Partial discharge identification method and device of power equipment, computer readable storage medium and electronic equipment

Abstract

The invention discloses a partial discharge identification method and device of power equipment, a computer readable storage medium and electronic equipment. The method comprises the following steps: acquiring a partial discharge pulse signal of power equipment in a first preset time window; determining at least one statistical parameter of the partial discharge pulse signal based on the partial discharge pulse signal; determining a statistical significance index of the partial discharge pulse signal based on the at least one statistical parameter; on the basis of a comparison result of the statistical significance index and a statistical significance index threshold value, a target recognition model is called to detect the partial discharge pulse signal, a partial discharge recognition result of the power equipment is obtained, and the partial discharge recognition result at least comprises the discharge type of partial discharge of the power equipment. And the confidence coefficient of the partial discharge identification result is used for evaluating the accuracy degree of the partial discharge identification result. According to the invention, the technical problem of inaccurate partial discharge identification of power equipment is solved.

Description

Technical Field

[0001] The present invention relates to the field of electric power technology, and in particular to a method and device for identifying partial discharge of electric power equipment, a computer-readable storage medium and an electronic device. Background Art

[0002] At present, power equipment, especially high-voltage electrical equipment such as gas-insulated metal-enclosed switchgear (GIS), may have insulation defects during the design, manufacturing, transportation and installation process due to the influence of various uncontrollable factors. These defects manifest as partial discharge (PD) phenomenon during the operation of power equipment, which is an early signal of aging and potential failure of electrical equipment. Partial discharge not only accelerates the degradation of insulating materials, but may also cause more serious equipment failures, posing a threat to the safety and stability of the power system. Therefore, accurately and timely identifying partial discharge and determining whether it is associated with insulation defects is the key to preventing power equipment failures and ensuring the smooth operation of the power system.

[0003] In the related art, the identification of partial discharge mainly relies on two technical means: phase resolved partial discharge (PRPD) image analysis and phase resolved pulse sequence (PRPS) image analysis. Among them, the PRPD image can intuitively show the relationship between the discharge pulse and the voltage phase, while the PRPS image emphasizes the periodicity and repeatability of the discharge pulse. However, these identification methods are easily affected by the aliasing of the partial discharge signal and the external interference signal, resulting in a decrease in the recognition accuracy, showing the application effect. In addition, since the partial discharge caused by many initial defects presents intermittent and statistical characteristics, it is difficult for traditional recognition algorithms to make accurate judgments within a limited time or cycle. Moreover, the PRPD image analysis method and the PRPS image analysis method often require professionals to interpret the PRPD and PRPS images based on experience and professional knowledge, which is not only inefficient, but also easily affected by subjective judgment, and there is a technical problem of inaccurate partial discharge identification of power equipment.

[0004] With respect to the above-mentioned technical problem of inaccurate identification of partial discharge of power equipment, no effective solution has been proposed so far. Summary of the invention

[0005] The embodiments of the present invention provide a method and device for identifying partial discharge of electric equipment, a computer-readable storage medium and an electronic device, so as to at least solve the technical problem of inaccurate identification of partial discharge of electric equipment.

[0006] According to one aspect of an embodiment of the present invention, a method for identifying partial discharge of an electric device is provided, the method comprising: obtaining a partial discharge pulse signal of the electric device within a first preset time window, wherein the partial discharge pulse signal is used to characterize the discharge characteristics of the electric device within the first preset time window; determining at least one statistical parameter of the partial discharge pulse signal based on the partial discharge pulse signal, wherein the statistical parameter is used to characterize the distribution characteristics of the partial discharge pulse signal; determining a statistical significance index of the partial discharge pulse signal based on at least one statistical parameter, wherein the statistical significance index is used to quantify the significance of the partial discharge pulse relative to background noise, and the background noise is used to indicate an interference signal of the partial discharge pulse signal; based on a comparison result between the statistical significance index and a threshold value of the statistical significance index, calling a target recognition model to detect the partial discharge pulse signal, and obtaining a partial discharge recognition result of the electric device, wherein the target recognition model comprises a phase-based statistical recognition model and a timing analysis model, and the partial discharge recognition result comprises at least a discharge type of the partial discharge of the electric device, and a confidence level of the partial discharge recognition result, and the confidence level is used to evaluate the accuracy of the partial discharge recognition result.

[0007] Optionally, based on the local discharge pulse signal, determining at least one statistical parameter of the local discharge pulse signal includes: preprocessing the local discharge pulse signal to obtain a preprocessed local discharge pulse signal, wherein the preprocessing is used to indicate filtering out high-frequency noise and low-frequency noise in the local discharge pulse signal; based on the preprocessed local discharge pulse signal, determining at least one statistical parameter of the preprocessed local discharge pulse signal, wherein the statistical parameter includes at least one of the following: signal amplitude, signal frequency, and signal duration of the preprocessed local discharge pulse signal.

[0008] Optionally, based on at least one statistical parameter, determining a statistical significance index of the partial discharge pulse signal includes: performing standardization processing on at least one statistical parameter to obtain a standardized value of at least one statistical parameter; and performing weighted summation on the standardized value of at least one statistical parameter to obtain a statistical significance index.

[0009] Optionally, weighted summation is performed on the standardized values ​​of at least one statistical parameter to obtain a statistical significance index, including: respectively determining weights corresponding to the standardized values ​​of at least one statistical parameter, wherein the weights are used to characterize the degree of influence of the statistical parameters on whether the local discharge pulse signal has statistical significance, and the weights are adjusted at least according to the discharge characteristics of the local discharge pulse signal and the local discharge pulse signals of the power equipment in the historical stage; weighted summation is performed on the standardized values ​​of at least one statistical parameter and the weights corresponding to the standardized values ​​to obtain a statistical significance index.

[0010] Optionally, based on the comparison result between the statistical significance index and the statistical significance index threshold, the target recognition model is called to detect the local discharge pulse signal to obtain the local discharge recognition result of the power equipment, including: in response to the comparison result indicating that the statistical significance index is greater than the statistical significance index threshold, the phase-based statistical recognition model is called to detect the local discharge pulse signal to obtain the first local discharge recognition result of the power equipment; in response to the comparison result indicating that the statistical significance index is not greater than the statistical significance index threshold, the timing analysis model is called to detect the local discharge pulse signal to obtain the second local discharge recognition result of the power equipment.

