Transformer state evaluation method of four-level hierarchical evaluation model based on health index

Abstract

The invention relates to a transformer state evaluation method, and particularly discloses a transformer state evaluation method of a four-level hierarchical evaluation model based on health indexes, which comprises the following five steps of: 1, constructing an ontology information layer health index model according to factory design, external environment, operation load and other information; 2, constructing a state information layer health index model according to the data of gas dissolved in oil, the data of oil components, the volume fraction of furfural and other information; 3, constructing a health index model of a correction information layer according to information such as a cooling mode, a historical fault frequency and an iron core grounding current; 4, constructing a comprehensive information layer health index model according to the health indexes obtained by the first three layers of models, and calculating a final health index; and 5, predicting the residual life of the transformer according to the final health index. According to the invention, the operation state of the transformer can be scientifically evaluated, and the purpose of residual life prediction is achieved.

Description

[0001] This invention relates to the field of transformer condition assessment research, specifically to a transformer condition assessment technology based on a four-level hierarchical assessment model with a health index. Background Technology

[0002] Power transformers, as indispensable key equipment in modern power systems, primarily operate on the principle of electromagnetic induction, undertaking the crucial mission of voltage transformation and power transmission. By converting high and low voltages to adapt to different electricity demands, they act as a bridge in the widespread transmission of electricity. However, in modern power systems, with the ever-increasing demand from the power grid, the health and lifespan management of transformers, as core equipment of the power grid, have become important concerns for the safe operation of the power system. Transformers not only affect the stability of power supply, but their failures often cause serious economic losses and social impacts.

[0003] Statistical data shows that the average service life of a power transformer is 20 to 30 years. However, the specific lifespan is affected by various factors, such as design and manufacturing processes, environmental conditions, load levels, the aging of internal insulation materials, and electrical and mechanical failures of equipment components. Due to the combined effects of these factors, the service life of a transformer exhibits significant uncertainty and individual differences, leading to a discrepancy between its actual and theoretical lifespan. Therefore, regularly assessing the health status and operational condition of power transformers is crucial for determining their lifespan and developing appropriate maintenance plans. Effective health status assessments can not only extend the service life of transformers but also provide strong assurance for the safe operation of the power system.

[0004] Given the complexity of the influence of various indicators on the lifespan of power transformers, establishing a mathematical model to establish the relationship between transformer evaluation indicators and remaining lifespan has become a research hotspot. Health index theory is an assessment method that quantifies the health status of equipment by analyzing its operating history and aging patterns. It not only intuitively reflects the current health status of the equipment but also provides a basis for predicting its remaining lifespan. This invention, based on health index theory, constructs health index models at different levels, combines expert experience and relevant standards, and uses weighted calculations to obtain the final health index, thereby achieving the goal of predicting the remaining lifespan of transformers. Summary of the Invention

[0005] A transformer condition assessment method based on a four-level hierarchical assessment model with a health index is proposed, which can scientifically assess the transformer's operating status and simultaneously predict its remaining life.

[0006] This invention employs the following technical solution: a transformer condition assessment method based on a four-level hierarchical assessment model with a health index, comprising the following steps:

[0007] The first step is to construct a health index model for the ontology information layer based on information such as factory design, external environment, and operating load.

[0008] The second step is to construct a state information layer health index model based on information such as dissolved gas data, oil composition data, and furfural volume fraction.

[0009] The third step is to construct a modified information layer health index model based on information such as cooling method, number of historical faults, and core grounding current.

[0010] The fourth step is to construct a comprehensive information layer health index model based on the health index obtained from the first three layers of the model, calculate the final health index, and predict the remaining life of the transformer based on the final health index.

[0011] Furthermore, the specific steps of the first step are as follows:

[0012] Step 1: Determine the aging factor B;

[0013] The formula for calculating health level is:

[0014]

[0015] Wherein, HI is the final health index, HI0 is the initial health index of the equipment, B is the aging coefficient, T1 is the year the equipment was put into operation, and T2 is the current year.

