Method for diagnosing the deterioration status of oil-filled transformers

Abstract

A method for diagnosing the deterioration state of an oil-filled transformer equipped with insulating oil and insulating paper, comprising: a step of preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and obtaining two or more groups of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; a step of adjusting the detected amounts of the compounds to a selection deterioration index smaller than the detected amounts of the compounds by multiplying or dividing the detected amounts of the compounds by a constant; a step of selecting one or more numerical values ​​as a deterioration index selected from a group consisting of the sum, mean, median, first quartile, third quartile, and interquartile range of the selection deterioration index based on the group of data; a step of creating a calibration curve relating the deterioration index and the average degree of polymerization based on two or more groups of data; and a step of diagnosing the deterioration state of the oil-filled transformer using the calibration curve.

Description

Technical Field

[0001] The present disclosure relates to a method for diagnosing the deterioration state of an oil-filled transformer.

Background Art

[0002] An oil-filled transformer has a structure in which a core and windings coated with insulating paper are immersed in insulating oil. The life of the transformer is considered to be partly due to the deterioration of the insulating paper, which generally cannot be rewound. The deterioration of the insulating paper occurs when surge current generated by an external short circuit during the operation of the transformer flows through the windings, and the electromagnetic force generated at that time applies a tensile force to the insulating paper coating the windings. As a result of the tensile force being applied to the insulating paper, the mechanical strength of the insulating paper decreases, and the windings coated with the insulating paper develop cracks or are destroyed.

[0003] The mechanical strength of the insulating paper is correlated with the average degree of polymerization of the insulating paper. However, it is difficult to collect the insulating paper from a transformer during operation in order to measure the average degree of polymerization of the insulating paper. Therefore, by measuring the average degree of polymerization of the insulating paper by another method, the mechanical strength of the insulating paper and further the life of the transformer can be known. Non-Patent Document 1 discloses that various compounds are generated when the insulating paper deteriorates, and the average degree of polymerization of the insulating paper can be measured from the detected amounts of these compounds. The average degree of polymerization of the insulating paper represents a measure of the length of cellulose molecules. As the deterioration of the insulating paper progresses, the average degree of polymerization of the insulating paper decreases, and the mechanical strength also decreases accordingly. Non-Patent Document 1 describes that the average degree of polymerization of the insulating paper is 400 to 600 as a guideline for the life of the transformer.

[0004] Conventionally, the relationship between the detected amount of a deterioration index compound and the average degree of polymerization has been obtained in advance, and the average degree of polymerization of the insulating paper has been determined from the detected amount of the deterioration index compound in the insulating oil collected from the transformer to be diagnosed. Non-Patent Documents 1 and 2 disclose, for example, furfural, carbon dioxide + carbon monoxide (CO2 + CO), water, acetone, and methanol as deterioration index compounds. These compounds are considered to be compounds generated by the decomposition of cellulose molecules in the insulating paper.

[0005] Patent Document 1 discloses a method for diagnosing the deterioration state of a transformer by using lignin-derived decomposition products as deterioration indicator compounds, since lignin, although a small component in insulating paper, contributes to the mechanical strength of the insulating paper. Lignin-derived decomposition products are compounds having a benzene ring structure, such as phenol, toluene, styrene, vanillin, and coniphenylaldehyde. Patent Document 1 states that the deterioration state of the transformer can be diagnosed from the total amount of the above compounds detected.

[0006] Depending on the type of insulating paper, some compounds may not be suitable for diagnosing the deterioration state of a transformer because they produce only small amounts of the aforementioned deterioration indicator compounds. Patent Document 2 discloses a method for diagnosing the deterioration of an insulating paper used in a transformer equipped with heat-resistant insulating paper that produces very little furfural, by using nitrogen-based compounds as deterioration indicator compounds. Non-Patent Document 2 discloses a method for diagnosing the deterioration of a transformer equipped with heat-resistant insulating paper using methanol.

[0007] Patent Document 3 discloses that, for a transformer equipped with heat-resistant insulating paper, it is possible to diagnose the deterioration of the transformer from the total amount of any two or more compounds among the five furan compounds: furfural, 5-methylfurfural, 5-hydroxymethylfurfural, 2-acetylfuran, and 2-furfuryl alcohol, and that it is particularly preferable to diagnose the deterioration of the transformer from the total amount of all five compounds.

[0008] Non-patent document 3 points out that methanol is not useful as a degradation indicator compound used for diagnosing the degradation of transformers due to its high volatility. [Prior art documents] [Patent Documents]

[0009] [Patent Document 1] Japanese Patent Publication No. 2014-062858 [Patent Document 2] Japanese Patent Publication No. 2020-085817 [Patent Document 3] Japanese Patent Publication No. 2024-043734 [Non-patent literature]

[0010] [Non-Patent Document 1] "Maintenance and Management of Oil-Immersed Transformers, Part IV: Deterioration Diagnosis of Oil-Immersed Transformers," Electrical Cooperative Research, Japan Electrical Cooperative Research Association, February 25, 1999, Vol. 54, No. 5 (Part 1), pp. 158-169. [Non-Patent Document 2] Oscar H. Arroyo et.al., “Relationships between Methanol Marker and Mechanical Performance of Electrical Insulation Papers for Power Transformers under Accelerated Thermal Aging”, IEEE Transactions on Dielectrics and Electrical Insulation Vol. 22, No. 6; December 2015, pp3625-3632 [Non-Patent Document 3] Estrela Mariana Prux von Steinkirch Souza et.al., “Evaluation of the Chemical Stability of Methanol Generated during Paper Degradation in Power Transformers”, IEEE Transactions on Dielectrics and Electrical Insulation Vol. 23, No. 5; October 2016, pp3209-3214 [Overview of the Initiative] [Problems that the invention aims to solve]

[0011] In transformer degradation diagnosis, the evaluation of the same compound as a degradation indicator compound can vary. For example, in the degradation diagnosis of a transformer equipped with heat-resistant insulating paper, Non-Patent Document 2 discloses that methanol can be used as a degradation indicator compound, while Non-Patent Document 3 states that methanol is not suitable as a degradation indicator compound because its high volatility makes reproducibility difficult to obtain.

[0012] It is a well-known problem that the average degree of polymerization of insulating paper, determined from a single degradation indicator compound, fluctuates depending on the amount of the degradation indicator compound detected, thus degrading the accuracy of transformer degradation diagnosis.

[0013] Several factors can cause fluctuations in the detected levels of degradation indicator compounds. Highly volatile compounds, such as methanol, acetone, and CO2+CO, may fluctuate in detection levels due to evaporation. Furthermore, furfural compounds, widely used in the degradation diagnosis of insulating paper, are produced in extremely small amounts from heat-resistant insulating paper, leading to fluctuations in detection levels. Additionally, the operating conditions of the transformer, such as the amount of oxygen and moisture in the transformer, can also cause fluctuations in detection levels.

[0014] Furthermore, in order to improve the above-mentioned problems, Patent Document 3 discloses the use of total values ​​and average values ​​for the detected amounts of multiple degradation indicator compounds. However, in many cases, the accuracy of transformer degradation diagnosis did not improve. This was thought to be because even if total values ​​and average values ​​for multiple degradation indicator compounds were calculated, these values ​​were influenced by the degradation indicator compound with the highest detected amount.

[0015] Patent Document 3 shows the relationship between the total amount of 5-methylfurfural (5MEF) and 5-hydroxymethylfurfural (5HMF) and the average degree of polymerization remaining in Figure 4, and the relationship between the total amount of 5MEF and 2-acetylfuran (2ACF) and the average degree of polymerization remaining in Figure 5. In the above case, even if 5HMF in Figure 4 and 2ACF in Figure 5 are swapped, the graphs in Figures 4 and 5 will show the same results. This indicates that the detected amounts of 5HMF and 2ACF are small and therefore do not affect the total amount of the detected compounds. Furthermore, in Patent Document 3, the change in the average degree of polymerization with respect to the total amount of the five furans is small, so the average degree of polymerization cannot be accurately determined.

[0016] Furthermore, while the sum of CO2 and CO is a well-known indicator compound for degradation, CO2 and CO are highly volatile and easily volatilize, making their detection levels prone to fluctuation. Also, the amount of CO2 produced is several tens of times greater than the amount of CO produced, meaning that the sum of CO2 and CO is almost entirely comprised of CO2. Therefore, there are challenges in using the sum of CO2 and CO as an indicator compound for degradation.

[0017] This disclosure is made in view of the above circumstances and aims to provide a method for accurately diagnosing the deterioration state of an oil-filled transformer. [Means for solving the problem]

[0018] The method of the present disclosure is a method for diagnosing the deterioration state of an oil-filled transformer including insulating oil and insulating paper, comprising the steps of preparing a reference insulating oil which is the same insulating oil as the above-mentioned insulating oil and a reference insulating paper which is the same insulating paper as the above-mentioned insulating paper, and obtaining two or more groups of a set of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; adjusting the detected amount of the compound to a selection deterioration index smaller than the detected amount of the compound by multiplying or dividing by a constant; selecting one or more numerical values selected from the group consisting of the total value, average value, median value, first quartile, third quartile, and interquartile range of the selection deterioration index based on the set of data as the deterioration index; creating a calibration curve regarding the deterioration index and the average degree of polymerization based on two or more groups of the set of data; and diagnosing the deterioration state of the oil-filled transformer using the calibration curve.