[0011] Optionally, after obtaining the second partial discharge identification result, the partial discharge identification method of the power equipment also includes: obtaining a partial discharge pulse signal of the power equipment within a second preset time length, wherein the second preset time length is greater than the first preset time length; determining a statistical significance index corresponding to the partial discharge pulse signal obtained within the second preset time length; in response to the statistical significance index corresponding to the partial discharge pulse signal obtained within the second preset time length being greater than a statistical significance index threshold, calling a phase-based statistical identification model to detect the partial discharge pulse signal of the power equipment obtained within the second preset time length to obtain a third partial discharge identification result of the power equipment; and using the third partial discharge identification result to update the second partial discharge identification result.

[0012] Optionally, the discharge type is at least one of the following: corona discharge, creeping discharge and suspension discharge.

[0013] According to another aspect of an embodiment of the present invention, a partial discharge identification device for an electric device is provided. The device comprises: an acquisition unit, configured to acquire a partial discharge pulse signal of the electric device within a first preset time window, wherein the partial discharge pulse signal is used to characterize the discharge characteristics of the electric device within the first preset time window; a first determination unit, configured to determine at least one statistical parameter of the partial discharge pulse signal based on the partial discharge pulse signal, wherein the statistical parameter is used to characterize the distribution characteristics of the partial discharge pulse signal; a second determination unit, configured to determine a statistical significance index of the partial discharge pulse signal based on at least one statistical parameter, wherein the statistical significance index is used to quantify the significance of the partial discharge pulse relative to background noise, and the background noise is used to indicate an interference signal of the partial discharge pulse signal; a calling unit, configured to call a target recognition model to detect the partial discharge pulse signal based on a comparison result between the statistical significance index and a statistical significance index threshold, and obtain a partial discharge identification result of the electric device, wherein the target recognition model comprises a phase-based statistical recognition model and a timing analysis model, and the partial discharge identification result comprises at least a discharge type of the partial discharge of the electric device and a confidence level of the partial discharge identification result, and the confidence level is used to evaluate the accuracy of the partial discharge identification result.

[0014] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored program, wherein when the program is executed by a processor, the device where the storage medium is located is controlled to execute the partial discharge identification method of the power equipment in the embodiment of the present invention.

[0015] According to another aspect of an embodiment of the present invention, a processor is further provided. The processor is used to run a program, wherein when the program is run, the method for identifying partial discharge of electric power equipment in the embodiment of the present invention is executed.

[0016] According to another aspect of the embodiments of the present invention, a computer program product is provided, which includes computer instructions, and when the computer instructions are executed by a processor, the method for identifying partial discharge of electric power equipment in the embodiments of the present invention is implemented.

[0017] In an embodiment of the present invention, a statistical significance determination mechanism is introduced, which can perform statistical analysis on the local discharge pulse signal of the power equipment, obtain the statistical significance index of the local discharge pulse signal of the power equipment, and effectively screen out significant discharge events, and then compare the statistical significance index with the statistical significance index threshold, and dynamically select the target recognition model according to the comparison result, so that the recognition process is more flexible and adaptive, which helps to adapt to the recognition needs of different discharge types and different working conditions, and can improve the accuracy of the local discharge recognition results of the power equipment. Moreover, the recognition result carries a confidence level, which helps the confidence probability reflect the degree of determination of the local discharge recognition result by the target recognition model, and helps on-site personnel or systems to consider the reliability of the local discharge recognition result when making decisions, thereby solving the technical problem of inaccurate local discharge recognition of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0019] Figure 1 is a flow chart of a method for identifying partial discharge of electric power equipment according to an embodiment of the present invention;

[0020] Figure 2 is a schematic diagram of a corona discharge according to an embodiment of the present application;

[0021] Figure 3 is a schematic diagram of a surface discharge according to an embodiment of the present application;

[0022] Figure 4 is a schematic diagram of a suspended discharge according to an embodiment of the present application;

[0023] Figure 5 is a schematic diagram of a model framework of a phase-based feature recognition model according to an embodiment of the present application;

[0024] Figure 6 is a schematic diagram of a confusion matrix of a phase-based feature recognition model on a test set according to an embodiment of the present application;

[0025] Figure 7 is a schematic diagram of a model framework of a time series recognition model according to an embodiment of the present application;

[0026] Figure 8 is a schematic diagram of a confusion matrix of a time series recognition model on a test set according to an embodiment of the present application;

[0027] Fig. 9is a schematic diagram of a partial discharge identification device for electric power equipment according to an embodiment of the present invention. DETAILED DESCRIPTION

[0028] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme 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 only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, functional component or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, functional components or devices.

[0030] Example 1

[0031] According to an embodiment of the present invention, an embodiment of a method for identifying partial discharge of an electrical device is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0032] Figure 1 is a flow chart of a method for identifying partial discharge of electric power equipment according to an embodiment of the present invention. Figure 1 As shown, the method may include the following steps:

[0033] Step S101, obtaining a partial discharge pulse signal of an electric power device within a first preset time window.

[0034] In the technical solution provided in the above step S101 of the present invention, the power equipment may be a high voltage electrical equipment in the power system, for example, a GIS. The first preset time window is used to indicate a selected continuous time range when monitoring the partial discharge pulse signal, for example, 200 milliseconds (ms), 500 milliseconds (ms) or a longer time period, which is not specifically limited here. The partial discharge pulse signal is used to characterize the discharge characteristics of the power equipment within the first preset time window.

[0035] In this embodiment, the partial discharge pulse signal of the power equipment within the first preset time window is obtained. The purpose is to provide raw data for subsequent statistical significance analysis and model identification by monitoring and recording the signals generated by the partial discharge phenomenon in the initial stage of equipment operation.

[0036] Step S102: determining at least one statistical parameter of the partial discharge pulse signal based on the partial discharge pulse signal.

[0037] In the technical solution provided in the above step S102 of the present invention, the statistical parameters are used to characterize the distribution characteristics of the partial discharge pulse signal, wherein the statistical parameters may include but are not limited to: the amplitude of the discharge pulse, the frequency of the discharge pulse, the duration of the discharge pulse, etc. It is a value extracted from the partial discharge pulse signal, used to describe the distribution characteristics and statistical properties of the signal, such as the average amplitude, frequency, duration, etc. of the pulse.

[0038] In this embodiment, after the partial discharge pulse signal of the power equipment is acquired, at least one statistical parameter may be extracted from the partial discharge pulse signal.

