Transformer fault diagnosis method and device, equipment and storage medium

Abstract

The embodiment of the invention discloses a transformer fault diagnosis method and device, equipment and a storage medium. The method comprises the following steps: constructing a simulation model of a transformer; collecting simulation operation data of the simulation model under different working conditions; taking the simulation operation data as a training sample, training a preset prediction model according to the training sample, and obtaining a transformer fault prediction model; and acquiring actual operation parameters of the transformer, and inputting the actual operation parameters into the transformer fault prediction model to obtain a fault type corresponding to the actual operation parameters output by the transformer fault prediction model. The simulation operation data generated through the simulation model is used for training the transformer fault prediction model so as to ensure the comprehensiveness of the training data; secondly, the actual operation data are input into the transformer fault prediction model for prediction, potential fault signs and fault possibility of the transformer can be identified, and maintenance and protection measures can be taken in advance, so that the fault time and the downtime of the transformer are reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of fault diagnosis, and in particular to a transformer fault diagnosis method and device, equipment and storage medium. Background Art

[0002] In the operation of power systems, transformers are key power equipment, and they play an important role in ensuring the safe and stable operation of power systems. However, under fault conditions, the spatial physical quantities of transformers and disconnectors are significantly different from those under normal conditions, which may have a significant impact on the operation of power systems.

[0003] Transformer turn-to-turn short circuit fault is one of the most frequent faults. Poor insulation material performance, poor heat dissipation, transformer overload and other reasons may lead to transformer turn-to-turn short circuit fault. Transformer turn-to-turn short circuit increases the current at the winding end, generates huge short circuit force in the winding, damages the winding insulation and structural parts, and even causes the winding to loosen, twist, deform, break, and even burn in severe cases. However, there is still no effective transformer short circuit fault analysis method. Summary of the invention

[0004] Based on this, it is necessary to propose a transformer fault diagnosis method and device, equipment and storage medium to address the above problems, so that the transformer fault diagnosis results can be more accurate and effective.

[0005] To achieve the above objectives, the present application provides a transformer fault diagnosis method in a first aspect, the method comprising:

[0006] Build a simulation model of the transformer;

[0007] Collecting simulation operation data of the simulation model under different working conditions;

[0008] The simulation operation data is used as a training sample, and a preset prediction model is trained according to the training sample to obtain a transformer fault prediction model;

[0009] The actual operating parameters of the transformer are collected, and the actual operating parameters are input into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model.

[0010] Furthermore, collecting simulation operation data of the simulation model under different working conditions specifically includes:

[0011] Perform a frequency sweep response test based on the simulation model to simulate the operating state of the transformer under different working conditions, and obtain initial operating data of the simulation model under different working conditions;

[0012] Obtaining historical operating data of the transformer;

[0013] Performing comparative analysis on the historical operation data and the initial operation data to obtain a comparative result;

[0014] Correcting the simulation model according to the comparison result to obtain a corrected simulation model;

[0015] The simulation operation data of the modified simulation model under different working conditions are collected.

[0016] Furthermore, the sweep frequency response test is performed based on the simulation model to simulate the operating state of the transformer under different working conditions to obtain the initial operating data of the simulation model under different working conditions, specifically including:

[0017] Perform a frequency sweep response test based on the simulation model to simulate the operating state of the transformer under different working conditions to obtain initial operating parameters of the simulation model under different working conditions, wherein the initial operating parameters include at least current, voltage and excitation parameters;

[0018] Calculating electrical characteristics of the transformer core according to the initial operating parameters, wherein the electrical characteristics at least include: variation characteristics of resistance, inductance and capacitance;

[0019] Observing the mechanical stress and hot spot temperature generated by the simulation model under the electrical characteristics, and determining the initial fault type of the simulation model under the initial operating parameters according to the mechanical stress and hot spot temperature of the simulation model;

[0020] The initial operation parameters and the initial fault types corresponding to the initial operation parameters constitute the initial operation data.

