Transformer fault diagnosis method, system and equipment and storage medium

Abstract

The invention discloses a transformer fault diagnosis method, system and device and a storage medium, and the method comprises the following steps: defining the fault type of a transformer, and obtaining the concentration value of dissolved gas in an oil sample of the transformer in operation; determining the uncertainty range of the concentration value of each dissolved gas, and generating a data set of uncertainty of all dissolved gases; forming a plurality of groups of dissolved gas content combinations by using the generated data set, performing fault diagnosis on all the combinations by using a Duval pentagonal method, calculating the proportion of each fault type as the probability of occurrence of the fault, and generating a feature vector containing the probabilities of a plurality of fault types; and establishing a fault diagnosis model, training the fault diagnosis model, and performing fault diagnosis and type prediction on the transformer by the trained fault diagnosis model according to the concentration value of the dissolved gas in the oil sample of the transformer in operation. The influence of the measurement uncertainty of the dissolved gas is eliminated, and the accuracy and reliability of fault diagnosis are improved.

Description

Technical Field

[0001] The present invention belongs to the field of power transformers, and relates to a transformer fault diagnosis method, system, device and storage medium. Background Art

[0002] Power transformers play a crucial role in the power grid, ensuring the efficient transmission and conversion of energy. Although transformer insulating oil is industrially refined, during use, the effects of electrical, thermal, and chemical stresses can lead to the formation of various decomposition products, and these impurities will reduce the service life and service reliability of the transformer. Early electrical and thermal faults in power transformers will cause dissolved gases to form in the oil. As an important means of power transformer fault diagnosis, the dissolved gas analysis (DGA) method can achieve early detection and location of transformer faults by analyzing the components and concentrations of dissolved gases in transformer oil. However, the timeliness of this method is relatively low and the detection speed is slow. In recent years, the application of artificial intelligence (AI) has greatly promoted and enriched new methods for power transformer fault diagnosis. Combining the DGA expert database with the AI model can improve the fault diagnosis rate and accuracy. Common models include support vector machines, random forests, neural networks, etc.

[0003] In addition, various measurement devices used at home and abroad currently have different detection accuracies and measurement thresholds, and these inherent characteristics and differences in the devices will cause certain errors in the measured values. However, existing DGA methods often ignore the impact of the inherent measurement uncertainty of the measurement device on the diagnostic results, resulting in limitations in the accuracy and reliability of the diagnostic results. And the data source with errors will also interfere with the fault diagnosis and affect the diagnostic accuracy. Summary of the Invention

[0004] The purpose of the present invention is to overcome the above-mentioned shortcomings of the prior art, and provide a transformer fault diagnosis method, system, device and storage medium, which eliminates the influence of the uncertainty of dissolved gas measurement and improves the accuracy and reliability of fault diagnosis.

[0005] To achieve the above object, the present invention adopts the following technical solutions: A transformer fault diagnosis method includes the following processes: Define the transformer fault types, and obtain the concentration values of dissolved gases in the transformer oil sample during operation; Determine the uncertainty range of each dissolved gas concentration value, and generate a dataset of all dissolved gas uncertainties; Use the generated dataset to form multiple groups of dissolved gas content combinations, use the Duval pentagon method to perform fault diagnosis on all combinations, calculate the proportion of each fault type, which is used as the probability of the occurrence of the fault, and generate a feature vector containing the probabilities of multiple fault types; Obtain the concentration values of dissolved gases in all oil samples for calculation to form multiple feature vectors; Establish a fault diagnosis model. Combine all feature vectors and the corresponding true fault type labels to form a training data set, and train the fault diagnosis model. The trained fault diagnosis model performs fault diagnosis and type prediction on the transformer according to the concentration values of dissolved gases in the transformer oil sample during operation.

[0006] Preferably, the transformer fault types include low-energy discharge, high-energy discharge, partial discharge, low-temperature overheating, medium-temperature overheating, and high-temperature overheating.

[0007] Preferably, the dissolved gases include one or more of H2, CH4, C2H2, C2H4, C2H6, CO, and CO2.

