A switch cabinet discharge fault detection method, system, device and medium

Abstract

The present application relates to a switch cabinet discharge fault detection method, system, equipment and medium, belonging to the technical field of power equipment fault detection, the fault detection method includes: receiving real-time monitoring data of the switch cabinet to be monitored and aligning the timestamp; performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data, generating the corrected electric field signal data and performing time-frequency domain analysis, extracting the discharge pulse density and high-frequency energy proportion; performing frequency filtering and waveform analysis on the vibration signal, calculating the impact intensity factor and the vibration energy entropy value; making an abnormal judgment on the discharge pulse density and the impact intensity factor, and marking the abnormal discharge according to the abnormal judgment result; based on the synchronous change correlation of the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, verifying the abnormal discharge marking result, and obtaining the discharge fault result. The present application can realize efficient detection and accurate diagnosis of switch cabinet discharge faults.

Description

Technical Field

[0001] The present application relates to the technical field of power equipment fault detection, and in particular to a switch cabinet discharge fault detection method, system, equipment and medium. Background Art

[0002] As the scale and complexity of power systems continue to grow, the safety and reliability of power equipment has become a vital research area. Switchgear is one of the most important equipment in the power system, responsible for controlling the on and off of circuits and protecting power equipment from abnormal conditions such as overcurrent and short circuit. Due to factors such as long-term operation, environmental changes, and equipment aging, discharge faults inside switchgear often occur, which may have a serious impact on the normal operation of power equipment and even cause accidents. Therefore, the early diagnosis and detection of discharge faults inside switchgear has become a difficult problem in power equipment management.

[0003] At present, common switch cabinet fault detection methods are difficult to meet the requirements of high precision and high reliability when facing complex environmental conditions and various fault types. For example, electric field signals are easily affected by changes in air conductivity in high humidity environments, resulting in deviations in the amplitude of the signal, which in turn affects the accuracy of fault diagnosis. Vibration signals are also interfered with by various factors (such as mechanical noise and environmental vibration), resulting in a reduced correlation between vibration characteristics and faults. Discharge faults are usually accompanied by high-frequency pulses and mechanical vibrations, and their frequency, intensity, and occurrence time may be closely related to changes in the state of the equipment. For these tiny, short-term electromagnetic discharge signals and vibration signals, traditional signal processing methods have difficulty extracting enough features for accurate diagnosis. Summary of the invention

[0004] In order to achieve efficient detection and accurate diagnosis of switch cabinet discharge faults, the present application provides a switch cabinet discharge fault detection method, system, device and medium.

[0005] In a first aspect, the present application provides a switch cabinet discharge fault detection method, which adopts the following technical solution:

[0006] A switch cabinet discharge fault detection method, the fault detection method comprising:

[0007] Receive real-time monitoring data of the switch cabinet to be monitored and perform time stamp alignment; the real-time monitoring data includes electric field signal data, vibration signal data and current humidity data;

[0008] Performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data;

[0009] Perform time-frequency domain analysis on the corrected electric field signal data to extract the discharge pulse density and high-frequency energy ratio;

[0010] Performing frequency filtering and waveform analysis on the vibration signal to calculate the impact intensity factor and the vibration energy entropy value;

[0011] Performing abnormality judgment on the discharge pulse density and the impact intensity factor, and marking abnormal discharge according to the abnormality judgment result to obtain an abnormal discharge marking result;

[0012] Based on the synchronous change correlation between the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, the abnormal discharge marking result is verified to obtain a discharge fault result.

[0013] By adopting the above technical solutions, based on the synchronous collection and processing of multi-source data, combined with dynamic amplitude compensation, time-frequency domain analysis, vibration signal analysis, anomaly detection and other technologies, efficient and accurate detection of discharge faults is achieved. The system can respond to environmental changes such as humidity and vibration in real time, accurately identify the occurrence of discharge faults, further reduce false alarms through multi-dimensional signal verification, and finally output accurate fault locations and warnings. This technical solution has strong robustness and adaptability, and can adapt to changes in different environmental conditions and fault modes.

[0014] Optionally, the step of performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data includes:

[0015] Verifying the validity of the current humidity data and the electric field signal data, and outputting valid humidity data and a synchronous electric field signal;

[0016] Compare the effective humidity data with a preset reference humidity value to calculate the humidity difference;

[0017] Based on a preset humidity compensation coefficient mapping table, a corresponding dynamic compensation coefficient and a compensation model are determined according to the humidity difference; the compensation model includes a linear compensation model and a nonlinear compensation model;

[0018] The synchronous electric field signal is corrected according to the dynamic compensation coefficient and the compensation model, and the amplitude is truncated to obtain corrected electric field signal data.

[0019] By adopting the above technical solution and combining humidity data with electric field signal data for synchronization and dynamic amplitude compensation, the interference of humidity on the electric field signal is effectively eliminated. The final corrected electric field signal more realistically reflects the discharge events in the electric field, providing high-precision data support for power equipment fault diagnosis and early warning systems, greatly improving the stability and accuracy of the system.

[0020] Optionally, after the step of obtaining the corrected electric field signal data, the step further includes:

[0021] Determining whether the effective humidity data is greater than a preset humidity threshold;

[0022] If yes, perform spectrum analysis on the corrected electric field signal data, extract continuous frequency bands in the spectrum whose amplitude exceeds the preset noise threshold, and obtain candidate noise frequency bands;

[0023] Matching the candidate noise frequency band based on a preset humidity noise feature library, and marking it as a noise concentration frequency band if the match is successful;

[0024] Filtering the noise concentration frequency band based on an adaptive band-stop filter to obtain filtered electric field signal data;

[0025] Calculate a first signal-to-noise ratio of the corrected electric field signal data outside the noise concentration frequency band;

[0026] Calculating a second signal-to-noise ratio of the filtered electric field signal data in a preset discharge characteristic frequency band;

[0027] The corrected electric field signal data and the filtered electric field signal data are weightedly fused based on the first signal-to-noise ratio and the second signal-to-noise ratio to obtain a final corrected electric field signal.

[0028] By adopting the above technical solution, when the humidity exceeds the standard, the influence of humidity on the electric field signal is analyzed by spectrum analysis and noise filtering, and the quality of the corrected signal is evaluated by calculating the signal-to-noise ratio to ensure the effectiveness and accuracy of the signal. Finally, the optimal electric field signal is generated by weighted fusion of the corrected signal and the filtered signal, thus providing high-quality and reliable data support for fault diagnosis of power equipment.

