Comprehensive monitoring and analysis method and system for partial discharge data of power equipment

Abstract

The invention discloses a comprehensive monitoring and analysis method and system for partial discharge data of power equipment, and relates to the technical field of partial discharge monitoring and analysis of the power equipment, and the method comprises the steps: S1, obtaining comprehensive monitoring data according to a preset monitoring frequency; s2, carrying out classification processing on original signals in the comprehensive monitoring data, and determining a current discharge category; s3, performing discharge condition analysis according to the current discharge type, the equipment data and the environment data to obtain a comprehensive discharge analysis result; when the comprehensive discharge analysis result is critical, sending the comprehensive discharge analysis result, obtaining the adjusted monitoring frequency, taking the adjusted monitoring frequency as a preset monitoring frequency, and executing S1; and when the comprehensive discharge analysis result is abnormal, obtaining maintenance data, and sending the comprehensive discharge analysis result and the maintenance data. According to the method, the comprehensiveness, accuracy and timeliness of analysis can be improved.

Description

Technical Field

[0001] This application relates to the technical field of partial discharge monitoring and analysis of power equipment, and particularly to a comprehensive monitoring and analysis method and system for partial discharge data of power equipment. Background Art

[0002] Partial discharge (PD) of power equipment refers to the discharge that occurs only in a local area of the internal insulation of the power equipment without penetrating the entire internal insulation of the equipment. Partial discharge will generate corresponding physical phenomena such as electricity, light, heat, and chemistry. By corroding the insulating medium, it causes further deterioration of the insulation and ultimately leads to equipment failure. Therefore, the monitoring and analysis of partial discharge data of power equipment are crucial for ensuring the safe operation of power equipment. However, the existing methods and systems for monitoring and analyzing partial discharge data of power equipment have the following problems:

[0003] (1) During the same period, a power equipment may have one type of discharge situation, or may have multiple different types of discharge situations simultaneously. However, the existing methods and systems for monitoring and analyzing partial discharge data of power equipment usually can only analyze the discharge data of a single type. Therefore, the analysis results obtained are relatively one-sided and have low accuracy.

[0004] (2) The existing methods and systems for monitoring and analyzing partial discharge data of power equipment usually do not fully consider the influence of different discharge types, equipment data, and environmental data on the analysis results of partial discharge situations. Their analysis is not comprehensive and has low accuracy.

[0005] (3) The existing methods and systems for monitoring and analyzing partial discharge data of power equipment cannot adjust the monitoring frequency in a timely manner according to real-time situations, and cannot guarantee the real-time nature of monitoring.

[0006] Therefore, there is an urgent need to develop a comprehensive monitoring and analysis method and system for partial discharge data of power equipment to solve the above problems. Summary of the Invention

[0007] The purpose of this application is to provide a comprehensive monitoring and analysis method and system for partial discharge data of power equipment, which can improve the comprehensiveness, accuracy, and timeliness of analysis.

[0008] To achieve the above object, the present application provides a comprehensive monitoring and analysis method for partial discharge data of power equipment, including the following steps: S1: Obtain comprehensive monitoring data according to a preset monitoring frequency, where the comprehensive monitoring data at least includes: original signals, equipment data, and environmental data; S2: Classify the original signals in the comprehensive monitoring data to determine the current discharge category, where the current discharge category at least includes one or more of: internal discharge, corona discharge, surface discharge, floating potential discharge, and poor contact discharge; S3: Analyze the discharge situation based on the current discharge category, equipment data, and environmental data to obtain a comprehensive discharge analysis result, where the comprehensive discharge analysis result is: normal, abnormal, or critical; S4: When the comprehensive discharge analysis result is normal, send the comprehensive discharge analysis result; when the comprehensive discharge analysis result is critical, send the comprehensive discharge analysis result, obtain the adjusted monitoring frequency, and use the adjusted monitoring frequency as the preset monitoring frequency to execute S1; when the comprehensive discharge analysis result is abnormal, obtain maintenance data, and send the comprehensive discharge analysis result and the maintenance data, where the maintenance data at least includes: discharge location, current discharge category, and risk index corresponding to each discharge category in the current discharge category.