[0021] Further, the historical operation data includes historical operation parameters and historical fault types corresponding to the historical operation parameters, and the comparison and analysis between the historical operation data and the initial operation data to obtain a comparison result specifically includes:

[0022] When the target initial fault type in the initial operation data and the target historical fault type in the historical operation data are the same fault type, the data difference between the target historical operation parameters corresponding to the target historical fault type and the target initial operation parameters corresponding to the target initial fault type is calculated, and the data difference is the comparison result.

[0023] Further, the simulation model is corrected according to the comparison result to obtain a corrected simulation model, which specifically includes:

[0024] Calculating a correction value of a simulation parameter in the simulation model according to the data difference;

[0025] The simulation model is corrected according to the correction value to obtain a corrected simulation model.

[0026] Furthermore, the method further comprises:

[0027] The transformer fault prediction model is combined with clustering according to the simulation operation data to obtain a plurality of clusters, each cluster corresponding to a fault type;

[0028] The collecting the actual operating parameters of the transformer, inputting the actual operating parameters into the transformer fault prediction model, and obtaining the fault type corresponding to the actual operating parameters output by the transformer fault prediction model specifically includes:

[0029] Calculating the Euclidean distance between the actual operating parameters of the transformer and each cluster, and determining the target cluster into which the actual operating parameters fall;

[0030] The fault type corresponding to the target cluster is used as the fault type corresponding to the actual operating parameter of the transformer.

[0031] Furthermore, calculating the Euclidean distance between the actual operating parameter of the transformer and each cluster and determining the target cluster into which the actual operating parameter falls specifically includes:

[0032] respectively calculating the Euclidean distance between the actual operating parameters of the transformer and the center of each cluster;

[0033] The target cluster where the simulation operating parameters corresponding to the minimum Euclidean distance belong is used as the target cluster where the actual operating parameters of the transformer fall into.

[0034] To achieve the above-mentioned purpose, the second aspect of the present application provides a transformer fault diagnosis device, the device comprising: a simulation unit, a model training unit and a fault prediction unit;

[0035] The simulation unit is used to construct a simulation model of the transformer;

[0036] The model training unit is used to collect simulation operation data of the simulation model under different working conditions;

[0037] The simulation operation data is used as a training sample, and a preset prediction model is trained according to the training sample to obtain a transformer fault prediction model;

[0038] The fault prediction unit is used to collect actual operating parameters of the transformer, and input the actual operating parameters into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model.

[0039] To achieve the above-mentioned purpose, the third aspect of the present application provides a computer-readable storage medium storing a computer program, characterized in that when the computer program is executed by a processor, the processor executes the steps of the method described in the first aspect.

[0040] To achieve the above-mentioned purpose, the fourth aspect of the present application provides a computer device, including a memory and a processor, characterized in that the memory stores a computer program, and when the computer program is executed by the processor, the processor executes the steps of the method described in the first aspect.

[0041] The embodiments of the present invention have the following beneficial effects:

[0042] The embodiment of the present invention proposes a method for diagnosing transformer faults, which includes: constructing a simulation model of the transformer; collecting simulation operation data of the simulation model under different working conditions; using the simulation operation data as training samples, training a preset prediction model according to the training samples, and obtaining a transformer fault prediction model; collecting actual operating parameters of the transformer, and inputting the actual operating parameters into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model. By building a simulation model of the transformer, the simulation operation data generated by the simulation model is used to train the transformer fault prediction model to ensure the comprehensiveness of the training data, so that the prediction results of the transformer fault prediction model are more accurate; secondly, inputting the collected actual operation data into the transformer fault prediction model for prediction can identify potential fault signs of the transformer and predict the possibility of transformer failure, which is helpful to take maintenance and protection measures in advance to reduce transformer failures and downtime. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] in:

[0045] Figure 1 A schematic diagram of a flow chart of a transformer fault diagnosis method according to an embodiment of the present invention;

[0046] Figure 2 is a structural block diagram of a transformer fault diagnosis device according to an embodiment of the present invention;

[0047] Figure 3 1 is a diagram showing the internal structure of a computer device in an embodiment of the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] Based on this, a transformer fault diagnosis method is proposed in an embodiment of the present invention, referring to Figure 1 , Figure 1 The figure is a flow chart of a transformer fault diagnosis method according to an embodiment of the present invention, and the method comprises:

[0050] Step 110: construct a simulation model of the transformer.