[0029] Optionally, the step of performing abnormality judgment on the discharge pulse density and the impact intensity factor, and marking abnormal discharge according to the abnormality judgment result to obtain the abnormal discharge marking result includes:

[0030] Based on a preset mapping table, determining a corresponding discharge pulse density threshold according to the current humidity data;

[0031] Determining whether the discharge pulse density exceeds the discharge pulse density threshold and the impact intensity factor exceeds a preset impact intensity threshold;

[0032] If so, an abnormal discharge mark is generated and recorded to obtain an abnormal discharge mark result.

[0033] By adopting the above technical solution, the discharge pulse density and impact intensity factor are judged abnormally, and combined with real-time humidity data and preset thresholds, potential discharge faults can be accurately identified and marked. By comprehensively analyzing the abnormal conditions of these two key indicators, the system effectively improves the accuracy of fault detection, reduces the possibility of false alarms and missed alarms, ensures that the equipment can be warned and processed at an early stage, and greatly improves the safety and stability of the equipment.

[0034] Optionally, based on the synchronous change correlation between the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, the abnormal discharge marking result is verified to obtain the discharge fault result, which includes:

[0035] According to the timestamp of the abnormal discharge marking result, align the time axis of the high-frequency energy proportion, the impact intensity factor and the vibration energy entropy value, and perform normalization processing;

[0036] Based on a preset sliding window length, segmenting the high-frequency energy proportion and the impact intensity factor;

[0037] Calculate the Pearson correlation coefficient between the high-frequency energy proportion and the impact intensity factor in each window;

[0038] Marking the relevance of each window according to the Pearson correlation coefficient, marking the window whose Pearson correlation coefficient exceeds a preset coefficient threshold as a strong correlation, and obtaining a sequence of relevance windows;

[0039] Calculating the mean value of the vibration energy entropy value for each window in the correlation window sequence, and marking the window whose mean value exceeds a preset entropy value range as a vibration entropy anomaly, to obtain a vibration entropy anomaly marking sequence;

[0040] A coverage check is performed based on the correlation window sequence and the vibration entropy anomaly mark sequence, and a window sequence with strong correlation and vibration entropy anomaly lasting more than a preset time is marked as a discharge fault segment to obtain a discharge fault result.

[0041] By adopting the above technical solutions, based on multi-level signal processing and analysis, combined with the synchronous changes of high-frequency energy proportion, impact intensity factor and vibration energy entropy value, possible discharge fault events can be accurately identified. Through these precise verification methods, it can help equipment managers to discover potential problems in time, take effective maintenance measures, prevent equipment damage or downtime, and improve system reliability and stability.

[0042] Optionally, after the step of obtaining the discharge fault result, the step further includes:

[0043] Generate maintenance reminder information for the switch cabinet to be monitored based on the discharge fault result;

[0044] Generate a maintenance work order based on the maintenance prompt information and push it to the maintenance terminal;

[0045] Receive maintenance progress feedback information sent by the maintenance terminal in real time, update the maintenance status, and generate a maintenance evaluation report after the maintenance is completed.

[0046] By adopting the above technical solutions, a comprehensive and systematic equipment maintenance process has been realized. Through automated data analysis and intelligent fault detection systems, the efficiency and accuracy of maintenance work can be greatly improved, equipment downtime can be reduced, and maintenance resource allocation can be optimized, thereby ensuring the long-term stable operation of equipment and reducing the risk of failures.

[0047] In a second aspect, the present application provides a switch cabinet discharge fault detection system, which adopts the following technical solution:

[0048] A switch cabinet discharge fault detection system, the fault detection system comprising:

[0049] A data monitoring module, used to receive real-time monitoring data of the switch cabinet to be monitored and perform time stamp alignment; the real-time monitoring data includes electric field signal data, vibration signal data and current humidity data;

[0050] an amplitude compensation module, configured to perform dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data;

[0051] The time-frequency domain analysis module is used to perform time-frequency domain analysis on the corrected electric field signal data to extract the discharge pulse density and high-frequency energy ratio;

[0052] A vibration signal processing module, used for performing frequency filtering and waveform analysis on the vibration signal, and calculating an impact intensity factor and a vibration energy entropy value;

[0053] An abnormality judgment module is used to make an abnormality judgment on the discharge pulse density and the impact intensity factor, and to mark an abnormal discharge according to the abnormality judgment result to obtain an abnormal discharge marking result;

[0054] The fault verification module is used to verify the abnormal discharge marking result based on the synchronous change correlation of the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value to obtain a discharge fault result.

[0055] Optionally, the fault detection system further includes:

[0056] A maintenance prompt module is used to generate maintenance prompt information of the switch cabinet to be monitored according to the discharge fault result;

[0057] A maintenance work order generation module, used to generate a maintenance work order based on the maintenance prompt information and push it to the maintenance terminal;

[0058] The maintenance status update module is used to receive maintenance progress feedback information sent by the maintenance terminal in real time and update the maintenance status;

[0059] The maintenance assessment module is used to generate a maintenance assessment report after maintenance is completed.

[0060] In a third aspect, the present application provides a computer device, which adopts the following technical solution:

[0061] A computer device comprises a memory, a processor and a computer program stored in the memory, wherein the processor executes the computer program to implement the steps of the method as described in the first aspect.

[0062] In a fourth aspect, the present application provides a computer-readable storage medium, which adopts the following technical solution:

[0063] A computer-readable storage medium stores a computer program that can be loaded by a processor and execute any one of the methods in the first aspect.

[0064] In summary, the present application includes at least one of the following beneficial technical effects: the system can respond to environmental changes such as humidity and vibration in real time, accurately identify the occurrence of discharge faults, further reduce false alarms through multi-dimensional signal verification, and ultimately output precise fault locations and warnings. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1 It is a first flow chart of a switch cabinet discharge fault detection method according to one of the embodiments of the present application.

[0066] Figure 2 This is a second flow chart of a switch cabinet discharge fault detection method according to one of the embodiments of the present application.

[0067] Figure 3 It is a third flow chart of a switch cabinet discharge fault detection method according to one of the embodiments of the present application.

