Partial discharge signal detection method, device and equipment and storage medium
By combining ultra-high frequency signals, ultrasonic signals, and current pulse signals, and using a deep learning model for partial discharge signal detection, the problem of high false alarm rate is solved, and higher detection accuracy is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-18
- Publication Date
- 2026-03-20
AI Technical Summary
Existing technologies have a high false alarm rate in partial discharge detection, and using only one detection method can easily lead to false alarms in complex field environments.
Partial discharge signal detection is performed using ultra-high frequency (UHF) signals, ultrasonic signals, and current pulse signals. The UHF signals are converted into time-frequency diagrams, and a joint time-domain waveform of ultrasonic signals and current pulse signals is constructed and input into a deep learning model for signal detection.
It reduced the false alarm rate of partial discharge signals and improved the accuracy of detection results.
Smart Images
Figure CN121703586A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of partial discharge detection technology, and in particular to a method, apparatus, device and storage medium for detecting partial discharge signals. Background Technology
[0002] Partial discharge is a discharge that occurs within a localized area of the insulation of electrical equipment under the influence of a sufficiently strong electric field. It is the cause of insulation breakdown in electrical equipment, and partial discharge detection is of great significance for ensuring the safe operation of electrical equipment. Current technologies generally employ only one detection method for partial discharge detection. However, the on-site environment for partial discharge detection is quite complex, and using only one detection method can easily lead to false alarms. Summary of the Invention
[0003] The purpose of this invention is to provide a method, apparatus, device, and storage medium for detecting partial discharge signals, which can reduce the false alarm rate of partial discharge signal detection.
[0004] To achieve the above objectives, embodiments of the present invention provide a method for detecting partial discharge signals, comprising: Input ultra-high frequency signals, ultrasonic signals, and current pulse signals; Convert the ultra-high frequency signal into a time-frequency diagram; Construct the joint time-domain waveform of the ultrasonic signal and the current pulse signal; The time-frequency graph and the joint time-domain waveform are input into a deep learning model to obtain the signal detection result.
[0005] As an improvement to the above scheme, the ultra-high frequency signal is acquired by an ultra-high frequency sensor, the ultrasonic signal is acquired by an optical fiber ultrasonic sensor, and the current pulse signal is acquired by a high frequency current transformer.
[0006] As an improvement to the above scheme, before converting the ultra-high frequency signal into a time-frequency diagram, the method further includes: Wavelet decomposition is performed on the UHF signal to extract the UHF signal in the target frequency band.
[0007] As an improvement to the above solution, the step of converting the ultra-high frequency signal into a time-frequency diagram includes: The ultra-high frequency signal is subjected to a short-time Fourier transform to obtain a time-frequency diagram.
[0008] As an improvement to the above scheme, before constructing the joint time-domain waveform of the ultrasonic signal and the current pulse signal, the method further includes: Input the vibration signal of the transformer core; A compensation signal is generated based on the vibration signal; The ultrasonic signal is compensated using the compensation signal.
[0009] As an improvement to the above scheme, before constructing the joint time-domain waveform of the ultrasonic signal and the current pulse signal, the method further includes preprocessing the ultrasonic signal using a blind source separation algorithm.
[0010] As an improvement to the above scheme, the deep learning model is a dual-channel convolutional neural network.
[0011] To achieve the above objectives, embodiments of the present invention also provide a partial discharge signal detection device, comprising: The signal input module is used to input ultra-high frequency signals, ultrasonic signals, and current pulse signals. The time-frequency conversion module is used to convert the ultra-high frequency signal into a time-frequency diagram; A waveform construction module is used to construct the joint time-domain waveform of the ultrasonic signal and the current pulse signal; The detection module is used to input the time-frequency diagram and the joint time-domain waveform into a deep learning model to obtain signal detection results.
[0012] To achieve the above objectives, embodiments of the present invention also provide a partial discharge signal detection device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the partial discharge signal detection method as described in any of the above embodiments.