[0051] Specifically, according to the structural dimensions and physical characteristics of the actual power transformer to be simulated, the physical field parameters and material properties are set in the finite element software to establish a complete and detailed three-dimensional simulation model.

[0052] In order to build a simulation model that is more in line with the actual transformer, the embodiment of the present invention determines the optimal voltage and current feeding method under different working conditions by comparing the voltage and current data of the simulation model under different feeding methods. For example, the voltage and current data corresponding to different working conditions under different feeding methods of the simulation model are obtained, and compared with the voltage and current of the actual transformer under different working conditions, and the feeding method corresponding to the voltage and current that is closer to the actual transformer is used as the optimal voltage and current feeding method of the actual transformer. The optimal voltage and current feeding method is used as the voltage and current feeding method of the simulation model. A three-dimensional simulation model is constructed through the transformer accessory structure, and the distribution relationship of the structural magnetic field under different working conditions is studied to establish a simulation model with actual physical functions.

[0053] Step 120: Collect simulation operation data of the simulation model under different working conditions.

[0054] After the simulation model is built, the actual operating state of the transformer is simulated based on the simulation model, the operating state of the transformer under different working conditions is continuously simulated, and the simulation operation data under different working conditions are collected. Since the actual operating state that can be collected by the transformer is limited and it is difficult to measure more abundant operating data, simulation can be performed based on the simulation model of the transformer to obtain more simulation operation data under more operating states, providing a large amount of information-rich training data for subsequent model training.

[0055] Step 130: Use the simulation operation data as training samples, train a preset prediction model according to the training samples, and obtain a transformer fault prediction model.

[0056] Specifically, the simulation operation data obtained in step 120 is used as a training sample, and the training sample includes operating parameters and fault types. The training sample is input into a preset prediction model for training to obtain a transformer fault prediction model including a curve relationship between operating parameters and fault types.

[0057] In an embodiment of the present invention, the preset prediction model adopts an adaptive neural fuzzy system (ANFIS), which is a new type of fuzzy inference system structure that organically combines fuzzy logic and neural networks. The adaptive neural fuzzy system is usually based on the Takagi-Sugeno model, and its network structure generally includes five layers, including an input layer and an output layer. The curve formula is derived according to the input-output system and data variables to obtain the curve relationship between the operating parameters and the fault type, so as to predict the transformer fault type.

[0058] Step 140: Collect actual operating parameters of the transformer, and input the actual operating parameters into a transformer fault prediction model to obtain a fault type corresponding to the actual operating parameters output by the transformer fault prediction model.

[0059] In the embodiment of the present invention, after the transformer fault prediction model is obtained, the actual condition of the transformer can be monitored to timely discover transformer abnormalities and fault phenomena.

[0060] Specifically, the actual operating parameters of the transformer are collected in real time, and the real-time operating parameters are input into the transformer fault prediction model. The transformer fault prediction model performs calculation and analysis based on the actual operating parameters to obtain the fault type of the transformer.

[0061] The present invention builds a simulation model of the transformer, simulates different operating states based on the simulation model, and obtains simulated operating data under different working conditions. The simulated operating data contains richer operating data information of the transformer, so that the prediction result of the prediction model trained based on the simulated operating data is more accurate; in addition, the actual operating data of the collected transformer is input into the transformer fault prediction model for prediction, which can identify potential fault signs or fault possibilities of the transformer, and help to take maintenance and protection measures in advance to reduce transformer failures and downtime.