[0068] Figure 4 It is a fourth flow chart of a switch cabinet discharge fault detection method according to one of the embodiments of the present application.

[0069] Figure 5 It is a fifth flow chart of a switch cabinet discharge fault detection method according to one of the embodiments of the present application.

[0070] Figure 6 It is a sixth flow chart of a switch cabinet discharge fault detection method according to one of the embodiments of the present application. DETAILED DESCRIPTION

[0071] In order to make the purpose, technical solutions and advantages of this application more clear, the following Figure 1-6 It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0072] The embodiment of the present application discloses a method for detecting discharge faults in a switch cabinet.

[0073] Reference Figure 1 , a switch cabinet discharge fault detection method, the fault detection method comprising:

[0074] Step S101, receiving real-time monitoring data of the switch cabinet to be monitored and aligning the timestamp; the real-time monitoring data includes electric field signal data, vibration signal data and current humidity data;

[0075] Among them, the discharge fault detection of the switch cabinet to be monitored relies on data from multiple signal sources, including electric field signals, vibration signals and humidity data. These signals come from different sensors, and the acquisition time is usually asynchronous. To ensure the accuracy of the analysis results, these different data sources need to be aligned through timestamps to ensure that the electric field, vibration and humidity data at each time point can correspond.

[0076] Specifically, the timestamp alignment method compares the data collection time of each sensor and performs interpolation or cutting so that all data can be aligned to the same time point. This process ensures that different types of data can be processed in parallel, providing a synchronous basis for subsequent analysis.

[0077] It can be understood that accurate time synchronization ensures the correct correlation of multimodal data, laying a data foundation for subsequent signal processing and fault detection. Data alignment of different sensors can avoid errors and deviations caused by time differences and improve the reliability of system analysis.

[0078] Step S102, performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data;

[0079] Humidity is an important factor affecting the electric field signal, especially in high humidity environments, where the electric field signal may be distorted due to changes in air conductivity. In order to eliminate the influence of humidity, the electric field signal can be dynamically compensated for amplitude using humidity data, so that the electric field signal can more accurately reflect the characteristics of the discharge fault.

[0080] Specifically, based on the difference between the real-time humidity data and the preset reference humidity value, a dynamic compensation model is used to adjust the amplitude of the electric field signal. The electric field signal can be corrected by linear or nonlinear methods. For example, if the humidity is higher than the preset threshold, the electric field signal is attenuated and corrected according to the compensation formula to restore the true characteristics of the signal. At the same time, a threshold judgment is made on the humidity signal. When the humidity exceeds the threshold, a larger compensation is performed to effectively eliminate the interference of humidity changes on the electric field signal, correct the amplitude of the signal, and make the electric field signal more truly reflect the characteristics of the discharge event.

[0081] Step S103, performing time-frequency domain analysis on the corrected electric field signal data to extract the discharge pulse density and high-frequency energy ratio;

[0082] Among them, the discharge event will generate high-frequency pulses in the electric field signal, and its frequency and density can reveal the characteristics of the discharge. Therefore, through time-frequency domain analysis technology (such as wavelet transform or short-time Fourier transform), the electric field signal is converted from the time domain to the frequency domain, so as to extract the density of the discharge pulse and the proportion of high-frequency energy.

[0083] Specifically, the time-frequency analysis method can be used to decompose the corrected electric field signal into multiple frequency bands, and calculate the high-frequency energy ratio and the density of the discharge pulse. The discharge pulse density is estimated by calculating the number of pulses that appear per unit time, while the high-frequency energy ratio is obtained by calculating the energy proportion of the high-frequency component in the signal.

[0084] It is understandable that through time-frequency analysis, the discharge pulse characteristics in the electric field signal can be accurately identified, especially the energy of the high-frequency component, which can serve as an effective basis for judging discharge faults. The analysis of high-frequency energy and pulse density helps to timely discover potential discharge faults in equipment.

[0085] Step S104, performing frequency filtering and waveform analysis on the vibration signal, and calculating the impact intensity factor and the vibration energy entropy value;

[0086] Among them, discharge faults are usually accompanied by mechanical vibrations of the equipment. By analyzing the characteristics of the vibration signal, it is possible to assist in identifying the occurrence of discharge events. Frequency filtering first removes the noise component in the vibration signal, and then calculates the impact intensity factor and vibration energy entropy value through waveform analysis. These two parameters can effectively reflect the intensity and complexity of the fault vibration.

[0087] Specifically, frequency filtering is to perform bandpass filtering on the vibration signal, retain the frequency components related to the discharge fault, and remove low-frequency mechanical noise and high-frequency irrelevant noise. The impact intensity factor can be calculated by the ratio of the kurtosis value (kurtosis) of the vibration signal to the root mean square value (RMS), reflecting the impact intensity of the vibration. Based on the frequency components of the vibration signal, the entropy value of the energy distribution of the signal can be calculated to measure the complexity of the vibration signal. The higher the entropy value, the more complex the structure of the vibration signal.

[0088] Step S105, performing abnormality judgment on the discharge pulse density and the impact intensity factor, and marking abnormal discharge according to the abnormality judgment result to obtain an abnormal discharge marking result;

[0089] Among them, combined with the discharge pulse density and the impact intensity factor, statistical analysis or rule judgment methods can be used to determine whether the signal exceeds the normal threshold. If these two indicators are abnormal at the same time, it can be judged as a suspected discharge event and abnormal discharge marking can be performed.

[0090] Specifically, it can be determined whether the discharge pulse density exceeds the set range based on the threshold corresponding to the current humidity data, and whether the impact intensity factor is greater than the preset impact intensity threshold. If both conditions are met, it is marked as a suspected discharge event. By judging the abnormality of these key indicators, it is possible to identify in real time whether a discharge failure has occurred, warn of potential problems in advance, and effectively avoid more serious equipment damage.

[0091] Step S106, based on the synchronous change correlation of the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, the abnormal discharge marking result is verified to obtain the discharge fault result.

[0092] Among them, by analyzing the synchronous change correlation between the high-frequency energy ratio and the impact intensity factor and the vibration energy entropy value, the abnormal marking results are further verified to reduce the false alarm rate. The synchronously changing signal may reveal the actual occurrence of the discharge fault, while the vibration energy entropy value helps to determine whether the vibration signal is related to the discharge event.