[0013] To achieve the above objectives, embodiments of the present invention also provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the partial discharge signal detection method as described in any of the above embodiments.
[0014] Compared with existing technologies, the partial discharge signal detection method, apparatus, device, and storage medium of this invention input ultra-high frequency signals, ultrasonic signals, and current pulse signals; convert the ultra-high frequency signals into a time-frequency diagram; construct a joint time-domain waveform of the ultrasonic signals and the current pulse signals; and input the time-frequency diagram and the joint time-domain waveform into a deep learning model to obtain the signal detection result. This invention, by fusing multi-source data for partial discharge signal detection, can reduce the false alarm rate of partial discharge signals and improve the accuracy of the detection results. Attached Figure Description
[0015] Figure 1 This is a flowchart of a partial discharge signal detection method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a partial discharge signal detection device according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a partial discharge signal detection device provided in an embodiment of the present invention. Detailed Implementation
[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0017] In the description of this application, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this application.
[0018] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.
[0019] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.
[0020] See Figure 1 This is a flowchart of a partial discharge signal detection method provided in an embodiment of the present invention, including: S1. Input ultra-high frequency signals, ultrasonic signals, and current pulse signals; S2. Convert the ultra-high frequency signal into a time-frequency diagram; S3. Construct the joint time-domain waveform of the ultrasonic signal and the current pulse signal; S4. Input the time-frequency diagram and the joint time-domain waveform into the deep learning model to obtain the signal detection result.
[0021] Specifically, in this invention, multi-source sensors are used to collect multi-source data for partial discharge detection. For example, in step S1, an ultra-high frequency (UHF) sensor can be used to collect UHF signals, an optical fiber ultrasonic sensor can be used to collect ultrasonic signals, and a high-frequency current transformer can be used to collect current pulse signals. For example, when a suspected discharge pulse is detected, for example, when the amplitude is >5mV (millivolt), dynamic filtering can be activated to shield 5GHz (3.5GHz / 4.9GHz) noise. Furthermore, the signals collected by the aforementioned sensors can be preprocessed to remove noise interference. Furthermore, the UHF signal is converted into a time-frequency diagram, a joint time-domain waveform is constructed based on the ultrasonic signal and the current pulse signal, and these signals are used as two input signals for a deep learning model. For example, the deep learning model can be pre-trained using a dataset. Furthermore, in step S4, the deep learning model generates signal detection results based on the two input data, wherein the signal detection results are "real discharge," "mechanical noise," and "external pulse interference." In some exemplary embodiments, a transformer discharge signal propagation attenuation model can also be established to further identify the authenticity of the partial discharge signal after a "real discharge" is detected.
[0022] Compared with the prior art, the embodiments of the present invention can reduce the false alarm rate of partial discharge signals and improve the accuracy of partial discharge signal detection results by fusing ultra-high frequency signals, ultrasonic signals and current pulse signals for partial discharge signal detection.
[0023] As one optional implementation, the ultra-high frequency signal is acquired by an ultra-high frequency sensor, the ultrasonic signal is acquired by an optical fiber ultrasonic sensor, and the current pulse signal is acquired by a high frequency current transformer.
[0024] For example, the embodiments of the present invention can be used for transformer partial discharge signal detection, such as a 500kV oil-immersed transformer, and can also be used for other equipment, which are not limited here. For ease of understanding, the following explanation uses transformer partial discharge detection as an example.
[0025] For example, an ultra-high frequency sensor is used to capture electromagnetic waves generated by discharge in transformer oil. It can be embedded in the flange of the transformer tank and its operating frequency can be set to 300MHz (megahertz) to 3GHz (gigahertz). A fiber optic ultrasonic sensor can adopt a Fabry-Perot interference structure and be attached to the root of the high-voltage bushing. Its operating frequency can be set to 40kHz (kilohertz) to 300kHz. A high-frequency current transformer is used to monitor the current pulse of the neutral point grounding wire and can clamp the neutral point grounding wire. Its operating frequency can be set to 1MHz to 30MHz.
