Transformer partial discharge fault diagnosis method and system

Abstract

The invention provides a transformer partial discharge fault diagnosis method and system, and belongs to the technical field of transformers. The method comprises the following steps: inputting an ultrasonic signal generated during partial discharge of a transformer into a partial discharge ultrasonic detection and analysis model to obtain current partial discharge data, and judging whether the current partial discharge data exceeds a preset partial discharge threshold value or not to obtain a primary judgment result; inputting the content data of the dissolved gas in the transformer oil into the acetylene characteristic gas detection and analysis model to obtain acetylene characteristic gas content data, and judging whether the acetylene characteristic gas content data exceeds a preset acetylene characteristic gas content threshold value or not to obtain a secondary judgment result; and determining a final judgment result of the partial discharge state of the transformer based on the primary judgment result and the secondary judgment result. The accuracy, sensitivity and anti-interference capability of transformer partial discharge real-time detection and fault diagnosis are improved, and the purpose of automatic and precise detection of transformer partial discharge is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of transformers, and in particular to a transformer partial discharge fault diagnosis method, a transformer partial discharge fault diagnosis system, a machine-readable storage medium and an electronic device. Background Art

[0002] In the power system, transformers are the core equipment for energy conversion and transmission, and their safe operation is crucial to the stability of the power grid. During the long-term operation of transformers, partial discharge (PD) may occur due to factors such as aging of insulation materials and uneven electric field distribution. This is an early sign of transformer insulation degradation. Timely and accurate detection of partial discharge in transformers is of great significance for preventing major failures and ensuring the safety of the power grid.

[0003] The existing technology for transformer partial discharge detection has problems such as poor anti-interference ability, low sensitivity, long detection cycle and inability to locate the discharge position. Therefore, how to improve the accuracy, sensitivity and anti-interference ability of transformer partial discharge real-time detection and fault diagnosis is an urgent problem to be solved. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a transformer partial discharge fault diagnosis method and system to at least solve the above-mentioned problem of how to improve the accuracy, sensitivity and anti-interference ability of real-time detection and fault diagnosis of transformer partial discharge.

[0005] In order to achieve the above object, the present invention provides a transformer partial discharge fault diagnosis method in a first aspect, comprising:

[0006] Obtain the ultrasonic signal generated by partial discharge of the transformer and the dissolved gas content data in the transformer oil;

[0007] The ultrasonic signal generated during partial discharge of the transformer is input into the trained partial discharge ultrasonic detection and analysis model to obtain the current partial discharge data, and determine whether the current partial discharge data exceeds the preset partial discharge threshold value to obtain a judgment result;

[0008] The dissolved gas content data in the transformer oil is input into the trained acetylene characteristic gas detection and analysis model to obtain the acetylene characteristic gas content data, and it is determined whether the acetylene characteristic gas content data exceeds a preset acetylene characteristic gas content threshold value to obtain a secondary judgment result;

[0009] Based on the primary judgment result and the secondary judgment result, a final judgment result of the partial discharge state of the transformer is determined.

[0010] Optionally, the acquisition rules of the ultrasonic signal generated during partial discharge of the transformer include:

[0011] The ultrasonic sensor is controlled to collect the ultrasonic signal generated when the transformer is partially discharged, and the ultrasonic signal generated when the transformer is partially discharged is preprocessed; wherein,

[0012] The preprocessing process includes:

[0013] The ultrasonic signal generated by partial discharge of the transformer is converted into a corresponding electrical signal, and the electrical signal is amplified and filtered in turn.

[0014] Optionally, the rules for obtaining the data of dissolved gas content in the transformer oil include:

[0015] The TDLAS system is controlled to detect the characteristic absorption spectral line data of the dissolved gas in the transformer oil, and the characteristic absorption spectral line data of the dissolved gas in the transformer oil are converted to obtain the dissolved gas content data in the transformer oil.

[0016] Optionally, the training rules of the above partial discharge ultrasonic detection and analysis model include:

[0017] Based on the recursive neural network algorithm, a partial discharge ultrasonic detection and analysis model is established;

[0018] Acquire multiple historical ultrasonic signals generated during partial discharge of transformers and historical partial discharge data corresponding to each historical ultrasonic signal as a first training sample set; wherein the historical partial discharge data includes discharge type, discharge position and discharge displacement;

[0019] The partial discharge ultrasonic detection and analysis model is trained using the first training sample set to obtain a trained partial discharge ultrasonic detection and analysis model.