[0093] In some embodiments, a statistical method (such as Pearson's correlation coefficient) can be used to analyze the synchronous changes between the high-frequency energy ratio and the impact intensity factor. If the correlation is high and the entropy value of the vibration signal indicates a complex energy pattern, the occurrence of a discharge fault can be confirmed.

[0094] In the above implementation, based on the synchronous acquisition and processing of multi-source data, combined with dynamic amplitude compensation, time-frequency domain analysis, vibration signal analysis, anomaly detection and other technologies, efficient and accurate detection of discharge faults is achieved. The system can respond to environmental changes such as humidity and vibration in real time, accurately identify the occurrence of discharge faults, further reduce false alarms through multi-dimensional signal verification, and finally output accurate fault locations and warnings. This technical solution has strong robustness and adaptability, and can adapt to changes in different environmental conditions and fault modes.

[0095] Reference Figure 2 As an implementation method of step S102, the step of performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data includes:

[0096] Step S201, verifying the validity of current humidity data and electric field signal data, and outputting valid humidity data and synchronous electric field signal;

[0097] Among them, validity verification includes timestamp deviation verification and legitimacy verification of current humidity data.

[0098] Specifically, check whether the timestamp deviation of humidity data and electric field signal data is within the preset allowable range. Since there may be slight errors in the acquisition time of humidity and electric field signals, the system needs to ensure that the two sets of data are collected at the same time point to ensure the reliability of the analysis results. At the same time, verify whether the humidity data is within the range of the sensor. Humidity sensors usually have a certain working range. If the humidity value exceeds this range, the data is invalid.

[0099] Step S202, comparing the effective humidity data with a preset reference humidity value, and calculating the humidity difference;

[0100] After obtaining valid humidity data, the system compares the real-time humidity value with the preset reference humidity value and calculates the humidity difference. The reference humidity value can be set based on historical data or environmental standards, or dynamically adjusted based on the characteristics of the equipment insulation material.

[0101] For example, if the reference humidity value is 60% and the real-time humidity is 70%, the humidity difference ΔH is +10%; if the real-time humidity is lower than the reference humidity, the humidity difference ΔH is a negative value.

[0102] It should be noted that the degree to which the humidity deviates from the reference value is quantified to provide a quantitative basis for the compensation intensity, and the direction of compensation is determined by the positive or negative value of ΔH (only increased humidity requires compensation).

[0103] Step S203, based on a preset humidity compensation coefficient mapping table, determining a corresponding dynamic compensation coefficient and a compensation model according to the humidity difference; the compensation model includes a linear compensation model and a nonlinear compensation model;

[0104] Among them, according to the humidity difference, the system will search the preset humidity compensation coefficient mapping table to determine the corresponding dynamic compensation coefficient. The mapping table includes the corresponding relationship between the humidity difference and the compensation coefficient. At the same time, the appropriate compensation model is selected according to the size of the temperature difference. If the humidity difference is less than the preset difference threshold, the linear compensation model can be selected; if the humidity difference is greater than the preset difference threshold, the nonlinear compensation model is selected. The nonlinear compensation model can more accurately handle situations with large humidity changes.

[0105] In some embodiments, if the humidity difference ΔH≤30%, a linear compensation model is used, and the specific formula is:

[0106] E comp =E raw ×(1−αΔH);

[0107] If ΔH>30%, the nonlinear compensation model is used, and the specific formula is:

[0108] E comp =E raw ×(1−αΔH)⋅e −βΔH ;

[0109] In the above formula, α is the compensation coefficient, E comp is the corrected electric field signal data, E raw is the synchronous electric field signal, and β is the preset material constant.

[0110] Step S204, correcting the synchronous electric field signal according to the dynamic compensation coefficient and the compensation model, and performing amplitude truncation to obtain corrected electric field signal data.

[0111] Among them, after selecting the appropriate compensation coefficient and compensation model, the system corrects the electric field signal according to the compensation coefficient, and at the same time, the amplitude of the corrected electric field signal data is truncated according to the upper and lower limits of the sensor's range to prevent over-compensation from causing signal overflow or negative value distortion. For example, if the corrected electric field signal is greater than the upper limit of the sensor's range, the electric field signal is forced to be the upper limit of the range; if the corrected electric field signal is less than the lower limit of the sensor's range, the electric field signal is forced to be the lower limit of the range.

[0112] In the above implementation, the humidity data and electric field signal data are combined for synchronization and dynamic amplitude compensation, which effectively eliminates the interference of humidity on the electric field signal. The final corrected electric field signal more realistically reflects the discharge events in the electric field, provides high-precision data support for the power equipment fault diagnosis and early warning system, and greatly improves the stability and accuracy of the system.

[0113] Reference Figure 3 As a further implementation, after obtaining the corrected electric field signal data in step S204, the method further includes:

[0114] Step S301, determine whether the effective humidity data is greater than a preset humidity threshold; if so, jump to step S302; if not, do not perform any operation;

[0115] After completing humidity compensation and correcting the electric field signal, the system will first determine whether the current humidity data exceeds the preset humidity threshold. The humidity threshold is usually set based on the environmental standards or historical data of the device. If the humidity exceeds the threshold, it indicates that the ambient humidity is high, and further analysis and noise suppression processing may be required to avoid unnecessary computing overhead in low humidity environments.

[0116] For example, assuming the preset humidity threshold is 70%, if the current humidity data is 80%, the system will determine that the humidity exceeds the standard and proceed to the next step; if the humidity data is lower than 70%, there is no need for noise analysis and filtering, and the next step can be directly processed.

[0117] Step S302, performing spectrum analysis on the corrected electric field signal data, extracting continuous frequency bands in the spectrum whose amplitude exceeds a preset noise threshold, and obtaining candidate noise frequency bands;

[0118] Specifically, if the humidity exceeds the standard, the system will perform spectrum analysis on the corrected electric field signal data. Spectral analysis converts the time domain signal into a frequency domain signal through Fourier transform (FFT), which facilitates the identification of noise frequency bands. In the spectrum, the system will extract the frequency bands whose amplitudes continuously exceed the noise threshold according to the preset noise threshold. These frequency bands usually represent the noise part of the signal. The noise threshold is set according to the common noise level in the equipment environment, which is usually obtained through experiments or historical data.