[0062] In the embodiment of the present invention, step 120, collecting simulation operation data of the simulation model under different working conditions, specifically includes:

[0063] Step 121. Perform a frequency sweep response test based on the simulation model to simulate the operating state of the transformer under different working conditions and obtain the initial operating data of the simulation model under different working conditions.

[0064] Specifically, a swept frequency response test is used to collect simulation operation data of the simulation model under different working conditions. Specifically, the swept frequency response test is to continuously change the frequency of the excitation signal within a certain frequency range and measure the response signal of the simulation model to analyze the frequency response characteristics of the simulation model.

[0065] In one embodiment of the present invention, a swept frequency response test is performed based on a simulation model to simulate the operating state of the transformer under different working conditions, and obtain initial operating data of the simulation model under different working conditions, specifically including:

[0066] Step 1211, perform a frequency sweep response test based on the simulation model, simulate the operating state of the transformer under different working conditions, and obtain the initial operating parameters of the simulation model under different working conditions. The initial operating parameters include at least current, voltage and excitation parameters.

[0067] In an embodiment of the present invention, a continuous simulation is performed on the operating state of the transformer under different operating conditions based on a simulation model, and a continuous periodic frequency response test is used to obtain the excitation parameters of the transformer under different frequency conditions, as well as the winding voltage and current containing multiple frequency components.

[0068] Step 1212, calculating the electrical characteristics of the transformer core according to the initial operating parameters, wherein the electrical characteristics at least include: change characteristics of resistance, inductance and capacitance.

[0069] The voltage, current and excitation parameters obtained through simulation model simulation are used to determine the changing characteristics of the transformer core and internal resistance, inductance and capacitance parameters.

[0070] Step 1213, observe the mechanical stress and hot spot temperature generated by the simulation model under the electrical characteristics, and determine the initial fault type of the simulation model under the initial operating parameters based on the mechanical stress and hot spot temperature of the simulation model.

[0071] Step 1214, the initial operating parameters and the initial fault types corresponding to the initial operating parameters constitute initial operating data.

[0072] Specifically, the location and size of mechanical stress and hot spot temperature in the simulation model under different electrical characteristics will be different. Therefore, the mechanical stress and hot spot temperature in the simulation model under different resistance, inductance, and capacitance parameter change characteristics can be observed. Then, the internal conditions of the transformer, the deformation of the winding, and the occurrence of mechanical failures can be determined based on the mechanical stress and hot spot temperature in the simulation model, and the failure conditions corresponding to different current, voltage, and excitation parameters can be determined, that is, the failure types corresponding to different operating parameters.

[0073] Step 122. Obtain the historical operation data of the transformer.

[0074] In the embodiment of the present invention, historical operation data of the actual transformer in normal working state and abnormal working state are collected, that is, historical operation data when a short-circuit fault occurs and when no short-circuit fault occurs.

[0075] Voltage can be collected through voltage sensors or voltage transformers; leakage inductance is usually collected by measuring the leakage inductance of the transformer, such as the short-circuit method and the open-circuit method; spatial magnetic field strength is measured by magnetic field measuring instruments, such as Tesla meters (Gauss meters) or flux meters, etc. Signal processing technology and feature extraction algorithms are used to obtain the current, voltage and excitation parameters of the transformer in normal working conditions and when a short-circuit fault occurs.

[0076] Step 123. Compare and analyze the historical operation data with the initial operation data to obtain a comparison result.

[0077] In order to improve the accuracy of the prediction results of the transformer fault prediction model, the embodiment of the present invention also modifies the transformer fault prediction model. Specifically, the historical operation data of the transformer is compared with the simulated operation data obtained by simulating the simulation model to obtain a comparison result, which reflects the gap between the simulation model and the actual transformer. The simulation model is modified according to the comparison result, so that the prediction results of the modified transformer fault prediction model are more accurate.