[0020] Optionally, the training rules of the above acetylene characteristic gas detection and analysis model include:

[0021] Based on the recursive neural network algorithm, an acetylene characteristic gas detection and analysis model is established;

[0022] Acquire a plurality of historical transformer oil dissolved gas content data and historical acetylene characteristic gas content data corresponding to each of the historical transformer oil dissolved gas content data as a second training sample set;

[0023] The acetylene characteristic gas detection and analysis model is trained using the second training sample set to obtain a trained acetylene characteristic gas detection and analysis model.

[0024] Optionally, the final judgment result of the partial discharge state of the transformer is determined based on the primary judgment result and the secondary judgment result, including:

[0025] If the secondary judgment result is that the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold, then the secondary judgment result is determined to be that the current partial discharge data exceeds the preset partial discharge threshold; if the secondary judgment result is that the acetylene characteristic gas content data does not exceed the preset acetylene characteristic gas content threshold, then the secondary judgment result is determined to be that the current partial discharge data does not exceed the preset partial discharge threshold;

[0026] When both the primary judgment result and the secondary judgment result are that the current partial discharge data exceeds the preset partial discharge threshold, the final judgment result of determining the partial discharge state of the transformer is that partial discharge occurs in the transformer;

[0027] When both the first judgment result and the second judgment result are that the current partial discharge data does not exceed the preset partial discharge threshold, the final judgment result of determining the partial discharge state of the transformer is that no partial discharge occurs in the transformer;

[0028] When the primary judgment result and the secondary judgment result are inconsistent, the partial discharge detection process is repeated until the primary judgment result and the secondary judgment result are consistent; wherein,

[0029] The PD detection process includes:

[0030] Re-acquire the ultrasonic signal generated by partial discharge of the transformer and the data of dissolved gas content in the transformer oil;

[0031] Based on the re-acquired ultrasonic signal generated when the transformer is partially discharged, the judgment result is updated once;

[0032] Based on the re-acquired data of dissolved gas content in transformer oil, the secondary judgment result is updated.

[0033] Optionally, the above partial discharge detection process also includes:

[0034] Initialize the execution times of the partial discharge detection process to zero;

[0035] After executing a partial discharge detection process, the number of executions of the partial discharge detection process is increased by one;

[0036] When the number of executions of the inspection process exceeds the preset threshold, a maintenance alarm signal is issued.

[0037] A second aspect of the present invention provides a transformer partial discharge fault diagnosis system, comprising:

[0038] A data acquisition module is used to acquire the ultrasonic signal generated by partial discharge of the transformer and the data of dissolved gas content in the transformer oil;

[0039] The ultrasonic detection and analysis module is used to input the ultrasonic signal generated during the partial discharge of the transformer into the trained partial discharge ultrasonic detection and analysis model to obtain the current partial discharge data, and determine whether the current partial discharge data exceeds the preset partial discharge threshold value to obtain a judgment result;

[0040] The acetylene characteristic gas detection and analysis module is used to input the dissolved gas content data in the transformer oil into the trained acetylene characteristic gas detection and analysis model to obtain the acetylene characteristic gas content data, and judge whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold value to obtain a secondary judgment result;

[0041] The transformer partial discharge final judgment module is used to determine the final judgment result of the transformer partial discharge state based on the primary judgment result and the secondary judgment result.

[0042] In a third aspect of the present invention, a machine-readable storage medium is provided, on which instructions are stored. When the instructions are executed by a processor, the processor is configured to execute the transformer partial discharge fault diagnosis method.

[0043] In a fourth aspect of the present invention, an electronic device is provided, the electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the transformer partial discharge fault diagnosis method when executing the computer program.