[0009] As above, where the sub-steps of classifying the original signals in the comprehensive monitoring data to determine the current discharge category are as follows: S21: Classify the ultra-high frequency electromagnetic wave signals in the original signals to determine the first discharge category; S22: Classify the ultrasonic signals in the original signals to determine the second discharge category; S23: Classify the high-frequency current pulse signals in the original signals to determine the third discharge category; S24: Use the first discharge category, the second discharge category, and the third discharge category as the current discharge category.

[0010] As above, where the sub-steps of classifying the ultra-high frequency electromagnetic wave signals in the original signals to determine the first discharge category are as follows: S211: Perform a first preprocessing on the ultra-high frequency electromagnetic wave signals in the original signals to obtain the preprocessed ultra-high frequency electromagnetic wave signals, where the first preprocessing at least includes: noise reduction processing and filtering processing; S212: Extract features from the preprocessed ultra-high frequency electromagnetic wave signals to obtain the first partial discharge features, where the first partial discharge features include: first time domain features, first frequency domain features, first time-frequency domain features, and statistical features; S213: Input the first partial discharge features into a pre-trained first classification model, and analyze the first partial discharge features by the first classification model to determine the first discharge category, where the first discharge category is: internal discharge, corona discharge, or surface discharge.

[0011] As described above, wherein, for classifying the ultrasonic signal in the original signal to determine the sub-steps of the second discharge category are as follows: S221: Perform a second preprocessing on the ultrasonic signal in the original signal to obtain the preprocessed ultrasonic signal, wherein the second preprocessing at least includes: denoising processing and normalization processing; S222: Extract features from the preprocessed ultrasonic signal to obtain the second partial discharge feature, wherein the second partial discharge feature includes: the second time domain feature, the second frequency domain feature, and the second time-frequency domain feature; S223: Input the second partial discharge feature into the pre-trained second classification model, and analyze the second partial discharge feature by the second classification model to determine the second discharge category, wherein the second discharge category is: internal discharge, surface discharge, or corona discharge.

[0012] As described above, wherein, for classifying the high-frequency current pulse signal in the original signal to determine the sub-steps of the third discharge category are as follows: S231: Perform a third preprocessing on the high-frequency current pulse signal in the original signal to obtain the preprocessed high-frequency current pulse signal, wherein the third preprocessing at least includes: filtering processing and normalization processing; S232: Extract features from the preprocessed high-frequency current pulse signal to obtain the third partial discharge feature, wherein the third partial discharge feature includes: the third time domain feature, the third frequency domain feature, the third time-frequency domain feature, and the second statistical feature; S233: Input the third partial discharge feature into the pre-trained third classification model, and analyze the third partial discharge feature by the third classification model to determine the third discharge category, wherein the third discharge category is: internal discharge, floating potential discharge, or poor contact discharge.

[0013] As described above, the sub - steps for analyzing the discharge situation according to the current discharge category, device data, and environmental data to obtain the comprehensive discharge analysis result are as follows: S31: Read the current discharge category, and randomly generate an analysis serial number for each discharge category in the current discharge category. The analysis serial numbers increase in random order; S32: Take the discharge category corresponding to the minimum value in the analysis serial numbers as the target discharge category, traverse the impact weight database according to the target discharge category, and determine the impact weight data packet with the same standard discharge category and target discharge category as the target weight data packet. Among them, the target weight data packet includes at least: discharge category hazard value, hazard degree weight, device factor weight, weight of each device index, environmental factor weight, and weight of each environmental index; S33: Calculate the target discharge evaluation value according to the target weight data packet, device data, and environmental data; S34: Judge the analysis serial number of the target discharge category according to the total number of discharge categories in the current discharge category. If the total number of discharge categories in the current discharge category is greater than the analysis serial number of the target discharge category, eliminate the analysis serial number of the target discharge category and execute S32; if the total number of discharge categories in the current discharge category is equal to the analysis serial number of the target discharge category, eliminate the analysis serial number of the target discharge category and execute S35; S35: Analyze all the target discharge evaluation values to obtain the comprehensive discharge analysis result.