[0016] The formula for calculating health level can be derived as follows:

[0017]

[0018] Step 2: Determine T′ exp ;

[0019] Let T′ exp =T2-T1,T′ exp This refers to the expected service life based on the original design life of the transformer and taking into account factors such as the operating load.

[0020] Transformers typically have an initial expected service life (T) specified at the time of manufacture, based on national standards and industry requirements. exp Generally, it's 30 years. Based on the original expected lifespan T... exp After correcting for it using the reserve factor, load factor, and ambient temperature, the result can be obtained.

[0021]

[0022] Among them, T′ exp f is the expected service life based on the transformer's original design life and taking into account factors such as operating load. s f is the reserve coefficient. ef1 is the environmental factor, and f2 is the load factor.

[0023] Step 3: Determine the ontology information health index HI1.

[0024] Combined with T′ exp The ontology information layer health index model, constructed from the definition of B, is as follows:

[0025]

[0026] Furthermore, the specific steps of the second step are as follows:

[0027] Step 1: Determine the Health Index (HI) of dissolved gases in the oil. 2a ;

[0028] Multiply the volume fraction levels V of the five gases (H2, CH4, C2H6, C2H4, and C2H2) by their corresponding weights C to calculate the score S for each gas component.

[0029] S i =V i ×C i

[0030] The dissolved gas health index (HI) in oil is obtained based on the calculated values ​​of various gas components. 2a :

[0031]

[0032] Step 2: Determine the Oil Health Index (HI) 2b ;

[0033] HI 2b The oil quality is mainly determined by four data points: trace water content, acid value, dielectric loss, and breakdown voltage. The specific principle is similar to that of HI. 2a The calculation method is exactly the same: multiply the corresponding oil quality index grade V by the corresponding weight C and then add them together.

[0034]

[0035] Step 3: Determine the furfural health index (HI) 2c ;

[0036] Furfural is an organic compound that decomposes during the thermal aging of transformer oil. The degree of polymerization (DP) of paper and the volume fraction of furfural (FFA) satisfy an empirical formula:

[0037] DP = -123.6 × ln(FFA) + 456.38

[0038] When the furfural value in the insulating oil is 5, corresponding to a polymerization degree DP = 250, the equipment insulation level reaches the design lower limit and must be decommissioned. In this case, the health index HI = 7. When the furfural value in the insulating oil is 0.01, corresponding to a polymerization degree DP = 1000, the equipment insulation material is in good condition and performs excellently; in this case, HI = 0.1 is set. From the above relationships and boundary conditions, we can deduce...

[0039] HI 2c =2.06×FFA 0.4815

[0040] Among them, HI 2c FFA stands for furfural health index, and FFA is the furfural value in insulating oil.

[0041] Step 4: Determine the health index HI2.

[0042] Combining HI 2a HI 2b and HI 2c Three health indices are used to construct a state information layer health index model:

[0043]

[0044] in, and HI 2a HI 2b and HI 2c The weight.

[0045] Furthermore, the specific steps of the third step are as follows:

[0046] Step 1: Determine the cooling method correction factor Cor1;

[0047] When the cooling method is oil-immersed self-cooling or oil-immersed air cooling, Cor1 = 1; when the cooling method is forced oil circulation, Cor1 = 0.96; when the cooling method is forced guided oil circulation, Cor1 = 0.95.

[0048] Step 2: Determine the historical failure count correction factor Cor2;

[0049] The number of failures in the past 5 years is n. When n=0, Cor2 is 0.96; when n=1, Cor2=1; when n=2~4, Cor2=1.05; when n=5~10, Cor2=1.2; when n>10, Cor2=1.5.

[0050] Step 3: Determine the core grounding current correction factor Cor3;

[0051] The iron core grounding current I, A. When I = 0 A, Cor3 = 1; when 0 A < I ≤ 0.1 A, Cor3 = 1.05; when 0.1 A < I ≤ 0.3 A, Cor3 = 1.1; when I > 0.3 A, Cor3 = 1.2.

