On-load tap-changer contact overheating fault diagnosis method and system

Abstract

The invention discloses an on-load tap-changer contact overheating fault diagnosis method and system, relates to the technical field of on-load tap-changers, and aims to solve the technical problem that an existing on-load tap-changer contact overheating fault diagnosis method is poor in accuracy, and the method comprises the following steps: S1, collecting historical data; s2, collecting a plurality of groups of historical contact temperature curves of a to-be-diagnosed contact, wherein each group of historical contact temperature curves corresponds to a historical tap gear change condition; S2, dividing a characteristic temperature interval; dividing each group of historical contact temperature curves into a plurality of characteristic temperature intervals; s3, training a binary classifier; for each historical tap gear change condition, a binary classifier is trained based on the characteristic temperature interval under the historical tap gear change condition, and the binary classifier is used for judging whether the current characteristic temperature interval of the to-be-diagnosed contact temperature curve belongs to an abnormal characteristic temperature interval or not; and S4, analyzing the curve to be diagnosed. The on-load tap-changer contact overheating fault diagnosis method has the advantage of improving the on-load tap-changer contact overheating fault diagnosis accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of on-load tap-changers, and more specifically, to a method and system for diagnosing overheating faults of on-load tap-changer contacts. Background Art

[0002] As an important part of a transformer in a power system, the on-load tap-changer is used to switch the tap under the condition that the transformer is loaded, so as to adjust the output voltage of the transformer and ensure the stable operation of the power system. However, during actual operation, the contacts of the on-load tap-changer are prone to overheating faults, which may cause contact burnout, insulation aging, and even power outages, seriously affecting the reliability and safety of the power system.

[0003] Currently, the main methods for diagnosing overheating faults of on-load tap-changer contacts are as follows:

[0004] 1. Infrared thermometry: Detect the temperature distribution on the surface of the contacts through an infrared thermal imager to determine whether the contacts are overheated. This method can intuitively display the temperature of the contacts, enable non-contact measurement, and is relatively easy to operate.

[0005] 2. Oil-dissolved gas analysis method: Analyze the components and contents of the gases dissolved in the transformer oil, and judge whether there is an overheating fault of the contacts according to the change characteristics of the gases. This method can reflect the internal fault conditions of the contacts to a certain extent and has a good diagnostic effect on early latent faults.

[0006] 3. Electrical parameter monitoring method: Monitor electrical parameters such as current and voltage of the on-load tap-changer, and judge the working state of the contacts according to the changes in the parameters. This method can obtain electrical information in real time and is convenient for dynamically monitoring the operating state of the switch.

[0007] Although the above methods play a certain role in diagnosing overheating faults of on-load tap-changer contacts, there are still some deficiencies:

[0008] Limited accuracy: Infrared thermometry is easily affected by environmental factors (such as light, occlusion, etc.), and the measurement results may be deviated; the oil-dissolved gas analysis method can only indirectly reflect the fault conditions of the contacts and cannot accurately judge the specific location and severity of the faults; the electrical parameter monitoring method may be insensitive to some minor overheating faults and is prone to missed judgments.

[0009] In view of this, we propose a method and system for diagnosing overheating faults of on-load tap-changer contacts. Summary of the Invention

[0010] The object of the present invention is to provide a method and system for diagnosing overheating faults of on-load tap-changer contacts, so as to solve the technical problem of poor accuracy of existing methods for diagnosing overheating faults of on-load tap-changer contacts.

[0011] To solve the above technical problems, the present invention provides the following technical solution: A method for diagnosing overheating faults of on-load tap-changer contacts, comprising the following steps:

[0012] S1. Historical data collection;

[0013] Collect a plurality of groups of historical contact temperature curves of the contact to be diagnosed, and each group of historical contact temperature curves corresponds to a historical tap position change situation

[0014] S2. Characteristic temperature interval division;

[0015] Divide each group of historical contact temperature curves into multiple characteristic temperature intervals;

[0016] S3. Binary classifier training;

[0017] For each historical tap position change situation, train a binary classifier based on the characteristic temperature intervals under this historical tap position change situation. The binary classifier is used to determine whether the current characteristic temperature interval of the contact temperature curve to be diagnosed belongs to an abnormal characteristic temperature interval;

[0018] S4. Analysis of the curve to be diagnosed;

[0019] Collect the contact temperature curve of the contact to be diagnosed under the change situation of the tap position to be diagnosed, divide the contact temperature curve to be diagnosed, and obtain multiple characteristic temperature intervals;

[0020] Input each characteristic temperature interval into the corresponding binary classifier. The binary classifier whose output is an abnormal characteristic temperature interval is marked as an abnormal contact temperature curve, and several contact temperature curves to be diagnosed marked as abnormal contact temperature curves are obtained. The contact temperature curves to be diagnosed of the abnormal contact temperature curves are diagnosed as abnormal contact temperature curves.