[0014] As described above, the expression of the target discharge evaluation value is:

[0015] Among them, Spg i is the target discharge evaluation value corresponding to the target discharge category with the analysis serial number i; is the hazard degree weight in the target weight data packet corresponding to the target discharge category with the analysis serial number i; Fys i is the discharge category hazard value in the target weight data packet corresponding to the target discharge category with the analysis serial number i; is the device factor weight in the target weight data packet corresponding to the target discharge category with the analysis serial number i; Szb n is the normalized value of the n - th device index in the device data; η n is the weight of the n - th device index in the target weight data packet corresponding to the target discharge category with the analysis serial number i, n ∈ [1, N], N is a natural number representing the total number of device indices in the device data; is the environmental factor weight in the target weight data packet corresponding to the target discharge category with the analysis serial number i; Hzb m is the normalized value of the m - th environmental index in the environmental data; λ mTo analyze the weight of the m-th environmental indicator in the target weight data packet corresponding to the target discharge category with serial number i, where m ∈ [1, M] and M is a natural number representing the total number of environmental indicators in the environmental data.

[0016] As described above, among them, the sub-steps for analyzing all target discharge evaluation values to obtain a comprehensive discharge analysis result are as follows: S351: Traverse the evaluation threshold database according to the target discharge category corresponding to each target discharge evaluation value, and determine the standard threshold table with the same standard discharge category and target discharge category as the target threshold table; S352: Analyze the corresponding target discharge evaluation value using the target threshold table to generate a sub-discharge result. If the target discharge evaluation value is within the normal threshold range of the target threshold table, the generated sub-discharge result is normal; if the target discharge evaluation value is within the critical threshold range of the target threshold table, the generated sub-discharge result is critical; if the target discharge evaluation value is within the abnormal threshold range of the target threshold table, the generated sub-discharge result is abnormal; S353: Generate a comprehensive discharge analysis result based on all sub-discharge results; if all sub-discharge results are normal, the generated comprehensive discharge analysis result is normal; if there is one or more abnormalities among all sub-discharge results, the generated comprehensive discharge analysis result is abnormal; if there is one or more criticals among all sub-discharge results and there are no abnormalities, the comprehensive discharge analysis result is critical.

[0017] As described above, among them, the sub-steps for obtaining the adjusted monitoring frequency are as follows: S41: Obtain historical monitoring data, and obtain a frequency index value based on the historical monitoring data and the target discharge evaluation value. Among them, the historical monitoring data includes: the current total number of monitoring times and the current total number of abnormalities; S42: Traverse the monitoring frequency data table according to the frequency index value, and determine the adjusted frequency value corresponding to the frequency index range to which the frequency index value belongs as the target adjusted frequency value; S43: Obtain the adjusted monitoring frequency according to the target adjusted frequency value and the preset monitoring frequency, and the adjusted monitoring frequency = preset monitoring frequency + target adjusted frequency value.

[0018] This application also provides a comprehensive monitoring and analysis system for partial discharge data of power equipment, including: a plurality of user terminals and a comprehensive monitoring and analysis center; among them, the user terminal: is used to send comprehensive monitoring data to the comprehensive monitoring and analysis center according to the preset monitoring frequency; receive the comprehensive discharge analysis result and maintenance data; the comprehensive monitoring and analysis center: is used to execute the above-mentioned comprehensive monitoring and analysis method for partial discharge data of power equipment.

[0019] The beneficial effects achieved by this application are as follows:

[0020] (1) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application can monitor and analyze various discharge data, and the analysis results are comprehensive and accurate.

[0021] (2) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application fully consider the influence of different discharge categories, equipment data, and environmental data on the analysis results of partial discharge conditions, effectively improving the comprehensiveness and accuracy of the analysis.

[0022] (3) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application can adjust the monitoring frequency in a timely manner according to real-time conditions, ensuring the real-time nature and accuracy of the monitoring.