[0052] Step 4: Determine the partial discharge correction factor Cor4;

[0053] When there is partial discharge, Cor4 = 1.2; when there is no partial discharge, Cor4 = 1.

[0054] Step 5: Determine the near - zone short - circuit correction factor Cor5;

[0055] When there is a near - zone short - circuit, Cor5 = 1.05; when there is no near - zone short - circuit, Cor5 = 1.

[0056] Step 6: Determine the correction information health index HI3.

[0057] If the relevant data of the above 5 correction factors are missing, no correction is made, that is, the corresponding correction factor is 1. Based on the 5 correction factors, the correction information layer health index model is constructed as:

[0058]

[0059] Further, the specific steps of the fourth step are as follows:

[0060] Step 1: Determine the final health index HI;

[0061] Based on the health index obtained from the first three - layer model, construct the comprehensive information layer health index model:

[0062] HI = w1×HI1 + w2×HI2 + w3×HI3

[0063] Where w1, w2, and w3 are the weights of HI1, HI2, and HI3 respectively.

[0064] Step 2: Determine the remaining life EOL of the transformer.

[0065] From the theoretical definition formula of the health index, the relationship between the final health index HI and the remaining service life of the transformer can be obtained. That is, when HI = 7, it is considered that the transformer reaches the life critical point. Therefore, the remaining service life can be obtained as:

[0066]

[0067] The beneficial effects of the present invention are:

[0068] This invention proposes a transformer condition assessment method based on a four-level hierarchical evaluation model using a health index. First, a health index model at the body information layer is constructed based on information such as factory design, external environment, and operating load. Second, a health index model at the state information layer is constructed based on information such as dissolved gas data, oil composition data, and furfural volume fraction. Third, a modified health index model at the core grounding current is constructed based on information such as cooling method, historical fault count, and core grounding current. Finally, a comprehensive health index model at the integrated information layer is constructed based on the health indices obtained from the first three layers, the final health index is calculated, and the remaining life of the transformer is predicted based on the final health index. The proposed method can scientifically assess the operating status of transformers and simultaneously achieve the goal of predicting remaining life. Attached Figure Description

[0069] Figure 1 This is a flowchart of the method involved in this invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. The specific embodiments described herein are merely illustrative and are not intended to limit the invention.

[0071] The first step is to construct a health index model for the ontology information layer based on information such as factory design, external environment, and operating load. The specific steps are as follows:

[0072] S101: Determine the aging factor B;

[0073] The formula for calculating health level is:

[0074]

[0075] Wherein, HI is the final health index, HI0 is the initial health index of the equipment, B is the aging coefficient, T1 is the year the equipment was put into operation, and T2 is the current year.

[0076] The correspondence between the transformer health index and health level is shown in Table 1.

[0077] Table 1. Correlation between Transformer Health Index and Health Level

[0078]

[0079] The formula for calculating health level can be derived as follows:

[0080]

[0081] S102: Determine T′ exp ;

[0082] Let T′exp =T2-T1,T′ exp This refers to the expected service life based on the transformer's original design life and taking into account factors such as operating load. The transformer uses T′. exp After the New Year, which signifies the end of life, the health index HI is equal to 7, and one must retire.

[0083]

[0084] Transformers typically have an initial expected service life (T) specified at the time of manufacture, based on national standards and industry requirements. exp Generally, it's 30 years. Based on the original expected lifespan T... exp After correcting for it using the reserve factor, load factor, and ambient temperature, the result can be obtained.

[0085]

[0086] Among them, T′ exp f is the expected service life based on the transformer's original design life and taking into account factors such as operating load. s f is the reserve coefficient. e f1 is the environmental factor, and f2 is the load factor.

[0087] The transformer reserve factor and load factor are shown in Tables 2 and 3.

[0088] Table 2 Transformer Reserve Coefficient

[0089]

[0090] Table 3 Transformer Load Factor

[0091]

[0092] The transformer load factor is the ratio of the transformer's average load to its rated capacity.