[0119] For example, assuming that the spectrum analysis result of the corrected electric field signal shows that the amplitude of the frequency band in the range of 50 Hz to 150 Hz continuously exceeds the noise threshold (eg, >1 kV / m), the frequency band will be extracted as a candidate noise frequency band.

[0120] Step S303, matching the candidate noise frequency band based on the preset humidity noise feature library, and marking it as a noise concentration frequency band if the match is successful;

[0121] Among them, the preset humidity noise feature library is established based on historical data and experimental analysis, and records the common noise frequency bands and their amplitude characteristics in the electric field signal under different humidity conditions. After extracting the continuous frequency bands whose amplitude exceeds the noise threshold, the system will match the frequency bands according to the preset humidity noise feature library. For example, when the humidity is 80%, the common noise frequency bands may be concentrated in the range of 100Hz to 200Hz. The system can use dynamic time warping (DTW) or correlation coefficient method to compare the similarity between the candidate noise frequency bands and the templates in the feature library. If the similarity is >90%, it is marked as a noise concentrated frequency band.

[0122] It can be understood that, by matching with the humidity noise feature library, the system can accurately identify the characteristics of the noise frequency band and distinguish them, avoiding misidentifying valid signals as noise, thereby providing accurate data support for subsequent noise processing.

[0123] Step S304, filtering the noise concentration frequency band based on an adaptive band-stop filter to obtain filtered electric field signal data;

[0124] Specifically, for the part marked as the noise-concentrated frequency band, the system will use an adaptive band-stop filter for filtering. The band-stop filter can effectively suppress the noise signal in a specific frequency band and retain the valid signals in other frequency bands.

[0125] Among them, the advantage of the adaptive band-stop filter is that it can automatically adjust the filtering parameters according to the signal characteristics to adapt to the noise characteristics of different frequency bands. The filter will automatically adjust the band-stop bandwidth and attenuation according to the characteristics of the frequency band where the noise is concentrated, thereby removing noise to the maximum extent.

[0126] Step S305, calculating a first signal-to-noise ratio of the corrected electric field signal data outside the noise concentration frequency band;

[0127] Among them, the first signal-to-noise ratio SNR1 of the corrected electric field signal is calculated in the full frequency band outside the noise concentration frequency band, reflecting the signal quality of the area not polluted by noise;

[0128] The calculation formula is: SNR 1 =Signal power / noise power;

[0129] It can be understood that outside the noise concentration frequency band, the signal power comes from the effective signal part, while the noise power is caused by external environmental noise or other irrelevant signals. The calculated first signal-to-noise ratio reflects the quality of the corrected signal.

[0130] Step S306, calculating a second signal-to-noise ratio of the filtered electric field signal data in a preset discharge characteristic frequency band;

[0131] The second signal-to-noise ratio SNR of the filtered electric field signal is calculated in the preset discharge characteristic frequency band. 2 , evaluate the impact of filtering on the effective signal; the preset discharge characteristic frequency band is usually set according to the frequency characteristics of the discharge event, for example, it may be concentrated in the range of 1kHz-10MHz.

[0132] The calculation formula is: SNR 2 =Signal power / noise power.

[0133] It can be understood that the second signal-to-noise ratio can quantify the effectiveness of the filtered signal in the discharge characteristic frequency band, thereby ensuring that the filtered signal can still effectively reflect the discharge characteristics and provide a more reliable data basis for fault detection.

[0134] Step S307: weightedly fuse the corrected electric field signal data and the filtered electric field signal data based on the first signal-to-noise ratio and the second signal-to-noise ratio to obtain a final corrected electric field signal.

[0135] According to the first signal-to-noise ratio and the second signal-to-noise ratio, the system performs weighted fusion on the corrected electric field signal data and the filtered electric field signal data. The weight coefficient can be calculated based on the size of the two signal-to-noise ratios. If the first signal-to-noise ratio is significantly higher than the second signal-to-noise ratio, the corrected electric field signal has a higher weight, and vice versa.

[0136] The specific calculation formula is: final corrected electric field signal = corrected electric field signal × weight 1 + filtered electric field signal × weight 2; among them, the weight coefficient is dynamically adjusted according to the signal-to-noise ratio to ensure that signals with higher signal-to-noise ratios have a greater impact on the final result.

[0137] In the above implementation, when the humidity exceeds the standard, spectrum analysis and noise filtering are performed on the effect of humidity on the electric field signal, and the signal-to-noise ratio is calculated to evaluate the quality of the corrected signal to ensure the effectiveness and accuracy of the signal. Finally, the optimal electric field signal is generated by weighted fusion of the corrected signal and the filtered signal, thereby providing high-quality and reliable data support for fault diagnosis of power equipment.

[0138] Reference Figure 4 As an implementation method of step S105, the discharge pulse density and the impact intensity factor are judged abnormally, and abnormal discharge is marked according to the abnormal judgment result. The step of obtaining the abnormal discharge marking result includes:

[0139] Step S401, based on a preset mapping table, determining a corresponding discharge pulse density threshold according to current humidity data;

[0140] Among them, according to the current humidity data, the system searches for the discharge pulse density threshold corresponding to the humidity. For example, in an environment with high humidity, the electric field signal may be affected by the humidity, resulting in an increase in the discharge pulse density under normal circumstances. Therefore, the humidity value will determine the normal range of the discharge pulse density. Assuming that the humidity is 60%, the normal range of the discharge pulse density is 3 to 8 times per second; if the humidity is 80%, the range may be extended to 5 to 10 times per second.

[0141] Step S402, determining whether the discharge pulse density exceeds the discharge pulse density threshold and the impact intensity factor exceeds the preset impact intensity threshold; if so, jumping to step S403; if not, not performing any operation;

[0142] Among them, by comparing the currently calculated discharge pulse density with the threshold range under the humidity, it is determined whether the discharge pulse density exceeds the normal range. At the same time, the system uses the preset impact intensity threshold to determine whether the calculated impact intensity factor exceeds the threshold. The preset impact intensity threshold can be set according to the device type, historical data and environmental standards. For example, the impact intensity threshold may be 10mV·ms. If the impact intensity factor exceeds this value, it means that there may be a strong discharge event.

[0143] Step S403, generating and recording an abnormal discharge mark to obtain an abnormal discharge mark result.

