Active early warning method and system for slight turn-to-turn short circuit of transformer based on digital-electric fusion

Abstract

The invention discloses a transformer slight turn-to-turn short circuit active early warning method and system based on digital-electric fusion, and the method comprises the steps: obtaining the state monitoring data of a transformer, carrying out the data clustering and cleaning of the state monitoring data according to a dynamic basis function optimization strategy, and obtaining the target state monitoring data; according to a space-time alignment algorithm, matching the operation condition data influencing the slight turn-to-turn short circuit of the transformer with the target state monitoring data; fusing the target state monitoring data matched with the operation condition data with turn-to-turn short circuit fault related electrical test data, analyzing the reliability of different electrical test items to which the electrical test data belong based on an MMRL network, and analyzing the correlation of slight turn-to-turn short circuit characterization of the electrical test data; the severity of the turn-to-turn short circuit is solved through logarithmic electric fusion data cross validation of the electromagnetic model of the transformer, and active early warning is conducted on the slight turn-to-turn short circuit of the transformer.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent diagnosis of electric power equipment, and in particular relates to a method and system for active early warning of slight turn-to-turn short circuit of transformer based on digital-electrical fusion. Background Art

[0002] The transformer is the hub and core equipment of the power system. Its failure will seriously affect the power supply reliability of the power system. Turn-to-turn short circuit is the most common internal fault of the transformer. Therefore, it is necessary to carry out diagnostic monitoring of minor turn-to-turn short circuits of the transformer and accurately identify the fault in the early stage of turn-to-turn short circuit.

[0003] The existing technology mainly relies on online oil chromatography and electrical testing to detect and determine transformer turn-to-turn short circuits after the transformer fault expands. However, due to the small fault current and hidden characteristics of minor turn-to-turn short circuits, diagnostic analysis can only be carried out after the transformer fault trips, and active early warning of minor turn-to-turn short circuits cannot be achieved. Therefore, a technical solution for accurately and reliably detecting minor turn-to-turn short circuits using multi-data fusion technology is urgently needed to solve the above technical problems.

[0004] The prior art Chinese invention patent application 202310228558.4 discloses a method and device for transformer turn-to-turn short-circuit fault assessment. The invention draws a preliminary conclusion that the transformer is faulty through disassembly inspection, and simulates the transformer turn-to-turn short-circuit fault assessment mathematical model combined with the transformer short-circuit turn-to-turn short-circuit simulation test to obtain the final diagnosis conclusion. However, the invention patent application fails to solve technical problems such as multivariate data fusion analysis and early active warning of minor turn-to-turn short circuits. Summary of the invention

[0005] The present invention provides a transformer slight inter-turn short circuit active early warning method and system based on digital and electrical fusion, which are used to solve the technical problems of slight inter-turn short circuit monitoring and early warning.

[0006] In a first aspect, the present invention provides a transformer slight inter-turn short circuit active warning method based on digital and electrical fusion, comprising: Acquire transformer status monitoring data, and perform data clustering and cleaning on the status monitoring data according to a dynamic basis function optimization strategy to obtain target status monitoring data, wherein the target status monitoring data does not contain noise signals; According to the time-space alignment algorithm, the operating condition data affecting the transformer minor turn-to-turn short circuit is matched with the target state monitoring data; The target state monitoring data matched with the operating condition data is fused with the electrical test data related to the turn-to-turn short circuit fault to obtain digital-electrical fusion data, and the reliability of different electrical test items to which the electrical test data belongs is analyzed based on the MMRL network, and the relevance of the slight turn-to-turn short circuit characterization of the electrical test data is analyzed; The transformer electromagnetic model is used to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

[0007] In a second aspect, the present invention provides a transformer slight turn-to-turn short circuit active warning system based on digital-electric fusion, comprising: A processing module is configured to obtain state monitoring data of the transformer, and perform data clustering and cleaning on the state monitoring data according to a dynamic basis function optimization strategy to obtain target state monitoring data, wherein the target state monitoring data does not contain a noise signal; A matching module configured to match the operating condition data affecting the transformer slight turn-to-turn short circuit with the target state monitoring data according to a time-space alignment algorithm; An analysis module is configured to fuse the target state monitoring data matched with the operating condition data with the turn-to-turn short-circuit fault-related electrical test data to obtain digital-electrical fusion data, and analyze the reliability of different electrical test items to which the electrical test data belongs based on the MMRL network and analyze the relevance of the slight turn-to-turn short-circuit characterization of the electrical test data; The early warning module is configured to apply the transformer electromagnetic model to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

[0008] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can perform the steps of the active early warning method for slight inter-turn short circuit of a transformer based on digital-electrical fusion according to any embodiment of the present invention.