[0008] Preferably, the process of determining the uncertainty range of the concentration value of each dissolved gas and generating a data set of uncertainties of all dissolved gases is as follows: The uncertainty range of the concentration value of dissolved gas i is , is the maximum deviation between the measured concentration value and the true value of dissolved gas i, C 2i is the concentration of dissolved gas i after considering the uncertainty, and m is a proportionality coefficient; adjust the proportionality coefficient m to obtain multiple concentration values generated by dissolved gas i; repeat this process for all dissolved gases to obtain a data set considering the uncertainties of all gases.

[0009] Preferably, the process of forming multiple combinations of dissolved gas contents using the generated data set, performing fault diagnosis on all combinations using the Duval pentagon method, calculating the proportion of each fault type, and using it as the probability of the occurrence of the fault, and generating a feature vector containing multiple fault type probabilities is as follows: Take different concentration values of different dissolved gases to form multiple gas content combinations, and use the Duval pentagon method to perform fault diagnosis one by one. Count the proportion of each fault type in all diagnostic results as the probability of the possible occurrence of the fault, and then obtain the fault type probability feature vector under this sample.

[0010] Preferably, use the machine learning algorithm of random forest to train the fault diagnosis model.

[0011] Furthermore, the process of training the fault diagnosis model using the machine learning algorithm of random forest is as follows: Take the feature vector as the input and the true fault type label as the output, initialize the number of parameter trees and the maximum depth, and use the method of cross-validation grid search to adjust the parameters of the fault diagnosis model.

[0012] A transformer fault diagnosis system includes: Definition module, used to define transformer fault types and obtain the concentration values of dissolved gases in the transformer oil sample during operation; Uncertainty module, used to determine the uncertainty range of each dissolved gas concentration value and generate a dataset of all dissolved gas uncertainties; Feature vector module, used to form multiple groups of dissolved gas content combinations using the generated dataset, perform fault diagnosis on all combinations using the Duval pentagon method, calculate the proportion of each fault type, which is used as the probability of the occurrence of the fault, and generate a feature vector containing the probabilities of multiple fault types; Multi-feature vector module, used to obtain the concentration values of dissolved gases in all oil samples for calculation and form multiple feature vectors; Fault diagnosis module, used to establish a fault diagnosis model, form a training dataset with all feature vectors and the corresponding true fault type labels, train the fault diagnosis model, and the trained fault diagnosis model performs fault diagnosis and type prediction on the transformer according to the concentration values of dissolved gases in the transformer oil sample during operation.

[0013] A computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the steps of the transformer fault diagnosis method are implemented.

[0014] A computer-readable storage medium, the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the transformer fault diagnosis method are implemented.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention incorporates the uncertainty of gas concentration measurement into the dissolved gas analysis, calculates the probabilities of each fault type using the Duval pentagon method, and combines an artificial intelligence model for fault diagnosis, eliminating the influence of misdiagnosis caused by measurement uncertainty, improving the accuracy and reliability of fault diagnosis, providing a new idea and method for the fault diagnosis of power transformers, and having a wide application prospect. Brief Description of the Drawings

[0016] Figure 1 It is a flowchart of the transformer fault diagnosis method of the present invention. Detailed Embodiments

[0017] The following details the embodiments of the present invention. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals represent the same or similar elements or elements with the same or similar functions from beginning to end. The embodiments described below with reference to the drawings are exemplary and are only used to explain the present invention and should not be construed as a limitation of the present invention.

[0018] In the description of the present invention, it should be understood that the orientation or positional relationship indicated by the terms "center", "longitudinal", "transverse", "length", "width", "thickness", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", "clockwise", "counterclockwise", etc. is based on the orientation or positional relationship shown in the drawings. It is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation to the present invention. In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly specifying the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of the said features. In the description of the present invention, "a plurality of" means two or more unless otherwise specifically defined.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms "mounted", "connected" and "coupled" shall be construed broadly. For example, it may be a fixed connection, a detachable connection or an integral connection; it may be a mechanical connection, an electrical connection or a connection capable of mutual communication; it may be directly connected or indirectly connected through an intermediate medium, and it may be the internal communication of two elements or the interaction relationship between two elements. The term "and / or" used herein includes any and all combinations of one or more of the related listed items. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. The terms used in the description of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0020] In the present invention, unless otherwise clearly defined and limited, the first feature being "on" or "under" the second feature may include the direct contact between the first and second features, or may include the situation where the first and second features are not in direct contact but in contact through additional features therebetween. Moreover, the first feature being "above", "over" and "on top of" the second feature includes that the first feature is directly above and obliquely above the second feature, or merely means that the horizontal height of the first feature is higher than that of the second feature. The first feature being "under", "below" and "beneath" the second feature includes that the first feature is directly below and obliquely below the second feature, or merely means that the horizontal height of the first feature is lower than that of the second feature.