[0019] The method in another embodiment of the present implementation is a method for diagnosing the deterioration state of an oil-filled transformer including insulating oil and insulating paper, comprising the steps of preparing a reference insulating oil which is the same insulating oil as the above-mentioned insulating oil and a reference insulating paper which is the same insulating paper as the above-mentioned insulating paper, and obtaining two or more groups of a set of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; adjusting the detected amount of the compound to a selection deterioration index by one or more adjustments selected from the group consisting of normalization and standardization; selecting one or more numerical values selected from the group consisting of the average value and the median value of the selection deterioration index based on the set of data as the deterioration index; creating a calibration curve regarding the deterioration index and the average degree of polymerization based on two or more groups of the set of data; and diagnosing the deterioration state of the oil-filled transformer based on the calibration curve.

[0020] A method in yet another aspect of the present disclosure is a method for diagnosing the deterioration state of an oil-immersed transformer including insulating oil and insulating paper, the method including: preparing a reference insulating oil that is the same insulating oil as the insulating oil and a reference insulating paper that is the same insulating paper as the insulating paper; obtaining two or more groups of a set of data including the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; adjusting the detected amounts of the compounds to selection deterioration indices by logarithmic transformation; selecting, as deterioration indices, one or more numerical values selected from the group consisting of the total value and the average value of the selection deterioration indices based on the set of data; creating a calibration curve regarding the deterioration indices and the average degree of polymerization based on the two or more groups of the set of data; and diagnosing the deterioration state of the oil-immersed transformer based on the calibration curve.

[0021] A method in yet another aspect of the present disclosure is a method for diagnosing the deterioration state of an oil-immersed transformer including insulating oil and insulating paper, the method including: preparing a reference insulating oil that is the same insulating oil as the insulating oil and a reference insulating paper that is the same insulating paper as the insulating paper; obtaining two or more groups of a set of data including the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; selecting, as a deterioration index, the geometric mean value of the detected amounts of the two or more compounds in the set of data; creating a calibration curve regarding the deterioration index and the average degree of polymerization based on the two or more groups of the set of data; and diagnosing the deterioration state of the oil-immersed transformer based on the calibration curve.

[0022] A further embodiment of the present disclosure is a method for diagnosing the deterioration state of an oil-filled transformer comprising insulating oil and insulating paper, comprising the steps of: preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and determining a compound contained in the reference insulating oil for which the coefficient of determination is 0.7 or more with respect to the detected amount of the compound and the average degree of polymerization of the reference insulating paper as a deterioration indicator compound; obtaining two or more groups of data consisting of the detected amount of the deterioration indicator compound and the average degree of polymerization of the reference insulating paper; selecting one or more numerical values ​​selected from the group consisting of the sum, mean, median, first quartile, third quartile, and interquartile range of the detected amount of the deterioration indicator compound based on the data from the groups, creating a calibration curve for the deterioration indicator and the average degree of polymerization based on the two or more groups of data from the groups, and diagnosing the deterioration state of the oil-filled transformer based on the calibration curve. [Effects of the Invention]

[0023] This disclosure provides a method for accurately diagnosing the deterioration state of an oil-filled transformer. [Brief explanation of the drawing]

[0024] [Figure 1] Figure 1 is a flowchart illustrating an example of a method for diagnosing the deterioration state of an oil-filled transformer according to the present disclosure. [Figure 2] Figure 2 is a conceptual diagram of a heating test apparatus used to create a calibration curve for diagnosing the degradation state of a transformer in this disclosure. [Figure 3] Figure 3 shows calibration curves relating the average detected amount and the average degree of polymerization for test examples 1 to 9 shown in Table 6. [Figure 4] Figure 4 shows calibration curves relating the average detected amount and the average degree of polymerization for test examples 10 to 24 shown in Table 7. [Figure 5] Figure 5 shows a calibration curve relating the degradation index and average degree of polymerization in Example 1, as shown in Table 10. [Figure 6]Figure 6 shows a calibration curve relating the degradation index and average degree of polymerization in Example 2, as shown in Table 13. [Figure 7] Figure 7 shows a calibration curve relating the degradation index and average degree of polymerization in Example 3, as shown in Table 16. [Figure 8] Figure 8 shows a calibration curve relating the degradation index and average degree of polymerization in Example 4, as shown in Table 19. [Figure 9] Figure 9 shows a calibration curve relating the degradation index and average degree of polymerization in Example 5, as shown in Table 23. [Figure 10] Figure 10 shows a calibration curve relating the degradation index and average degree of polymerization in Example 6, as shown in Table 26. [Figure 11] Figure 11 shows a calibration curve relating the degradation index and average degree of polymerization in Example 7, as shown in Table 27. [Figure 12] Figure 12 shows a calibration curve relating the detected amounts of acetone and methylpyrazine in Tables 1 and 2 to the average degree of polymerization. [Figure 13] Figure 13 shows the relationship between the degradation index and the average degree of polymerization in Example 8, as shown in Table 29. [Figure 14] Figure 14 shows a calibration curve relating the average detected amount and average degree of polymerization of the compound in Reference Example 1 shown in Table 30. [Figure 15] Figure 15 shows a calibration curve relating the degradation index and average degree of polymerization in Example 9, as shown in Table 31. [Figure 16] Figure 16 is a plot of the nonpolarity parameters and boiling points for each compound. [Figure 17] Figure 17 is a plot of the polarity parameters and boiling points for each compound. [Figure 18] Figure 18 shows a graph of the compounds listed in Table 28, with Ro on the X-axis and Rc on the Y-axis. [Figure 19] Figure 19 is a graph showing the (Ro2+Rc2)1 / 2 of the compounds listed in Table 28 on the X-axis and the coefficient of determination on the Y-axis. [Figure 20]Figure 20 shows the relationship between the mean value of the coefficient of determination and the coefficient of determination. [Modes for carrying out the invention]

[0025] The embodiments of this disclosure will be described below with reference to the drawings. Note that in the drawings of this disclosure, dimensions such as length, width, thickness, and depth have been modified as appropriate for clarity and simplification, and do not represent actual dimensions.

[0026] In this disclosure, the diagnosis of the deterioration state of the oil-filled transformer to be diagnosed is achieved by estimating the average degree of polymerization of the insulating paper contained in the oil-filled transformer from the detected amounts of two or more compounds contained in the insulating oil contained in the oil-filled transformer. By estimating the average degree of polymerization of the insulating paper, the remaining lifespan of the oil-filled transformer can be estimated, making it possible to plan for the replacement of the oil-filled transformer at an appropriate time.

[0027] Embodiment 1. The method in this embodiment is a method for diagnosing the deterioration state of an oil-filled transformer equipped with insulating oil and insulating paper, and includes the steps of: preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and obtaining two or more groups of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; adjusting the detected amounts of the compounds to a selectable deterioration index smaller than the detected amounts of the compounds by multiplying or dividing the detected amounts of the compounds by a constant; selecting one or more numerical values ​​as a deterioration index selected from a group consisting of the sum, mean, median, first quartile, third quartile, and interquartile range of the selectable deterioration index based on the data from the group; creating a calibration curve relating the deterioration index and the average degree of polymerization based on two or more groups of data from the group; and diagnosing the deterioration state of the oil-filled transformer using the calibration curve.

[0028] The oil-filled transformer to be diagnosed is equipped with insulating oil and insulating paper. In the oil-filled transformer, it is preferable that the insulating paper is immersed in the insulating oil. In this specification, the oil-filled transformer to be diagnosed is also simply referred to as a transformer.

[0029] The type of insulating oil used in transformers is not particularly limited and includes, for example, mineral oil, electrically insulating oils such as alkylbenzene, and vegetable oils. Mineral oil used as an insulating oil is nonpolar.

[0030] Insulating paper contains cellulose as a constituent component, and cellulose is a highly polar molecule. Examples of insulating paper include kraft paper and heat-resistant insulating paper. Depending on the purpose, insulating paper may also contain air, copper, and pressboard. If the insulating paper is heat-resistant insulating paper, it contains a heat-resistant treatment agent. Examples of heat-resistant treatment agents include nitrogen-containing compounds such as dicyandiamide.

[0031] The method in this embodiment 1 will be described with reference to Figure 1. Figure 1 is a flowchart showing an example of a method for diagnosing the deterioration state of an oil-filled transformer according to this disclosure. The method in this embodiment 1 includes the following acquisition step S1, adjustment step S2, selection step S3, creation step S4, and diagnosis step S5. The details of each step in this embodiment 1 will be described below.

[0032] (Acquisition process) In this process, a reference insulating oil, which is the same insulating oil as the insulating oil, and a reference insulating paper, which is the same insulating paper as the insulating paper, are prepared, and two or more sets of data consisting of the detection amounts of two or more compounds contained in the reference insulating oil (hereinafter also referred to as "degradation indicator compounds") and the average degree of polymerization of the reference insulating paper are obtained.