[0009] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program instructions are executed by a processor, the processor executes the steps of the active early warning method for minor turn-to-turn short circuit of a transformer based on digital-electric fusion according to any embodiment of the present invention.

[0010] The active early warning method and system for transformer minor inter-turn short circuit based on digital and electrical integration of the present application integrates monitoring data and electrical test data, deeply analyzes the history of the two types of data and the changing patterns between the data, identifies potential transformer faults in advance, and combines electromagnetic simulation models to provide early warning for minor inter-turn short circuit faults. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0012] Figure 1 A flowchart of a transformer slight turn-to-turn short-circuit active early warning method based on digital-electric fusion provided in one embodiment of the present invention; Figure 2 A structural block diagram of a transformer slight turn-to-turn short-circuit active early warning system based on digital-electrical fusion provided in one embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0013] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0014] See also Figure 1 , which shows a flow chart of a method for active early warning of a transformer slight inter-turn short circuit based on digital and electrical integration of the present application.

[0015] like Figure 1 As shown, the active early warning method for transformer slight inter-turn short circuit based on digital-electric fusion specifically includes the following steps: Step S101, acquiring transformer status monitoring data, and performing data clustering and cleaning on the status monitoring data according to a dynamic basis function optimization strategy to obtain target status monitoring data, wherein the target status monitoring data does not contain noise signals.

[0016] In this step, the status monitoring data includes: transformer oil chromatogram online monitoring data, voltage recording data, current recording data, oil surface temperature data, winding temperature data, ultrasonic partial discharge monitoring data, differential current harmonic characteristic data, arrester online monitoring data, meteorological data and vibration data; It should be noted that the state monitoring data is clustered and cleaned according to the dynamic basis function optimization strategy to obtain the target state monitoring data including: The dynamic basis function is used to optimize the state monitoring data, the optimal decomposition basis function is adaptively selected based on the signal spectrum entropy value, and the number of decomposition layers is dynamically determined by the hierarchical correlation analysis algorithm. The expression of the number of decomposition layers is: ; In the formula, is the number of decomposition layers, is the highest effective frequency of the signal, It is the noise-dominated frequency band; The noise components in the state monitoring data are stripped by using a coupled threshold generation network, wherein the input layer of the network is used to fuse the signal statistical characteristics with the frequency domain energy distribution, and the output layer of the network is used to generate a frequency band adaptive dynamic threshold curve to replace the global fixed threshold; The high frequency components in the state monitoring data are subjected to piecewise nonlinear compression according to the frequency-band progressive attenuation operator, wherein the expression of the frequency-band progressive attenuation operator is: ; In the formula, is the frequency-band progressive attenuation operator, is the high frequency component amplitude, is the high frequency attenuation rate; A bidirectional residual calibration mechanism is introduced to fuse and reconstruct the Nth layer low-frequency components and the compressed high-frequency components to obtain the target state monitoring data.

[0017] Step S102: Matching the operating condition data affecting the transformer slight turn-to-turn short circuit with the target state monitoring data according to the time-space alignment algorithm.

[0018] In this step, the operating condition data that affects the transformer's slight inter-turn short circuit includes: short circuit impact data, lightning strike data, abnormal switch switching data, and DC bias magnetic data.

[0019] According to the time-space alignment algorithm, the short-circuit impact data, lightning strike data, switch abnormal switching data and DC bias magnetic data are dynamically associated with the internal transient waveform of the transformer in the target state monitoring data.

[0020] The event source location for: , Encode external events, is the event abnormal deviation, is the time when the event occurs; the set of monitoring points is ,in is the number of monitoring sensors, i is the monitoring data sequence, To encode the monitoring data, To monitor the abnormal deviation of data, is a set element of monitoring points. Calculate the spatial impact weight distribution of the operating condition data affecting the transformer minor turn-to-turn short circuit: ; In the formula, is the field intensity attenuation coefficient; Build time delay compensation ; is the communication transmission rate; Generate the time-space matrix H of monitoring data: ; In the formula, is the length of the time series, To monitor the characteristic values ​​of different dimensions of data (including oil temperature, ultrasonic partial discharge, three-phase voltage / current, zero-sequence voltage / current, and online oil chromatography), is the time decay coefficient.