[0021] The following disclosure provides many different embodiments or examples for implementing different structures of the present invention. To simplify the disclosure of the present invention, the components and settings of specific examples are described below. Of course, they are merely examples and are not intended to limit the present invention. In addition, the present invention may repeat reference numerals and / or reference letters in different examples. Such repetition is for the purpose of simplification and clarity, and in itself does not indicate the relationship between the various embodiments and / or settings discussed. In addition, the present invention provides examples of various specific processes and materials, but those of ordinary skill in the art can be aware of the application of other processes and / or the use of other materials.

[0022] Embodiment 1: As Figure 1 shown, the transformer fault diagnosis method described in this embodiment includes the following processes: Define the transformer fault types and obtain the concentration values of the dissolved gases in the transformer oil sample during operation.

[0023] Determine the uncertainty range of each dissolved gas concentration value and generate a dataset of all dissolved gas uncertainties.

[0024] Use the generated dataset to form multiple groups of dissolved gas content combinations, use the Duval pentagon method to perform fault diagnosis on all combinations, calculate the proportion of each fault type, which is used as the probability of the occurrence of the fault, and generate a feature vector containing the probabilities of multiple fault types.

[0025] Obtain the concentration values of the dissolved gases in all oil samples for calculation and form multiple feature vectors.

[0026] Establish a fault diagnosis model, form a training dataset by combining all feature vectors and the corresponding true fault type labels, train the fault diagnosis model, and the trained fault diagnosis model performs fault diagnosis and type prediction on the transformer according to the concentration values of the dissolved gases in the transformer oil sample during operation.

[0027] Embodiment 2: As Figure 1 shown, the transformer fault diagnosis method considering measurement uncertainty described in this embodiment includes the following steps: Step 1: Define the transformer fault types, collect the oil sample of the power transformer during operation as a sample, and use a gas chromatograph to measure the concentration of the dissolved gases in the oil; Step 2: Conduct uncertainty modeling, determine the uncertainty range of each dissolved gas concentration value, and generate a dataset considering the uncertainties of all gases; Step 3: Use the generated dataset to form multiple combinations of dissolved gas contents, and use the Duval pentagon method to diagnose faults for all combinations. Calculate the proportion of each fault type, which is the probability of the occurrence of the fault, and generate a feature vector containing the probabilities of multiple fault types. Step 4: Repeat Steps 1 to 3 for all oil sample specimens to form multiple feature vectors of fault type probabilities. Step 5: Combine the feature vectors of fault type probabilities of all samples and the corresponding true fault type labels to form a training dataset, and use the machine learning algorithm of random forest for training to carry out fault diagnosis and type prediction based on the fault probability distribution.

[0028] In a preferred embodiment, in Step 1, define the transformer fault types as D1 (low-energy discharge), D2 (high-energy discharge), PD (partial discharge), T1 (low-temperature overheating), T2 (medium-temperature overheating), and T3 (high-temperature overheating), and collect oil samples from the operating transformer.

[0029] The dissolved gases include one or more of H2, CH4, C2H2, C2H4, C2H6, CO, and CO2.

[0030] In a preferred embodiment, use a gas chromatograph to measure the concentrations of dissolved H2, CH4, C2H2, C2H4, C2H6, CO, and CO2 in the sample oil.

[0031] Specifically, the measured concentrations of dissolved gases in a group of oil sample specimens are shown in the following table.

[0032]

[0033] In a preferred embodiment, in Step 2, let U be the measurement uncertainty of the gas chromatograph, and take 5%. be the measured dissolved gas concentration, be the maximum deviation between the measured concentration value and the true value of dissolved gas i, be the concentration of dissolved gas i after considering the uncertainty.