[0033] The reference insulating oil is the same insulating oil used in the transformer being diagnosed. The reference insulating oil contains two or more degradation indicator compounds. The degradation indicator compounds contained in the reference insulating oil refer to compounds that have decomposed due to heating of the insulating paper, and may be liquid or gaseous at room temperature. Examples of degradation indicator compounds that are liquid at room temperature include compounds derived from cellulose, compounds derived from sulfur, compounds derived from heat treatment agents, and compounds derived from lignin. Examples of cellulose-derived compounds include methanol, acetone, acetic acid, furfural, furfuryl alcohol, methylfuran, and dimethylcyclohexanol. Examples of sulfur-derived compounds include dimethyl sulfone and dimethyl sulfide. Examples of heat treatment agent-derived compounds include methylpyrazine, acetamide, pyrrole, 3-methylpyrrole, 1,3-diazine, and 1-methylpyrrole. Examples of lignin-derived compounds include phenol and methylphenol. Examples of degradation indicator compounds in gases at room temperature include carbon dioxide, carbon monoxide, methane, ethane, and propane. The reference insulating oil may contain two or more of the above degradation indicator compounds, and may contain three or more.

[0034] Degradation indicator compounds contained in reference insulating oil can be detected by various methods, and can be detected in the presence of air and moisture depending on the purpose of detection. For example, detection methods include pretreatment of the recovered reference insulating oil followed by quantification. Examples of pretreatment methods for reference insulating oil include solvent extraction, headspace extraction, solid-phase extraction, solid-phase micro-extraction, and the method described in Japanese Patent Publication No. 9-72892. Methods for quantifying degradation indicator compounds include quantitative analysis using a gas chromatograph-mass spectrometer (GC / MS).

[0035] The average degree of polymerization of reference insulating paper represents the length of cellulose molecules, which are the main constituent component. Degradation of insulating paper is caused by the decomposition of cellulose, a constituent component of insulating paper, due to the effects of heating, oxygen, and moisture during transformer operation. Therefore, degradation of the insulating paper leads to a decrease in its average degree of polymerization, which in turn leads to the degradation of the transformer. The average degree of polymerization of reference insulating paper can be measured in accordance with the Japan Electrical Manufacturers' Association standard JEM1455: Method for measuring the average degree of polymerization of insulating paper for transformers (1991).

[0036] If the insulating paper of the transformer to be diagnosed is heat-resistant insulating paper, the degradation indicator compounds in the group of data obtained in acquisition step S1 may consist only of compounds derived from heat-resistant treatment agents such as methylpyrazine, acetamide, pyrrole, 3-methylpyrrole, 1,3-diazine, and 1-methylpyrrole. In this case, the number of compounds whose detection amount is measured in acquisition step S1 is reduced, which facilitates processing in subsequent steps. Among the above heat-resistant treatment agent-derived compounds, a group of data may be obtained excluding compounds that show a small tendency to increase in detection amount with the degradation of the insulating paper; for example, acetamide may be excluded.

[0037] The set of data obtained in acquisition step S1 consists of the detected amounts of two or more degradation indicator compounds and the average degree of polymerization of the reference insulating paper. The set of data can be obtained by measuring the reference insulating oil and reference insulating paper obtained under certain conditions according to the measurement method described above. There are two or more of the above degradation indicator compounds, and from the viewpoint of improving the accuracy of transformer degradation diagnosis, three or more are preferable, five or more are more preferable, and seven or more are even preferable.

[0038] In the data acquisition process S1, two or more groups of data are acquired, and from the viewpoint of improving the accuracy of transformer degradation diagnosis, it is preferable to acquire three or more groups, more preferably five or more groups, even more preferably seven or more groups, and even more preferably ten or more groups. The data in the two or more acquired groups may be acquired under different conditions, for example, under different heating conditions, or under conditions at room temperature (e.g., 25°C) before heating.

[0039] With respect to the data set obtained in step S1, from the viewpoint of improving the accuracy of transformer degradation diagnosis, the average degree of polymerization of the reference insulating paper is preferably 800 or less, more preferably 600 or less, and even more preferably 400 or less. In order to bring the average degree of polymerization of the reference insulating paper within the above numerical range, the reference insulating paper may be heated in reference insulating oil, and the temperature may be 50°C to 300°C, 100°C to 200°C, or 120°C to 150°C. Furthermore, the conditions under which the reference insulating paper is heated in reference insulating oil may be the temperature at which the actual heating of the transformer to be diagnosed is expected, or a temperature at which accelerated testing can be performed. Furthermore, the number of days for which the reference insulating paper is heated in reference insulating oil may be 300 days or less, 200 days or less, or 100 days or less. There is no particular lower limit to the number of days for which the reference insulating paper is heated in reference insulating oil.

[0040] The detection of reference insulating oil and reference insulating paper can be performed, for example, as follows: Place the reference insulating oil and reference insulating paper in any proportion into a stainless steel container and seal it. Heat this container at any temperature, and at any intervals of any number of days, detect the degradation indicator compounds in the reference insulating oil and measure the average degree of polymerization of the reference insulating paper according to the method described above.

[0041] (adjustment process) In this process, the detected amount of the degradation indicator compound is adjusted to a selectable degradation indicator smaller than the detected amount of the degradation indicator compound by multiplying or dividing it by a constant.

[0042] In this embodiment, the selective degradation index is a numerical value adjusted to reduce the detected amount of the degradation indicator compound. The selective degradation index can be adjusted by multiplying or dividing the detected amount of the degradation indicator compound by a constant. According to adjustment step S2, the difference in the detected amount for each degradation indicator compound can be reduced. The selective degradation index may be adjusted by one or more constants, or two or more constants, from the detected amount of a given degradation indicator compound, as described below.

[0043] The constant should be able to reduce the amount of degradation indicator compound detected. When multiplying the amount of degradation indicator compound detected by the constant, the constant must be greater than 0 and less than 1. When dividing the amount of degradation indicator compound detected by the constant, the constant must be greater than 1. The constant may be, for example, 0.1, 0.5, 10, or 100. The constant is not 1.

[0044] The constant may be a different value for each degradation indicator compound, or it may be the same value for all degradation indicator compounds. If the constant is a different value for each degradation indicator compound, for example, the constant may be 0.1 for methanol and 0.5 for acetone, and the value obtained from the amount of methanol detected in one group of data from two or more groups may be used as the constant, and the value obtained from the amount of acetone detected in one group of data from two or more groups may be used as the constant. The constant may also be the same value for all degradation indicator compounds. For example, the amount of all degradation indicator compounds detected may be divided by a constant of 0.1.

[0045] From the viewpoint of further improving the accuracy of transformer degradation diagnosis, the constant is preferably selected from one or more numerical values ​​chosen from a group consisting of the sum, mean, maximum, minimum, and median values ​​of the detected amounts of degradation indicator compounds in a group of two or more sets of data. For example, the mean value of the detected amounts of degradation indicator compounds in a group of two or more sets of data can be obtained as follows. Two detected amounts of a certain degradation indicator compound can be obtained from reference insulating oil heated under two different conditions, and the above mean value can be obtained by averaging these detected amounts. In this specification, "mean value" refers to the arithmetic mean unless otherwise specified.

[0046] In adjustment step S2, the amount of degradation indicator compound detected may be the measured value as is, a rounded value to facilitate the adjustment of the selection degradation indicator, a value converted to the concentration in the reference insulating oil, or a value corrected by the detection sensitivity.

[0047] The mean value of the coefficient of variation (CV value) in a group of two or more sets of data is preferably 1.7 or less, and more preferably 1.3 or less, from the viewpoint of further improving the accuracy of transformer degradation diagnosis. The CV value is obtained by dividing the standard deviation of two or more selectable degradation indicators in a group of data by the mean value of two or more selectable degradation indicators in a group of data. One CV value can be obtained for each group of data, and two or more can be obtained depending on the number of types of adjusted selectable degradation indicators.

[0048] (Selection process) In this process, one or more numerical values ​​are selected as a degradation index from a group consisting of the sum, mean, median, first quartile, third quartile, and interquartile range of a set of data.

[0049] The degradation index is a numerical value selected using a selection degradation index based on a set of data. Therefore, the degradation index is less affected by differences in the detection amount of each degradation index compound than when the degradation index is selected using the detection amount of each degradation index compound, allowing for highly accurate determination of the average degree of polymerization of the insulating paper in the transformer being diagnosed from the calibration curve in preparation step S4. The degradation index can be selected from a group consisting of the sum of two or more selection degradation indices in a set of data, the mean, the median, the first quartile, the third quartile, and the interquartile range. Based on a single set of data, one or more numerical values ​​may be selected as the degradation index, or two or more numerical values ​​may be selected.

[0050] (Creation process) In this process, a calibration curve for degradation indicators and average degree of polymerization is created based on data from two or more groups.

[0051] A calibration curve can be created by plotting the degradation index obtained based on data from two or more groups and the average degree of polymerization of the reference insulating paper. The degradation index is a numerical value selected from the amount of degradation index compound detected in acquisition step S1 by adjustment step S2 and selection step S3. The average degree of polymerization of the reference insulating paper is the same as the average degree of polymerization of the reference insulating paper for the group of data obtained in acquisition step S1. More than one calibration curve may be created, and from the viewpoint of improving the accuracy of transformer degradation diagnosis, it is preferable to create two or more calibration curves using two or more selected degradation indices.