[0044] Through the above technical scheme, a transformer partial discharge fault diagnosis method and system are provided. The ultrasonic signal generated during the partial discharge of the transformer is analyzed by the trained partial discharge ultrasonic detection and analysis model to determine the current partial discharge data, and judge whether the current partial discharge data exceeds the preset partial discharge threshold value, and obtain a primary judgment result. The acetylene characteristic gas detection and analysis model is trained to identify and process the dissolved gas content data in the transformer oil with the acetylene characteristic gas content as the target function to obtain the acetylene characteristic gas content data, thereby determining whether the current partial discharge data exceeds the preset partial discharge threshold value by judging whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold value, and obtaining a secondary judgment result, thereby combining the primary judgment result and the secondary judgment result to determine the final judgment result of the partial discharge state of the transformer. The method and system combine the ultrasonic partial discharge detection process with the acetylene characteristic gas content detection process, and simultaneously capture the ultrasonic signal generated by the partial discharge of the transformer and the acetylene characteristic gas content data in the transformer oil, and respectively use the trained partial discharge ultrasonic detection and analysis model and the trained acetylene characteristic gas detection and analysis model to intelligently analyze the ultrasonic signal generated during the partial discharge of the transformer and the dissolved gas content data in the transformer oil. By analyzing the ultrasonic signal and the acetylene content to determine the location and intensity of partial discharge, the sensitivity and accuracy of partial discharge detection can be greatly improved, thereby improving the accuracy, sensitivity and anti-interference ability of real-time detection and fault diagnosis of partial discharge of the transformer, achieving the purpose of automated and precise detection of partial discharge of the transformer, and also achieving the purpose of intelligent analysis of transformer discharge faults, thus improving the guarantee for reliable operation of the transformer.

[0045] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0047] Figure 1 It is a flow chart of a transformer partial discharge fault diagnosis method provided by one embodiment of the present invention;

[0048] Figure 2 It is a structural schematic diagram of a transformer partial discharge fault diagnosis device provided by an embodiment of the present invention;

[0049] Figure 3 It is a flow chart of another transformer partial discharge fault diagnosis method provided by an embodiment of the present invention;

[0050] Figure 4 It is a block diagram of a transformer partial discharge fault diagnosis system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0051] The specific implementation of the embodiment of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the embodiment of the present invention, and is not used to limit the embodiment of the present invention.

[0052] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0053] The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with embodiments.

[0054] Figure 1 FIG. 1 is a flow chart of a transformer partial discharge fault diagnosis method provided by an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a transformer partial discharge fault diagnosis method, comprising:

[0055] S110: Acquire the ultrasonic signal generated during partial discharge of the transformer and the dissolved gas content data in the transformer oil;

[0056] Figure 2 FIG. 1 is a schematic diagram of a transformer partial discharge fault diagnosis device provided by an embodiment of the present invention. Figure 2 As shown, a transformer partial discharge fault diagnosis device includes a transformer partial discharge ultrasonic detection system (i.e. Figure 2 Ultrasonic detection system in), acetylene characteristic gas detection system (i.e. Figure 2 TDLAS detection system in the backend server diagnostic system (i.e. Figure 2 The transformer partial discharge ultrasonic detection system consists of an ultrasonic sensor, a signal acquisition and processing unit, an A / D converter, and a data acquisition control module. The ultrasonic sensor is directly arranged on the wall of the transformer box, and the ultrasonic sensor is closely attached to the wall of the transformer box through an acoustic coupling agent.

[0057] In some implementations of the present embodiment, the acquisition rules of the ultrasonic signal generated by the partial discharge of the transformer include: controlling the ultrasonic sensor to collect the ultrasonic signal generated by the partial discharge of the transformer, and preprocessing the ultrasonic signal generated by the partial discharge of the transformer; wherein the preprocessing process includes: converting the ultrasonic signal generated by the partial discharge of the transformer into a corresponding electrical signal, and amplifying and filtering the electrical signal in sequence.

[0058] Specifically, the method captures the ultrasonic signal generated by the partial discharge of the transformer by an ultrasonic sensor installed on the outer wall of the transformer oil tank, thereby realizing the detection of partial discharge of the transformer. The method uses the ultrasonic signal generated by the partial discharge of the transformer to determine whether there is a partial discharge phenomenon of the transformer. The ultrasonic signal generated by the partial discharge of the transformer is received by the ultrasonic sensor and converted into an electrical signal. After pre-processing steps such as amplification and filtering, the intensity, location and discharge type of the partial discharge of the transformer are determined according to the intensity, propagation time and frequency characteristics of the ultrasonic signal. The method is less affected by electrical interference, has accurate positioning and is easy to operate, and is suitable for on-site testing.

[0059] In some implementations of the present embodiment, the acquisition rules of the above-mentioned dissolved gas content data in the transformer oil include: controlling the TDLAS system to detect the characteristic absorption spectrum line data of the dissolved gas in the transformer oil, and converting the characteristic absorption spectrum line data of the dissolved gas in the transformer oil to obtain the dissolved gas content data in the transformer oil.