[0034] Specifically:

[0035]

[0036] Specifically, m is a proportionality coefficient,

[0037] In a preferred embodiment, the uncertainty range of the concentration value of dissolved gas i is . By adjusting the proportionality coefficient m (in this embodiment, m takes 0, 0.1, 0.3, 0.5, 0.7, 0.9, 1 respectively), multiple concentration values that the dissolved gas i may generate are obtained. This step is repeated for all the dissolved gases, and then a data set considering the uncertainties of all the dissolved gases is obtained.

[0038] In a preferred embodiment, in step 3, different concentration values of different dissolved gases are taken to form multiple combinations of dissolved gas contents. The Duval pentagon method is used to carry out fault diagnosis one by one, and the proportion of each fault type in all the diagnosis results is counted as the probability that the fault may occur. Then, the fault type probability feature vector under this sample is obtained.

[0039] Specifically, in the fault type probability feature vector, the probability of being determined as D1 fault is 4.08%, the probability of being determined as D2 fault is 4.08%, the probability of being determined as PD fault is 77.55%, the probability of being determined as T1 fault is 8.16%, the probability of being determined as T2 fault is 0.00%, and the probability of being determined as T3 fault is 6.13%.

[0040] In a preferred embodiment, in step 4, steps 1 to 3 are repeated 12,000 times, and finally 12,000 groups of fault type probability feature vectors are obtained and used for training and prediction by machine learning algorithms.

[0041] In a preferred embodiment, in step 5, a fault diagnosis model is established. The fault type probability feature vectors of all samples are used as input features, and the true fault type labels are used as output features. The random forest algorithm is used to train the fault diagnosis model.

[0042] Preferably, the random forest algorithm framework includes: Taking the feature vector as the input and the true fault type label as the output, initializing the number of parameter trees (n_estimators) to 100 and the maximum depth (max_depth) to None, and using the method of cross-validation grid search to adjust the parameters of the fault diagnosis model.

[0043] Adopting the trained fault diagnosis model, according to the concentration values of the dissolved gases in the transformer oil sample to be diagnosed during operation, the fault diagnosis and type prediction of the transformer to be diagnosed are carried out. The data without considering measurement uncertainties and the data considering measurement uncertainties are respectively input into the random forest algorithm, and the recognition accuracies obtained are shown in the following table.

[0044]

[0045] The present invention takes into account the impact of measurement uncertainty on data accuracy. By introducing uncertainty modeling and fault probability calculation, the accuracy of power transformer fault identification is greatly improved.

[0046] The following is an apparatus embodiment of the present invention, which can be used to implement the method embodiment of the present invention. For details not disclosed in the apparatus embodiment, please refer to the method embodiment of the present invention.

[0047] In another embodiment of the present invention, a transformer fault diagnosis system is provided. The transformer fault diagnosis system can be used to implement the above-mentioned transformer fault diagnosis method. Specifically, the transformer fault diagnosis system includes a definition module, an uncertainty module, a feature vector module, a multi-feature vector module, and a fault diagnosis module.

[0048] Among them, the definition module is used to define the types of transformer faults and obtain the concentration values of dissolved gases in the transformer oil sample during operation.

[0049] The uncertainty module is used to determine the uncertainty range of each dissolved gas concentration value and generate a dataset of all dissolved gas uncertainties.

[0050] The feature vector module is used to form multiple groups of dissolved gas content combinations by using the generated dataset, perform fault diagnosis on all combinations using the Duval pentagon method, calculate the proportion of each fault type, which is used as the probability of the occurrence of the fault, and generate a feature vector containing the probabilities of multiple fault types.

[0051] The multi-feature vector module is used to obtain the concentration values of dissolved gases in all oil samples for calculation and form multiple feature vectors.

[0052] The fault diagnosis module is used to establish a fault diagnosis model, form a training dataset by combining all feature vectors and the corresponding true fault type labels, train the fault diagnosis model, and the trained fault diagnosis model performs fault diagnosis and type prediction on the transformer according to the concentration values of dissolved gases in the transformer oil sample during operation.