[0052] From the viewpoint of improving the accuracy of transformer degradation diagnosis, the coefficient of determination of the calibration curve is preferably 0.7 or higher, more preferably 0.8 or higher, even more preferably 0.9 or higher, and even more preferably 0.95 or higher. The coefficient of determination of the calibration curve can be calculated from the numerical values ​​of the degradation index and the average degree of polymerization and the regression equation of the calibration curve. The coefficient of determination of the calibration curve is a measure that expresses the degree of agreement between the measured values ​​of the degradation index and the average degree of polymerization and the calibration curve using the least squares method. The coefficient of determination is "R 2 The coefficient of determination is expressed as being between 0 and 1. The closer the coefficient of determination is to 1, the better the match between the measured value and the calibration curve is considered to be.

[0053] (Diagnostic process) In this process, the deterioration status of the oil-filled transformer is diagnosed based on a calibration curve.

[0054] By estimating the average degree of polymerization of the insulating paper in the transformer from the created calibration curve, the deterioration state of the transformer under inspection can be diagnosed. If the average degree of polymerization of the insulating paper in the transformer is between 400 and 600, the transformer can be diagnosed as deteriorated.

[0055] The average degree of polymerization of the insulating paper in a transformer can be estimated, for example, by following the procedure below. The insulating oil in the transformer being diagnosed contains two or more degradation indicator compounds. Two or more samples of insulating oil are obtained from transformers heated under different heating conditions, and the detection amounts of two or more degradation indicator compounds are obtained from these insulating oils. Then, from the insulating oil of transformers heated under the same heating conditions, degradation indicators are selected as a group of data in the same manner as the process for selecting the degradation indicators described above. Finally, by fitting the obtained degradation indicators to the regression equation obtained from the calibration curve, the average degree of polymerization of the insulating paper can be estimated.

[0056] Embodiment 2. Another form of this implementation is a method for diagnosing the deterioration state of an oil-filled transformer equipped with insulating oil and insulating paper, comprising the steps of: preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and obtaining two or more sets of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; adjusting the detected amounts of the compounds to a selectable deterioration index by one or more adjustments selected from the group consisting of normalization and standardization; selecting one or more numerical values ​​selected from the group consisting of the average value and median value of the selectable deterioration index based on the data from the set as a deterioration index; creating a calibration curve relating to the deterioration index and the average degree of polymerization based on two or more sets of data from the set; and diagnosing the deterioration state of the oil-filled transformer based on the calibration curve.

[0057] The method in this second embodiment will be described with reference to Figure 1. The method in this second embodiment includes the acquisition step S1, adjustment step S2, selection step S3, creation step S4, and diagnostic step S5, similar to the method in the first embodiment. The steps in this second embodiment will be described below, but a detailed explanation of the same method as in the first embodiment will not be repeated.

[0058] (adjustment process) In this process, the detected amount of degradation indicator compound is adjusted to a selectable degradation indicator by one or more adjustments selected from the group consisting of normalization and standardization.

[0059] In this embodiment, the selective degradation index is a numerical value obtained by adjusting the detected amount of the degradation index compound to a certain range. The selective degradation index can be adjusted by normalization or standardization as described below. According to adjustment step S2, the difference in the detected amount of each degradation index compound can be made into a numerical value within a certain range.

[0060] For normalization, it is preferable to use the difference between the minimum and maximum values ​​of the detection amount of the degradation indicator compound in one group of data from two or more groups. The detection amount of the degradation indicator compound can be normalized by the following formula (4) and the selection degradation index can be adjusted. According to formula (4), the selection degradation index is adjusted to be between α and β for any number α and β greater than 0.

[0061]

number

[0062] For standardization, it is preferable to use the mean and standard deviation of the detection amounts of the degradation indicator compound in one of two or more data sets. The detection amounts of the degradation indicator compound can be standardized by the following formula (5) to adjust the selection degradation index. According to formula (5), for a given degradation indicator compound, the mean of the selection degradation index for two or more sets will be any number γ, where γ is greater than 0.

[0063]

number

[0064] (Selection process) In this process, one or more numerical values ​​are selected as degradation indicators from a group consisting of the mean and median values ​​of a set of data for selection of degradation indicators.

[0065] The degradation index is adjusted from a selection of degradation indices that have been normalized or standardized to a certain range by normalizing the detected amounts of all degradation index compounds. Therefore, compared to selecting a degradation index from the detected amounts of degradation index compounds, the differences in the detected amounts of each degradation index compound fall within a certain range, allowing for highly accurate determination of the average degree of polymerization of the insulating paper in the transformer being diagnosed from the calibration curve in preparation step S4. The degradation index can be selected from a group consisting of the mean and median values ​​of two or more selection of degradation indices in a set of data. The selected degradation index may be further logarithmically transformed before use.

[0066] Embodiment 3. A method in yet another form of the present disclosure is a method for diagnosing the deterioration state of an oil-filled transformer comprising insulating oil and insulating paper, comprising the steps of: preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and obtaining two or more groups of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; adjusting the detected amounts of the compounds to a selectable deterioration index by logarithmically transforming them; selecting one or more numerical values ​​as a deterioration index from a group consisting of the sum and average values ​​of the selectable deterioration index based on the data from the groups; creating a calibration curve relating the deterioration index and the average degree of polymerization based on two or more groups of data from the groups; and diagnosing the deterioration state of the oil-filled transformer based on the calibration curve.

[0067] The method in this third embodiment will be described with reference to Figure 1. The method in this third embodiment includes an acquisition step S1, an adjustment step S2, a selection step S3, a creation step S4, and a diagnostic step S5, similar to the method in the first embodiment. The steps in this third embodiment will be described below, but a detailed explanation of the same method as in the first embodiment will not be repeated.

[0068] (adjustment process) In this process, the amount of degradation indicator compound detected is adjusted to a selectable degradation indicator by logarithmically transforming it.

[0069] In this embodiment, the selective degradation index is a numerical value adjusted to reduce the amount of degradation index compound detected. The selective degradation index can be adjusted by logarithmic transformation of the amount of degradation index compound detected. According to adjustment step S2, the difference in the amount of each degradation index compound detected can be reduced, for example, to within one order of magnitude.

[0070] For logarithmic transformation, the base of the logarithm must be greater than 0. The logarithmic transformation may be performed using, for example, a common logarithm with base 10, or a natural logarithm with base e (Napier's number). When the detected amount of the degradation indicator compound is 0, the selection degradation index is adjusted by replacing it with the minimum detected amount of the degradation indicator compound in a different set of data for the same degradation indicator compound, and then performing a logarithmic transformation.

[0071] (Selection process) In this process, one or more numerical values ​​are selected as degradation indicators from a group consisting of the sum and mean values ​​of the above-mentioned degradation indicators based on a set of data.

[0072] The degradation index is a numerical value selected using a selection degradation index based on a set of data. Therefore, compared to selecting the degradation index from the detected amount of each degradation index compound, the influence of differences in the detected amount of each degradation index compound is reduced, allowing for highly accurate determination of the average degree of polymerization of the insulating paper in the transformer being diagnosed from the calibration curve in the preparation process S4. The degradation index can be selected from a group consisting of the mean and median values ​​of two or more selection degradation indices in a set of data.

[0073] Embodiment 4. A method in yet another form of the present disclosure is a method for diagnosing the deterioration state of an oil-filled transformer comprising insulating oil and insulating paper, comprising the steps of: preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and obtaining two or more groups of data consisting of the detected amounts of two or more compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper; selecting the synergistic mean of the detected amounts of the two or more compounds in the data group as a deterioration index; creating a calibration curve relating the deterioration index and the average degree of polymerization based on two or more groups of data; and diagnosing the deterioration state of the oil-filled transformer based on the calibration curve.

[0074] The method in this fourth embodiment will be described with reference to Figure 1. The method in this fourth embodiment includes an acquisition step S1, a selection step S3, a creation step S4, and a diagnostic step S5. Each step in this fourth embodiment will be described below, but a detailed explanation of the same method as in the first embodiment will not be repeated.

[0075] (Selection process) In this process, the geometric mean of the detected amounts of two or more degradation indicator compounds in a group of data is selected as the degradation index.

[0076] In this embodiment, the degradation index can be selected from the detection amounts of two or more degradation index compounds in a group of data using the following formula (6).

[0077]

number

[0078] Embodiment 5. A further embodiment of the present disclosure is a method for diagnosing the deterioration state of an oil-filled transformer comprising insulating oil and insulating paper, comprising the steps of: preparing a reference insulating oil which is the same insulating oil as the insulating oil and a reference insulating paper which is the same insulating paper as the insulating paper, and determining a compound contained in the reference insulating oil for which the coefficient of determination is 0.7 or more with respect to the detected amount of the compound and the average degree of polymerization of the reference insulating paper as a deterioration indicator compound; obtaining two or more groups of data consisting of the detected amount of the deterioration indicator compound and the average degree of polymerization of the reference insulating paper; selecting one or more numerical values ​​selected from the group consisting of the sum, mean, median, first quartile, third quartile, and interquartile range of the detected amount of the deterioration indicator compound based on the data from the groups, creating a calibration curve for the deterioration indicator and the average degree of polymerization based on the two or more groups of data from the groups, and diagnosing the deterioration state of the oil-filled transformer based on the calibration curve.

[0079] The method in this 5th embodiment will be described with reference to Figure 1. The method in this 5th embodiment includes a decision step S0, an acquisition step S1, a selection step S3, a creation step S4, and a diagnosis step S5. Each step in this 5th embodiment will be described below, but a detailed explanation of the same method as in the method in the 1st embodiment will not be repeated.