[0039] In this step, determining the statistical parameters of the partial discharge pulse signal is the process of converting the original partial discharge pulse signal into quantifiable indicators. These statistical parameters can reveal the distribution characteristics of the discharge pulse signal and provide key information for subsequent significance judgment and accurate identification of insulation defects in GIS equipment.

[0040] Step S103: determining a statistical significance index of the partial discharge pulse signal based on at least one statistical parameter.

[0041] In the technical solution provided in the above step S103 of the present invention, the statistical significance index is used to quantify the significance of the partial discharge pulse relative to the background noise, and the background noise is used to indicate the interference signal of the partial discharge pulse signal.

[0042] In this embodiment, at least one acquired statistical parameter is calculated by a mathematical formula or algorithm to obtain a statistically significant index of a partial discharge pulse signal. Since there are various electronic noises, electromagnetic interference, etc. in the operating environment of the power equipment, these may be mistakenly identified as partial discharge pulse signals by the monitoring system. Background noise refers to the statistical characteristics of the interference signal received by the monitoring system when there is no partial discharge activity. By calculating the statistical significance index, a numerical value can be obtained to quantify the significance of the discharge pulse signal relative to the background noise. Among them, the standardized value of each statistical parameter can be calculated by the Z score method, and then the calculated standardized value is weighted averaged according to the weight corresponding to the statistical parameter, and the statistical significance index of the partial discharge pulse signal can be obtained.

[0043] Optionally, by calculating a statistical significance index based on the statistical parameters of the partial discharge pulse signal, the difference between the partial discharge pulse signal and the background noise can be quantified, helping to determine which signals are statistically significant partial discharge signals, thereby providing a basis for subsequent partial discharge identification, reducing misjudgment and improving identification accuracy.

[0044] Step S104: Based on the comparison result between the statistical significance index and the statistical significance index threshold, the target recognition model is called to detect the partial discharge pulse signal to obtain the partial discharge recognition result of the power equipment.

[0045] In the technical solution provided in the above step S104 of the present invention, the target recognition model includes a phase-based statistical recognition model and a timing analysis model, and the partial discharge recognition result includes at least the discharge type of the partial discharge of the power equipment and the confidence of the partial discharge recognition result, wherein the discharge type includes at least: corona discharge, surface discharge and suspended discharge, and the confidence is used to evaluate the accuracy of the partial discharge recognition result.

[0046] In this embodiment, after obtaining the statistical significance index, the statistical significance index can be compared with the statistical significance index threshold to obtain a comparison result, and then the target recognition model to be called is determined according to the comparison result. Among them, when the statistical significance index exceeds the statistical significance index threshold, it indicates that the partial discharge pulse signal is statistically significantly different from the background noise and has the characteristics of abnormal discharge activity.

[0047] Optionally, if the comparison result of the statistical significance index and the threshold value of the statistical significance index indicates that the partial discharge pulse signal has statistical significance, a special model can be called to further analyze and identify the partial discharge pulse signal. Among them, the target recognition model can be a phase-based statistical recognition model or a timing analysis model, wherein the phase-based statistical recognition model is a convolutional neural network (CNN) algorithm based on the PRPD spectrum, and the timing analysis model is an algorithm based on the long short-term memory neural network model (LSTM), depending on the level of the statistical significance index and the characteristics of the signal. The phase-based statistical recognition model or the timing analysis model is based on advanced technologies such as deep learning, and can extract complex feature patterns from the partial discharge pulse signal and identify specific discharge types.

[0048] Optionally, through the detection of the target recognition model, a specific partial discharge recognition result can be generated, indicating whether there is a partial discharge phenomenon in the power equipment, as well as the possible type and location of the discharge. In addition, the partial discharge recognition result can also carry a confidence probability related to the partial discharge recognition result. The confidence probability reflects the degree of certainty of the target recognition model on the partial discharge recognition result, which helps field personnel or systems consider the reliability of the partial discharge recognition result when making decisions. Moreover, through automated processing and analysis by the model, the influence of manual intervention and subjective judgment is reduced. This helps to improve the efficiency and accuracy of recognition and reduce operation and maintenance costs.

[0049] Optionally, after obtaining the partial discharge identification results, the specific location of the insulation defect in the power equipment can be determined based on the discharge type and discharge location in the partial discharge identification results, which helps power equipment maintenance personnel to discover and locate insulation defects at an early stage, take preventive measures, avoid the occurrence of faults, and thus ensure the safe and stable operation of the power system.

[0050] In the above steps S101 to S104, a statistical significance determination mechanism is introduced, which can perform statistical analysis on the partial discharge pulse signal of the power equipment, obtain the statistical significance index of the partial discharge pulse signal of the power equipment, and effectively screen out significant discharge events, and then compare the statistical significance index with the statistical significance index threshold, and dynamically select the target recognition model according to the comparison result, so that the recognition process is more flexible and adaptive, which helps to adapt to the recognition needs of different discharge types and different working conditions, and can improve the accuracy of the partial discharge recognition results of the power equipment. Moreover, the recognition results carry confidence, which helps the confidence probability reflect the degree of determination of the partial discharge recognition results by the target recognition model, and helps on-site personnel or systems to consider the reliability of the partial discharge recognition results when making decisions, thereby solving the technical problem of inaccurate partial discharge recognition of power equipment.

[0051] The above method of this embodiment is further introduced below.

[0052] As an optional implementation manner, step S102, based on the local discharge pulse signal, determines at least one statistical parameter of the local discharge pulse signal, including: preprocessing the local discharge pulse signal to obtain a preprocessed local discharge pulse signal, wherein the preprocessing is used to indicate filtering out high-frequency noise and low-frequency noise in the local discharge pulse signal; based on the preprocessed local discharge pulse signal, determining at least one statistical parameter of the preprocessed local discharge pulse signal, wherein the statistical parameter includes at least one of the following: signal amplitude, signal frequency, and signal duration of the preprocessed local discharge pulse signal.

[0053] In this embodiment, since in the actual environment, the partial discharge pulse signal is often accompanied by various noises, including high-frequency noise (such as electromagnetic interference, radio frequency signal) and low-frequency noise (such as power supply noise, vibration noise). By pre-processing the partial discharge pulse signal, these high-frequency noise and low-frequency noise are filtered out, and a purer and easier to analyze signal can be obtained. Optionally, the commonly used pre-processing techniques include filtering (such as low-pass filtering, high-pass filtering, band-pass filtering), signal denoising (such as wavelet denoising, median filtering), and signal enhancement (such as gain adjustment in time domain or frequency domain). Through pre-processing, the signal-to-noise ratio (SNR) of the partial discharge pulse signal can be improved, that is, the ratio of signal intensity to noise intensity, so that subsequent feature extraction and analysis are more accurate.