[0026] For example, in some implementations, the input signal is preprocessed before step S2.
[0027] As one optional implementation, before converting the ultra-high frequency signal into a time-frequency diagram, the method further includes: Wavelet decomposition is performed on the UHF signal to extract the UHF signal in the target frequency band.
[0028] Specifically, wavelet decomposition is performed on the input UHF signal to extract the UHF signal in the target frequency band. For example, the characteristic frequency band of discharge in transformer oil is 150MHz~1.5GHz, therefore, the target frequency band is 150MHz~1.5GHz.
[0029] As one optional implementation, in step S2, converting the ultra-high frequency signal into a time-frequency diagram includes: The ultra-high frequency signal is subjected to a short-time Fourier transform (STFT) to obtain a time-frequency diagram.
[0030] As one optional implementation, before constructing the joint time-domain waveform of the ultrasonic signal and the current pulse signal, the method further includes: Input the vibration signal of the transformer core; A compensation signal is generated based on the vibration signal; The ultrasonic signal is compensated using the compensation signal.
[0031] It is worth noting that the vibration signal of the transformer core can cause mechanical noise coupling. Therefore, in order to improve the accuracy of the detection results, this embodiment of the invention also generates a compensation signal based on the vibration signal of the transformer core itself, so as to remove the coupling effect of mechanical noise from the acquired ultrasonic signal. For example, the vibration signal of the transformer core can be acquired using an accelerometer.
[0032] As one optional implementation, before constructing the joint time-domain waveform of the ultrasonic signal and the current pulse signal, the ultrasonic signal is further preprocessed using a blind source separation algorithm.
[0033] For example, in some implementations, a blind source separation algorithm can also be used to separate the vibration noise of the iron core from the ultrasonic signal. For example, the blind source separation algorithm can be the FastICA (Fast Independent Component Analysis) algorithm.
[0034] As one alternative implementation, the deep learning model is a two-channel convolutional neural network.
[0035] For example, in some implementations, to further improve the accuracy of partial discharge signal detection, PRPD (Phase-Resolved Partial Discharge) can be used to determine the authenticity of the partial discharge signal. For example, for transformer partial discharge signal detection, when the confidence level of the "real discharge" classification result output by the deep learning model is greater than 90% and the PRPD phase distribution conforms to the internal discharge characteristics of the transformer (clustering at 1 / 2 and 3 / 2 of the power frequency cycle), the "real discharge" detection result is then output.
[0036] Compared with existing technologies, the partial discharge signal detection method of this invention involves inputting ultra-high frequency signals, ultrasonic signals, and current pulse signals; converting the ultra-high frequency signals into a time-frequency diagram; constructing a joint time-domain waveform of the ultrasonic signals and the current pulse signals; and inputting the time-frequency diagram and the joint time-domain waveform into a deep learning model to obtain the signal detection result. This invention, by fusing multiple data sources for partial discharge signal detection, can reduce the false alarm rate of partial discharge signals and improve the accuracy of the detection results.
[0037] See Figure 2 This invention also provides a partial discharge signal detection device 10, comprising: Signal input module 11 is used to input ultra-high frequency signals, ultrasonic signals and current pulse signals; Time-frequency conversion module 12 is used to convert the ultra-high frequency signal into a time-frequency diagram; Waveform construction module 13 is used to construct the joint time-domain waveform of the ultrasonic signal and the current pulse signal; The detection module 14 is used to input the time-frequency diagram and the joint time-domain waveform into the deep learning model to obtain the signal detection result.
[0038] The partial discharge signal detection device provided in this embodiment of the invention can realize all the process steps of the partial discharge signal detection method described in the above embodiments. The functions and technical effects of each module and unit in the device are the same as the functions and technical effects of the partial discharge signal detection method described in the above embodiments. The specific implementation method will not be described in detail here.