[0093]

[0094] Environmental coefficient f e From indoor and outdoor coefficient f e1 and temperature coefficient f e2 Joint decision

[0095] f e =f e1 ×f e2

[0096] The indoor-outdoor coefficient is shown in Table 4-4, and the highest temperature coefficient of the substation in the past year is shown in Tables 4 and 5.

[0097] Table 4 Indoor-Outdoor Coefficient

[0098]

[0099] Table 5: Highest Temperature Coefficient of Substation in the Past Year

[0100]

[0101] S103: Determine the ontology information health index HI1.

[0102] Combined with T′ exp The ontology information layer health index model, constructed from the definition of B, is as follows:

[0103]

[0104] The second step involves constructing a state information layer health index model based on data such as dissolved gas in the oil, oil composition data, and furfural volume fraction. The specific steps are as follows:

[0105] S201: Determining the Health Index (HI) of Dissolved Gases in Oil 2a ;

[0106] Based on the volume fractions of the five gases H2, CH4, C2H6, C2H4, and C2H2, the corresponding volume fraction levels V are calculated, and the specific stratification is shown in Table 6.

[0107] Table 6 Oil Chromatography Gas Classification

[0108]

[0109] Multiply the volume fraction levels V of the five gases (H2, CH4, C2H6, C2H4, and C2H2) by their corresponding weights C to calculate the score S for each gas component.

[0110] S i =V i ×C i

[0111] The weights of the gases in the oil chromatography are shown in Table 7.

[0112] Table 7 Oil Chromatography Gas Weights

[0113]

[0114] The dissolved gas health index (HI) in oil is obtained based on the calculated values ​​of various gas components. 2a :

[0115]

[0116] S202: Determine the Oil Health Index (HI) 2b ;

[0117] HI 2bThe oil quality is mainly determined by four data points: trace water content, acid value, dielectric loss, and breakdown voltage. The specific principle is similar to that of HI. 2a The calculation methods are exactly the same, and the classification table and weight table are shown in Tables 8 and 9.

[0118] Table 8 Oil Quality Data Grading

[0119]

[0120] Table 9. Weighting of Oil Quality Data

[0121]

[0122] The result is obtained by multiplying the corresponding oil quality index grade V by the corresponding weight C and then adding them together:

[0123]

[0124] S203: Determining the Furfural Health Index (HI) 2c ;

[0125] Furfural is an organic compound that decomposes during the thermal aging of transformer oil. The degree of polymerization (DP) of paper and the volume fraction of furfural (FFA) satisfy an empirical formula:

[0126] DP = -123.6 × ln(FFA) + 456.38

[0127] The relationship between the degree of polymerization of paper insulation, furfural in oil, insulation life and health index is shown in Table 10.

[0128] Table 10 Furfural Content Grading Table

[0129]

[0130] Based on the relationships in Table 10, a power function relationship can be constructed. Using the least squares method, the specific functional relationship (Rf) between furfural and the health index can be calculated. 2 =0.9961) is:

[0131] HI 2c =2.06×FFA 0.4815

[0132] Among them, HI 2c FFA stands for furfural health index, and FFA is the furfural value in insulating oil.

[0133] S204: Determine the health index HI2 of the status information.

[0134] Combining HI 2a HI 2b and HI 2c Three health indices are used to construct a state information layer health index model:

[0135]

[0136] Among them, and are the weights of HI 2a , HI 2b and HI 2c respectively.

[0137] Combining expert experience and relevant standards, the weights of the three are determined as shown in Table 11.