[0021] Preferably, the specific process of the binary classifier training includes:

[0022] S31. Curve classification;

[0023] Based on the characteristic temperature intervals of the historical temperature curves of the contact to be diagnosed, obtain the characteristic temperature intervals corresponding to several historical tap position change situations. Based on the characteristic temperature intervals corresponding to several historical tap position change situations, divide the historical contact temperature curves into abnormal contact temperature curves and normal contact temperature curves;

[0024] S32. Dimension expansion;

[0025] Multidimensionally expand each characteristic temperature interval to obtain a number of new characteristic temperature intervals;

[0026] S33. Feature extraction;

[0027] Extract the sample features of each new characteristic temperature interval based on the normal contact temperature curve and the abnormal contact temperature curve;

[0028] S34. Classifier training;

[0029] Train a binary classifier based on the sample features to obtain several binary classifiers

[0030] Preferably, the step of dividing the historical contact temperature curve into an abnormal contact temperature curve and a normal contact temperature curve is specifically as follows:

[0031] S311. Calculation of temperature rise coefficient;

[0032] Obtain the characteristic temperature rise coefficient α of each characteristic temperature interval of the historical contact temperature curve, where T start is the temperature at the starting moment of the characteristic temperature interval, T end is the temperature at the ending moment of the characteristic temperature interval, and Δt is the duration of the characteristic temperature interval;

[0033] S312. Preliminary classification;

[0034] Mark the characteristic temperature interval with a characteristic temperature rise coefficient greater than the first threshold α 1 as an abnormal contact temperature curve, and mark the characteristic temperature interval with a characteristic temperature rise coefficient less than the second threshold α 2 as a normal contact temperature curve, where α 2 < α 1 ;

[0035] S313. Pending processing;

[0036] Mark the characteristic temperature interval with a characteristic temperature rise coefficient equal to the second threshold α 2 as a pending contact temperature curve;

[0037] S314. Persistence calculation;

[0038] Calculate the persistence β of each characteristic temperature interval of the historical contact temperature curve. The calculation formula is where ΔT i is the temperature change value at the i-th adjacent moment within the characteristic temperature interval, and N is the number of temperature data points within the characteristic temperature interval minus 1;

[0039] S314. Continuous classification;

[0040] Mark the characteristic temperature interval with a persistence greater than the third threshold β 1The characteristic temperature range is marked as the normal contact temperature curve, and the persistence less than the fourth threshold β 2 The characteristic temperature range is marked as the abnormal contact temperature curve, where β 2 <β 1 .

[0041] Preferably, the step of performing multi-dimensional expansion on each characteristic temperature range is specifically as follows:

[0042] S321. Dimension division;

[0043] Each characteristic temperature range is divided into n dimensions according to the dimension. Each dimension corresponds to a binary classifier, and the binary classifier corresponding to each dimension is used to determine whether the corresponding dimension within the characteristic temperature range is normal;

[0044] One dimension of the characteristic temperature range is determined by the time interval from the corresponding moment in the historical tap position change situation corresponding to each data point within the corresponding characteristic temperature range to the corresponding moment in the historical tap position change situation corresponding to the next data point;

[0045] S322. Multi-dimensional expansion;

[0046] Perform multi-dimensional expansion on the characteristic temperature range corresponding to the historical tap position change situation to obtain a number of new characteristic temperature ranges. The new characteristic temperature ranges are determined by the dimension corresponding to each characteristic temperature range and the tap position change situation corresponding to each characteristic temperature range. One two-dimensionality corresponds to a binary classifier, where n k represents the number of dimensions divided for the kth characteristic temperature range, and m is the total number of characteristic temperature ranges.

[0047] Preferably, S4 further includes: obtaining the historical tap position change situation corresponding to the abnormal characteristic temperature range through the output results of multiple binary classifiers, and calculating the similarity S between the tap position change situation to be diagnosed and each historical tap position change situation. The calculation formula is where x i is the tap position value at the ith moment of the tap position change situation to be diagnosed, y i is the tap position value at the ith moment of the historical tap position change situation, and L is the total number of moments;

[0048] Output the historical tap position change situation with the highest similarity to the tap position change situation to be diagnosed, and the corresponding contact overheating fault type and severity level of the abnormal characteristic temperature range.