[0023] (4) When the comprehensive discharge analysis result in this application is critical, the comprehensive monitoring and analysis method and system for partial discharge data of power equipment can obtain predicted discharge data, and assist the user terminal to know in advance the time and location where the equipment may fail through the predicted discharge data, facilitating the advance arrangement of maintenance plans, thereby avoiding serious consequences such as production interruption and safety accidents caused by sudden equipment failures; and, it can perform maintenance and repair in a timely manner before the equipment is severely damaged, thereby reducing the wear and aging of the equipment caused by excessive or abnormal discharge and extending the service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] In order to more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments recorded in this application. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0025] Figure 1 It is a schematic structural diagram of an embodiment of the comprehensive monitoring and analysis system for partial discharge data of power equipment;

[0026] Figure 2 It is a flowchart of an embodiment of the comprehensive monitoring and analysis method for partial discharge data of power equipment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0027] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0028] As Figure 1 shown, this application provides a comprehensive monitoring and analysis system for partial discharge data of power equipment, including: a plurality of user terminals 1 and a comprehensive monitoring and analysis center 2.

[0029] Among them, the user terminal 1: is used to send comprehensive monitoring data to the comprehensive monitoring and analysis center 2 according to a preset monitoring frequency; receive the comprehensive discharge analysis result and maintenance data.

[0030] The comprehensive monitoring and analysis center 2: is used to execute the comprehensive monitoring and analysis method of the partial discharge data of power equipment described below.

[0031] Furthermore, the comprehensive monitoring and analysis center 2 at least includes: an acquisition unit, a classification unit, a first analysis unit, a second analysis unit, a third analysis unit, and a storage unit.

[0032] Among them, the acquisition unit: acquires comprehensive monitoring data according to a preset monitoring frequency, where the comprehensive monitoring data at least includes: original signals, equipment data, and environmental data; receives and sends the comprehensive discharge analysis result, the adjusted monitoring frequency, and maintenance data.

[0033] The classification unit: classifies the original signals in the comprehensive monitoring data to determine the current discharge category, where the current discharge category at least includes: one or more of internal discharge, corona discharge, surface discharge, floating potential discharge, and poor contact discharge.

[0034] The first analysis unit: analyzes the discharge situation based on the current discharge category, equipment data, and environmental data to obtain the comprehensive discharge analysis result, where the comprehensive discharge analysis result is: normal, abnormal, or critical.

[0035] The second analysis unit: when the comprehensive discharge analysis result is critical, acquires the adjusted monitoring frequency and uses the adjusted monitoring frequency as the preset monitoring frequency.

[0036] The third analysis unit: when the comprehensive discharge analysis result is abnormal, acquires maintenance data.

[0037] The storage unit: is used to store the influence weight database, the evaluation threshold database, the monitoring frequency data table, and the risk level database.

[0038] Among them, the influence weight database includes: multiple influence weight data packets, one influence weight table corresponds to one standard discharge category, and each influence weight data packet at least includes: discharge category hazard value, hazard degree weight, equipment factor weight, weight of each equipment index, environmental factor weight, and weight of each environmental index.

[0039] The standard discharge category at least includes: internal discharge, corona discharge, surface discharge, floating potential discharge, and poor contact discharge. The standard discharge category can directly reflect the hazard degree of the discharge.

[0040] Specifically, the hazard value of the discharge category is a pre-assigned score used to characterize the degree of hazard of the discharge category to the device. Since different discharge categories have different degrees of hazard to the device, the hazard values corresponding to different standard discharge categories are different.

[0041] Among them, the evaluation threshold database includes multiple standard threshold tables. One standard threshold table corresponds to one standard discharge category, and each standard threshold table includes: a normal threshold range, a critical threshold range, and an abnormal threshold range.

[0042] Specifically, the expression of the normal threshold range is: [Zyz min , Zyz max , where Zyz min is the minimum threshold when the discharge condition is normal; Zyz max is the maximum threshold when the discharge condition is normal.