[0054] Optionally, the determination of the statistical parameters of the pre-processed partial discharge pulse signal is the basis for the subsequent identification of whether there is partial discharge in the power equipment. It should be noted that the statistical parameters include but are not limited to: signal amplitude, signal frequency and signal duration, which are the key characteristics of partial discharge activity. Among them, the signal amplitude reflects the energy of the partial discharge pulse signal; the signal frequency indicates the number of discharge events per unit time, which is related to the activity of the partial discharge; the signal duration describes the time span of a single discharge event, which is crucial for understanding the physical process and mechanism of partial discharge. These parameters are usually determined by calculating the mean value, standard deviation, frequency component distribution, etc. of the partial discharge pulse signal, thereby providing a quantitative basis for subsequent pattern recognition and defect judgment.

[0055] In this step, the purity of the partial discharge pulse signal is ensured through preprocessing, which is a key step to avoid misidentification and improve recognition accuracy. The determination of statistical parameters is a quantitative description of the signal characteristics from a mathematical and statistical perspective, providing intuitive and quantitative features for subsequent significance analysis and model recognition. Through this process, it can be ensured that the subsequent target recognition model can receive high-quality input signals, thereby more accurately identifying possible partial discharge phenomena and insulation defects in power equipment, which is of great significance to the safe operation of the power system and equipment maintenance.

[0056] As an optional implementation manner, step S103, based on at least one statistical parameter, determines the statistical significance index of the partial discharge pulse signal, including: performing standardization processing on the at least one statistical parameter to obtain a standardized value of the at least one statistical parameter; performing weighted summation on the standardized value of the at least one statistical parameter to obtain the statistical significance index.

[0057] In this embodiment, the significance of the partial discharge pulse signal relative to the background noise is quantified through standardization processing and weighted summation, thereby providing a scientific basis for subsequent defect identification and early warning.

[0058] Optionally, standardization is a data preprocessing technique used to eliminate the impact of the dimension of statistical parameters and convert them to a unified scale for comparison and analysis. In the identification of partial discharge pulse signals, standardization usually uses the Z-score method, where the Z-score calculation formula can be expressed by the following formula:

[0059]

[0060] Among them, Z is used to indicate the standardized value, X is used to indicate the actual value of the statistical parameter, μ is used to indicate the mean value of the statistical parameter, and σ is used to indicate the standard deviation of the statistical parameter. Through standardization, the amplitude, frequency, duration and other parameters of the partial discharge pulse signal can be converted into dimensionless standardized values. The size of these values ​​reflects the degree of deviation between the parameter value and the average value, that is, whether the parameter value deviates significantly from the normal range.

[0061] Optionally, after converting the statistical parameters into standardized values, multiple standardized statistical parameter values ​​may be fused by weighted summation to generate a comprehensive statistical significance index.

[0062] In this step, the influence of the original parameter value dimension is eliminated by standardizing the statistical parameters, so that different statistical parameters can be compared and analyzed on the same scale, which is crucial for comprehensively evaluating the significance of partial discharge pulse signals. The weighted summation takes into account the contribution of different parameters in identifying partial discharge. By combining the standardized values ​​of multiple parameters, the abnormality of partial discharge pulse signals can be more comprehensively reflected, improving the accuracy and reliability of identification.

[0063] As an optional implementation, weighted summation is performed on the standardized values ​​of at least one statistical parameter to obtain a statistical significance index, including: respectively determining weights corresponding to the standardized values ​​of at least one statistical parameter, wherein the weights are used to characterize the degree of influence of the statistical parameters on whether the local discharge pulse signal has statistical significance, and the weights are adjusted at least according to the discharge characteristics of the local discharge pulse signal and the local discharge pulse signals of the power equipment in the historical stage; weighted summation is performed on the standardized values ​​of at least one statistical parameter and the weights corresponding to the standardized values ​​to obtain a statistical significance index.

[0064] In this embodiment, after obtaining the annotated numerical value corresponding to each statistical parameter, the weight corresponding to each statistical parameter can be determined according to the relative importance of each statistical parameter in identifying the partial discharge of the power equipment, and then the standardized value corresponding to each statistical parameter and the weight are weighted summed to obtain the statistical significance index of the partial discharge pulse signal.

[0065] For example, the calculation formula for weighted summation of the annotated values ​​of each statistical parameter is:

[0066] Z=ω1Z A +ω2Z F +ω3Z D

[0067] Among them, S is used to indicate the statistical significance index, ω1, ω2 and ω3 are the weights of each statistical parameter, and Z A , ZF and Z D are the Z scores of amplitude, frequency, and duration, respectively, where the weights ω1, ω2, and ω3 can be adjusted according to the characteristics of the partial discharge pulse and historical data to optimize the accuracy of the statistical significance indicator.

[0068] Optionally, the weights may be determined based on historical data and expert knowledge, and adjusted by analyzing the contribution of different statistical parameters in identifying different types of partial discharges, to ensure that the target recognition model focuses more on those statistical parameters that are more sensitive to discharge recognition.

[0069] Optionally, after determining the statistical significance index, if the value of the statistical significance index is greater than (exceeds) a preset statistical significance threshold, it can be considered that the partial discharge pulse signal is statistically significant, that is, the discharge signal is not caused by random noise or interference, but actually exists in the power equipment and may be related to the insulation defects of the equipment. The level of this index also indirectly reflects the severity of the partial discharge phenomenon and the urgency of further diagnosis.

[0070] Optionally, by performing a weighted summation calculation on the standardized values ​​of the statistical parameters, the degree of abnormality of the partial discharge pulse signal can be comprehensively evaluated from multiple perspectives, thereby improving the accuracy and reliability of defect identification. By introducing the concept of weights, the importance of each parameter can be dynamically adjusted according to historical data and equipment characteristics, making the model more adaptable to specific monitoring environments and types of power equipment, and enhancing the flexibility and intelligence of the identification method. The final statistical significance index provides a key quantitative basis for the subsequent defect identification model selection and status assessment, which helps power system maintenance personnel to promptly discover and deal with potential insulation problems and ensure the safe and stable operation of power equipment.