[0049] A on-load tap-changer contact overheating fault diagnosis system includes a server. The server includes a memory, a processor, and executable instructions stored on the memory and executable on the processor.

[0050] Preferably, it further includes:

[0051] A collection module, configured to collect multiple groups of historical contact temperature curves of the contact to be diagnosed, and collect the contact temperature curve of the contact to be diagnosed under the condition of the change of the tap position to be diagnosed;

[0052] A characteristic temperature interval division module, configured to divide the historical contact temperature curve and the contact temperature curve to be diagnosed into multiple characteristic temperature intervals;

[0053] A classifier training module, configured to train a binary classifier for each historical tap position change condition based on the characteristic temperature intervals under the corresponding historical tap position change condition, to obtain several binary classifiers, and the binary classifier is used to determine whether the current characteristic temperature interval of the contact temperature curve to be diagnosed belongs to an abnormal characteristic temperature interval;

[0054] A diagnosis module, configured to input each characteristic temperature interval of the contact temperature curve to be diagnosed into several binary classifiers, and the binary classifier whose output is an abnormal characteristic temperature interval is marked as an abnormal contact temperature curve, and the abnormal contact temperature curve is the abnormal contact temperature curve.

[0055] Compared with the prior art, the beneficial effects of the present invention are:

[0056] 1. By collecting multiple groups of historical contact temperature curves of the contact to be diagnosed and training a binary classifier for each historical tap position change condition, the present invention can fully consider the influence of the tap position change on the contact temperature, so as to more accurately judge whether the characteristic temperature interval of the contact temperature curve to be diagnosed is abnormal, effectively improving the accuracy of fault diagnosis. For example, in practical applications, for some slight overheating faults caused by tap position switching, the present invention can detect them more sensitively, avoiding the occurrence of missed judgments, and solving the problem of poor accuracy of the existing diagnosis method for overheating faults of on-load tap changer contacts.

[0057] 2. In the dimension expansion step of the present invention, each characteristic temperature interval is expanded multidimensionally to obtain multiple new characteristic temperature intervals. This multi-dimensional analysis method can more comprehensively reflect the working state and heat generation situation of the contact, overcoming the limitation of the prior art of diagnosing from a single angle. By comprehensively considering information in different dimensions, it is possible to more accurately identify abnormal characteristic temperature intervals and improve the reliability of fault diagnosis.

[0058] 3. The present invention outputs the results of multiple binary classifiers, obtains the historical tap position change situation corresponding to the abnormal characteristic temperature range, and outputs the contact overheating fault type and severity corresponding to the historical tap position change situation with the highest similarity to the tap position change situation to be diagnosed and the corresponding abnormal characteristic temperature range. This enables the operation and maintenance personnel to quickly and accurately understand the specific situation of the fault, take targeted maintenance measures in a timely manner, and reduce the impact of the fault on the power system. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a schematic diagram of the system framework of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0060] To facilitate the understanding of the technical solution of the present invention by those skilled in the art, the technical solution of the present invention will be further described below with reference to the accompanying drawings of the specification.

[0061] Embodiment 1. The present invention provides a method for diagnosing the contact overheating fault of an on-load tap-changer, including the following steps:

[0062] S1. Historical data collection;

[0063] Collect a number of historical contact temperature curves of the contact to be diagnosed, and each group of historical contact temperature curves corresponds to a historical tap position change situation

[0064] S2. Division of characteristic temperature ranges;

[0065] Divide each group of historical contact temperature curves into multiple characteristic temperature ranges;

[0066] S3. Binary classifier training;

[0067] For each historical tap position change situation, train a binary classifier based on the characteristic temperature range under this historical tap position change situation. The binary classifier is used to determine whether the current characteristic temperature range of the contact temperature curve to be diagnosed belongs to the abnormal characteristic temperature range;

[0068] S4. Analysis of the curve to be diagnosed;

[0069] Collect the contact temperature curve of the contact to be diagnosed under the tap position change situation to be diagnosed, divide the contact temperature curve to be diagnosed, and obtain multiple characteristic temperature ranges;

[0070] Input each characteristic temperature range into the corresponding binary classifier. The binary classifier whose output is the abnormal characteristic temperature range is marked as the abnormal contact temperature curve, and several contact temperature curves to be diagnosed marked as the abnormal contact temperature curve are obtained. The contact temperature curve to be diagnosed of the abnormal contact temperature curve is diagnosed as the abnormal contact temperature curve.