[0043] The expression of the critical threshold range is: [Lyz min , Lyz max , where Lyz min is the minimum threshold when the discharge condition is normal but an abnormality is about to occur; Lyz max is the maximum threshold when the discharge condition is normal but an abnormality is about to occur.

[0044] The expression of the abnormal threshold range is: [Gyz min , Gyz max , where Gyz min is the minimum threshold when the discharge condition is abnormal; Gyz max is the maximum threshold when the discharge condition is abnormal.

[0045] Among them, the monitoring frequency data table includes multiple frequency index ranges, and one frequency index range corresponds to one adjustment frequency value.

[0046] Among them, the risk level database includes multiple risk level data packets. One risk level data packet corresponds to one standard discharge category, and each risk level data packet includes multiple risk level ranges, and one risk level range corresponds to one risk level.

[0047] Furthermore, based on artificial intelligence technology or deep learning technology, multiple actual monitoring and analysis data and / or multiple experimental data are analyzed and processed, so as to update the impact weight database, the evaluation threshold database, the monitoring frequency data table, and the risk level database in real time, so as to ensure the comprehensiveness, timeliness, and accuracy of the impact weight database, the evaluation threshold database, the monitoring frequency data table, and the risk level database.

[0048] Further, the comprehensive monitoring and analysis center 2 further includes: a prediction unit; wherein, the prediction unit: when the comprehensive discharge analysis result is critical, obtains prediction discharge data and sends the prediction discharge data to the user terminal.

[0049] As Figure 2 shown, the present application provides a method for comprehensively monitoring and analyzing partial discharge data of a power equipment, including the following steps:

[0050] S1: Obtain comprehensive monitoring data according to a preset monitoring frequency, wherein the comprehensive monitoring data at least includes: an original signal, equipment data, and environmental data.

[0051] Among them, the original signal at least includes: a UHF electromagnetic wave signal, an ultrasonic wave signal, and a high-frequency current pulse signal.

[0052] The equipment data can directly reflect the equipment state, and the equipment data at least includes: a plurality of equipment indicators, such as: equipment voltage, equipment current, equipment temperature, and insulation performance.

[0053] The environmental data is an external factor affecting the discharge, and the environmental data at least includes: a plurality of environmental indicators, such as: temperature data, humidity data, pollution degree, and altitude.

[0054] S2: Classify the original signal in the comprehensive monitoring data to determine the current discharge category, wherein the current discharge category at least includes: one or more of internal discharge, corona discharge, surface discharge, floating potential discharge, and poor contact discharge.

[0055] Further, the sub-steps of classifying the original signal in the comprehensive monitoring data to determine the current discharge category are as follows:

[0056] S21: Classify the UHF electromagnetic wave signal in the original signal to determine the first discharge category.

[0057] Further, the sub-steps of classifying the UHF electromagnetic wave signal in the original signal to determine the first discharge category are as follows:

[0058] S211: Perform a first preprocessing on the UHF electromagnetic wave signal in the original signal to obtain a preprocessed UHF electromagnetic wave signal, wherein the first preprocessing at least includes: noise reduction processing and filtering processing.

[0059] Specifically, performing noise reduction processing on the UHF electromagnetic wave signal in the original signal to obtain a noise-reduced UHF electromagnetic wave signal, and performing filtering processing on the noise-reduced UHF electromagnetic wave signal can improve the quality of the UHF electromagnetic wave signal.

[0060] Among them, noise reduction processing can be achieved by using existing technologies, such as wavelet noise reduction, empirical mode decomposition (EMD) noise reduction, or noise reduction based on statistical models.

[0061] Filtering processing can be achieved by using existing technologies, such as low-pass filtering, band-pass filtering, or adaptive filtering.

[0062] S212: Extract features from the preprocessed ultra-high frequency electromagnetic wave signal to obtain the first partial discharge feature, where the first partial discharge feature includes: the first time-domain feature, the first frequency-domain feature, the first time-frequency domain feature, and the statistical feature.

