correlation degree value in combination with the obtained grey correlation coefficient, and uploading the data information to the priority determination module; The priority determination module is used for dividing the attributes of the power transformer into health grades according to the obtained grey correlation degree value according to the received data information, comparing and determining the priority of each health grade, and uploading the data information to the fuzzy control module; The fuzzy control module is used for associating the determined health grades with the grey correlation degree value, constructing a fuzzy logic control rule based on a fuzzy control scheme according to the received data information, and uploading the data information to the state determination module; The state determination module is used for determining the health state of the actual power transformer by using the obtained fuzzy logic control rule according to the received data information.

Images