[0039] See Figure 3 This invention also provides a partial discharge signal detection device 20, including a processor 21, a memory 22, and a computer program stored in the memory 22 and configured to be executed by the processor 21. When the processor 21 executes the computer program, it implements the steps described in the partial discharge signal detection method embodiments above, for example... Figure 1The steps S1 to S4 described above; or, when the processor 21 executes the computer program, it implements the functions of each module in the above-described device embodiments.
[0040] The partial discharge signal detection device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The partial discharge signal detection device may include, but is not limited to, a processor and memory. Those skilled in the art will understand that the schematic diagram is merely an example of a partial discharge signal detection device and does not constitute a limitation on the device. It may include more or fewer components than illustrated, or combine certain components, or use different components. For example, the partial discharge signal detection device may also include input / output devices, network access devices, buses, etc.
[0041] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the partial discharge signal detection device, connecting all parts of the device via various interfaces and lines.
[0042] The memory can be used to store the computer program and / or modules. The processor implements various functions of the partial discharge signal detection device by running or executing the computer program and / or modules stored in the memory and by calling the data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function, etc.; the data storage area may store data created according to the use of the controller, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
[0043] If the module integrated into the partial discharge signal detection device is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.
[0044] Compared with existing technologies, the partial discharge signal detection device, equipment, and storage medium of this invention input ultra-high frequency signals, ultrasonic signals, and current pulse signals; convert the ultra-high frequency signals into a time-frequency diagram; construct a joint time-domain waveform of the ultrasonic signals and the current pulse signals; and input the time-frequency diagram and the joint time-domain waveform into a deep learning model to obtain the signal detection result. This invention, by fusing multiple data sources for partial discharge signal detection, can reduce the false alarm rate of partial discharge signals and improve the accuracy of the detection results.
[0045] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A method for detecting partial discharge signals, characterized in that, include: Input ultra-high frequency signals, ultrasonic signals, and current pulse signals; Convert the ultra-high frequency signal into a time-frequency diagram; Construct the joint time-domain waveform of the ultrasonic signal and the current pulse signal; The time-frequency graph and the joint time-domain waveform are input into a deep learning model to obtain the signal detection result.
2. The partial discharge signal detection method as described in claim 1, characterized in that, The ultra-high frequency signal is acquired by an ultra-high frequency sensor, the ultrasonic signal is acquired by an optical fiber ultrasonic sensor, and the current pulse signal is acquired by a high frequency current transformer.
3. The partial discharge signal detection method as described in claim 1, characterized in that, Before converting the ultra-high frequency signal into a time-frequency diagram, the method further includes: Wavelet decomposition is performed on the UHF signal to extract the UHF signal in the target frequency band.
4. The partial discharge signal detection method as described in claim 1, characterized in that, The step of converting the ultra-high frequency signal into a time-frequency diagram includes: The ultra-high frequency signal is subjected to a short-time Fourier transform to obtain a time-frequency diagram.
5. The partial discharge signal detection method as described in claim 1, characterized in that, Before constructing the joint time-domain waveform of the ultrasonic signal and the current pulse signal, the method further includes: Input the vibration signal of the transformer core; A compensation signal is generated based on the vibration signal; The ultrasonic signal is compensated using the compensation signal.
6. The partial discharge signal detection method as described in claim 1, characterized in that, Before constructing the joint time-domain waveform of the ultrasonic signal and the current pulse signal, the method further includes preprocessing the ultrasonic signal using a blind source separation algorithm.
7. The partial discharge signal detection method as described in claim 1, characterized in that, The deep learning model is a dual-channel convolutional neural network.
8. A partial discharge signal detection device, characterized in that, include: The signal input module is used to input ultra-high frequency signals, ultrasonic signals, and current pulse signals. The time-frequency conversion module is used to convert the ultra-high frequency signal into a time-frequency diagram; A waveform construction module is used to construct the joint time-domain waveform of the ultrasonic signal and the current pulse signal; The detection module is used to input the time-frequency diagram and the joint time-domain waveform into a deep learning model to obtain signal detection results.
9. A partial discharge signal detection device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the partial discharge signal detection method as described in any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the partial discharge signal detection method as described in any one of claims 1 to 7.