[0021] Step S103, the target state monitoring data matched with the operating condition data is merged with the electrical test data related to the turn-to-turn short circuit fault to obtain digital electrical fusion data, and the reliability of different electrical test items belonging to the electrical test data is analyzed based on the MMRL network, and the correlation of the slight turn-to-turn short circuit characterization of the electrical test data is analyzed.

[0022] In this step, the electrical test items include: frequency response impedance test, no-load test, short-circuit impedance test, winding capacitance test, voltage ratio test and DC resistance test, among which the no-load test and voltage ratio test are turn-to-turn short-circuit strong correlation tests, and the frequency response impedance test, short-circuit impedance test, winding capacitance test and DC resistance test are turn-to-turn short-circuit weak correlation tests.

[0023] The reliability of different electrical test items of electrical test data analyzed based on MMRL network includes: The Pearson correlation coefficient of the electrical test data and the target state monitoring data is expressed as: ; In the formula, is the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data, is the electrical test sequence number, is the kth eigenvector of the electrical test, is the kth eigenvector corresponding to the time-space matrix of the monitoring data, for The variance of for The variance of For the turn-to-turn short-circuit strong correlation test, if the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data is <0.5, the physical correlation between the kth electrical test data and the target state monitoring data is judged to be insufficient and marked as low reliability, otherwise it is high reliability; For the inter-turn short-circuit weak correlation test, if the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data is <0.2, the physical correlation between the kth electrical test data and the target state monitoring data is judged to be insufficient and marked as low reliability, otherwise it is high reliability.

[0024] Furthermore, the correlation of analyzing the electrical test data for slight turn-to-turn short circuit characterization includes: Extract high-order derivative extreme points and cumulative deviation index for turn-to-turn short-circuit strong correlation test; Extract local curvature variation, sliding window variance and long-term drift rate for inter-turn short circuit weak correlation test; Generate electrical test feature vectors of uniform dimension ,in, , , , , , They are no-load current offset, voltage ratio deviation, frequency response curvature, impedance variance, capacitance differential extreme value, and resistance drift rate; The electrical test feature vector The weight coefficients of each electrical test feature dimension are generated by learning the parameter matrix, and the correlation of the slight turn-to-turn short circuit characterization of the electrical test data with small offset is quantified. The expression is: ; In the formula, To learn the parameter matrix, characterize the correlation weights between each test data and the turn-to-turn short-circuit fault, is the Sigmoid function, which maps the weights to the (0,1) interval.

[0025] Step S104, applying the transformer electromagnetic model to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

[0026] In this step, the transformer electromagnetic model is used to cross-validate the logarithmic electric fusion data to solve the severity of the inter-turn short circuit, including: Construct the electromagnetic model of the transformer under different short-circuit turns of the transformer, input the reliability electrical test data into the electromagnetic model of the transformer, generate the corresponding simulation data, and calculate the simulation confidence, the expression is: ; In the formula, is the simulation confidence, is the difference tolerance threshold, is the degree of difference; When the simulation confidence <0.9, trigger parameter inversion optimization and establish the objective function , using particle swarm optimization algorithm to adjust the short-circuit position number of the transformer electromagnetic model , short-circuit turns and short-circuit turn leakage flux , iterate until convergence and obtain updated simulation data; Compare the updated simulation data with the target state monitoring data and calculate the difference, which is expressed as: ; In the formula, For the difference, For updated simulation data, is the target state monitoring data, is the 2-norm of the difference between the updated simulation data and the target state monitoring data.

[0027] Furthermore, active early warning for minor transformer turn-to-turn short circuits includes: Using the weighted evidence theory, the confidence of the target state monitoring data, the Pearson correlation coefficient of the electrical test data and the target state monitoring data, and the difference are integrated into the comprehensive fault probability, which is expressed as follows: ; In the formula, is the comprehensive failure probability, is the confidence weight coefficient of the target state monitoring data, Matching weights for the transformer electromagnetic model, is the difference tolerance threshold, is the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data, is the difference between the transformer electromagnetic model and the kth electrical test, is the number of short-circuit turns, is the total number of turns of the inter-turn short-circuit winding; When the comprehensive failure probability ≥ When the fault occurs, the diagnostic test is triggered, new test data is generated, the new test data is input into the transformer electromagnetic model, the short circuit position and degree are inverted, the corrected simulation data is output, and the corrected simulation data is compared with the new test data again to obtain the updated difference degree; The comprehensive failure probability is corrected according to the updated difference degree to obtain an updated comprehensive failure probability; When 0.6≤updated comprehensive failure probability<0.8, record the abnormality and increase the monitoring frequency; When 0.8≤updated comprehensive failure probability<0.9, a risk warning prompt is issued, fault recording monitoring and partial discharge and oil chromatography retest are started, and a power outage maintenance test plan is generated; When the updated comprehensive failure probability is ≥ 0.9, an active warning is issued, a power outage maintenance instruction is triggered, and the fault point is located in combination with the simulation results.