[0071] As an optional implementation method, step S104, based on the comparison result of the statistical significance index and the statistical significance index threshold, calls the target recognition model to detect the local discharge pulse signal to obtain the local discharge recognition result of the power equipment, including: in response to the comparison result indicating that the statistical significance index is greater than the statistical significance index threshold, calls the phase-based statistical recognition model to detect the local discharge pulse signal to obtain the first local discharge recognition result of the power equipment; in response to the comparison result indicating that the statistical significance index is not greater than the statistical significance index threshold, calls the timing analysis model to detect the local discharge pulse signal to obtain the second local discharge recognition result of the power equipment.

[0072] In this embodiment, the statistical significance index of the partial discharge pulse signal is compared with the preset statistical significance index threshold, and then the corresponding defect recognition model is called, which is an important mechanism for realizing the intelligent recognition of GIS insulation defects. This process dynamically selects the model to adapt to the significance of the signal, thereby improving the accuracy and efficiency of recognition.

[0073] Optionally, after obtaining the statistical significance index, the calculated statistical significance index can be compared with a preset statistical significance index threshold to obtain a comparison result. If the statistical significance index is greater than the statistical significance index threshold, this indicates that the local discharge pulse signal is statistically significantly different from the background noise and has a stable and recognizable pattern. At this time, a phase-based statistical recognition model (for example, a convolutional neural network CNN algorithm based on a PRPD spectrum) can be called for further analysis. The phase-based statistical recognition model is good at extracting patterns from the phase and statistical features of the local discharge pulse signal and identifying specific types of insulation defects. The call of the phase-based statistical recognition model, based on the significance of the signal, can more accurately locate and identify the defect type in the power equipment and obtain a first local discharge recognition result. Among them, the first local discharge result can be regarded as a true value result, which can be used to issue a fault warning instruction based on local discharge monitoring.

[0074] Optionally, if the statistical significance index is not greater than the statistical significance index threshold, this may mean that the local discharge activity in the local discharge pulse signal exhibits weak statistical characteristics, or the local discharge pulse signal is severely interfered by noise, and a clear conclusion cannot be immediately obtained through the phase-based statistical recognition model. At this time, a timing analysis model (for example, an LSTM-based algorithm) can be called to analyze the signal. The timing analysis model can capture the sequence characteristics of the local discharge pulse signal that change over time. Even when the local discharge pulse signal characteristics are weak or the noise is large, the discharge type can be identified by analyzing the time series pattern of the signal to obtain a second local discharge recognition result.

[0075] In this step, the dynamic model selection based on signal significance enables the use of a more direct and efficient phase-based statistical recognition model for identification when the signal features are obvious, and the use of a time series analysis model to identify the signal using the time series information of the signal when the signal features are weak or there is a lot of noise. Such a mechanism not only improves the accuracy of identification, but also enhances the adaptability and robustness of the system, and can effectively cope with complex and changing signal environments. At the same time, by giving the identification results under different models (the first and second partial discharge identification results), the system can provide more comprehensive diagnostic information, which helps the operation and maintenance personnel to comprehensively judge the health status of the power equipment, take maintenance measures in a timely manner, and ensure the safe and stable operation of the power system.

[0076] As an optional implementation mode, after obtaining the second partial discharge identification result, the partial discharge identification method of the electric equipment also includes: obtaining a partial discharge pulse signal of the electric equipment within a second preset time length, wherein the second preset time length is greater than the first preset time length; determining a statistical significance index corresponding to the partial discharge pulse signal obtained within the second preset time length; in response to the statistical significance index corresponding to the partial discharge pulse signal obtained within the second preset time length being greater than a statistical significance index threshold, calling a phase-based statistical identification model to detect the partial discharge pulse signal of the electric equipment obtained within the second preset time length to obtain a third partial discharge identification result of the electric equipment; and using the third partial discharge identification result to update the second partial discharge identification result.

[0077] In this embodiment, when the statistical significance index does not reach the statistical significance index threshold, the partial discharge identification result obtained by calling the timing identification algorithm can be regarded as a probabilistic result. Since the partial discharge phenomenon may be hidden and intermittent, it may not be possible to fully capture the characteristics of this phenomenon only through signals within a short time window. Based on this, in order to improve the accuracy of partial discharge identification of power equipment, the identification results can be further verified and optimized by analyzing the discharge pulse signal within a longer time window to ensure its accuracy and reliability.

[0078] For example, after obtaining the second partial discharge identification result, a partial discharge pulse signal within a second preset time length is obtained, wherein the second preset time length is usually longer than the first preset time length, and then the aforementioned steps of determining the partial discharge identification result are cyclically executed, and the statistical significance index is further calculated until the statistical significance index of the power equipment is greater than the statistical significance index threshold, indicating that the discharge phenomenon has a sufficiently high statistical significance and can be considered as a real discharge event, rather than background noise or accidental interference. At this time, the phase-based statistical identification model (such as a convolutional neural network model based on the PRPD spectrum) will be called to perform a deep analysis of the signal within this time window, identify the possible partial discharge type and location in the power equipment, and obtain a third partial discharge identification result. At this time, the third partial discharge identification result can also be regarded as a true value result, which can be used to issue a fault warning instruction based on partial discharge monitoring.

[0079] Optionally, after obtaining the third partial discharge recognition result, the second partial discharge recognition result is updated using the third partial discharge recognition result. This process involves dynamic updating and optimization of the recognition result, and by comparing and fusing the recognition results in different time windows, the accuracy and stability of the recognition can be improved.

[0080] Optionally, by expanding the monitoring time window, increasing the amount of data and time dimension, the accuracy and confidence of partial discharge identification can be further improved. Especially when dealing with intermittent or variable discharge phenomena, it can provide more comprehensive signal feature analysis, help reduce misjudgments and missed judgments, and ensure that the health status monitoring and diagnosis of power equipment are more accurate and reliable.

[0081] As an optional implementation, the discharge type is at least one of the following: corona discharge, creeping discharge and suspension discharge.

[0082] Optionally, Figure 2 is a schematic diagram of a corona discharge according to an embodiment of the present application, Figure 3 is a schematic diagram of a surface discharge according to an embodiment of the present application, Figure 4 is a schematic diagram of a suspended discharge according to an embodiment of the present application.

[0083] The technical solution of the embodiment of the present invention is illustrated below in conjunction with preferred implementation modes.