[0063] Specifically, for the first time-domain feature: calculate the peak value, effective value, kurtosis, and skewness of the preprocessed ultra-high frequency electromagnetic wave signal. The peak value can reflect the intensity of the discharge, the effective value can reflect the overall energy level, and the kurtosis and skewness can characterize the distribution characteristics of the preprocessed ultra-high frequency electromagnetic wave signal.

[0064] The first frequency-domain feature: Convert the preprocessed ultra-high frequency electromagnetic wave signal to the frequency domain through fast Fourier transform, and extract the center frequency, bandwidth, and amplitude of each frequency component. Different discharge categories have different characteristic distributions in the frequency domain.

[0065] The first time-frequency domain feature: Use time-frequency analysis methods (such as wavelet transform, short-time Fourier transform, etc.) to obtain the energy distribution characteristics of the preprocessed ultra-high frequency electromagnetic wave signal on the time-frequency plane, such as the energy concentration area and change trend in the time-frequency diagram.

[0066] The statistical feature: Calculate the statistics of the preprocessed ultra-high frequency electromagnetic wave signal, such as the probability density function and power spectral density, etc., and analyze its distribution law and statistical characteristics to distinguish different discharge categories.

[0067] S213: Input the first partial discharge feature into the pre-trained first classification model, and the first classification model analyzes the first partial discharge feature to determine the first discharge category, where the first discharge category is: internal discharge, corona discharge, or surface discharge.

[0068] Specifically, the first classification model is a pre-trained deep learning model (such as a CNN (Convolutional Neural Network)-LSTM (Long Short-Term Memory Network) hybrid neural network), support vector machine, artificial intelligence, or decision tree.

[0069] S22: Classify the ultrasonic signal in the original signal to determine the second discharge category.

[0070] Furthermore, the sub-steps of classifying the ultrasonic signal in the original signal to determine the second discharge category are as follows:

[0071] S221: Perform second preprocessing on the ultrasonic signal in the original signal to obtain the preprocessed ultrasonic signal, where the second preprocessing at least includes: denoising processing and normalization processing.

[0072] Specifically, performing denoising processing on the ultrasonic signal in the original signal to obtain the denoised ultrasonic signal can remove the random noise in the signal and smooth the signal. Performing normalization processing on the denoised ultrasonic signal to normalize the denoised ultrasonic signal to a preset interval can eliminate the influence of different amplitudes on subsequent processing.

[0073] Among them, the denoising processing of the ultrasonic signal in the original signal can be achieved by using existing denoising processing methods, such as: mean filtering, median filtering.

[0074] The normalization processing of the denoised ultrasonic signal can be achieved by using existing technologies.

[0075] S222: Extract features from the preprocessed ultrasonic signal to obtain the second partial discharge feature, where the second partial discharge feature includes: second time domain feature, second frequency domain feature, and second time-frequency domain feature.

[0076] Specifically, the second time domain feature: Calculate the peak value, effective value, kurtosis, and skewness of the preprocessed ultrasonic signal. The peak value can reflect the intensity of the discharge, the effective value can reflect the overall energy level, and the kurtosis and skewness can characterize the distribution characteristics of the preprocessed ultrasonic signal.

[0077] The second frequency domain feature: Convert the preprocessed ultrasonic signal to the frequency domain through fast Fourier transform, and extract the center frequency, bandwidth, and amplitude of each frequency component. Different discharge categories have different characteristic distributions in the frequency domain.

[0078] The second time-frequency domain feature: Use time-frequency analysis methods (such as: wavelet transform, short-time Fourier transform, etc.) to obtain the energy distribution characteristics of the preprocessed ultrasonic signal on the time-frequency plane, such as the energy concentration area and change trend in the time-frequency diagram.

[0079] S223: Input the second partial discharge feature into the pre-trained second classification model, and the second classification model analyzes the second partial discharge feature to determine the second discharge category, where the second discharge category is: internal discharge, surface discharge, or corona discharge.

[0080] Specifically, the second classification model is a model with a feature selection mechanism such as logistic regression, artificial intelligence, or decision tree.