[0144] Among them, abnormal discharge marks can include key information such as the time of event occurrence, humidity data, discharge pulse density, and impact intensity factor, which are convenient for subsequent analysis and early warning. The mark can be stored in the database for real-time monitoring systems or maintenance personnel to view. By regularly checking and analyzing these marks, potential discharge faults can be discovered in time and corresponding preventive measures can be taken.

[0145] It is understandable that by combining the judgment of the two key indicators, the system can more accurately identify whether an abnormal discharge event has occurred. Only when both conditions are met will it be marked as a suspected discharge event, reducing the false alarm rate.

[0146] In the above implementation, the abnormality judgment of the discharge pulse density and the impact intensity factor is carried out, and the potential discharge fault can be accurately identified and marked by combining the real-time humidity data and the preset threshold. By comprehensively analyzing the abnormal conditions of these two key indicators, the system effectively improves the accuracy of fault detection, reduces the possibility of false alarms and missed alarms, ensures that the equipment can be warned and processed at an early stage, and greatly improves the safety and stability of the equipment.

[0147] Reference Figure 5As an implementation method of step S106, based on the synchronous change correlation of the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, the abnormal discharge marking result is verified, and the step of obtaining the discharge fault result includes:

[0148] Step S501, aligning the time axis of the high-frequency energy proportion, the impact intensity factor and the vibration energy entropy value according to the timestamp of the abnormal discharge marking result, and performing normalization processing;

[0149] Among them, in the raw data collected by the system, the high-frequency energy proportion, impact intensity factor and vibration energy entropy value usually come from different sensors and signal sources, and may have different timestamps or time resolutions. Therefore, these signals need to be aligned according to the timestamps of the abnormal discharge marking results.

[0150] Specifically, the purpose of the alignment operation is to ensure that the time points of different signals correspond one to one during analysis, so that the data at each time point can be effectively compared and avoid analysis errors caused by time inconsistency. Once the signal time axis is aligned, normalization is performed to eliminate the impact of different dimensions and numerical ranges on subsequent analysis. For example, the numerical ranges of high-frequency energy proportion and impact intensity factor may vary greatly. The normalization operation can convert them to a unified range (such as [0,1]) for unified comparison.

[0151] Step S502, segmenting the high-frequency energy proportion and the impact intensity factor based on a preset sliding window length;

[0152] Among them, the sliding window method is a commonly used signal processing technology, especially suitable for processing time series data. In this step, the sliding window is used to divide the continuous high-frequency energy proportion and impact intensity factor data into several segments, each of which represents the data characteristics within a period of time.

[0153] Specifically, it is very important to choose an appropriate sliding window length (e.g., 1 second, 5 seconds, etc.), which determines the temporal resolution of data analysis. Different window sizes may be required in different application scenarios. For example, a longer window can smooth out short-term fluctuations, while a shorter window helps capture rapidly changing signals. Each segment of the sliding window can be considered an independent sample, which facilitates the calculation of the Pearson correlation coefficient and the analysis of signal correlation in different time periods.

[0154] Step S503, calculating the Pearson correlation coefficient between the high frequency energy proportion and the impact intensity factor in each window;

[0155] Among them, the Pearson correlation coefficient is a standard method to measure the linear relationship between two variables, and its value range is -1 to 1. The closer the coefficient is to 1, the stronger the positive correlation between the two variables; the closer it is to -1, the stronger the negative correlation; and the closer it is to 0, the no linear relationship.

[0156] Specifically, in this step, the Pearson correlation coefficient is used to quantify the synchronous change correlation between the high-frequency energy proportion and the impact intensity factor. By calculating the Pearson correlation coefficient within each sliding window, the system can determine whether the relationship between the two indicators is strong during these time periods.

[0157] Step S504, marking the relevance of each window according to the Pearson correlation coefficient, marking the window whose Pearson correlation coefficient exceeds a preset coefficient threshold as strongly correlated, and obtaining a sequence of correlated windows;

[0158] Specifically, the system sets a threshold (e.g., 0.8) to determine whether the correlation between the high-frequency energy ratio and the impact intensity factor is strong enough. If the Pearson correlation coefficient exceeds this threshold, it is considered that the signal changes in this time period have a strong correlation, indicating that a discharge event may have occurred.

[0159] Among them, strongly correlated windows are those windows in which the change trends of the high-frequency energy proportion and the impact intensity factor are consistent within a specific time period. These windows will be focused on in subsequent analysis to determine whether there is a discharge fault.

[0160] Step S505, calculating the mean value of the vibration energy entropy value for each window in the correlation window sequence, and marking the window whose mean value exceeds the preset entropy value range as a vibration entropy anomaly, to obtain a vibration entropy anomaly marking sequence;

[0161] Among them, the vibration energy entropy value is an important indicator to describe the complexity of the signal, and is usually used to measure the uncertainty of the vibration signal. The higher the entropy value of the signal, the more complex its energy distribution is, which may be related to abnormal phenomena such as discharge events.

[0162] In this step, the mean value of the vibration signal energy entropy value of each correlation window is calculated and compared with the preset entropy value range. If the mean value exceeds the preset range, it means that the vibration signal of the window is more complex, which may indicate the presence of a vibration mode related to the discharge event.

[0163] Step S506, performing a coverage check based on the correlation window sequence and the vibration entropy anomaly mark sequence, marking the window sequence with strong correlation and vibration entropy anomaly lasting more than a preset time as a discharge fault segment, and obtaining a discharge fault result.

[0164] If the high-frequency energy ratio is strongly correlated with the impact intensity factor and the vibration entropy is abnormal and lasts for more than a preset time period, the period will be marked as a "discharge fault period". The preset time period refers to the time threshold for the continuous occurrence of strong correlation and vibration entropy abnormality. For example, if the strong correlation and vibration entropy abnormality last for more than 5 seconds in a certain period of time, the system will mark it as a discharge fault period.

[0165] In the embodiment of the present application, through coverage checking, the system can accurately identify the actual discharge fault segment, reduce false alarms caused by short-term fluctuations or abnormal data, and improve the detection accuracy of discharge faults.

[0166] In the above implementation, based on multi-level signal processing and analysis, combined with the synchronous changes of high-frequency energy proportion, impact intensity factor and vibration energy entropy value, possible discharge fault events can be accurately identified. Through these precise verification methods, it can help equipment managers to discover potential problems in time, take effective maintenance measures, prevent equipment damage or downtime, and improve system reliability and stability.