[0071] In an embodiment of the present invention, the specific process of binary classifier training includes:

[0072] S31. Curve classification;

[0073] Based on each characteristic temperature interval of the historical temperature curve of the contact to be diagnosed, obtain the characteristic temperature intervals corresponding to several historical tap position change situations. Based on the characteristic temperature intervals corresponding to several historical tap position change situations, divide the historical contact temperature curve into an abnormal contact temperature curve and a normal contact temperature curve;

[0074] S32. Dimension expansion;

[0075] Perform multi-dimensional expansion on each characteristic temperature interval to obtain several new characteristic temperature intervals;

[0076] S33. Feature extraction;

[0077] Extract the sample features of each new characteristic temperature interval based on the normal contact temperature curve and the abnormal contact temperature curve;

[0078] S34. Classifier training;

[0079] Perform binary classifier training based on the sample features to obtain several binary classifiers

[0080] In an embodiment of the present invention, the step of dividing the historical contact temperature curve into an abnormal contact temperature curve and a normal contact temperature curve is specifically as follows:

[0081] S311. Calculation of temperature rise coefficient;

[0082] Obtain the characteristic temperature rise coefficient α of each characteristic temperature interval of the historical contact temperature curve, where T start is the temperature at the start time of the characteristic temperature interval, T end is the temperature at the end time of the characteristic temperature interval, and Δt is the duration of the characteristic temperature interval;

[0083] S312. Preliminary classification;

[0084] Mark the characteristic temperature interval with a characteristic temperature rise coefficient greater than the first threshold α 1 as an abnormal contact temperature curve, and mark the characteristic temperature interval with a characteristic temperature rise coefficient less than the second threshold α 2 as a normal contact temperature curve, where α 2 <α 1 ;

[0085] S313. Pending processing;

[0086] Mark the characteristic temperature interval with a characteristic temperature rise coefficient equal to the second threshold α 2 as a pending contact temperature curve;

[0087] S314. Continuous calculation;

[0088] Calculate the persistence β of each characteristic temperature interval of the historical contact temperature curve. The calculation formula is where ΔT i is the temperature change value at the i-th adjacent moment within the characteristic temperature interval, and N is the number of temperature data points within the characteristic temperature interval minus 1;

[0089] S314. Continuous classification;

[0090] Mark the characteristic temperature intervals with persistence greater than the third threshold β 1 as normal contact temperature curves, and mark the characteristic temperature intervals with persistence less than the fourth threshold β 2 as abnormal contact temperature curves, where β 2 < β 1 .

[0091] In the embodiments of the present invention, the step of performing multi-dimensional expansion on each characteristic temperature interval is specifically as follows:

[0092] S321. Dimension division;

[0093] Divide each characteristic temperature interval into m dimensions according to the dimension. Each dimension corresponds to a binary classifier, and the binary classifier corresponding to each dimension is used to determine whether the corresponding dimension within the characteristic temperature interval is normal;

[0094] One dimension of the characteristic temperature interval is determined by the time interval from the corresponding moment in the historical tap position change situation corresponding to each data point within the corresponding characteristic temperature interval to the corresponding moment in the historical tap position change situation corresponding to the next data point;

[0095] S322. Multi-dimensional expansion;

[0096] Perform multi-dimensional expansion on the characteristic temperature interval corresponding to the historical tap position change situation to obtain several new characteristic temperature intervals. The new characteristic temperature intervals are determined by the dimension corresponding to each characteristic temperature interval and the tap position change situation corresponding to each characteristic temperature interval. One two-dimensionality corresponds to a binary classifier, where n k represents the number of dimensions divided for the k-th characteristic temperature interval, and m is the total number of characteristic temperature intervals.

[0097] In the embodiments of the present invention, S4 further includes: obtaining the historical tap position change situation corresponding to the abnormal characteristic temperature interval through the output results of multiple binary classifiers, and calculating the similarity S between the tap position change situation to be diagnosed and each historical tap position change situation. The calculation formula is where x i$y_i$ is the tap position value of the tap position change to be diagnosed at the $i$-th moment. i $L_i$ is the tap position value of the historical tap position change at the $i$-th moment, and $L$ is the total number of moments.

[0098] Output the historical tap position change with the highest similarity to the tap position change to be diagnosed, and the contact overheating fault type and severity corresponding to the abnormal characteristic temperature range.