[0060] Specifically, the acetylene characteristic gas detection system (i.e. Figure 2 The TDLAS detection system in the transformer oil detection system adopts the tunable diode laser absorption spectroscopy detection (TDLAS) technology. The acetylene characteristic gas detection system includes a laser controller, a laser, a hollow fiber gas chamber, a photodetector and a signal processing module; the signal processing module includes a preamplifier, a phase-locked amplifier, an A / D converter, an I / O converter and other devices for converting and sorting the detection signal (i.e., the characteristic absorption spectrum data of the dissolved gas in the transformer oil). The narrow line width and tunable characteristics of the laser are used to scan the characteristic absorption spectrum of all components of the gas, thereby realizing the qualitative and quantitative measurement of the gas. This method can determine whether there is partial discharge inside the transformer and the severity of the discharge by collecting and analyzing the dissolved gas content data in the transformer oil (it should be noted that in the process of partial discharge, common generated gases include hydrogen (H2), methane (CH4), ethylene (C2H4), acetylene (C2H2), etc., and the content of specific gases (such as acetylene, ethylene, hydrogen, etc.) can be analyzed to diagnose whether there is partial discharge inside the transformer). This method is sensitive to the discovery of early latent faults.

[0061] S120: Inputting the ultrasonic signal generated when the transformer is partially discharged into the trained partial discharge ultrasonic detection and analysis model to obtain current partial discharge data, and determining whether the current partial discharge data exceeds a preset partial discharge threshold to obtain a determination result;

[0062] Specifically, the ultrasonic signal generated during partial discharge of the transformer is analyzed through the trained partial discharge ultrasonic detection and analysis model to determine the current partial discharge data, and judge whether the current partial discharge data exceeds the preset partial discharge threshold, and obtain a judgment result. The judgment result represents the result of judging whether the current partial discharge data exceeds the preset partial discharge threshold based on the ultrasonic signal generated during partial discharge of the transformer.

[0063] S130: Inputting the dissolved gas content data in the transformer oil into the trained acetylene characteristic gas detection and analysis model to obtain the acetylene characteristic gas content data, and judging whether the acetylene characteristic gas content data exceeds a preset acetylene characteristic gas content threshold value to obtain a secondary judgment result;

[0064] Specifically, the trained acetylene characteristic gas detection and analysis model uses the acetylene characteristic gas content as the target function to identify and process the dissolved gas content data in the transformer oil to obtain the acetylene characteristic gas content data, thereby determining whether the current partial discharge data exceeds the preset partial discharge threshold by judging whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold, and obtaining a secondary judgment result, which represents the result of judging whether the current partial discharge data exceeds the preset partial discharge threshold based on the acetylene characteristic gas content data. Among them, acetylene is a characteristic gas of high-energy discharge, and the change of acetylene characteristic gas content can effectively reflect the internal discharge fault of the transformer.

[0065] S140: Determine a final judgment result of the partial discharge state of the transformer based on the primary judgment result and the secondary judgment result.

[0066] Specifically, the method analyzes the ultrasonic signal generated during the partial discharge of the transformer through the trained partial discharge ultrasonic detection and analysis model to determine the current partial discharge data, and judge whether the current partial discharge data exceeds the preset partial discharge threshold, and obtains a primary judgment result. The trained acetylene characteristic gas detection and analysis model uses the acetylene characteristic gas content as the target function to identify and process the dissolved gas content data in the transformer oil to obtain the acetylene characteristic gas content data, thereby determining whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold, and determining whether the current partial discharge data exceeds the preset partial discharge threshold, and obtaining a secondary judgment result, thereby combining the primary judgment result and the secondary judgment result to determine the final judgment result of the partial discharge state of the transformer. The method combines the ultrasonic partial discharge detection process with the acetylene characteristic gas content detection process, and simultaneously captures the ultrasonic signal generated by the partial discharge of the transformer and the acetylene characteristic gas content data in the transformer oil, and respectively uses the trained partial discharge ultrasonic detection and analysis model and the trained acetylene characteristic gas detection and analysis model to intelligently analyze the ultrasonic signal generated during the partial discharge of the transformer and the dissolved gas content data in the transformer oil. By analyzing the ultrasonic signal and the acetylene content to determine the location and intensity of partial discharge, the sensitivity and accuracy of partial discharge detection can be greatly improved, thereby improving the accuracy, sensitivity and anti-interference ability of real-time detection and fault diagnosis of partial discharge of the transformer, achieving the purpose of automated and precise detection of partial discharge of the transformer, and also achieving the purpose of intelligent analysis of transformer discharge faults, thus improving the guarantee for reliable operation of the transformer.