[0078] In one embodiment of the present invention, the historical operation data includes historical operation parameters and historical fault types corresponding to the historical operation parameters. A comparison analysis is performed on the historical operation data and the initial operation data to obtain a comparison result, which specifically includes: when the target initial fault type in the initial operation data and the target historical fault type in the historical operation data are the same fault type, calculating the data difference between the target historical operation parameters corresponding to the target historical fault type and the target initial operation parameters corresponding to the target initial fault type, and the data difference is the comparison result.

[0079] Specifically, the historical operation data of the transformer obtained includes historical operation parameters and historical fault types corresponding to the historical operation parameters. For example, the historical fault types may include no fault, turn-to-turn short circuit, winding grounding, and phase-to-phase short circuit.

[0080] There are differences between the operating parameters corresponding to each fault type. The target historical operating parameters when the transformer has the target fault type can be compared with the target initial operating parameters when the simulation model simulates the target fault type to determine the difference between the two, such as calculating the data difference between the two operating parameters. The transformer fault prediction model can be corrected based on the data difference. For example, the historical operating parameters of the transformer and the simulated operating parameters of the simulation model are obtained when a turn-to-turn short circuit occurs, and the data difference between the historical operating parameters and the simulated operating parameters is calculated.

[0081] Step 124, modify the simulation model according to the comparison result to obtain a modified simulation model.

[0082] In one embodiment of the present invention, the simulation model is corrected according to the comparison result to obtain a corrected simulation model, which specifically includes:

[0083] Step 1241, calculate the correction value of the simulation parameter in the simulation model according to the data difference.

[0084] Step 1242, correct the simulation model according to the correction value to obtain a corrected simulation model.

[0085] Specifically, the correction value is determined according to the data difference between the simulation model and the operating data when the transformer has the same fault. The correction value is the correction value of the physical parameters in the simulation model, such as the correction value of the resistance, inductance and other parameters. The simulation model is corrected according to the correction value, and the corrected simulation model obtained is closer to the actual transformer. Therefore, the simulation operation data obtained based on the corrected simulation model is more accurate.

[0086] Step 125, collect simulation operation data of the modified simulation model under different working conditions.

[0087] After building a more accurate revised simulation model, the actual operating state of the transformer is simulated based on the revised simulation model, and the operating state of the transformer under different working conditions is continuously simulated, and the simulation operation data under different working conditions are collected. Since the revised simulation operation data is more accurate, the accuracy of the prediction results of the transformer fault prediction model trained based on the revised simulation operation data is improved.

[0088] In one embodiment of the present invention, the transformer fault diagnosis method further includes: the transformer fault prediction model is clustered according to the simulation operation data to obtain a plurality of clusters, each cluster corresponding to a fault type.

[0089] In an embodiment of the present invention, the transformer is classified according to various typical fault types that may actually occur, and the simulation operation data is combined with a clustering method to obtain several clusters. The cluster whose center is closest to the origin of the coordinate axis represents that the transformer is in a completely healthy state, and the cluster whose center is farthest from the origin of the coordinate axis represents the most serious fault type and requires immediate intervention.

[0090] Based on this, the actual operating parameters of the transformer are collected and input into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model, including:

[0091] Step 141. Calculate the Euclidean distance between the actual operating parameters of the transformer and each cluster, and determine the target cluster into which the actual operating parameters fall.

[0092] Specifically, the distance between the actual operating parameters and each cluster is determined by calculating the Euclidean distance between the actual operating parameters of the transformer and each cluster, and then the cluster into which the actual operating parameters fall is determined.

[0093] In one embodiment of the present invention, calculating the Euclidean distance between the actual operating parameters of the transformer and each cluster and determining the target cluster into which the actual operating parameters fall specifically includes:

[0094] Step 1411, calculate the Euclidean distance between the actual operating parameters of the transformer and the center of each cluster respectively.

[0095] Step 1412, taking the target cluster where the simulation operating parameters corresponding to the minimum Euclidean distance belong as the target cluster where the actual operating parameters of the transformer fall.