[0028] See also Figure 2 , which shows a structural block diagram of a transformer slight inter-turn short circuit active warning system based on digital and electrical integration of the present application.

[0029] like Figure 2 As shown, the transformer slight inter-turn short circuit active warning system 200 includes a processing module 210 , a matching module 220 , an analysis module 230 and a warning module 240 .

[0030] The processing module 210 is configured to obtain transformer state monitoring data, and perform data clustering and cleaning on the state monitoring data according to a dynamic basis function optimization strategy to obtain target state monitoring data, wherein the target state monitoring data does not contain noise signals; A matching module 220 is configured to match the operating condition data affecting the transformer minor turn-to-turn short circuit with the target state monitoring data according to a time-space alignment algorithm; The analysis module 230 is configured to fuse the target state monitoring data matched with the operating condition data with the turn-to-turn short circuit fault-related electrical test data to obtain digital-electrical fusion data, and analyze the reliability of different electrical test items to which the electrical test data belongs based on the MMRL network and analyze the relevance of the slight turn-to-turn short circuit characterization of the electrical test data; The early warning module 240 is configured to apply the transformer electromagnetic model to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

[0031] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0032] In other embodiments, the present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor executes the transformer slight turn-to-turn short circuit active warning method based on digital-electric fusion in any of the above method embodiments; As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows: Acquire transformer status monitoring data, and perform data clustering and cleaning on the status monitoring data according to a dynamic basis function optimization strategy to obtain target status monitoring data, wherein the target status monitoring data does not contain noise signals; According to the time-space alignment algorithm, the operating condition data affecting the transformer minor turn-to-turn short circuit is matched with the target state monitoring data; The target state monitoring data matched with the operating condition data is fused with the electrical test data related to the turn-to-turn short circuit fault to obtain digital-electrical fusion data, and the reliability of different electrical test items to which the electrical test data belongs is analyzed based on the MMRL network, and the relevance of the slight turn-to-turn short circuit characterization of the electrical test data is analyzed; The transformer electromagnetic model is used to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

[0033] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the transformer slight inter-turn short circuit active early warning system based on digital-electric fusion, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the transformer slight inter-turn short circuit active early warning system based on digital-electric fusion via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0034] Figure 3is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 In the example, the bus connection is used. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the active early warning method for transformer slight inter-turn short circuit based on digital and electrical fusion of the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the active early warning system for transformer slight inter-turn short circuit based on digital and electrical fusion. The output device 340 may include display devices such as display screens.

[0035] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0036] As an implementation mode, the electronic device is applied to a transformer slight inter-turn short circuit active warning system based on digital-electric fusion, and is used for a client, and includes: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: Acquire transformer status monitoring data, and perform data clustering and cleaning on the status monitoring data according to a dynamic basis function optimization strategy to obtain target status monitoring data, wherein the target status monitoring data does not contain noise signals; According to the time-space alignment algorithm, the operating condition data affecting the transformer minor turn-to-turn short circuit is matched with the target state monitoring data; The target state monitoring data matched with the operating condition data is fused with the electrical test data related to the turn-to-turn short circuit fault to obtain digital-electrical fusion data, and the reliability of different electrical test items to which the electrical test data belongs is analyzed based on the MMRL network, and the relevance of the slight turn-to-turn short circuit characterization of the electrical test data is analyzed; The transformer electromagnetic model is used to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