[0067] In the above implementation process, the method uses tunable diode laser absorption spectroscopy technology to assist in judging the partial discharge problem of the transformer. Without affecting the real-time detection of the ultrasonic detection method, it solves the problems of poor anti-interference ability and low sensitivity of the ultrasonic detection method during application. By combining the ultrasonic partial discharge detection process with the acetylene characteristic gas content detection process, the problems of missed judgment and wrong judgment of the internal discharge fault of the transformer are avoided, and the accuracy of the partial discharge test of the transformer is improved.

[0068] In some implementations of the present embodiment, the training rules of the above-mentioned partial discharge ultrasonic detection and analysis model include: establishing a partial discharge ultrasonic detection and analysis model based on a recursive neural network algorithm; obtaining historical ultrasonic signals generated by multiple historical transformer partial discharges and historical partial discharge data corresponding to each historical ultrasonic signal as a first training sample set; wherein the historical partial discharge data includes a discharge type, a discharge position, and a discharge displacement; and using the first training sample set to train the partial discharge ultrasonic detection and analysis model to obtain a trained partial discharge ultrasonic detection and analysis model.

[0069] Specifically, the method constructs a partial discharge ultrasonic detection and analysis model based on a recursive neural network algorithm, and uses historical ultrasonic signals generated during multiple historical transformer partial discharges and historical partial discharge data corresponding to each historical ultrasonic signal to train the partial discharge ultrasonic detection and analysis model. The trained partial discharge ultrasonic detection and analysis model can analyze the ultrasonic signals generated during transformer partial discharge, which can greatly improve the accuracy, sensitivity and anti-interference ability of real-time detection of transformer partial discharge.

[0070] In some implementations of this embodiment, the training rules of the acetylene characteristic gas detection and analysis model include: establishing an acetylene characteristic gas detection and analysis model based on a recursive neural network algorithm; obtaining multiple historical transformer oil dissolved gas content data and historical acetylene characteristic gas content data corresponding to each historical transformer oil dissolved gas content data as a second training sample set; using the second training sample set to train the acetylene characteristic gas detection and analysis model to obtain a trained acetylene characteristic gas detection and analysis model. Thus, the collected transformer oil dissolved gas content data is intelligently analyzed using a recursive neural network algorithm, which can greatly improve the accuracy, sensitivity and anti-interference ability of real-time detection of transformer partial discharge.

[0071] In the above implementation process, the method uses a recursive neural network algorithm to extract characteristic parameters (i.e., historical partial discharge data corresponding to each historical ultrasonic signal and historical acetylene characteristic gas content data corresponding to each historical transformer oil dissolved gas content data) from the collected and processed information (i.e., historical ultrasonic signals generated during multiple historical transformer partial discharges and historical transformer oil dissolved gas content data), establishes a detection and analysis model for partial discharge sample data (i.e., a trained partial discharge ultrasonic detection and analysis model and a trained acetylene characteristic gas detection and analysis model), and performs feature classification processing on the extracted characteristic parameters (i.e., current partial discharge data and acetylene characteristic gas content data), thereby realizing classification recognition and automatic detection of transformer partial discharge.

[0072] Among them, in the partial discharge ultrasonic detection and analysis model: the recursive neural network algorithm uses the local discharge characteristics to automatically identify the discharge signal, judge the discharge type, automatically determine the discharge position, discharge displacement and other parameters, and can use three-dimensional space to mark the discharge point.

[0073] In addition, in the acetylene characteristic gas detection and analysis model: the tunable diode laser absorption spectroscopy detection system (TDLAS system) detects all component gases in transformer oil (standard C2H2, CH4, C2H4, C2H6, CO, CO2, H2, etc.), and the recursive neural network algorithm uses acetylene (C2H2) characteristic gas as the target function to process and identify the collected data, and finally determines whether there is a partial discharge fault.