[0084] At present, power equipment, especially high-voltage electrical equipment such as gas-insulated metal-enclosed switchgear (GIS), may have insulation defects during the design, manufacturing, transportation and installation process due to the influence of various uncontrollable factors. These defects manifest as partial discharge (PD) phenomena during the operation of power equipment, which is an early signal of aging and potential failure of electrical equipment. Partial discharge not only accelerates the degradation of insulating materials, but may also cause more serious equipment failures, posing a threat to the safety and stability of the power system. Therefore, accurately and timely identifying partial discharge and determining whether it is associated with insulation defects is the key to preventing power equipment failures and ensuring the smooth operation of the power system.

[0085] In the related technology, partial discharge identification mainly relies on two technical means: PRPD image analysis and PRPS image analysis. Among them, PRPD images can intuitively show the relationship between discharge pulses and voltage phases, while PRPS images emphasize the periodicity and repeatability of discharge pulses. However, these identification methods are easily affected by the aliasing of partial discharge signals and external interference signals, resulting in a decrease in recognition accuracy, showing the application effect. In addition, due to the intermittent and statistical characteristics of partial discharges caused by many initial defects, traditional recognition algorithms are difficult to make accurate judgments within a limited time or period. Moreover, the PRPD image analysis method and the PRPS image analysis method often require professionals to interpret PRPD and PRPS images based on experience and professional knowledge, which is not only inefficient, but also easily affected by subjective judgment, and there is a technical problem of inaccurate partial discharge identification of power equipment.

[0086] However, the embodiment of the present application provides a GIS insulation defect identification method based on the statistical significance judgment mechanism of local discharge pulses. First, the local discharge pulses of the GIS are tracked and monitored, and the statistical significance of the monitored local discharge pulses is judged. According to the statistical significance processing logic, the phase-based statistical recognition model and the timing analysis model are respectively called, and the credibility of the model output results is given at the same time. The statistical significance judgment mechanism can perform statistical analysis on the discharge pulse data, screen out significant discharge events, help reduce the impact of noise and random errors on the recognition results, and improve the accuracy of recognition. By combining the phase-based statistical recognition model and the timing analysis model, the discharge pulse can be comprehensively analyzed from multiple angles. Compared with the traditional single model recognition, it can capture the discharge characteristics more comprehensively and improve the accuracy of recognition.

[0087] In the present application, after monitoring the local discharge pulse signal of the power equipment within a time window, the statistical parameters of the local discharge pulse within the window can be calculated, wherein the statistical parameters include but are not limited to: discharge amplitude, discharge frequency, and discharge duration, wherein the discharge amplitude is the most intuitive parameter reflecting the severity of the local discharge. The average value of all pulse amplitudes in the positive half cycle within the measurement time range of an integer multiple of the power frequency cycle is recorded as Aav+; the average value of all pulse amplitudes in the positive half cycle is recorded as Aav-; the discharge frequency is the most intuitive parameter reflecting the activity of the local discharge. The number of discharge pulses occurring in the positive half cycle within the unit measurement time t is recorded as N+; the number of discharge pulses occurring in the negative half cycle is recorded as N-. The duration of the discharge pulse is used to indicate the degree of dispersion of the time intervals between the discharge pulses within the measurement time relative to the average value. When the variance is small, it indicates that the distribution between the discharge pulses is more uniform; when the variance is large, it indicates that the discharge pulses are more dispersed in time sequence.

[0088] Optionally, for the statistical parameter set consisting of the discharge pulse amplitude, discharge pulse frequency and discharge pulse duration extracted above, the mean and standard deviation of the statistical parameter set are calculated, and then the standardized value of each statistical parameter is calculated using the Z-score method.

[0089] Optionally, the Z scores of the statistical parameters are weighted and summed to obtain a statistical significance index. Among them, the statistical significance index of the discharge pulses collected under the three types of discharge is obtained as follows:

[0090] Discharge type <![CDATA[ω1]]> <![CDATA[ω2]]> <![CDATA[ω3]]> Z Corona discharge 0.4 0.5 0.1 5.24 Surface discharge 0.5 0.3 0.2 5.62 Suspension discharge 0.2 0.5 0.3 2.45

[0091] Among them, ω1 is used to characterize the weight corresponding to the discharge pulse amplitude, ω2 is used to characterize the weight corresponding to the discharge pulse frequency, and ω3 is used to characterize the weight corresponding to the discharge duration. Z is used to characterize the calculated statistical significance index.

[0092] Optionally, assuming that the threshold value of the statistical significance index is 3, it can be obtained that among the above data, the corona discharge and creeping discharge data are statistically significant, while the suspension discharge pulse data are not statistically significant.

[0093] Optionally, for a discharge data set that meets statistical significance, a phase-based feature recognition model based on a CNN deep network can be used to identify partial discharges, where: Figure 5 is a schematic diagram of a model framework of a phase-based feature recognition model according to an embodiment of the present application, such as Figure 5 As shown in the figure, the phase-based feature recognition model consists of a convolutional layer, a pooling layer and a fully connected layer. After the input data is processed by the convolutional layer, the pooling layer and the fully connected layer, the recognition result of the partial discharge can be obtained.

[0094] Optionally, the collected data is divided into a training set and a test set in an 8:2 ratio, wherein the training set is used to train the above model, wherein: Figure 6 is a schematic diagram of a confusion matrix of a phase-based feature recognition model on a test set according to an embodiment of the present application, such as Figure 6 As shown, by calculating the true positive examples TP (120), false positive examples FP (7), true negative examples TN (3), and false negative examples TP (141) in the confusion matrix, the recognition accuracy of the model is obtained, wherein the calculation formula for the recognition accuracy of the model can be: (TP+TN) / (TP+TN+FP+FN), and the recognition accuracy of the model calculated by this formula is: 96.31%.

[0095] Optionally, the discharge data collected in real time is judged as surface discharge, and a fault warning instruction for partial discharge monitoring is issued to the system. For the discharge data set that does not meet the statistical significance, a time series recognition model based on LSTM unit is established, in which: Figure 7 It is a schematic diagram of a model framework of a time series recognition model according to an embodiment of the present application.

[0096] Optionally, the collected data is divided into a training set and a test set according to an 8:2 ratio, wherein the training set is used to train the above model (time series recognition model), wherein: Figure 8 It is a schematic diagram of a confusion matrix of a time series recognition model on a test set according to an embodiment of the present application. Based on the aforementioned calculation method, the recognition accuracy of the model is: 91.14%.

[0097] Optionally, convert the statistical significance into a confidence probability: (P 置信 It is used to represent the confidence probability, S is used to represent the statistical significance index, S 阈值The probability of identifying the data in the current acquisition window as suspended discharge is 97.62%.