[0099] Embodiment 2: As Figure 1 shown, a on-load tap-changer contact overheating fault diagnosis system includes a server, which includes a memory, a processor, and executable instructions stored on the memory and executable on the processor.

[0100] In an embodiment of the present invention, an intelligent inspection device is further included. The intelligent inspection device includes a sensor, a sampling circuit, and a microprocessor; the sensor is installed on the on-load tap-changer contact, the sensor is connected to the microprocessor through the sampling circuit; the sensor is used to collect the current, temperature, and tap position of the on-load tap-changer, the sampling circuit is used to digitize the current, temperature, and tap position information, and the microprocessor sends information such as the tap position, current, temperature, and time of the tap-changer to the host computer; the host computer includes a memory and a processor, the memory is used to store the historical contact temperature curve collected, the processor is used to divide each historical contact temperature curve into multiple characteristic temperature ranges, train multiple binary classifiers, and the host computer also includes an on-load tap-changer contact overheating fault diagnosis system.

[0101] In an embodiment of the present invention, the following are further included:

[0102] A collection module, which is used to collect multiple groups of historical contact temperature curves of the contact to be diagnosed, and collect the contact temperature curve of the contact to be diagnosed under the change of the tap position to be diagnosed;

[0103] A characteristic temperature range division module, which is used to divide the historical contact temperature curve and the contact temperature curve to be diagnosed into multiple characteristic temperature ranges;

[0104] A classifier training module, which is used to train a binary classifier for each historical tap position change situation based on the characteristic temperature range under the historical tap position change situation, and obtain several binary classifiers. The binary classifier is used to judge whether the current characteristic temperature range of the contact temperature curve to be diagnosed belongs to the abnormal characteristic temperature range;

[0105] A diagnosis module, which is used to input each characteristic temperature range of the contact temperature curve to be diagnosed into several binary classifiers. The binary classifier whose output is the abnormal characteristic temperature range is marked as the abnormal contact temperature curve, and the abnormal contact temperature curve is the abnormal contact temperature curve.

[0106] The embodiments disclosed in the present invention are preferred embodiments, but not limited thereto. Those of ordinary skill in the art can easily understand the spirit of the present invention based on the above embodiments and make different extensions and changes. As long as they do not depart from the spirit of the present invention, they are within the protection scope of the present invention.

Claims

1. A method for diagnosing an on-load tap changer contact overheating fault, characterized in that: The following steps are involved: S1. Historical data collection; Collect several groups of historical contact temperature curves of the contacts to be diagnosed, each group of historical contact temperature curves corresponds to a historical tap position change situation S2, characteristic temperature interval division; Divide each group of historical contact temperature curves into multiple characteristic temperature intervals; S3, binary classifier training; For each historical tap position change situation, a binary classifier is trained based on the characteristic temperature interval under the historical tap position change situation, and the binary classifier is used to determine whether the current characteristic temperature interval of the contact temperature curve to be diagnosed belongs to an abnormal characteristic temperature interval; S4, analysis of the curve to be diagnosed; Collecting the temperature curve of the contact to be diagnosed when the tap position to be diagnosed changes, dividing the temperature curve of the contact to be diagnosed, and obtaining multiple characteristic temperature intervals; Each characteristic temperature interval is input into the corresponding binary classifier, and the binary classifier whose output is an abnormal characteristic temperature interval is marked as an abnormal contact temperature curve, and several contact temperature curves to be diagnosed marked as abnormal contact temperature curves are obtained, and the contact temperature curves to be diagnosed of the abnormal contact temperature curves are diagnosed as abnormal contact temperature curves.

2. A method for diagnosing an on-load tap changer contact overheating fault according to claim 1, characterized in that: The specific process of the binary classifier training includes: S31, curve classification; Based on each characteristic temperature interval of the historical temperature curve of the contact to be diagnosed, characteristic temperature intervals corresponding to several historical tapping gear position changes are obtained, and based on the characteristic temperature intervals corresponding to several historical tapping gear position changes, the historical contact temperature curve is divided into an abnormal contact temperature curve and a normal contact temperature curve; S32, dimensional expansion; Expand each characteristic temperature interval multidimensionally to obtain several new characteristic temperature intervals; S33, feature extraction; Extracting new characteristic temperature interval sample features based on normal contact temperature curve and abnormal contact temperature curve; S34, classifier training; Binary classifiers are trained based on sample features to obtain several binary classifiers.