[0096] Specifically, the collected actual operating parameters of the transformer are used as input parameters of the transformer fault prediction model, and the Euclidean distance between the input parameters and the simulated operating parameters corresponding to the center of each cluster is calculated. The Euclidean distance is used to quantify the similarity between the actual operating parameters of the transformer and the simulated operating parameters. Optionally, the Euclidean distance between the two can be calculated by the following formula:

[0097]

[0098] Where x is the actual operating parameter, y is the simulated operating parameter, n is the total number of operating parameters, and i is any one of the n operating parameters.

[0099] The calculated Euclidean distance is normalized using the minimum-maximum normalization method to prevent attributes with a larger range from having a greater impact on the result. The calculation method is as follows:

[0100]

[0101] All normalized Euclidean distances are compared to determine the minimum Euclidean distance, and the target cluster corresponding to the minimum Euclidean distance is taken as the cluster into which the actual operating data falls. That is to say, under the actual operating parameters, the fault type corresponding to the target cluster is the transformer fault.

[0102] Step 142: The fault type corresponding to the target cluster is used as the fault type corresponding to the actual operating parameters of the transformer.

[0103] The embodiment of the present invention determines the state of the transformer through the actual operating parameters of the transformer and the Euclidean distance between each cluster, thereby indicating the severity of the transformer fault and the urgency of intervention.

[0104] In one embodiment of the present invention, an online monitoring background is set up to monitor the fault status of the transformer, realize the triggering, collection, transmission, storage and management of data, and perform rapid analysis on the transformer. The fault type of the transformer can be confirmed in real time, so as to discover the abnormal status of the transformer in real time and timely.

[0105] In the embodiment of the present invention, a transformer fault diagnosis device is also provided, which can be referred to Figure 2 , Figure 2 2 is a structural block diagram of a transformer fault diagnosis device in an embodiment of the present invention. The device includes: a simulation unit 201, a model training unit 202 and a fault prediction unit 203.

[0106] The simulation unit 201 is used to construct a simulation model of the transformer.

[0107] The model training unit 202 is used to collect simulation operation data of the simulation model under different working conditions.

[0108] The simulation operation data is used as training samples, and the preset prediction model is trained according to the training samples to obtain the transformer fault prediction model.

[0109] The fault prediction unit 203 is used to collect actual operating parameters of the transformer, and input the actual operating parameters into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model.

[0110] The transformer fault diagnosis device proposed in the present invention builds a simulation model of the transformer, and uses the simulation operation data generated by the simulation model to train the transformer fault prediction model to ensure the comprehensiveness of the training data, so that the prediction results of the transformer fault prediction model are more accurate; secondly, the collected actual operation data is input into the transformer fault prediction model for prediction, which can identify potential fault signs of the transformer and predict the possibility of transformer failure, which is helpful to take maintenance and protection measures in advance to reduce transformer failures and downtime.

[0111] Figure 3 FIG. 1 shows an internal structure diagram of a computer device in one embodiment of the present invention. The computer device may be a terminal or a system. Figure 3 As shown, the computer device includes a processor, a memory and a network interface connected via a system bus. The memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor may implement each step in the above method embodiment. Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0112] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor executes each step in the above method embodiment.

[0113] In one embodiment, a computer-readable storage medium is proposed, which stores a computer program. When the computer program is executed by a processor, the processor executes each step in the above method embodiment. A person of ordinary skill in the art can understand that all or part of the processes in the above embodiment method can be completed by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium, and the program can include the processes of the embodiments of the above methods when executed. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0114] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0115] The above embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several modifications and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for diagnosing transformer faults, characterized in that: The method comprises: Build a simulation model of the transformer; Collecting simulation operation data of the simulation model under different working conditions; The simulation operation data is used as a training sample, and a preset prediction model is trained according to the training sample to obtain a transformer fault prediction model; The actual operating parameters of the transformer are collected, and the actual operating parameters are input into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model.