[0080] (Decision process) In this process, a reference insulating oil, which is the same insulating oil as the insulating oil, and a reference insulating paper, which is the same insulating paper as the insulating paper, are prepared. From the compounds contained in the reference insulating oil, compounds whose coefficient of determination is 0.7 or higher with respect to the detected amount of the compound and the average degree of polymerization of the reference insulating paper are determined as degradation indicator compounds.

[0081] In this embodiment, the degradation indicator compound refers to a compound contained in the reference insulating oil whose coefficient of determination is 0.7 or higher with respect to the detected amount of the compound and the average degree of polymerization of the reference insulating oil. The above degradation indicator compound is a correlated compound in which the detected amount tends to increase in the transformer being diagnosed, i.e., with respect to the deterioration of the insulating paper, i.e., with respect to the decrease in the average degree of polymerization of the insulating paper. Therefore, by going through the determination step S0 before the acquisition step S1, the coefficient of determination of the calibration curve created in the creation step S4 is improved, and the accuracy of the transformer degradation diagnosis can be improved. The above coefficient of determination is 0.7 or higher, and from the viewpoint of improving the accuracy of the transformer degradation diagnosis, 0.8 or higher is preferable, and 0.9 or higher is more preferable. The coefficient of determination of the calibration curve can be calculated from the numerical values ​​of the detected amount of the compound and the average degree of polymerization of the reference insulating paper, and the regression equation of the calibration curve.

[0082] The detected amount of the compound and the average degree of polymerization of the reference insulating paper may be the measured values ​​as they are, or they may be rounded to facilitate the calculation of the coefficient of determination, or they may be corrected values ​​based on the detection sensitivity.

[0083] In the determination step S0, from the viewpoint of making the calibration curve easier to create, it is preferable to determine the degradation indicator compound from the Hansen solubility parameter and boiling point of the compound contained in the reference insulating oil. The Hansen solubility parameter is a value used to predict the solubility of a substance, and for example, values ​​reported in literature such as A, B, C, and D, as well as values ​​obtained from Hansen Solubility Parameter Software (HSPiP), can be used. HSPiP can calculate the Hansen solubility parameter from the structure of the compound. In addition, Hansen solubility parameters can also be searched on the following homepage A. Document A: Wesley L. Archer et. Al., "Determination of Hansen Solubility Parameters for Selected Cellulose Ether Derivatives", Industrial & Engineering Chemistry Research, Vol. 30, No. 5; 1991, pp2292-2298 Document B: Charles M. Hansen, Tim S. Poulsen, "Hansen Solubility Parameters - Biological Materials", Taylor & Francis Group, LLC; 2007, pp269-292 Document C: Martina Levin, Per Redelius “Determination of Three-Dimensional Solubility Parameters and Solubility Spheres for Naphthenic Mineral Oils”, Energy & Fuels, Vol. 22, No.5; 2008, pp3395-3401 Document D: Martina Levin, Per Redelius “Determining the Hansen Solubility Parameter of Three Corrosion Inhibitors and the Correlation with Mineral Oil”, Energy & Fuels, Vol. 26; 2012, pp7243-7250 Homepage A: HSP Basics, [online], [Accessed November 6, 2024], Internet,<URL: https: / / www.stevenabbott.co.uk / practical-solubility / hsp-basics.php>

[0084] The Hansen solubility parameter varies depending on the compound. The Hansen solubility parameter δ is composed of the London dispersion force δd, the dipole force δp, and the hydrogen bonding force δh, and is expressed by the following formula (7). The nonpolar parameter NP in the Hansen solubility parameter δ is expressed by the following formula (8). The polar parameter P in the Hansen solubility parameter δ is expressed by the following formula (9). Furthermore, the compound may have a boiling point of 500°C or less, or 400°C or less, or 300°C or less.

[0085]

number

[0086]

number

[0087]

number

[0088] The Hansen solubility parameter and boiling point of a compound are unique values ​​for each compound. Therefore, it is sufficient to identify the physical properties of the compound from its Hansen solubility parameter and boiling point, and to estimate that the coefficient of determination for the detected amount of the compound and the average degree of polymerization of the reference insulating oil is 0.7 or higher. Using this method, even if a correlated compound is discovered that shows an increasing trend in the detected amount as the average degree of polymerization of the insulating paper decreases, it is possible to estimate that the coefficient of determination is 0.7 or higher from the Hansen solubility parameter and boiling point.

[0089] Furthermore, in the determination step S0, it is preferable to determine the degradation indicator compound by satisfying the following equations (1) to (3) in order to facilitate the creation of a calibration curve.

[0090]

number

number

number

[0091] The Hansen solubility parameters for the reference insulating paper can be those reported in the above-mentioned literature A, B, C, and D, as well as those obtained from the Hansen Solubility Parameter Software (HSPiP). The Hansen solubility parameters for the reference insulating oil can be those reported in the above-mentioned literature A, B, C, and D, as well as those obtained from the Hansen Solubility Parameter Software (HSPiP). The Hansen solubility parameters for the compound can be those reported in the above-mentioned literature A, B, C, and D, as well as those obtained from the Hansen Solubility Parameter Software (HSPiP).

[0092] (Acquisition process) In this process, two or more sets of data are obtained, each consisting of the amount of degradation indicator compound detected and the average degree of polymerization of the reference insulating paper.

[0093] (Selection process) In this process, one or more numerical values ​​are selected as a degradation index from a group consisting of the sum, mean, median, first quartile, third quartile, and interquartile range of the degradation index compound detected based on a set of data.

[0094] The degradation index is a numerical value selected using the detected amount of a degradation index compound based on a set of data. Therefore, rather than selecting the degradation index from correlated and uncorrelated compounds, a calibration curve can be created from the detected amounts of correlated degradation index compounds, allowing for highly accurate determination of the average degree of polymerization of the insulating paper in the transformer being diagnosed. The degradation index can be selected from a group consisting of the sum of the detected amounts of the degradation index compound in a set of data, the mean, the median, the first quartile, the third quartile, and the interquartile range. Based on a set of data, one or more numerical values ​​may be selected as the degradation index, and two or more numerical values ​​may also be selected. [Examples]

[0095] The present disclosure will be further described below with reference to examples and comparative examples, but the present disclosure is not limited to these examples.

[0096] (Heat treatment of the sample) Figure 2 is a conceptual diagram of a heating test apparatus used to create a calibration curve for diagnosing the degradation state of a transformer in this disclosure. The heating test apparatus 1 comprises a container 10, an inlet valve 11, and a vacuum valve 12. The container 10 is a 500 ml stainless steel container that can be sealed. The inlet valve 11 is a valve for introducing degassed reference insulating oil 3 into the container 10. The vacuum valve 12 is a valve for creating a vacuum in the container 10. Samples can be obtained from reference insulating paper 2 and reference insulating oil 3.

[0097] The following describes the method for obtaining the samples necessary to acquire a set of data consisting of the detection amounts of two or more degradation indicator compounds contained in the reference insulating oil and the average degree of polymerization of the reference insulating paper. Container 10 was sealed with 6g of pre-dried reference insulating paper 2. Next, the inside of container 10 was evacuated using a vacuum valve 12, and after closing the vacuum valve 12, container 10 was filled with reference insulating oil 3 through the inlet valve 11. Then, with container 10 sealed by closing the inlet valve 11, it was heated at the following temperature and time to obtain samples containing reference insulating oil and reference insulating paper under the conditions shown in Test Examples 1 to 24 below. For Test Examples 1 to 9, heat-resistant insulating paper with dicyandiamide as the heat-resistant treatment agent was used as reference insulating paper 2. For Test Examples 10 to 24, kraft paper was used as reference insulating paper 2. Mineral oil (JIS C 2320 Type 1 No. 4 oil) was used as reference insulating oil 3. • Test example 1 is a sample with a heating time of 0 days. • Test example 2 is a sample that was heated at 130°C for 10 days. • Test example 3 is a sample that was heated at 130°C for 30 days. • Test example 4 is a sample that was heated at 130°C for 60 days. • Test example 5 is a sample that was heated at 130°C for 120 days. • Test example 6 is a sample that was heated at 150°C for 10 days. • Test example 7 is a sample that was heated at 150°C for 30 days. • Test example 8 is a sample that was heated at 150°C for 60 days. • Test example 9 is a sample that was heated at 150°C for 120 days. • Test example 10 is a sample that was heated at 150°C for 3 days. • Test example 11 is a sample that was heated at 150°C for 7 days. • Test example 12 is a sample that was heated at 150°C for 14 days. • Test example 13 is a sample that was heated at 150°C for 30 days. • Test example 14 is a sample that was heated at 150°C for 60 days. • Test example 15 is a sample that was heated at 165°C for one day. • Test example 16 is a sample that was heated at 165°C for 3 days. • Test example 17 is a sample that was heated at 165°C for 5 days. • Test example 18 is a sample that was heated at 165°C for 7 days. • Test example 19 is a sample that was heated at 165°C for 14 days. • Test example 20 is a sample that was heated at 180°C for 0.5 days. • Test example 21 is a sample that was heated at 180°C for one day. • Test example 22 is a sample that was heated at 180°C for 2 days. • Test example 23 is a sample that was heated at 180°C for 3 days. • Test example 24 is a sample that was heated at 180°C for 5 days.