[0167] Reference Figure 6 As a further implementation, after the step of obtaining the discharge fault result, the method further includes:

[0168] Step S601, generating maintenance prompt information of the switch cabinet to be monitored according to the discharge fault result;

[0169] Among them, after completing the analysis and marking of the discharge fault results, maintenance prompt information for a specific switch cabinet can be generated according to the diagnosis results of the discharge fault to remind relevant personnel to perform necessary inspections and maintenance.

[0170] Specifically, maintenance reminder information usually includes the following: the time and specific location of fault identification (such as switch cabinet number), detailed information of the discharge fault (such as the time period when the discharge occurred, the duration of the fault section, etc.), equipment and components that may be affected (such as electrical contacts, insulating materials, etc.), the expected impact of the fault and the priority of maintenance (such as whether it will cause equipment shutdown or other chain failures), and recommended maintenance measures or repair methods (such as checking the insulation layer, replacing damaged parts, etc.).

[0171] It should be noted that the generation of maintenance reminder information needs to consider multiple factors, such as the severity of the fault, the specific time of occurrence, the duration of the fault segment, the relevant equipment components, the expected repair time, the health status of the equipment, etc. These factors will help maintenance personnel to understand the specific situation of the fault and its urgency, and optimize the deployment of maintenance resources and the arrangement of maintenance plans.

[0172] Step S602, generating a maintenance work order based on the maintenance prompt information and pushing it to the maintenance terminal;

[0173] After the maintenance reminder information is generated, the system needs to convert it into an actual maintenance work order to ensure that the maintenance personnel can respond quickly. The work order is not just a simple task instruction, it should include all the necessary details so that the maintenance personnel can fully understand the nature of the problem and the actions that need to be taken.

[0174] Specifically, the system will automatically generate a maintenance work order based on the fault diagnosis results, and push the work order content to the maintenance team responsible for the area or equipment. The work order may include the following: equipment information where the fault occurred (such as switch cabinet number, location, etc.), fault diagnosis information (such as type of discharge fault, cause analysis, impact range, etc.), urgency and processing priority (such as high priority or low priority), tools and spare parts required (such as replacement of specific parts, testing instruments, etc.).

[0175] Step S603: receiving maintenance progress feedback information sent by the maintenance terminal in real time, updating the maintenance status, and generating a maintenance evaluation report after the maintenance is completed.

[0176] Specifically, after the maintenance work begins, the system will track and monitor the maintenance progress in real time. By receiving maintenance progress feedback information, the system can update the maintenance status in time and provide the latest progress information to the equipment manager. When the maintenance work is completed, the system will generate a detailed maintenance assessment report based on the maintenance personnel's work report to record the fault handling process, maintenance results and equipment recovery status.

[0177] In the above implementation, a comprehensive and systematic equipment maintenance process is realized through automated data analysis and intelligent fault detection systems, which can greatly improve the efficiency and accuracy of maintenance work, reduce equipment downtime, and optimize maintenance resource allocation, thereby ensuring the long-term stable operation of the equipment and reducing the risk of failures.

[0178] The embodiment of the present application also discloses a switch cabinet discharge fault detection system.

[0179] A switch cabinet discharge fault detection system, the fault detection system comprising:

[0180] The data monitoring module is used to receive the real-time monitoring data of the switch cabinet to be monitored and perform time stamp alignment; the real-time monitoring data includes electric field signal data, vibration signal data and current humidity data;

[0181] An amplitude compensation module, used for performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data;

[0182] The time-frequency domain analysis module is used to perform time-frequency domain analysis on the corrected electric field signal data to extract the discharge pulse density and high-frequency energy ratio;

[0183] The vibration signal processing module is used to perform frequency filtering and waveform analysis on the vibration signal, and calculate the impact intensity factor and vibration energy entropy value;

[0184] An abnormality judgment module is used to judge the abnormality of the discharge pulse density and the impact intensity factor, mark the abnormal discharge according to the abnormality judgment result, and obtain the abnormal discharge marking result;

[0185] The fault verification module is used to verify the abnormal discharge marking result based on the synchronous change correlation of the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value to obtain the discharge fault result.

[0186] As a further implementation of the fault detection system, the fault detection system further includes:

[0187] A maintenance prompt module is used to generate maintenance prompt information of the switch cabinet to be monitored according to the discharge fault result;

[0188] A maintenance work order generation module is used to generate a maintenance work order based on maintenance prompt information and push it to the maintenance terminal;

[0189] The maintenance status update module is used to receive maintenance progress feedback information sent by the maintenance terminal in real time and update the maintenance status;

[0190] The maintenance assessment module is used to generate a maintenance assessment report after maintenance is completed.

[0191] The switch cabinet discharge fault detection system of the embodiment of the present application can implement any of the above-mentioned switch cabinet discharge fault detection methods, and the specific working process of each module in the switch cabinet discharge fault detection system can refer to the corresponding process in the above-mentioned method embodiment.

[0192] In the several embodiments provided in this application, it should be understood that the provided methods and systems can be implemented in other ways. For example, the system embodiments described above are only illustrative; for example, the division of a certain module is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed.

[0193] The embodiment of the present application also discloses a computer device.

[0194] The computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the switch cabinet discharge fault detection method as described above is implemented.

[0195] The embodiment of the present application also discloses a computer-readable storage medium.

[0196] A computer-readable storage medium stores a computer program that can be loaded by a processor and executed by any one of the switch cabinet discharge fault detection methods described above.

[0197] Among them, computer-readable storage media can be any tangible medium that contains or stores a program, which can be used by or in conjunction with an instruction execution system, apparatus or device; the program code contained on the computer-readable medium can be transmitted using any appropriate medium, including but not limited to wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0198] It should be noted that in the above embodiments, the description of each embodiment has different emphases, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0199] The above are all preferred embodiments of the present application, and are not intended to limit the protection scope of the present application. Any feature disclosed in this specification (including the abstract and drawings), unless otherwise stated, can be replaced by other equivalent or alternative features with similar purposes. That is, unless otherwise stated, each feature is only an example of a series of equivalent or similar features.