[0138] Table 11 Weight Table of the Insulation Status Information Item Layer [[ID=二十]] [[ID=二十一]]

[0139] [[ID=二十二]] [[ID=二十三]] [[ID=二十四]]

[0140] [[ID=二十五]]Thus, the state information health index HI2 can be obtained: [[ID=二十六]] [[ID=二十七]]

[0141] [[ID=二十八]]HI2 = 0.4024 × HI[[ID=二十九]] 2a [[ID=三十]]+ 0.3288 × HI[[ID=三十一]] 2b [[ID=三十二]]+ 0.2688 × HI[[ID=三十三]] 2c [[ID=三十四]] [[ID=三十五]]

[0142] [[ID=三十六]]Step 3: Based on information such as the cooling method, the number of historical faults, and the iron core grounding current, construct a correction information layer health index model. The specific steps are as follows: [[ID=三十七]] [[ID=三十八]]

[0143] [[ID=三十九]]S301: Determine the cooling method correction coefficient Cor1; [[ID=四十]] [[ID=四十一]]

[0144] [[ID=四十二]]When the cooling method is oil-immersed self-cooling or oil-immersed air-cooling, Cor1 = 1; when the cooling method is forced oil circulation, Cor1 = 0.96; when the cooling method is forced directed oil circulation, Cor1 = 0.95. [[ID=四十三]] [[ID=四十四]]

[0145] [[ID=四十五]]S302: Determine the historical fault number correction coefficient Cor2; [[ID=四十六]] [[ID=四十七]]

[0146] [[ID=四十八]]The number of faults in the past 5 years is n. When n = 0, Cor2 = 0.96; when n = 1, Cor2 = 1; when n = 2 - 4, Cor2 = 1.05; when n = 5 - 10, Cor2 = 1.2; when n > 10, Cor2 = 1.5. [[ID=四十九]] [[ID=五十]]

[0147] [[ID=五十一]]S303: Determine the iron core grounding current correction coefficient Cor3; [[ID=五十二]] [[ID=五十三]]

[0148] [[ID=五十四]]The iron core grounding current is I, A. When I = 0A, Cor3 = 1; when 0A < I ≤ 0.1A, Cor3 = 1.05; when 0.1A < I ≤ 0.3A, Cor3 = 1.1; when I > 0.3A, Cor3 = 1.2. [[ID=五十五]] [[ID=五十六]]

[0149] [[ID=五十七]]S304: Determine the partial discharge correction coefficient Cor4; [[ID=五十八]]<s

[0150] When partial discharge is present, Cor4 = 1.2; when partial discharge is absent, Cor4 = 1.

[0151] S305: Determine the near-zone short-circuit correction factor Cor5;

[0152] When a near-circuit short circuit exists, Cor5 = 1.05; when a near-circuit short circuit does not exist, Cor5 = 1.

[0153] S306: If relevant data is missing for the above five correction coefficients, no correction is made, i.e., the corresponding correction coefficient is 1. Based on the five correction coefficients, the corrected information layer health index model is constructed as follows:

[0154]

[0155] The fourth step involves constructing a comprehensive information layer health index model based on the health indices obtained from the first three layers of the model, calculating the final health index, and then predicting the remaining lifespan of the transformer based on the final health index. The specific steps are as follows:

[0156] S401: Determine the final Health Index (HI);

[0157] Based on the health indices obtained from the first three layers of the model, a comprehensive information layer health index model is constructed:

[0158] HI = w1 × HI1 + w2 × HI2 + w3 × HI3

[0159] Among them, w1, w2 and w3 are the weights of HI1, HI2 and HI3, respectively.

[0160] Based on expert experience and relevant standards, the weights are determined as shown in Table 12.

[0161] Table 12 Weights of Health Indices at Each Level

[0162]

[0163] Therefore, the State Information Health Index (HI) can be obtained:

[0164] HI=0.3445HI1+0.4637HI2+0.1918HI3

[0165] S402: Determine the remaining life (EOL) of the transformer.

[0166] Based on the theoretical definition of the health index, the relationship between the final health index HI and the remaining service life of the transformer can be obtained. That is, when HI = 7, the transformer is considered to have reached the critical point of its service life. Therefore, the remaining service life can be obtained as follows:

[0167]

[0168] Example:

[0169] Taking a 500kV oil-immersed transformer that was put into operation in January 2001 as an example, condition assessment and remaining life prediction were carried out. A routine outage pre-test maintenance was conducted in June 2005, and the relevant data are shown in Tables 13 to 16.