[0098] (Analysis of degradation indicator compounds) Degradation indicator compounds in liquid form at room temperature were analyzed using a solid-phase microextraction apparatus (SPME) and a gas chromatograph / mass spectrometer (GC / MS). The specific analytical method is described below. First, 2 g of insulating oil and a stirring bar were placed in a 10 ml vial and the vial was stoppered. A hole was made in the stopper large enough for the SPME syringe to pass through. The vial was placed on a stirrer and heated to 50°C at a rotation speed of 200 rpm. In this state, the SPME syringe was inserted into the space above the oil inside the vial and adsorbed for 60 minutes. The adsorbent for SPME was polydimethylsiloxane. Next, SPME was inserted into the inlet of the GC / MS to obtain a chromatogram. By comparing the chromatograms of insulating oil heated at the above temperature and time, compounds that changed over time were searched for, and their area values ​​(unitless) were obtained. In this disclosure, unless otherwise specified, the amount of detected degradation indicator compounds in liquid form at room temperature is expressed as the area value.

[0099] The detected degradation indicator compounds at room temperature were methanol, acetone, acetic acid, dimethyl sulfone, dimethyl sulfide, furfural, furfuryl alcohol, methylpyrazine, phenol, methylphenol, acetamide, pyrrole, 3-methylpyrrole, 1,3-diazine, 1-methylpyrrole, methylfuran, and dimethylcyclohexanol.

[0100] Degradation indicator compounds in the form of gases at room temperature were analyzed in accordance with the Japan Petroleum Institute standard JPI-5R-51-98, "Method for sampling and analysis of free and dissolved gases of gases and insulating oil from oil-filled electrical equipment" (1998). In this disclosure, unless otherwise specified, the detected amount of degradation indicator compounds in the form of gases at room temperature is expressed as concentration (ppm).

[0101] The degradation indicator compounds detected at room temperature that were liquid were carbon dioxide, carbon monoxide, methane, ethane, and propane.

[0102] (Analysis of average degree of polymerization) The average degree of polymerization of insulating paper was analyzed in accordance with the Japan Electrical Manufacturers' Association standard JEM1455 "Method for Measuring the Average Degree of Polymerization of Insulating Paper for Transformers" (1991). The average degree of polymerization of insulating paper under the conditions shown in Test Examples 1 to 24 above was analyzed.

[0103] Test Examples 1-24 were analyzed according to the analysis method described above. Test Examples 1-24 are treated as a single set of data based on the analysis described above. Table 1 shows the amount of liquid degradation indicator compounds detected at room temperature and the average degree of polymerization of the insulating paper for Test Examples 1-5, and Table 2 shows the amount of liquid degradation indicator compounds detected at room temperature and the average degree of polymerization of the reference insulating paper for Test Examples 6-9. Table 3 shows the amount of gaseous degradation indicator compounds detected at room temperature and the average degree of polymerization of the reference insulating paper for Test Examples 10-14, Table 4 shows the amount of gaseous degradation indicator compounds detected at room temperature and the average degree of polymerization of the reference insulating paper for Test Examples 15-19, and Table 5 shows the amount of gaseous degradation indicator compounds detected at room temperature and the average degree of polymerization of the reference insulating paper for Test Examples 20-24.

[0104] [Table 1]

[0105] [Table 2]

[0106] [Table 3]

[0107] [Table 4]

[0108] [Table 5]

[0109] The above degradation indicator compounds showed an increasing trend in detection levels with increasing test time. The basic statistical quantities, which represent the detection levels of the degradation indicator compounds, were calculated using spreadsheet software (Microsoft Excel). The detection levels of the degradation indicator compounds in the table are the area values ​​rounded to the nearest tens, hundreds, or thousands place.

[0110] (Transformer deterioration diagnosis) For Test Examples 1 to 24, we determined whether the calibration curves created based on the following examples and comparative examples were suitable for diagnosing the deterioration state of oil-filled transformers. Specifically, we determined that if the coefficient of determination of the created calibration curve was 0.7 or higher, it was possible to diagnose the deterioration state well.

[0111] Furthermore, it was determined that a better diagnosis of the degradation state could be achieved if the average CV value for each test example used in creating the calibration curve was 1.7 or less. The CV value is a numerical value obtained based on the numerical values ​​used in creating the calibration curve. The numerical values ​​used in creating the calibration curve differ depending on the example and comparative example, but for example, they may be raw data of the detected amount of the degradation indicator compound, or they may be a selection degradation index.

[0112] <Comparative Example 1> In Comparative Example 1, the average value of the detected amount of the degradation indicator compound in a group of data was calculated for Test Examples 1-9 in Tables 1 and 2, and a calibration curve was created relating this average value to the average degree of polymerization of the reference insulating paper. Table 6 shows the relationship between the average value of the detected amount and the average degree of polymerization for Test Examples 1-9. Figure 3 is a diagram showing the calibration curve relating the average value of the detected amount and the average degree of polymerization for Test Examples 1-9 shown in Table 6.

[0113] [Table 6]

[0114] The coefficient of determination of the calibration curve in Figure 3 was 0.3357. For Comparative Example 1, the coefficient of determination of the created calibration curve was not good. Furthermore, the CV values ​​obtained based on Tables 1 and 2 ranged from 1.25 to 3.4, with an average value of 2.12. Therefore, even using the calibration curve in Figure 3, the average degree of polymerization of the insulating paper could not be sufficiently estimated, resulting in poor accuracy in diagnosing the deterioration state of the transformer.

[0115] <Comparative Example 2> In Comparative Example 2, for Test Examples 10-24 in Tables 3-5, the average value of the detected amount of the degradation indicator compound in a group of data was calculated, and a calibration curve was created relating this average value to the average degree of polymerization of the reference insulating paper. Table 7 shows the relationship between the average value of the detected amount and the average degree of polymerization for Test Examples 10-24. Figure 4 is a diagram showing the calibration curve relating the average value of the detected amount and the average degree of polymerization for Test Examples 10-24 shown in Table 7.

[0116] [Table 7]

[0117] The coefficient of determination of the calibration curve in Figure 4 was 0.5955. For Comparative Example 2, the coefficient of determination of the created calibration curve was not good. Furthermore, the CV values ​​obtained based on Tables 3 to 5 ranged from 1.32 to 1.76, with an average value of 1.55. Therefore, even using the calibration curve in Figure 4, the average degree of polymerization of the insulating paper could not be sufficiently estimated, resulting in poor accuracy in diagnosing the deterioration state of the transformer.

[0118] For reference, the coefficient of determination of the calibration curve between the detected amount of carbon dioxide and the average degree of polymerization of the reference degree in Test Examples 10-24 in Tables 3-5 was 0.5312, which was almost the same as the coefficient of determination of the calibration curve in Figure 4. This suggests that, because carbon dioxide accounts for a large proportion of the detected amount of gaseous degradation indicator compounds at room temperature, the proportion of carbon dioxide is directly reflected even when averaging as in Comparative Example 2.

[0119] <Example 1> In Example 1, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection of degradation indicators: The detection amounts of each degradation indicator compound in Tables 1 and 2 were adjusted by dividing them by the average detection amount of the degradation indicator compound for Test Examples 1-9 by a constant. The constant differs for each degradation indicator compound. • Degradation Index: The average value of the selection degradation index for a group of data was selected. A degradation index was selected for each of the 1-9 test examples. Table 8 shows the degradation index used for selection in Example 1 for Test Examples 1-5. Table 9 shows the degradation index used for selection in Example 1 for Test Examples 6-9. Table 10 shows the relationship between the degradation index and the average degree of polymerization in Example 1. Figure 5 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 1 as shown in Table 10.

[0120] [Table 8]

[0121] [Table 9]

[0122] [Table 10]

[0123] The coefficient of determination of the calibration curve in Figure 5 was 0.9711. For Example 1, the coefficient of determination of the created calibration curve was good. The CV values ​​obtained based on Tables 8 and 9 ranged from 0.29 to 2.14, with an average value of 0.84. Therefore, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated using the calibration curve in Figure 5, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0124] Furthermore, although not explained using tables and figures, calibration curves were created between the degradation index obtained by selecting the median, first quartile, third quartile, or interquartile range of the selectable degradation index shown in Tables 8 and 9, and the average degree of polymerization, and the coefficient of determination was examined. The coefficient of determination for the calibration curve between the degradation index and the average degree of polymerization, obtained by selecting the median of the selectable degradation index shown in Tables 8 and 9, was 0.9948. The coefficient of determination for the calibration curve between the degradation index and the average degree of polymerization, obtained by selecting the first quartile of the selectable degradation index shown in Tables 8 and 9, was 0.9548. The coefficient of determination for the calibration curve between the degradation index and the average degree of polymerization, obtained by selecting the third quartile of the selectable degradation index shown in Tables 8 and 9, was 0.9498. The coefficient of determination for the calibration curve between the degradation index and the average degree of polymerization, obtained by selecting the interquartile range of the selectable degradation index shown in Tables 8 and 9, was 0.7076. Therefore, regardless of which option was selected, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated, and the degradation state of the transformer being diagnosed could be diagnosed with high accuracy.