Claims

1. A switch cabinet discharge fault detection method, characterized in that: The fault detection method comprises: Receive real-time monitoring data of the switch cabinet to be monitored and perform time stamp alignment; the real-time monitoring data includes electric field signal data, vibration signal data and current humidity data; Performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data; Perform time-frequency domain analysis on the corrected electric field signal data to extract the discharge pulse density and high-frequency energy ratio; Performing frequency filtering and waveform analysis on the vibration signal to calculate the impact intensity factor and the vibration energy entropy value; Performing abnormality judgment on the discharge pulse density and the impact intensity factor, and marking abnormal discharge according to the abnormality judgment result to obtain an abnormal discharge marking result; Based on the synchronous change correlation between the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, the abnormal discharge marking result is verified to obtain a discharge fault result.

2. A switch cabinet discharge fault detection method according to claim 1, characterized in that: The step of performing dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data comprises: Verifying the validity of the current humidity data and the electric field signal data, and outputting valid humidity data and a synchronous electric field signal; Compare the effective humidity data with a preset reference humidity value to calculate the humidity difference; Based on a preset humidity compensation coefficient mapping table, a corresponding dynamic compensation coefficient and a compensation model are determined according to the humidity difference; the compensation model includes a linear compensation model and a nonlinear compensation model; The synchronous electric field signal is corrected according to the dynamic compensation coefficient and the compensation model, and the amplitude is truncated to obtain corrected electric field signal data.

3. A switch cabinet discharge fault detection method according to claim 2, characterized in that: After the step of obtaining the corrected electric field signal data, the method further includes: Determining whether the effective humidity data is greater than a preset humidity threshold; If yes, perform spectrum analysis on the corrected electric field signal data, extract continuous frequency bands in the spectrum whose amplitude exceeds the preset noise threshold, and obtain candidate noise frequency bands; Matching the candidate noise frequency band based on a preset humidity noise feature library, and marking it as a noise concentration frequency band if the match is successful; Filtering the noise concentration frequency band based on an adaptive band-stop filter to obtain filtered electric field signal data; Calculate a first signal-to-noise ratio of the corrected electric field signal data outside the noise concentration frequency band; Calculating a second signal-to-noise ratio of the filtered electric field signal data in a preset discharge characteristic frequency band; The corrected electric field signal data and the filtered electric field signal data are weightedly fused based on the first signal-to-noise ratio and the second signal-to-noise ratio to obtain a final corrected electric field signal.

4. A switch cabinet discharge fault detection method according to claim 1, characterized in that: The steps of making an abnormal judgment on the discharge pulse density and the impact intensity factor, marking an abnormal discharge according to the abnormal judgment result, and obtaining the abnormal discharge marking result include: Based on a preset mapping table, determining a corresponding discharge pulse density threshold according to the current humidity data; Determining whether the discharge pulse density exceeds the discharge pulse density threshold and the impact intensity factor exceeds a preset impact intensity threshold; If so, an abnormal discharge mark is generated and recorded to obtain an abnormal discharge mark result.

5. A switch cabinet discharge fault detection method according to claim 1, characterized in that: Based on the synchronous change correlation between the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value, the abnormal discharge marking result is verified to obtain the discharge fault result, which includes: According to the timestamp of the abnormal discharge marking result, align the time axis of the high-frequency energy proportion, the impact intensity factor and the vibration energy entropy value, and perform normalization processing; Based on a preset sliding window length, segmenting the high-frequency energy proportion and the impact intensity factor; Calculate the Pearson correlation coefficient between the high-frequency energy proportion and the impact intensity factor in each window; Marking the relevance of each window according to the Pearson correlation coefficient, marking the window whose Pearson correlation coefficient exceeds a preset coefficient threshold as a strong correlation, and obtaining a sequence of relevance windows; Calculating the mean value of the vibration energy entropy value for each window in the correlation window sequence, and marking the window whose mean value exceeds a preset entropy value range as a vibration entropy anomaly, to obtain a vibration entropy anomaly marking sequence; A coverage check is performed based on the correlation window sequence and the vibration entropy anomaly mark sequence, and a window sequence with strong correlation and vibration entropy anomaly lasting more than a preset time is marked as a discharge fault segment to obtain a discharge fault result.

6. A switch cabinet discharge fault detection method according to any one of claims 1 to 5, characterized in that: After the step of obtaining the discharge fault result, the method further includes: Generate maintenance reminder information for the switch cabinet to be monitored based on the discharge fault result; Generate a maintenance work order based on the maintenance prompt information and push it to the maintenance terminal; Receive maintenance progress feedback information sent by the maintenance terminal in real time, update the maintenance status, and generate a maintenance evaluation report after the maintenance is completed.

7. A switch cabinet discharge fault detection system, characterized in that: The fault detection system comprises: A data monitoring module, used to receive real-time monitoring data of the switch cabinet to be monitored and perform time stamp alignment; the real-time monitoring data includes electric field signal data, vibration signal data and current humidity data; an amplitude compensation module, configured to perform dynamic amplitude compensation on the electric field signal data based on the aligned current humidity data to generate corrected electric field signal data; The time-frequency domain analysis module is used to perform time-frequency domain analysis on the corrected electric field signal data to extract the discharge pulse density and high-frequency energy ratio; A vibration signal processing module, used for performing frequency filtering and waveform analysis on the vibration signal, and calculating an impact intensity factor and a vibration energy entropy value; An abnormality judgment module is used to make an abnormality judgment on the discharge pulse density and the impact intensity factor, and to mark an abnormal discharge according to the abnormality judgment result to obtain an abnormal discharge marking result; The fault verification module is used to verify the abnormal discharge marking result based on the synchronous change correlation of the high-frequency energy proportion and the impact intensity factor and the vibration energy entropy value to obtain a discharge fault result.

8. A switch cabinet discharge fault detection system according to claim 7, characterized in that: The fault detection system also includes: A maintenance prompt module is used to generate maintenance prompt information of the switch cabinet to be monitored according to the discharge fault result; A maintenance work order generation module, used to generate a maintenance work order based on the maintenance prompt information and push it to the maintenance terminal; The maintenance status update module is used to receive maintenance progress feedback information sent by the maintenance terminal in real time and update the maintenance status; The maintenance assessment module is used to generate a maintenance assessment report after maintenance is completed.

9. A computer device, characterized in that: The method comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

10. A computer-readable storage medium, characterized in that: A computer program is stored which can be loaded by a processor and execute the method according to any one of claims 1 to 6.

Images