[0037] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0038] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A transformer slight inter-turn short circuit active warning method based on digital and electrical fusion, characterized in that: include: Acquire transformer status monitoring data, and perform data clustering and cleaning on the status monitoring data according to a dynamic basis function optimization strategy to obtain target status monitoring data, wherein the target status monitoring data does not contain noise signals; According to the time-space alignment algorithm, the operating condition data affecting the transformer minor turn-to-turn short circuit is matched with the target state monitoring data; The target state monitoring data matched with the operating condition data is fused with the electrical test data related to the turn-to-turn short circuit fault to obtain digital-electrical fusion data, and the reliability of different electrical test items to which the electrical test data belongs is analyzed based on the MMRL network, and the relevance of the slight turn-to-turn short circuit characterization of the electrical test data is analyzed; The transformer electromagnetic model is used to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

2. According to claim 1, a transformer slight turn-to-turn short circuit active warning method based on digital and electrical fusion is characterized in that: The state monitoring data includes: transformer oil chromatogram online monitoring data, voltage recording data, current recording data, oil surface temperature data, winding temperature data, ultrasonic partial discharge monitoring data, differential current harmonic characteristic data, arrester online monitoring data, meteorological data and vibration data; The data clustering and cleaning of the state monitoring data according to the dynamic basis function optimization strategy to obtain the target state monitoring data includes: The dynamic basis function is used to optimize the state monitoring data, the optimal decomposition basis function is adaptively selected based on the signal spectrum entropy value, and the number of decomposition layers is dynamically determined by the hierarchical correlation analysis algorithm. The expression of the number of decomposition layers is: ； In the formula, is the number of decomposition layers, is the highest effective frequency of the signal, It is the noise-dominated frequency band; The noise components in the state monitoring data are stripped by using a coupled threshold generation network, wherein the input layer of the network is used to fuse the signal statistical characteristics with the frequency domain energy distribution, and the output layer of the network is used to generate a frequency band adaptive dynamic threshold curve to replace the global fixed threshold; The high frequency components in the state monitoring data are subjected to piecewise nonlinear compression according to the frequency-band progressive attenuation operator, wherein the expression of the frequency-band progressive attenuation operator is: ； In the formula, is the frequency-band progressive attenuation operator, is the high frequency component amplitude, is the high frequency attenuation rate; A bidirectional residual calibration mechanism is introduced to fuse and reconstruct the Nth layer low-frequency components and the compressed high-frequency components to obtain the target state monitoring data.

3. The active early warning method for transformer slight turn-to-turn short circuit based on digital-electric fusion according to claim 1 is characterized in that: The operating condition data affecting the transformer's slight inter-turn short circuit includes: short circuit impact data, lightning strike data, abnormal switch switching data, and DC bias magnetic data; The matching of the operating condition data affecting the transformer minor turn-to-turn short circuit with the target state monitoring data according to the time-space alignment algorithm includes: According to the time-space alignment algorithm, the short-circuit impact data, lightning strike data, switch abnormal switching data and DC bias magnetic data are dynamically associated with the internal transient waveform of the transformer in the target state monitoring data.

4. The active early warning method for transformer slight turn-to-turn short circuit based on digital-electric fusion according to claim 1 is characterized in that: The electrical test items include: frequency response impedance test, no-load test, short-circuit impedance test, winding capacitance test, voltage ratio test and DC resistance test, among which, no-load test and voltage ratio test are turn-to-turn short-circuit strong correlation tests, and frequency response impedance test, short-circuit impedance test, winding capacitance test and DC resistance test are turn-to-turn short-circuit weak correlation tests; The reliability of different electrical test items of electrical test data analyzed based on the MMRL network includes: The Pearson correlation coefficient of the electrical test data and the target state monitoring data is expressed as: ； In the formula, is the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data, is the electrical test sequence number, is the kth eigenvector of the electrical test, is the kth eigenvector corresponding to the time-space matrix of the monitoring data, for The variance of for The variance of For the turn-to-turn short-circuit strong correlation test, if the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data is <0.5, the physical correlation between the kth electrical test data and the target state monitoring data is judged to be insufficient and marked as low reliability, otherwise it is high reliability; For the inter-turn short-circuit weak correlation test, if the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data is <0.2, the physical correlation between the kth electrical test data and the target state monitoring data is judged to be insufficient and marked as low reliability, otherwise it is high reliability.