2. The method according to claim 1, characterized in that The collecting of simulation operation data of the simulation model under different working conditions specifically includes: Perform a frequency sweep response test based on the simulation model to simulate the operating state of the transformer under different working conditions, and obtain initial operating data of the simulation model under different working conditions; Obtaining historical operating data of the transformer; Performing comparative analysis on the historical operation data and the initial operation data to obtain a comparative result; Correcting the simulation model according to the comparison result to obtain a corrected simulation model; The simulation operation data of the modified simulation model under different working conditions are collected.

3. The method according to claim 2, characterized in that The sweep frequency response test is performed based on the simulation model to simulate the operating state of the transformer under different working conditions to obtain the initial operating data of the simulation model under different working conditions, specifically including: Perform a frequency sweep response test based on the simulation model to simulate the operating state of the transformer under different working conditions to obtain initial operating parameters of the simulation model under different working conditions, wherein the initial operating parameters include at least current, voltage and excitation parameters; Calculating electrical characteristics of the transformer core according to the initial operating parameters, wherein the electrical characteristics at least include: variation characteristics of resistance, inductance and capacitance; Observing the mechanical stress and hot spot temperature generated by the simulation model under the electrical characteristics, and determining the initial fault type of the simulation model under the initial operating parameters according to the mechanical stress and hot spot temperature of the simulation model; The initial operation parameters and the initial fault types corresponding to the initial operation parameters constitute the initial operation data.

4. The method according to claim 3, characterized in that The historical operation data includes historical operation parameters and historical fault types corresponding to the historical operation parameters, and the comparison and analysis between the historical operation data and the initial operation data to obtain the comparison result specifically includes: When the target initial fault type in the initial operation data and the target historical fault type in the historical operation data are the same fault type, the data difference between the target historical operation parameters corresponding to the target historical fault type and the target initial operation parameters corresponding to the target initial fault type is calculated, and the data difference is the comparison result.

5. The method according to claim 4, characterized in that The step of correcting the simulation model according to the comparison result to obtain a corrected simulation model specifically includes: Calculating a correction value of a simulation parameter in the simulation model according to the data difference; The simulation model is corrected according to the correction value to obtain a corrected simulation model.

6. The method according to claim 1, characterized in that The method further comprises: The transformer fault prediction model is combined with clustering according to the simulation operation data to obtain a plurality of clusters, each cluster corresponding to a fault type; The collecting the actual operating parameters of the transformer, inputting the actual operating parameters into the transformer fault prediction model, and obtaining the fault type corresponding to the actual operating parameters output by the transformer fault prediction model specifically includes: Calculating the Euclidean distance between the actual operating parameters of the transformer and each cluster, and determining the target cluster into which the actual operating parameters fall; The fault type corresponding to the target cluster is used as the fault type corresponding to the actual operating parameter of the transformer.

7. The method according to claim 6, characterized in that The calculating the Euclidean distance between the actual operating parameter of the transformer and each cluster, and determining the target cluster into which the actual operating parameter falls, specifically includes: respectively calculating the Euclidean distance between the actual operating parameters of the transformer and the center of each cluster; The target cluster where the simulation operating parameters corresponding to the minimum Euclidean distance belong is used as the target cluster where the actual operating parameters of the transformer fall into.

8. A transformer fault diagnosis device, characterized in that: The device comprises: a simulation unit, a model training unit and a fault prediction unit; The simulation unit is used to construct a simulation model of the transformer; The model training unit is used to collect simulation operation data of the simulation model under different working conditions; The simulation operation data is used as a training sample, and a preset prediction model is trained according to the training sample to obtain a transformer fault prediction model; The fault prediction unit is used to collect actual operating parameters of the transformer, and input the actual operating parameters into the transformer fault prediction model to obtain the fault type corresponding to the actual operating parameters output by the transformer fault prediction model.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

10. A computer device comprising a memory and a processor, characterized in that: The memory stores a computer program, and when the computer program is executed by the processor, the processor is caused to perform the steps of the method according to any one of claims 1 to 7.

Images