[0074] Please refer to Figure 3 , Figure 3 It is a flow chart of another transformer partial discharge fault diagnosis method provided by an embodiment of the present invention. In some embodiments of the present embodiment, the above-mentioned determination of the final judgment result of the transformer partial discharge state based on the primary judgment result and the secondary judgment result includes: if the secondary judgment result is that the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold, then the secondary judgment result is determined to be that the current partial discharge data exceeds the preset partial discharge threshold; if the secondary judgment result is that the acetylene characteristic gas content data does not exceed the preset acetylene characteristic gas content threshold, then the secondary judgment result is determined to be that the current partial discharge data does not exceed the preset partial discharge threshold; when both the primary judgment result and the secondary judgment result are that the current partial discharge data exceeds the preset partial discharge threshold, the final judgment result of the transformer partial discharge state is determined to be Partial discharge occurs in the transformer; when both the primary judgment result and the secondary judgment result are that the current partial discharge data does not exceed the preset partial discharge threshold, the final judgment result of determining the partial discharge state of the transformer is that no partial discharge occurs in the transformer; when the primary judgment result and the secondary judgment result are inconsistent, the partial discharge detection process is repeated until the primary judgment result and the secondary judgment result are consistent; wherein the partial discharge detection process includes: reacquiring the ultrasonic signal generated when the transformer is partially discharged and the data on the dissolved gas content in the transformer oil; updating the primary judgment result based on the reacquired ultrasonic signal generated when the transformer is partially discharged; and updating the secondary judgment result based on the reacquired data on the dissolved gas content in the transformer oil.

[0075] In some implementations of the present embodiment, the above-mentioned partial discharge detection process also includes: initializing the number of executions of the partial discharge detection process to zero; after executing the partial discharge detection process once, increasing the number of executions of the partial discharge detection process by one; when the number of executions of the partial discharge detection process exceeds a preset number threshold, issuing a maintenance alarm signal.

[0076] Specifically, the method combines the ultrasonic partial discharge detection process with the tunable diode laser absorption spectroscopy detection (TDLAS) technology to detect the partial discharge of the transformer. The specific process is as follows: the background server collects the data of the transformer partial discharge ultrasonic detection system and the TDLAS detection system, extracts characteristic parameters (i.e., current partial discharge data and acetylene characteristic gas content data) from the collected information (i.e., the ultrasonic signal generated when the transformer is partially discharged and the dissolved gas content data in the transformer oil) through the recursive neural network algorithm, and judges whether it exceeds the threshold of the possibility of partial discharge according to the characteristic parameters. After judging the possibility of partial discharge of the transformer based on the ultrasonic signal generated when the transformer is partially discharged (i.e., judging the possibility of partial discharge of the transformer by the ultrasonic detection system), a secondary judgment is made in combination with the acetylene (C2H2) concentration change trend collected by the TDLAS system (i.e., the TDLAS detection system) to further eliminate the possibility of signal interference. If the ultrasonic detection system and the TDLAS detection system simultaneously judge that there is a partial discharge, it is considered that partial discharge occurs, and the discharge position and discharge intensity are located according to the detection results of the ultrasonic detection system and the TDLAS system. If both the ultrasonic detection system and the TDLAS detection system fail to detect partial discharge, it is considered that the transformer has no partial discharge. In addition, the method also provides a self-test function. When the detection results of the ultrasonic detection system and the TDLAS detection system are inconsistent, the self-test process is entered, the partial discharge detection process is repeated and the counter is started; each time the detection is repeated, the counter value increases accordingly. When the preset number threshold is exceeded, the system maintenance alarm is triggered to remind the maintenance personnel to perform maintenance and calibration on the system in a timely manner.

[0077] Figure 4 FIG. 1 is a block diagram of a transformer partial discharge fault diagnosis system provided by an embodiment of the present invention. Figure 4 As shown, an embodiment of the present invention provides a transformer partial discharge fault diagnosis system, comprising:

[0078] A data acquisition module is used to acquire the ultrasonic signal generated by partial discharge of the transformer and the data of dissolved gas content in the transformer oil;

[0079] The ultrasonic detection and analysis module is used to input the ultrasonic signal generated during the partial discharge of the transformer into the trained partial discharge ultrasonic detection and analysis model to obtain the current partial discharge data, and determine whether the current partial discharge data exceeds the preset partial discharge threshold value to obtain a judgment result;

[0080] The acetylene characteristic gas detection and analysis module is used to input the dissolved gas content data in the transformer oil into the trained acetylene characteristic gas detection and analysis model to obtain the acetylene characteristic gas content data, and judge whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold value to obtain a secondary judgment result;

[0081] The transformer partial discharge final judgment module is used to determine the final judgment result of the transformer partial discharge state based on the primary judgment result and the secondary judgment result.