3. The method for diagnosing an on-load tap changer contact overheating fault according to claim 2, characterized in that: The specific steps of dividing the historical contact temperature curve into abnormal contact temperature curve and normal contact temperature curve are as follows: S311, calculation of temperature rise coefficient; Obtain the characteristic temperature rise coefficient α of each characteristic temperature interval of the historical contact temperature curve, Among them, T start is the temperature at the beginning of the characteristic temperature interval, T end is the temperature at the end of the characteristic temperature interval, and Δt is the duration of the characteristic temperature interval; S312, preliminary classification; The characteristic temperature interval with a characteristic temperature rise coefficient greater than a first threshold value α1 is marked as an abnormal contact temperature curve, and the characteristic temperature interval with a characteristic temperature rise coefficient less than a second threshold value α2 is marked as a normal contact temperature curve, wherein α2<α1; S313, pending processing; Marking a characteristic temperature interval where the characteristic temperature rise coefficient is equal to the second threshold value α2 as a pending contact temperature curve; S314, continuous calculation; Calculate the persistence β of each characteristic temperature interval of the historical contact temperature curve. The calculation formula is: Where, ΔT i is the temperature change value at the i-th adjacent moment in the characteristic temperature interval, and N is the number of temperature data points in the characteristic temperature interval minus 1; S315, continuous classification; The characteristic temperature interval whose persistence is greater than the third threshold value β1 is marked as a normal contact temperature curve, and the characteristic temperature interval whose persistence is less than the fourth threshold value β2 is marked as an abnormal contact temperature curve, wherein β2<β1.

4. A method for diagnosing an on-load tap changer contact overheating fault according to claim 3, characterized in that: The steps of multi-dimensionally expanding each characteristic temperature range are specifically as follows: S321, Dimensional division; Each characteristic temperature interval is divided into n dimensions according to the dimension, and each dimension corresponds to a binary classifier. The binary classifier corresponding to each dimension is used to determine whether the corresponding dimension in the characteristic temperature interval is normal; One dimension of the characteristic temperature interval is determined by the time interval from the corresponding moment in the historical tapping gear position change situation corresponding to each data point in the corresponding characteristic temperature interval to the corresponding moment in the historical tapping gear position change situation corresponding to the next data point; S322, multi-dimensional expansion; The characteristic temperature intervals corresponding to the historical tap position changes are multi-dimensionally expanded to obtain several new characteristic temperature intervals, which are determined by the dimensions corresponding to each characteristic temperature interval and the tap position changes corresponding to each characteristic temperature interval. One two-dimensional corresponds to one binary classifier, where: n k It represents the number of dimensions into which the kth characteristic temperature interval is divided, and m is the total number of characteristic temperature intervals.

5. The method for diagnosing an on-load tap changer contact overheating fault according to claim 4, characterized in that: The step S4 also includes: obtaining the historical tap position changes corresponding to the abnormal characteristic temperature interval through the output results of multiple binary classifiers, and calculating the similarity S between the tap position changes to be diagnosed and the historical tap position changes, and the calculation formula is: Among them, x i is the gear value of the tap gear change to be diagnosed at the i-th moment, y i is the gear value of the historical tap gear change at the i-th moment, and L is the total number of moments; The contact overheating fault type and severity corresponding to the historical tap position change situation that is most similar to the tap position change situation to be diagnosed and the corresponding abnormal characteristic temperature range are output.

6. An on-load tap changer contact overheat fault diagnosis system applied to the on-load tap changer contact overheat fault diagnosis method according to claim 5, characterized in that: A server is included, the server including a memory, a processor, and executable instructions stored on the memory and executable on the processor.

7. The on-load tap changer contact overheat fault diagnosis system according to claim 6, characterized in that: Also includes: The acquisition module is used to acquire multiple groups of historical contact temperature curves of the contacts to be diagnosed, and to acquire the temperature curves of the contacts to be diagnosed when the tapping gear position to be diagnosed changes; A characteristic temperature interval division module is used to divide the historical contact temperature curve and the contact temperature curve to be diagnosed into multiple characteristic temperature intervals; A classifier training module is used to train a binary classifier based on the characteristic temperature interval under each historical tap position change situation, and obtain several binary classifiers, wherein the binary classifier is used to determine whether the current characteristic temperature interval of the contact temperature curve to be diagnosed belongs to an abnormal characteristic temperature interval; The diagnostic module is used to input each characteristic temperature interval of the contact temperature curve to be diagnosed into several binary classifiers. The binary classifier outputting the abnormal characteristic temperature interval is marked as an abnormal contact temperature curve, and the abnormal contact temperature curve is the abnormal contact temperature curve.

Images