[0081] S23: Classify the high-frequency current pulse signal in the original signal to determine the third discharge category.

[0082] Furthermore, the high-frequency current pulse signals in the original signal are classified, and the sub-steps for determining the third discharge category are as follows:

[0083] S231: Perform a third preprocessing on the high-frequency current pulse signals in the original signal to obtain the preprocessed high-frequency current pulse signals. The third preprocessing includes at least: filtering processing and normalization processing.

[0084] Specifically, perform filtering processing on the high-frequency current pulse signals in the original signal to obtain the high-frequency current pulse signals in the filtering, and perform normalization processing on the high-frequency current pulse signals in the filtering to obtain the preprocessed high-frequency current pulse signals.

[0085] Among them, the filtering processing of the high-frequency current pulse signals in the original signal can be achieved by using existing technologies. For example: digital filtering technology.

[0086] The normalization processing of the high-frequency current pulse signals in the filtering can be achieved by using existing technologies.

[0087] S232: Extract features from the preprocessed high-frequency current pulse signals to obtain the third partial discharge features. The third partial discharge features include: the third time-domain features, the third frequency-domain features, the third time-frequency domain features, and the second statistical features.

[0088] Specifically, the third time-domain features: Calculate the peak value, effective value, kurtosis, and skewness of the preprocessed high-frequency current pulse signals. The peak value can reflect the intensity of the discharge, the effective value can reflect the overall energy level, and the kurtosis and skewness can characterize the distribution characteristics of the preprocessed high-frequency current pulse signals.

[0089] The third frequency-domain features: Convert the preprocessed high-frequency current pulse signals to the frequency domain through fast Fourier transform, and extract the center frequency, bandwidth, and amplitudes of each frequency component. Different discharge categories have different characteristic distributions in the frequency domain.

[0090] The third time-frequency domain features: Use time-frequency analysis methods (such as: wavelet transform, short-time Fourier transform, etc.) to obtain the energy distribution features of the preprocessed high-frequency current pulse signals on the time-frequency plane, such as the energy concentration region and change trend in the time-frequency diagram, etc.

[0091] The second statistical features: Calculate the statistics of the preprocessed high-frequency current pulse signals, such as: probability density function and power spectral density, etc., and analyze their distribution laws and statistical characteristics to distinguish different discharge categories.

[0092] S233: Input the third partial discharge feature into a pre-trained third classification model, and analyze the third partial discharge feature by the third classification model to determine the third discharge category, where the third discharge category is: internal discharge, floating potential discharge, or poor contact discharge.

[0093] Specifically, the third classification model is a classifier such as a support vector machine, decision tree, random forest, or a deep learning model such as a convolutional neural network or recurrent neural network.

[0094] S24: Take the first discharge category, the second discharge category, and the third discharge category as the current discharge category.

[0095] Specifically, the discharge situation of the power equipment may be a type of discharge situation, or different types of discharge situations may exist simultaneously. For example: When there are internal air gaps in the insulation layer of a high-voltage cable resulting in internal discharge, surface discharge may occur at the terminal head of the high-voltage cable due to electric field concentration. In a humid and polluted environment, surface discharge is likely to occur on the surface of electrical equipment, and at the same time, due to the influence of humidity on the insulation performance, internal discharge may occur inside the electrical equipment. The internal electric field distribution of the power equipment is uneven, corona discharge may occur in areas with high electric field intensity, and internal discharge or surface discharge may be triggered at weak insulation areas with severe electric field distortion.

[0096] S3: Analyze the discharge situation based on the current discharge category, equipment data, and environmental data to obtain a comprehensive discharge analysis result, where the comprehensive discharge analysis result is: normal, abnormal, or critical.

[0097] Furthermore, the sub-steps of analyzing the discharge situation based on the current discharge category, equipment data, and environmental data to obtain a comprehensive discharge analysis result are as follows:

[0098] S31: Read the current discharge category, and randomly generate an analysis serial number for each discharge category in the current discharge category. The analysis serial numbers increase in random order.