[0170] Table 13 Basic Information Table

[0171]

[0172] According to the assessment model, the aging factor B = 0.8797, and the environmental factor f e =1.15, load factor f1 = 1.25, reserve factor f s =1, T2-T1=4.5, therefore we have:

[0173]

[0174] Table 14 Dissolved Gas Data in Oil

[0175]

[0176] Table 15 Data on oil components and furfural volume fraction

[0177]

[0178]

[0179] According to the evaluation model, HI can be obtained. 2a =2.143, HI 2b =1.092, HI 2c =0.1256, therefore:

[0180] HI2 = 0.4024 × HI 2a +0.3288×HI 2b +0.2688×HI 2c =1.255

[0181] Table 16 Correction Information Data

[0182]

[0183] According to the evaluation model, we can obtain:

[0184]

[0185] The final health index of the transformer is:

[0186] HI=0.3445HI1+0.4637HI2+0.1918HI3=1.396

[0187] Substituting into the formula, we can obtain the remaining service life as:

[0188]

[0189] The transformer has been in operation for 4.5 years, and overall, its indicators are trending towards normal. Calculations show that this transformer will reach the end of its lifespan in approximately 18.3 years. This lifespan assessment indicates that the service life of this power transformer is approximately 22.8 years, which is basically consistent with the statistical distribution of the lifespan of 500kV transformers in the scrapped transformer lifespan. Considering the actual environment in which the equipment is located, the transformer lifespan assessment is consistent with the actual operating conditions of the equipment. Therefore, it can be proven that this assessment model can scientifically assess the transformer's operating status and achieve the purpose of predicting its remaining lifespan.

Claims

1. A transformer condition assessment method based on a four-level hierarchical assessment model using a health index, characterized in that, It includes the following steps: The first step is to construct a health index model for the ontology information layer based on information such as factory design, external environment, and operating load. The second step is to construct a health index model for the status information layer based on information such as dissolved gas data in oil, oil quality component data, and furfural volume fraction. The third step is to construct a health index model for the correction information layer based on information such as cooling method, historical failure times, and core grounding current. The fourth step is to construct a health index model for the comprehensive information layer based on the health indices obtained from the first three layers, calculate the final health index, and predict the remaining life of the transformer based on the final health index.

2. The transformer condition assessment method based on a four-level hierarchical assessment model with a health index as described in claim 1, characterized in that, The specific steps of the first step are as follows: Step 1: Determine the aging coefficient B. The health level calculation formula is: Where, HI is the final health index, HI0 is the initial health index of the equipment, B is the aging coefficient, T1 is the year when the equipment was put into operation, and T2 is the current year. It can be deduced from the health level calculation formula that: Where, HI is the final health index, HI0 is the initial health index of the equipment, B is the aging coefficient, T1 is the year when the equipment was put into operation, and T'2 is the current year. Step 2: Determine T′ exp ; Let T′ exp =T2-T1,T′ exp The expected service life is based on the original design life of the transformer and takes into account factors such as the operating load. Based on the original expected service life T exp After correcting for it using the reserve factor, load factor, and ambient temperature, the result can be obtained. Among them, T′ exp f is the expected service life based on the transformer's original design life and taking into account factors such as operating load. s f is the reserve coefficient. e f1 is the environmental factor, and f2 is the load factor. Step 3: Determine the ontology information health index HI1. Combined with T′ exp The ontology information layer health index model, constructed from the definition of B, is as follows:

3. The transformer condition assessment method based on a four-level hierarchical assessment model with a health index as described in claim 1, characterized in that, The specific steps of the second step are as follows: Step 1: Determine the Health Index (HI) of dissolved gases in the oil. 2a ; Multiply the volume fraction grades V of 5 gases, namely H2, CH4, C2H6, C2H4, and C2H2, by the corresponding weights C to calculate the scores S of various gas components: S i =V i ×C i The dissolved gas health index (HI) in oil is obtained based on the calculated values ​​of various gas components. 2a : Step 2: Determine the Oil Health Index (HI) 2b ; HI 2b The oil quality is mainly determined by four data points: trace water content, acid value, dielectric loss, and breakdown voltage. The specific principle is similar to that of HI. 2a The calculation method is exactly the same: multiply the corresponding oil quality index grade V by the corresponding weight C and then add them together. Step 3: Determine the furfural health index (HI) 2c ; Furfural is an organic substance decomposed when transformer oil is thermally aged. The empirical formula between the degree of polymerization (DP) of paper and the furfural volume fraction (FFA) is: DP = -123.6×ln(FFA) + 456.38 When the furfural value in the insulating oil is 5, the corresponding degree of polymerization DP = 250, and the insulation level of the equipment reaches the design lower limit and must be taken out of service. At this time, the health index HI = 7; when the furfural value in the insulating oil is equal to 0.01, the corresponding degree of polymerization DP = 1000, and the insulation material condition of the equipment is good and the performance is excellent. At this time, HI = 0.1 is set. From the above relationship and boundary conditions, it can be deduced that HI 2c =2.06×FFA 0.4815 Among them, HI 2c FFA stands for furfural health index, and FFA is the furfural value in insulating oil. Step 4: Determine the status information health index HI2. Combining HI 2a HI 2b and HI 2c Three health indices are used to construct a state information layer health index model: in, and HI 2a HI 2b and HI 2c The weight.

4. The transformer condition assessment method based on a four-level hierarchical assessment model with a health index as described in claim 1, characterized in that, The specific steps of the third step are as follows: Step 1: Determine the cooling method correction coefficient Cor1. When the cooling method is oil-immersed self-cooling or oil-immersed air-cooling, Cor1 = 1; when the cooling method is forced oil circulation, Cor1 = 0.96; when the cooling method is forced guided oil circulation, Cor1 = 0.

95. Step 2: Determine the historical failure times correction coefficient Cor2. The number of failures in the recent 5 years is n. When n = 0, Cor2 = 0.96; when n = 1, Cor2 = 1; when n = 2 - 4, Cor2 = 1.05; when n = 5 - 10, Cor2 = 1.2; when n > 10, Cor2 = 1.

5. Step 3: Determine the core grounding current correction coefficient Cor3. The core grounding current is I, A. When I = 0A, Cor3 = 1; when 0A < I ≤ 0.1A, Cor3 = 1.05; when 0.1A < I ≤ 0.3A, Cor3 = 1.1; when I > 0.3A, Cor3 = 1.

2. Step 4: Determine the partial discharge correction coefficient Cor4. When partial discharge is present, Cor4 = 1.2; when partial discharge is not present, Cor4 = 1. Step 5: Determine the near-zone short-circuit correction factor Cor5; When a near-circuit short circuit exists, Cor5 = 1.05; when a near-circuit short circuit does not exist, Cor5 = 1. Step 6: Determine the revised Information Health Index (HI3); If any of the above five correction coefficients lack relevant data, no correction is applied, meaning the corresponding correction coefficient is 1. Based on these five correction coefficients, the corrected information layer health index model is constructed as follows:

5. The transformer condition assessment method based on a four-level hierarchical assessment model with a health index as described in claim 1, characterized in that, The specific steps of the fourth step are as follows: Step 1: Determine the final Health Index (HI); Based on the health indices obtained from the first three layers of the model, a comprehensive information layer health index model is constructed: HI = w1 × HI1 + w2 × HI2 + w3 × HI3 Where w1, w2, and w3 are the weights of HI1, HI2, and HI3, respectively; Step 2: Determine the transformer's remaining life (EOL); Based on the theoretical definition of the health index, the relationship between the final health index HI and the remaining service life of the transformer can be obtained. That is, when HI = 7, the transformer is considered to have reached the critical point of its service life. Therefore, the remaining service life can be obtained as follows:

Images