[0125] <Example 2> In Example 2, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection of degradation indicators: The detection amounts of each degradation indicator compound in Tables 1 and 2 were adjusted by dividing the median detection amount of the degradation indicator compound for Test Examples 1-9 by a constant. The constant differs for each degradation indicator compound. • Degradation Index: The median of the selection degradation index for a group of data was selected. A degradation index was selected for each of the 9 test examples. Table 11 shows the degradation index used for selection in Example 2 for Test Examples 1-5. Table 12 shows the degradation index used for selection in Example 2 for Test Examples 6-9. Table 13 shows the relationship between the degradation index and the average degree of polymerization in Example 2. Figure 6 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 2 as shown in Table 13.

[0126] [Table 11]

[0127] [Table 12]

[0128] [Table 13]

[0129] The coefficient of determination of the calibration curve in Figure 6 was 0.9928. For Example 2, the coefficient of determination of the created calibration curve was good. The CV values ​​obtained based on Tables 11 and 12 ranged from 0.59 to 2.56, with an average value of 1.03. Therefore, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated using the calibration curve in Figure 6, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0130] Furthermore, although not explained using tables and figures, a calibration curve was created relating the degradation index obtained by replacing the constant with the maximum or minimum value for the selection degradation index in Example 2 to the average degree of polymerization, and the coefficient of determination was investigated. The detection amount of each degradation indicator compound in Tables 1 and 2 was adjusted for the selection degradation index by dividing the maximum value of the degradation indicator compound by a constant for Test Examples 1 to 9. The coefficient of determination of the calibration curve between the selected degradation index and the average degree of polymerization, where the median of the selection degradation index in a group of data was 0.9705, was obtained. The detection amount of each degradation indicator compound in Tables 1 and 2 was adjusted for the selection degradation index by dividing the minimum value of the degradation indicator compound by a constant for Test Examples 1 to 9. The coefficient of determination of the calibration curve between the selected degradation index and the average degree of polymerization, where the median of the selection degradation index in a group of data was 0.9618. Therefore, in both cases, the accuracy was improved to the extent that the average degree of polymerization of the insulating paper could be estimated, and the degradation state of the transformer being diagnosed could be diagnosed with high accuracy.

[0131] <Example 3> In Example 3, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection degradation index: The detection amounts of the degradation index compounds in Tables 1 and 2 were adjusted by normalizing them. Normalization was performed according to equation (4), with α=10 and β=100. Therefore, the normalized selection degradation index was adjusted to be between 10 and 100. • Degradation Index: The average value of the selection degradation index for a group of data was selected. A degradation index was selected for each of the 1-9 test examples. Table 14 shows the degradation index used for selection in Example 3 for Test Examples 1-5. Table 15 shows the degradation index used for selection in Example 3 for Test Examples 6-9. Table 16 shows the relationship between the degradation index and the average degree of polymerization in Example 3. Figure 7 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 3 as shown in Table 16.

[0132] [Table 14]

[0133] [Table 15]

[0134] [Table 16]

[0135] The coefficient of determination of the calibration curve in Figure 7 was 0.9591. For Example 3, the coefficient of determination of the created calibration curve was good. The CV values ​​obtained based on Tables 14 and 15 ranged from 0.06 to 1.06, with an average value of 0.54. Therefore, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated using the calibration curve in Figure 7, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0136] Furthermore, although not explained using tables and figures, the coefficient of determination of the calibration curve between the degradation index obtained by selecting the median of the selectable degradation indices shown in Tables 14 and 15 and the average degree of polymerization was 0.9538. Therefore, the accuracy was improved to the point where the average degree of polymerization of the insulating paper could be estimated regardless of which index was selected, and the degradation state of the transformer being diagnosed could be diagnosed with high accuracy.

[0137] <Example 4> In Example 4, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection of degradation index: The detection amounts of each degradation index compound in Tables 1 and 2 were adjusted by standardization. Standardization was performed according to equation (5) with γ = 1.5. • Degradation Index: The average value of the selection degradation index for a group of data was selected. A degradation index was selected for each of the 1-9 test examples. Table 17 shows the degradation index used for selection in Example 4 for Test Examples 1-5. Table 18 shows the degradation index used for selection in Example 4 for Test Examples 6-9. Table 19 shows the relationship between the degradation index and the average degree of polymerization in Example 4. Figure 8 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 4 as shown in Table 19.

[0138] [Table 17]

[0139] [Table 18]

[0140] [Table 19]

[0141] The coefficient of determination of the calibration curve in Figure 8 was 0.9467. For Example 4, the coefficient of determination of the created calibration curve was good. The CV values ​​obtained based on Tables 17 and 18 ranged from 0.21 to 0.84, with an average value of 0.40. Therefore, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated using the calibration curve in Figure 8, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0142] Furthermore, although not explained using tables and figures, the coefficient of determination of the calibration curve between the degradation index obtained by selecting the median of the selectable degradation indices shown in Tables 17 and 18 and the average degree of polymerization was 0.9425. Therefore, the accuracy was improved to the point where the average degree of polymerization of the insulating paper could be estimated regardless of which index was selected, and the degradation state of the transformer being diagnosed could be diagnosed with high accuracy.

[0143] <Example 5> In Example 5, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection of degradation indicators: The detection amounts of each degradation indicator compound in Tables 3 to 5 were adjusted by standardization. Standardization was performed according to equation (5) with γ = 1.5. • Degradation Index: The average value of the selected degradation index was chosen from a group of data. A degradation index was selected for each of the 10-24 test examples. Table 20 shows the degradation index for selection of Example 5 for Test Examples 10-14. Table 21 shows the degradation index for selection of Example 5 for Test Examples 15-19. Table 22 shows the degradation index for selection of Example 5 for Test Examples 20-24. Table 23 shows the relationship between the degradation index and the average degree of polymerization in Example 5. Figure 9 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 5 as shown in Table 23.

[0144] [Table 20]

[0145] [Table 21]

[0146] [Table 22]

[0147] [Table 23]

[0148] The coefficient of determination of the calibration curve in Figure 9 was 0.8504. For Example 5, the coefficient of determination of the created calibration curve was good. The CV values ​​obtained based on Tables 20 to 22 ranged from 0.20 to 0.74, with an average value of 0.42. Therefore, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated using the calibration curve in Figure 9, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0149] <Example 6> In Example 6, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection of degradation index: The detection amounts of each degradation index compound in Tables 1 and 2 were adjusted by taking the common logarithm. • Degradation Index: The average value of the selection degradation index for a group of data was selected. A degradation index was selected for each of the 1-9 test examples. Table 24 shows the degradation index for selection of Example 6 for Test Examples 1-5. Table 25 shows the degradation index for selection of Example 6 for Test Examples 6-9. Table 26 shows the relationship between the degradation index and the average degree of polymerization in Example 6. Figure 10 is a diagram showing the calibration curve between the degradation index and the average degree of polymerization in Example 6 shown in Table 26.

[0150] [Table 24]

[0151] [Table 25]

[0152] [Table 26]

[0153] The coefficient of determination of the calibration curve in Figure 10 was 0.9223. For Example 6, the coefficient of determination of the created calibration curve was good. The CV values ​​obtained based on Tables 24 and 25 ranged from 0.11 to 0.20, with an average value of 0.16. Therefore, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated using the calibration curve in Figure 10, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0154] <Example 7> In Example 7, a degradation index was obtained as follows, and a calibration curve was created relating the degradation index to the average degree of polymerization of the reference insulating paper. • Degradation index: The geometric mean of the detected amounts of degradation index compounds in the data sets in Tables 1 and 2 was selected. A degradation index was selected for each of the Test Examples 1-9. Table 27 shows the relationship between the degradation index and the average degree of polymerization in Example 7. Figure 11 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 7, as shown in Table 27.

[0155] [Table 27]

[0156] The coefficient of determination of the calibration curve in Figure 11 was 0.9818. For Example 7, the coefficient of determination of the created calibration curve was good. Therefore, the accuracy improved to the point where it was possible to estimate the average degree of polymerization of the insulating paper of the transformer being diagnosed.

[0157] <Example 8> In Example 8, a degradation indicator compound and a degradation indicator were obtained as follows, and a calibration curve was created relating the degradation indicator to the average degree of polymerization of the reference insulating paper. • Degradation indicator compounds: Compounds whose coefficient of determination for the calibration curve relating the detected amount and average degree of polymerization for each compound shown in Tables 1 and 2 is 0.7 or higher were designated as degradation indicator compounds. • Degradation index: The average value of the detected amount of the degradation index compound in a group of data was selected. A degradation index was selected for each of the test examples 1 to 9. Table 28 shows the coefficient of determination obtained from the calibration curves relating the detected amount and average degree of polymerization for each compound shown in Tables 1 and 2. In Example 8, the detected amounts of the degradation indicator compounds can be obtained from Tables 1 and 2. Figure 12 shows the calibration curves relating the detected amounts and average degree of polymerization for acetone and methylpyrazine from Tables 1 and 2. In Example 8, acetone is a compound with a coefficient of determination of less than 0.7, and methylpyrazine is a degradation indicator compound with a coefficient of determination of 0.7 or more. Table 29 shows the relationship between the degradation index and the average degree of polymerization in Example 8. Figure 13 shows the relationship between the degradation index and the average degree of polymerization in Example 8 as shown in Table 29.