[0098] The beneficial effects of this application are analyzed below.

[0099] In this application, a statistical significance determination mechanism is introduced, which can perform statistical analysis on the discharge pulse data and screen out significant discharge events, which helps to reduce the impact of noise and random errors on the recognition results and improve the accuracy of recognition.

[0100] Furthermore, the combination of phase-based statistical recognition model and timing analysis model can comprehensively analyze the discharge pulse from multiple angles. Compared with the traditional single model recognition, this comprehensive judgment method can capture the discharge characteristics more comprehensively and improve the recognition accuracy.

[0101] Furthermore, the model is dynamically selected according to the statistical significance results of the discharge pulse, making the identification process more flexible and adaptive. This helps to adapt to the needs of GIS insulation defect identification under different discharge types and different working conditions. At the same time, the proposed model has high scalability, and more analysis models and algorithms can be introduced according to actual needs to further improve the accuracy and adaptability of identification.

[0102] Furthermore, computer algorithms can be used to automate processing and analysis, reducing the impact of human intervention and subjective judgment. This helps improve the efficiency and accuracy of identification and reduce operation and maintenance costs. By analyzing the characteristics and laws of discharge pulse signals, it can guide the improvement and optimization of equipment design and improve the reliability and safety of equipment.

[0103] According to an embodiment of the present invention, a partial discharge identification device for electric power equipment is also provided. It should be noted that the partial discharge identification device for electric power equipment can be used to execute the partial discharge identification method for electric power equipment in the embodiment.

[0104] Fig. 9 FIG. 1 is a schematic diagram of a partial discharge identification device for electric power equipment according to an embodiment of the present invention. Fig. 9 As shown, the partial discharge identification device 900 of the electric power equipment may include: an acquisition unit 901 , a first determination unit 902 , a second determination unit 903 and a calling unit 904 .

[0105] An acquisition unit 901 is used to acquire a partial discharge pulse signal of the power equipment within a first preset time window, wherein the partial discharge pulse signal is used to characterize a discharge feature of the power equipment within the first preset time window;

[0106] A first determining unit 902 is used to determine at least one statistical parameter of the partial discharge pulse signal based on the partial discharge pulse signal, wherein the statistical parameter is used to characterize the distribution characteristics of the partial discharge pulse signal;

[0107] A second determining unit 903 is used to determine a statistical significance index of the partial discharge pulse signal based on at least one statistical parameter, wherein the statistical significance index is used to quantify the significance of the partial discharge pulse relative to the background noise, and the background noise is used to indicate an interference signal of the partial discharge pulse signal;

[0108] The calling unit 904 is used to call the target recognition model to detect the partial discharge pulse signal based on the comparison result between the statistical significance index and the statistical significance index threshold, so as to obtain the partial discharge recognition result of the power equipment, wherein the target recognition model includes a phase-based statistical recognition model and a timing analysis model, and the partial discharge recognition result includes at least the discharge type of the partial discharge of the power equipment and the confidence of the partial discharge recognition result, and the confidence is used to evaluate the accuracy of the partial discharge recognition result.

[0109] Optionally, the first determination unit 902 is further used to: preprocess the local discharge pulse signal to obtain a preprocessed local discharge pulse signal, wherein the preprocessing is used to indicate filtering out high-frequency noise and low-frequency noise in the local discharge pulse signal; based on the preprocessed local discharge pulse signal, determine at least one statistical parameter of the preprocessed local discharge pulse signal, wherein the statistical parameter includes at least one of the following: signal amplitude, signal frequency, and signal duration of the preprocessed local discharge pulse signal.

[0110] Optionally, the second determination unit 903 is further used to: preprocess the local discharge pulse signal to obtain a preprocessed local discharge pulse signal, wherein the preprocessing is used to indicate filtering out high-frequency noise and low-frequency noise in the local discharge pulse signal; based on the preprocessed local discharge pulse signal, determine at least one statistical parameter of the preprocessed local discharge pulse signal, wherein the statistical parameter includes at least one of the following: signal amplitude, signal frequency, and signal duration of the preprocessed local discharge pulse signal.

[0111] Optionally, the second determination unit 903 is also used to: determine the weights corresponding to the standardized values ​​of at least one statistical parameter, respectively, wherein the weights are used to characterize the degree of influence of the statistical parameters on whether the local discharge pulse signal has statistical significance, and the weights are adjusted at least according to the discharge characteristics of the local discharge pulse signal and the local discharge pulse signals of the power equipment in the historical stage; perform weighted summation of the standardized values ​​of at least one statistical parameter and the weights corresponding to the standardized values ​​to obtain a statistical significance index.

[0112] Optionally, the calling unit 904 is also used to: in response to the comparison result indicating that the statistical significance index is greater than the statistical significance index threshold, call the phase-based statistical identification model to detect the local discharge pulse signal to obtain a first local discharge identification result of the power equipment; in response to the comparison result indicating that the statistical significance index is not greater than the statistical significance index threshold, call the timing analysis model to detect the local discharge pulse signal to obtain a second local discharge identification result of the power equipment.

[0113] Optionally, the local discharge identification device 900 of the electric power equipment is also used to: obtain a local discharge pulse signal of the electric power equipment within a second preset time length, wherein the second preset time length is greater than the first preset time length; determine a statistical significance index corresponding to the local discharge pulse signal obtained within the second preset time length; in response to the statistical significance index corresponding to the local discharge pulse signal obtained within the second preset time length being greater than a statistical significance index threshold, call a phase-based statistical identification model to detect the local discharge pulse signal of the electric power equipment obtained within the second preset time length, and obtain a third local discharge identification result of the electric power equipment; and use the third local discharge identification result to update the second local discharge identification result.

[0114] In this embodiment, a statistical significance determination mechanism is introduced, which can perform statistical analysis on the local discharge pulse signal of the power equipment, obtain the statistical significance index of the local discharge pulse signal of the power equipment, and effectively screen out significant discharge events, and then compare the statistical significance index with the statistical significance index threshold, and dynamically select the target recognition model according to the comparison result, so that the recognition process is more flexible and adaptive, which helps to adapt to the recognition needs of different discharge types and different working conditions, and can improve the accuracy of the local discharge recognition results of the power equipment. Moreover, the recognition results carry confidence, which helps the confidence probability reflect the degree of determination of the local discharge recognition results by the target recognition model, and helps on-site personnel or systems consider the reliability of the local discharge recognition results when making decisions, thereby solving the technical problem of inaccurate local discharge recognition of power equipment.