[0053] In another embodiment of the present invention, a terminal device is provided. The terminal device includes a processor and a memory. The memory is used to store a computer program, and the computer program includes program instructions. The processor is used to execute the program instructions stored in the computer storage medium. The processor may be a central processing unit (CPU), or may also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. It is the computing core and control core of the terminal, and is suitable for implementing one or more instructions. Specifically, it is suitable for loading and executing one or more instructions to implement the corresponding method flow or corresponding function. The processor described in the embodiment of the present invention can be used for the operation of the transformer fault diagnosis method, including: defining the transformer fault types, and obtaining the concentration values of the dissolved gases in the transformer oil sample during operation; determining the uncertainty range of each dissolved gas concentration value, and generating a data set of all dissolved gas uncertainties; forming multiple groups of dissolved gas content combinations by using the generated data set, using the Duval pentagon method to perform fault diagnosis on all combinations, calculating the proportion of each fault type, as the probability of occurrence of the fault, and generating a feature vector containing the probabilities of multiple fault types; obtaining the concentration values of the dissolved gases in all oil samples for calculation, and forming multiple feature vectors; establishing a fault diagnosis model, forming a training data set by combining all feature vectors and the corresponding true fault type labels, training the fault diagnosis model, and the trained fault diagnosis model performs fault diagnosis and type prediction on the transformer according to the concentration values of the dissolved gases in the transformer oil sample during operation.

[0054] In another embodiment, the present invention further provides a computer-readable storage medium (Memory). The computer-readable storage medium is a memory device in a terminal device and is used to store programs and data. It can be understood that the computer-readable storage medium here can include both the built-in storage medium in the terminal device and, of course, the extended storage medium supported by the terminal device. The computer-readable storage medium provides a storage space, and the operating system of the terminal is stored in this storage space. Moreover, one or more instructions suitable for being loaded and executed by a processor are stored in this storage space. These instructions can be one or more computer programs (including program codes). It should be noted that the computer-readable storage medium here can be a high-speed RAM memory or a non-volatile memory, such as at least one disk memory.

[0055] One or more instructions stored in the computer-readable storage medium can be loaded and executed by a processor to implement the corresponding steps of the transformer fault diagnosis method in the above embodiments; one or more instructions in the computer-readable storage medium are loaded and executed by the processor to perform the following steps: define the transformer fault types, and obtain the concentration values of dissolved gases in the transformer oil sample during operation; determine the uncertainty range of each dissolved gas concentration value, and generate a data set of the uncertainties of all dissolved gases; use the generated data set to form multiple combinations of dissolved gas contents, and use the Duval pentagon method to perform fault diagnosis on all combinations, calculate the proportion of each fault type, which is used as the probability of the occurrence of this fault, and generate a feature vector containing the probabilities of multiple fault types; obtain the concentration values of dissolved gases in all oil samples for calculation to form multiple feature vectors; establish a fault diagnosis model, form a training data set by combining all feature vectors and the corresponding true fault type labels, train the fault diagnosis model, and the trained fault diagnosis model performs fault diagnosis and type prediction on the transformer according to the concentration values of dissolved gases in the transformer oil sample during operation.

[0056] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0057] This application is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the flows and / or blocks in the flowchart and / or block diagram can also be implemented. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0058] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that implement the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0059] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one or more of the flows Figure 1 one or more of the flows and / or blocks Figure 1 or means for implementing the functions specified in one or more of the blocks.

[0060] The serial numbers of the above embodiments of the present application are only for description and do not represent the superiority or inferiority of the embodiments.

[0061] In the above embodiments of the present application, the descriptions of the respective embodiments have their own focuses. For parts not detailed in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0062] In several embodiments provided by the present application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only illustrative. For example, the division of the units can be a logical function division, and there can be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the couplings or direct couplings or communication connections shown or discussed with each other can be through some interfaces, and the indirect couplings or communication connections of the units or modules can be in electrical or other forms.

[0063] The unit described as a separating component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or may be distributed over multiple units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0064] The above description is only a preferred embodiment of the present application. It should be noted that for those of ordinary skill in the art of this technology, without departing from the principle of the present application, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present application.

[0065] It should be understood that the above description is for illustrative purposes and not for limitation. Many embodiments and many applications other than the examples provided will be apparent to those skilled in the art upon reading the above description. Therefore, the scope of this patent should not be determined with reference to the above description, but should be determined with reference to the full scope of the foregoing claims and the equivalents thereof. For the sake of completeness, all articles and references including patent applications and published announcements are incorporated herein by reference. The omission of any aspect of the subject matter disclosed herein in the foregoing claims is not intended to abandon such subject matter, nor should the applicant be regarded as not considering such subject matter as part of the disclosed inventive subject matter.