[0158] [Table 28]

[0159] [Table 29]

[0160] The coefficient of determination of the calibration curve for the detected amount of acetone in Figure 12 was 0.0257, and the coefficient of determination of the calibration curve for the detected amount of methylpyrazine was 0.9955. From Figure 12, methylpyrazine was a better degradation indicator compound than acetone for estimating the average degree of polymerization of insulating paper. The coefficient of determination of the calibration curve in Figure 13 was 0.9908. For Example 8, the coefficient of determination of the created calibration curve was good. Based on Tables 1 and 2, the CV values ​​obtained from the detected amounts of compounds with a coefficient of determination of 0.7 or higher in Table 28 ranged from 0.15 to 1.07, with an average value of 0.61. Therefore, the accuracy was improved to the extent that the average degree of polymerization of insulating paper could be estimated using the calibration curve in Figure 13, and the degradation state of the transformer being diagnosed could be diagnosed with high accuracy.

[0161] <Reference example 1> In Reference Example 1, a calibration curve was prepared in the same manner as in Comparative Example 1, except that only the detected amounts of compounds derived from the heat-resistant treatment agents methylpyrazine, acetamide, pyrrole, 3-methylpyrrole, 1,3-diazine, and 1-methylpyrrole were used among the degradation indicator compounds in Comparative Example 1. Table 30 shows the relationship between the average detected amount of the compounds and the average degree of polymerization in Reference Example 1. Figure 14 is a diagram showing the calibration curve relating the average detected amount of the compounds and the average degree of polymerization in Reference Example 1 shown in Table 30.

[0162] [Table 30]

[0163] The coefficient of determination of the calibration curve in Figure 14 was 0.9556. Based on Tables 1 and 2, the CV values ​​obtained from the detected amounts of compounds derived from the heat-resistant treatment agent ranged from 0.85 to 1.79, with an average value of 1.29. However, values ​​deviating from the calibration curve were observed in the dotted line portion of Figure 13.

[0164] <Example 9> In Example 9, a selection degradation index and a degradation index were obtained as follows, and a calibration curve was created relating the degradation index and the average degree of polymerization of the reference insulating paper. • Selection degradation index: This index was adjusted by normalizing the detection amounts of each compound listed in Reference Example 1. Normalization was performed according to equation (4), with α=10 and β=100. Therefore, the normalized selection degradation index was adjusted to be between 10 and 100. • Degradation Index: The average value of the selection degradation index for a group of data was selected. A degradation index was selected for each of the 1-9 test examples. Table 31 shows the relationship between the degradation index and the average degree of polymerization in Example 9. Figure 15 is a calibration curve showing the relationship between the degradation index and the average degree of polymerization in Example 9, as shown in Table 31.

[0165] [Table 31]

[0166] The coefficient of determination of the calibration curve in Figure 15 was 0.9645. For Example 9, the coefficient of determination of the created calibration curve was good. Based on Tables 14 and 15, the CV values ​​obtained from the compounds listed in Reference Example 1 ranged from 0.0 to 0.76, with an average value of 0.38, which was better than the calibration curve in Reference Example 1. Therefore, using the calibration curve in Figure 15, the accuracy improved to the point where the average degree of polymerization of the insulating paper could be estimated from the amount of compounds detected only from the heat-resistant insulating paper, and the deterioration state of the transformer being diagnosed could be diagnosed with high accuracy.

[0167] <Example 10> In Example 10, degradation indicator compounds were determined from the Hansen solubility parameters and boiling points of compounds contained in the reference insulating oil. Figure 16 is a plot of the nonpolarity parameters and boiling points for each compound. Figure 17 is a plot of the polarity parameters and boiling points for each compound. The nonpolarity parameter is obtained from equation (8). The polarity parameter is obtained from equation (9). In Figures 16 and 17, ○ represents a plot of a compound with a coefficient of determination of 0.7 or higher, and ● represents a plot of a compound with a coefficient of determination of less than 0.7.

[0168] From Figure 16, for compounds with a boiling point of 300°C or lower, if the boiling point is higher than 7 times the nonpolarity parameter raised to the power of -9, the coefficient of determination for that compound is 0.7 or higher. From Figure 17, for compounds with a boiling point of 300°C or lower, if the boiling point is higher than 1700 times the polarity parameter raised to the power of 6.7, the coefficient of determination for that compound is 0.7 or higher.

[0169] <Example 11> In Example 11, the Hansen solubility parameters of the compound, reference insulating paper, and reference insulating oil were determined by selecting compounds that satisfy equations (1) to (3) as degradation indicator compounds. Figure 18 shows a graph with Ro on the X axis and Rc on the Y axis for the compounds shown in Table 28. The Hansen solubility parameters (δdc, δpc, δhc) for the insulating paper were (26.1, 7.9, 17.3). The Hansen solubility parameters (δdo, δpo, δho) for the insulating oil were (16.6, 0.0, 5.0). The Hansen solubility parameters for each compound were obtained from the above-mentioned references A, B, C, D, and Homepage A, as well as from the Hansen solubility parameter software (HSPiP). In Figure 18, compounds plotted between the dashed line and the dotted line are compounds with a coefficient of determination of 0.7 or higher. Figure 19 shows the (Ro) of the compounds shown in Table 28. 2 +Rc 2 ) 1 / 2 This figure shows a graph with the x-axis representing the coefficient of determination and the y-axis representing the coefficient of determination. According to Figure 19, (Ro 2 +Rc 2 ) 1 / 2It was found that compounds satisfying the condition <25 have a coefficient of determination of 0.7 or higher.

[0170] <Reference example 2> Example 12 summarizes the relationship between the mean CV value and the coefficient of determination of the calibration curve, as described in previous examples. Figure 20 shows the relationship between the mean CV value and the coefficient of determination. From Figure 20, it was found that a good coefficient of determination could not be obtained when the mean CV value exceeded 1.5.

[0171] The embodiments and examples disclosed herein should be considered in all respects to be illustrative and not restrictive. The scope of this disclosure is indicated by the claims rather than the foregoing description, and all modifications are intended to be within the meaning and scope of the equivalents of the claims. [Explanation of symbols]

[0172] S0 Decision process, S1 Acquisition process, S2 Adjustment process, S3 Selection process, S4 Production process, S5 Diagnostic process, 1 Heating test apparatus, 10 Container, 11 Inlet valve, 12 Vacuum valve, 2 Reference insulating paper, 3 Reference insulating oil.

Claims

1. A method for diagnosing the deterioration state of an oil-filled transformer equipped with insulating oil and insulating paper, A step of preparing a reference insulating oil which is the same insulating oil as the aforementioned insulating oil and a reference insulating paper which is the same insulating paper as the aforementioned insulating paper, and determining a compound among the compounds contained in the reference insulating oil for which the coefficient of determination with respect to the detected amount of the compound and the average degree of polymerization of the reference insulating paper is 0.7 or higher as a degradation indicator compound, A step of obtaining two or more sets of data consisting of the amount of the degradation indicator compound detected and the average degree of polymerization of the reference insulating paper, A step of selecting one or more numerical values ​​as a degradation index, selected from a group consisting of the total amount, mean, median, first quartile, third quartile, and interquartile range of the degradation index compound based on the aforementioned group of data, A step of creating a calibration curve for the degradation index and the average degree of polymerization based on data from two or more groups of the aforementioned group, The process includes diagnosing the deterioration state of the oil-filled transformer based on the calibration curve, A method for determining the degradation indicator compound from the Hansen solubility parameter and boiling point of the compound in the aforementioned determination step.

2. A method for diagnosing the deterioration state of an oil-filled transformer equipped with insulating oil and insulating paper, A step of preparing a reference insulating oil which is the same insulating oil as the aforementioned insulating oil and a reference insulating paper which is the same insulating paper as the aforementioned insulating paper, and determining a compound among the compounds contained in the reference insulating oil for which the coefficient of determination with respect to the detected amount of the compound and the average degree of polymerization of the reference insulating paper is 0.7 or higher as a degradation indicator compound, A step of obtaining two or more sets of data consisting of the amount of the degradation indicator compound detected and the average degree of polymerization of the reference insulating paper, A step of selecting one or more numerical values ​​as a degradation index, selected from a group consisting of the total amount, mean, median, first quartile, third quartile, and interquartile range of the degradation index compound based on the aforementioned group of data, A step of creating a calibration curve for the degradation index and the average degree of polymerization based on data from two or more groups of the aforementioned group, The process includes diagnosing the deterioration state of the oil-filled transformer based on the calibration curve, A method for determining the degradation indicator compound in the aforementioned determination step by satisfying the following formulas (1) to (3). [Math 1] [Math 2] [Math 3] [In formulas (1) to (3), δd represents the London dispersion power of the Hansen solubility parameter in the compound, δp represents the inter-dipole force of the Hansen solubility parameter in the compound, δh represents the hydrogen bonding force of the Hansen solubility parameter in the compound, δdc represents the London dispersion force of the Hansen solubility parameter in the reference insulating paper, δpc represents the inter-dipole force of the Hansen solubility parameter in the reference insulating paper, δhc represents the hydrogen bonding force of the Hansen solubility parameter in the reference insulating paper, and δdo represents the London dispersion force of the Hansen solubility parameter in the reference insulating oil. δpo represents the inter-dipole force of the Hansen solubility parameter in the reference insulating oil, δho represents the hydrogen bonding force of the Hansen solubility parameter in the reference insulating oil, and Rc represents the distance of the Hansen solubility parameter between the reference insulating paper and the compound. Ro represents the distance of the Hansen solubility parameter between the reference insulating oil and the compound.

Images