[0082] Specifically, the system analyzes the ultrasonic signal generated during the partial discharge of the transformer through the trained partial discharge ultrasonic detection and analysis model to determine the current partial discharge data, and judge whether the current partial discharge data exceeds the preset partial discharge threshold, and obtain a primary judgment result. The acetylene characteristic gas detection and analysis model that has been trained uses the acetylene characteristic gas content as the target function to identify and process the dissolved gas content data in the transformer oil to obtain the acetylene characteristic gas content data, thereby determining whether the current partial discharge data exceeds the preset partial discharge threshold by judging whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold, and obtaining a secondary judgment result, thereby combining the primary judgment result and the secondary judgment result to determine the final judgment result of the partial discharge state of the transformer. The system combines the ultrasonic partial discharge detection process with the acetylene characteristic gas content detection process, and simultaneously captures the ultrasonic signal generated by the partial discharge of the transformer and the acetylene characteristic gas content data in the transformer oil, and respectively uses the trained partial discharge ultrasonic detection and analysis model and the trained acetylene characteristic gas detection and analysis model to intelligently analyze the ultrasonic signal generated during the partial discharge of the transformer and the dissolved gas content data in the transformer oil. By analyzing the ultrasonic signal and the acetylene content to determine the location and intensity of partial discharge, the sensitivity and accuracy of partial discharge detection can be greatly improved, thereby improving the accuracy, sensitivity and anti-interference ability of real-time detection and fault diagnosis of partial discharge of the transformer, achieving the purpose of automated and precise detection of partial discharge of the transformer, and also achieving the purpose of intelligent analysis of transformer discharge faults, thus improving the guarantee for reliable operation of the transformer.

[0083] The optional implementation modes of the embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above implementation modes. Within the technical concept of the embodiments of the present invention, various simple modifications can be made to the technical scheme of the embodiments of the present invention, and these simple modifications all belong to the protection scope of the embodiments of the present invention.

[0084] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. To avoid unnecessary repetition, the embodiments of the present invention will not further describe various possible combinations.

[0085] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including a number of instructions to enable a single-chip microcomputer, a chip or a processor to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0086] In addition, various implementation modes of the embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed by the embodiments of the present invention.

Claims

1. A method for diagnosing transformer partial discharge faults, It is characterized in that include: Obtain the ultrasonic signal generated by partial discharge of the transformer and the dissolved gas content data in the transformer oil; The ultrasonic signal generated during partial discharge of the transformer is input into the trained partial discharge ultrasonic detection and analysis model to obtain the current partial discharge data, and determine whether the current partial discharge data exceeds the preset partial discharge threshold value to obtain a judgment result; The dissolved gas content data in the transformer oil is input into the trained acetylene characteristic gas detection and analysis model to obtain the acetylene characteristic gas content data, and it is determined whether the acetylene characteristic gas content data exceeds a preset acetylene characteristic gas content threshold value to obtain a secondary judgment result; Based on the primary judgment result and the secondary judgment result, a final judgment result of the partial discharge state of the transformer is determined.

2. The transformer partial discharge fault diagnosis method according to claim 1, It is characterized in that The acquisition rules of the ultrasonic signal generated by the partial discharge of the transformer include: The ultrasonic sensor is controlled to collect the ultrasonic signal generated when the transformer is partially discharged, and the ultrasonic signal generated when the transformer is partially discharged is preprocessed; wherein, The pre-treatment process includes: The ultrasonic signal generated by partial discharge of the transformer is converted into a corresponding electrical signal, and the electrical signal is amplified and filtered in turn.

3. The transformer partial discharge fault diagnosis method according to claim 1, It is characterized in that The acquisition rules of the dissolved gas content data in the transformer oil include: The TDLAS system is controlled to detect the characteristic absorption spectral line data of the dissolved gas in the transformer oil, and the characteristic absorption spectral line data of the dissolved gas in the transformer oil are converted to obtain the dissolved gas content data in the transformer oil.