5. The method for active early warning of transformer slight turn-to-turn short circuit based on digital-electric fusion according to claim 4 is characterized in that: The relevance of the analytical electrical test data for characterization of minor turn-to-turn shorts includes: Extract high-order derivative extreme points and cumulative deviation index for turn-to-turn short-circuit strong correlation test; Extract local curvature variation, sliding window variance and long-term drift rate from the inter-turn short circuit weak correlation test; Generate electrical test feature vectors of uniform dimension ,in, , , , , , They are no-load current offset, voltage ratio deviation, frequency response curvature, impedance variance, capacitance differential extreme value, and resistance drift rate; The electrical test feature vector The weight coefficients of each electrical test feature dimension are generated by learning the parameter matrix, and the correlation of the slight turn-to-turn short circuit characterization of the electrical test data with small offset is quantified. The expression is: ； In the formula, To learn the parameter matrix, characterize the correlation weights between each test data and the turn-to-turn short-circuit fault, is the Sigmoid function, which maps the weights to the (0,1) interval.

6. The active early warning method for transformer slight turn-to-turn short circuit based on digital-electric fusion according to claim 1 is characterized in that: The application of transformer electromagnetic model logarithmic electric fusion data cross validation to solve the severity of inter-turn short circuit includes: Construct the electromagnetic model of the transformer under different short-circuit turns of the transformer, input the reliability electrical test data into the electromagnetic model of the transformer, generate the corresponding simulation data, and calculate the simulation confidence, the expression is: ; In the formula, is the simulation confidence, is the difference tolerance threshold, is the degree of difference; When the simulation confidence <0.9, trigger parameter inversion optimization and establish the objective function , using particle swarm optimization algorithm to adjust the short-circuit position number of the transformer electromagnetic model , short-circuit turns and short-circuit turn leakage flux , iterate until convergence and obtain updated simulation data; Compare the updated simulation data with the target state monitoring data and calculate the difference, which is expressed as: ; In the formula, For the difference, For updated simulation data, is the target state monitoring data, is the 2-norm of the difference between the updated simulation data and the target state monitoring data.

7. The active early warning method for transformer slight turn-to-turn short circuit based on digital-electric fusion according to claim 1 is characterized in that: The active early warning for a minor transformer turn-to-turn short circuit includes: Using the weighted evidence theory, the confidence of the target state monitoring data, the Pearson correlation coefficient of the electrical test data and the target state monitoring data, and the difference are integrated into the comprehensive fault probability, which is expressed as follows: ; In the formula, is the comprehensive failure probability, is the confidence weight coefficient of the target state monitoring data, Matching weights for the transformer electromagnetic model, is the difference tolerance threshold, is the Pearson correlation coefficient between the kth electrical test data and the target state monitoring data, is the difference between the transformer electromagnetic model and the kth electrical test, is the number of short-circuit turns, is the total number of turns of the inter-turn short-circuit winding; When the comprehensive failure probability ≥ When the fault occurs, the diagnostic test is triggered, new test data is generated, the new test data is input into the transformer electromagnetic model, the short circuit position and degree are inverted, the corrected simulation data is output, and the corrected simulation data is compared with the new test data again to obtain the updated difference degree; The comprehensive failure probability is corrected according to the updated difference degree to obtain an updated comprehensive failure probability; When 0.6≤updated comprehensive failure probability<0.8, record the abnormality and increase the monitoring frequency; When 0.8≤updated comprehensive failure probability<0.9, a risk warning prompt is issued, fault recording monitoring and partial discharge and oil chromatography retest are started, and a power outage maintenance test plan is generated; When the updated comprehensive failure probability is ≥ 0.9, an active warning is issued, a power outage maintenance instruction is triggered, and the fault point is located in combination with the simulation results.

8. An active early warning system for transformer minor turn-to-turn short circuit based on digital and electrical fusion, characterized in that: include: A processing module is configured to obtain state monitoring data of the transformer, and perform data clustering and cleaning on the state monitoring data according to a dynamic basis function optimization strategy to obtain target state monitoring data, wherein the target state monitoring data does not contain a noise signal; A matching module configured to match the operating condition data affecting the transformer minor turn-to-turn short circuit with the target state monitoring data according to a time-space alignment algorithm; An analysis module is configured to fuse the target state monitoring data matched with the operating condition data with the turn-to-turn short-circuit fault-related electrical test data to obtain digital-electrical fusion data, and analyze the reliability of different electrical test items to which the electrical test data belongs based on the MMRL network and analyze the relevance of the slight turn-to-turn short-circuit characterization of the electrical test data; The early warning module is configured to apply the transformer electromagnetic model to cross-validate the digital electrical fusion data to solve the severity of the inter-turn short circuit, and actively warn of slight inter-turn short circuit of the transformer.

9. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the method described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images