[0115] According to an embodiment of the present invention, a computer-readable storage medium is further provided. The storage medium includes a stored program, wherein the program executes the partial discharge identification method of the power equipment in the embodiment.

[0116] According to an embodiment of the present invention, a processor is further provided. The processor is used to run a program. When the program is run, the method for identifying partial discharge of electric power equipment in the embodiment is executed.

[0117] According to another aspect of the embodiments of the present invention, a computer program product is provided, which includes computer instructions, and when the computer instructions are executed by a processor, the method for identifying partial discharge of electric power equipment in the embodiments is implemented.

[0118] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0119] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0120] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0121] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed over multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0122] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0123] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent functional component, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software functional component, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the methods of each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0124] The above are only preferred embodiments of the present invention. It should be pointed out that, for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for identifying partial discharge of electric power equipment, characterized in that: include: Acquire a partial discharge pulse signal of the electric device within a first preset time window, wherein the partial discharge pulse signal is used to characterize the discharge characteristics of the electric device within the first preset time window; Based on the partial discharge pulse signal, determining at least one statistical parameter of the partial discharge pulse signal, wherein the statistical parameter is used to characterize a distribution feature of the partial discharge pulse signal; Determining a statistical significance index of the partial discharge pulse signal based on at least one of the statistical parameters, wherein the statistical significance index is used to quantify the significance of the partial discharge pulse relative to background noise, and the background noise is used to indicate an interference signal of the partial discharge pulse signal; Based on the comparison result between the statistical significance index and the statistical significance index threshold, the target recognition model is called to detect the local discharge pulse signal to obtain the local discharge recognition result of the electric equipment, wherein the target recognition model includes a phase-based statistical recognition model and a timing analysis model, and the local discharge recognition result includes at least the discharge type of the local discharge of the electric equipment and the confidence of the local discharge recognition result, and the confidence is used to evaluate the accuracy of the local discharge recognition result.

2. The method according to claim 1, characterized in that Based on the partial discharge pulse signal, determining at least one statistical parameter of the partial discharge pulse signal comprises: Preprocessing the partial discharge pulse signal to obtain a preprocessed partial discharge pulse signal, wherein the preprocessing is used to indicate filtering out high-frequency noise and low-frequency noise in the partial discharge pulse signal; Based on the preprocessed partial discharge pulse signal, at least one statistical parameter of the preprocessed partial discharge pulse signal is determined, wherein the statistical parameter includes at least one of the following: signal amplitude, signal frequency, and signal duration of the preprocessed partial discharge pulse signal.

3. The method according to claim 1, characterized in that Determining a statistical significance index of the partial discharge pulse signal based on at least one of the statistical parameters includes: Performing standardization processing on at least one of the statistical parameters to obtain a standardized value of at least one of the statistical parameters; The standardized value of at least one of the statistical parameters is weighted and summed to obtain the statistical significance index.

4. The method according to claim 3, characterized in that The weighted summation of the standardized values ​​of at least one of the statistical parameters is performed to obtain the statistical significance index, including: Determine a weight corresponding to a standardized value of at least one of the statistical parameters, respectively, wherein the weight is used to characterize whether the statistical parameter has a statistically significant influence on the partial discharge pulse signal, and the weight is adjusted at least according to the discharge characteristics of the partial discharge pulse signal and the partial discharge pulse signal of the power equipment in a historical stage; The standardized value of at least one of the statistical parameters and the weight corresponding to the standardized value are weighted summed to obtain the statistical significance index.

5. The method according to claim 1, characterized in that Based on the comparison result between the statistical significance index and the statistical significance index threshold, calling the target recognition model to detect the partial discharge pulse signal to obtain the partial discharge recognition result of the power equipment, including: In response to the comparison result indicating that the statistical significance index is greater than the statistical significance index threshold, calling the phase-based statistical identification model to detect the partial discharge pulse signal to obtain a first partial discharge identification result of the power equipment; In response to the comparison result indicating that the statistical significance index is not greater than the statistical significance index threshold, the timing analysis model is called to detect the partial discharge pulse signal to obtain a second partial discharge identification result of the electric power equipment.

6. The method according to claim 5, characterized in that After obtaining the second partial discharge identification result, the method further includes: Acquire a partial discharge pulse signal of the electric power equipment within a second preset time period, wherein the second preset time period is greater than the first preset time period; Determining a statistical significance index corresponding to the partial discharge pulse signal acquired within the second preset time length; In response to the statistical significance index corresponding to the partial discharge pulse signal obtained within the second preset time length being greater than the statistical significance index threshold, calling the phase-based statistical identification model to detect the partial discharge pulse signal of the electric equipment obtained within the second preset time length, and obtaining a third partial discharge identification result of the electric equipment; The second partial discharge identification result is updated using the third partial discharge identification result.

7. The method according to claim 1, characterized in that The discharge type is at least one of the following: corona discharge, creeping discharge and suspension discharge.

8. A partial discharge identification device for electric power equipment, characterized in that: include: An acquisition unit, configured to acquire a partial discharge pulse signal of the electric device within a first preset time window, wherein the partial discharge pulse signal is used to characterize a discharge feature of the electric device within the first preset time window; a first determining unit, configured to determine at least one statistical parameter of the partial discharge pulse signal based on the partial discharge pulse signal, wherein the statistical parameter is used to characterize a distribution feature of the partial discharge pulse signal; a second determining unit, configured to determine a statistical significance index of the partial discharge pulse signal based on at least one of the statistical parameters, wherein the statistical significance index is used to quantify the significance of the partial discharge pulse relative to background noise, and the background noise is used to indicate an interference signal of the partial discharge pulse signal; A calling unit is used to call a target recognition model to detect the partial discharge pulse signal based on the comparison result between the statistical significance index and the statistical significance index threshold, so as to obtain the partial discharge recognition result of the electric equipment, wherein the target recognition model includes a phase-based statistical recognition model and a timing analysis model, and the partial discharge recognition result includes at least the discharge type of the partial discharge of the electric equipment and the confidence of the partial discharge recognition result, and the confidence is used to evaluate the accuracy of the partial discharge recognition result.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed by a processor, the device where the storage medium is located is controlled to execute the method according to any one of claims 1 to 7.

10. An electronic device, characterized in that: include: A memory storing an executable program; A processor, configured to run the program, wherein the program executes the method according to any one of claims 1 to 7 when running.

Images