4. The transformer partial discharge fault diagnosis method according to claim 1, It is characterized in that The training rules of the partial discharge ultrasonic detection and analysis model include: Based on the recursive neural network algorithm, a partial discharge ultrasonic detection and analysis model is established; Acquire multiple historical ultrasonic signals generated during partial discharge of transformers and historical partial discharge data corresponding to each historical ultrasonic signal as a first training sample set; wherein the historical partial discharge data includes discharge type, discharge position and discharge displacement; The partial discharge ultrasonic detection and analysis model is trained using the first training sample set to obtain a trained partial discharge ultrasonic detection and analysis model.

5. The transformer partial discharge fault diagnosis method according to claim 1, It is characterized in that The training rules of the acetylene characteristic gas detection and analysis model include: Based on the recursive neural network algorithm, an acetylene characteristic gas detection and analysis model is established; Acquire a plurality of historical transformer oil dissolved gas content data and historical acetylene characteristic gas content data corresponding to each of the historical transformer oil dissolved gas content data as a second training sample set; The acetylene characteristic gas detection and analysis model is trained using the second training sample set to obtain a trained acetylene characteristic gas detection and analysis model.

6. The transformer partial discharge fault diagnosis method according to claim 1, It is characterized in that The method of determining the final judgment result of the partial discharge state of the transformer based on the primary judgment result and the secondary judgment result includes: If the secondary judgment result is that the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold, then the secondary judgment result is determined to be that the current partial discharge data exceeds the preset partial discharge threshold; if the secondary judgment result is that the acetylene characteristic gas content data does not exceed the preset acetylene characteristic gas content threshold, then the secondary judgment result is determined to be that the current partial discharge data does not exceed the preset partial discharge threshold; When both the primary judgment result and the secondary judgment result are that the current partial discharge data exceeds the preset partial discharge threshold, the final judgment result of determining the partial discharge state of the transformer is that partial discharge occurs in the transformer; When both the first judgment result and the second judgment result are that the current partial discharge data does not exceed the preset partial discharge threshold, the final judgment result of determining the partial discharge state of the transformer is that no partial discharge occurs in the transformer; When the primary judgment result and the secondary judgment result are inconsistent, the partial discharge detection process is repeated until the primary judgment result and the secondary judgment result are consistent; wherein, The partial discharge detection process includes: Re-acquire the ultrasonic signal generated by partial discharge of the transformer and the data of dissolved gas content in the transformer oil; Based on the re-acquired ultrasonic signal generated when the transformer is partially discharged, the judgment result is updated once; Based on the re-acquired data of dissolved gas content in transformer oil, the secondary judgment result is updated.

7. The transformer partial discharge fault diagnosis method according to claim 6, It is characterized in that The partial discharge detection process also includes: Initialize the execution times of the partial discharge detection process to zero; After executing a partial discharge detection process, the number of executions of the partial discharge detection process is increased by one; When the number of executions of the inspection process exceeds the preset threshold, a maintenance alarm signal is issued.

8. A transformer partial discharge fault diagnosis system, It is characterized in that include: A data acquisition module is used to acquire the ultrasonic signal generated by partial discharge of the transformer and the data of dissolved gas content in the transformer oil; The ultrasonic detection and analysis module is used to input the ultrasonic signal generated during the partial discharge of the transformer into the trained partial discharge ultrasonic detection and analysis model to obtain the current partial discharge data, and determine whether the current partial discharge data exceeds the preset partial discharge threshold value to obtain a judgment result; The acetylene characteristic gas detection and analysis module is used to input the dissolved gas content data in the transformer oil into the trained acetylene characteristic gas detection and analysis model to obtain the acetylene characteristic gas content data, and judge whether the acetylene characteristic gas content data exceeds the preset acetylene characteristic gas content threshold value to obtain a secondary judgment result; The transformer partial discharge final judgment module is used to determine the final judgment result of the transformer partial discharge state based on the primary judgment result and the secondary judgment result.

9. A machine-readable storage medium having instructions stored thereon, It is characterized in that When the instruction is executed by a processor, the processor is configured to execute the transformer partial discharge fault diagnosis method as claimed in any one of claims 1 to 7.

10. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, It is characterized in that When the processor executes the computer program, the transformer partial discharge fault diagnosis method according to any one of claims 1 to 7 is implemented.

Images