Claims

1. A transformer fault diagnosis method, characterized in that: The process includes: Define transformer fault types and obtain the concentration of dissolved gas in transformer oil samples during operation; Determine the uncertainty range for each dissolved gas concentration value and generate a dataset of all dissolved gas uncertainties; The generated data set is used to form multiple groups of dissolved gas content combinations, and the Duval pentagon method is used to diagnose faults in all combinations. The proportion of each fault type is calculated as the probability of the fault occurring, and a feature vector containing the probabilities of multiple fault types is generated. The concentration values ​​of dissolved gases in all oil samples are obtained for calculation to form multiple feature vectors; A fault diagnosis model is established, and all feature vectors and corresponding real fault type labels are combined into a training data set to train the fault diagnosis model. The trained fault diagnosis model performs fault diagnosis and type prediction on the transformer based on the concentration value of dissolved gas in the transformer oil sample during operation.

2. The transformer fault diagnosis method according to claim 1, characterized in that: Transformer fault types include low-energy discharge, high-energy discharge, partial discharge, low-temperature overheating, medium-temperature overheating and high-temperature overheating.

3. The transformer fault diagnosis method according to claim 1, characterized in that: The dissolved gas includes one or more of H2, CH4, C2H2, C2H4, C2H6, CO and CO2.

4. The transformer fault diagnosis method according to claim 1, characterized in that: The process of determining the uncertainty range of each dissolved gas concentration value and generating a data set of all dissolved gas uncertainties is as follows: The uncertainty range of the concentration value of dissolved gas i is , is the maximum deviation between the measured concentration of dissolved gas i and the true value, C 2i is the concentration of dissolved gas i after considering the uncertainty, and m is the proportional coefficient; the proportional coefficient m is adjusted to obtain multiple concentration values ​​produced by dissolved gas i; this process is repeated for all dissolved gases to obtain a data set that considers the uncertainty of all gases.

5. The transformer fault diagnosis method according to claim 1, characterized in that: The generated data set is used to form multiple groups of dissolved gas content combinations. The Duval pentagon method is used to diagnose faults in all combinations. The proportion of each fault type is calculated as the probability of the fault occurring. The process of generating a feature vector containing the probabilities of multiple fault types is as follows: Different concentration values ​​of different dissolved gases are taken to form multiple groups of gas content combinations. The Duval pentagon method is used to carry out fault diagnosis one by one. The proportion of each fault type in all diagnostic results is counted as the probability of the fault occurring, and then the fault type probability feature vector under the sample is obtained.

6. The transformer fault diagnosis method according to claim 1, characterized in that: The random forest machine learning algorithm is used to train the fault diagnosis model.

7. The transformer fault diagnosis method according to claim 6, characterized in that: The process of training the fault diagnosis model using the random forest machine learning algorithm is as follows: taking the feature vector as input, the true fault type label as output, initializing the number and maximum depth of parameter trees, and using the cross-validation grid search method to adjust the fault diagnosis model parameters.

8. A transformer fault diagnosis system, characterized in that: include: A definition module is used to define the transformer fault type and obtain the concentration value of dissolved gas in the transformer oil sample in operation; Uncertainty module, used to determine the uncertainty range for each dissolved gas concentration value and generate a dataset of all dissolved gas uncertainties; The feature vector module is used to form multiple groups of dissolved gas content combinations using the generated data set, perform fault diagnosis on all combinations using the Duval pentagon method, calculate the proportion of each fault type as the probability of the fault occurring, and generate a feature vector containing the probabilities of multiple fault types; The multi-feature vector module is used to obtain the concentration values ​​of dissolved gases in all oil samples for calculation to form multiple feature vectors; The fault diagnosis module is used to establish a fault diagnosis model. All feature vectors and corresponding real fault type labels form a training data set to train the fault diagnosis model. The trained fault diagnosis model performs fault diagnosis and type prediction on the transformer based on the concentration value of dissolved gas in the transformer oil sample during operation.

9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the transformer fault diagnosis method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the transformer fault diagnosis method according to any one of claims 1 to 7 are implemented.

Images