[0099] Specifically, the current discharge category includes A types of discharge categories. The analysis serial number of the discharge category with a random order of one is 1, the analysis serial number of the discharge category with a random order of two is 2,..., and the analysis serial number of the discharge category with a random order of A is A.

[0100] S32: Take the discharge category corresponding to the minimum value in the analysis serial numbers as the target discharge category, traverse the impact weight database according to the target discharge category, and determine the impact weight data packet with the same standard discharge category and target discharge category as the target weight data packet, where the target weight data packet at least includes: discharge category hazard value, hazard degree weight, equipment factor weight, weight of each equipment index, environmental factor weight, and weight of each environmental index.

[0101] S33: Calculate the target discharge evaluation value according to the target weight data packet, device data, and environmental data.

[0102] Furthermore, the expression of the target discharge evaluation value is:

[0103]

[0104] Among them, Spg i is the target discharge evaluation value corresponding to the target discharge category with analysis serial number i; is the hazard degree weight in the target weight data packet corresponding to the target discharge category with analysis serial number i; Fys i is the discharge category hazard value in the target weight data packet corresponding to the target discharge category with analysis serial number i; is the device factor weight in the target weight data packet corresponding to the target discharge category with analysis serial number i; Szb n is the normalized value of the nth device index in the device data; η n is the weight of the nth device index in the target weight data packet corresponding to the target discharge category with analysis serial number i, n ∈ [1, N], N is a natural number representing the total number of device indices in the device data; μ i 3 is the environmental factor weight in the target weight data packet corresponding to the target discharge category with analysis serial number i; Hzb m is the normalized value of the mth environmental index in the environmental data; λ m is the weight of the mth environmental index in the target weight data packet corresponding to the target discharge category with analysis serial number i, m ∈ [1, M], M is a natural number representing the total number of environmental indices in the environmental data.

[0105] Specifically, normalize the nth device index in the device data through the existing technology, map the nth device index to the interval [0, 10], and obtain the normalized value of the nth device index.

[0106] Normalize the mth environmental index in the environmental data through the existing technology, map the mth environmental index to the interval [0, 10], and obtain the normalized value of the mth environmental index.

[0107] Among them, η n and λ m The specific values of are set according to the actual situation.

[0108] S34: Judge the analysis serial number of the target discharge category according to the total number of discharge categories in the current discharge category. If the total number of discharge categories in the current discharge category is greater than the analysis serial number of the target discharge category, then eliminate the analysis serial number of the target discharge category and execute S32; if the total number of discharge categories in the current discharge category is equal to the analysis serial number of the target discharge category, then eliminate the analysis serial number of the target discharge category and execute S35.

[0109] S35: Analyze all target discharge evaluation values to obtain a comprehensive discharge analysis result.

[0110] Furthermore, the sub-steps for analyzing all target discharge evaluation values to obtain a comprehensive discharge analysis result are as follows:

[0111] S351: Traverse the evaluation threshold database according to the target discharge category corresponding to each target discharge evaluation value, and determine the standard threshold table with the same standard discharge category and target discharge category as the target threshold table.

[0112] S352: Analyze the corresponding target discharge evaluation value using the target threshold table to generate a sub-discharge result. If the target discharge evaluation value is within the normal threshold range of the target threshold table, the generated sub-discharge result is normal; if the target discharge evaluation value is within the critical threshold range of the target threshold table, the generated sub-discharge result is critical; if the target discharge evaluation value is within the abnormal threshold range of the target threshold table, the generated sub-discharge result is abnormal.

[0113] S353: Generate a comprehensive discharge analysis result based on all sub-discharge results; if all sub-discharge results are normal, the generated comprehensive discharge analysis result is normal; if there is one or more abnormalities among all sub-discharge results, the generated comprehensive discharge analysis result is abnormal; if there is one or more critical values among all sub-discharge results and no abnormalities, the comprehensive discharge analysis result is critical.

[0114] S4: When the comprehensive discharge analysis result is normal, send the comprehensive discharge analysis result; when the comprehensive discharge analysis result is critical, send the comprehensive discharge analysis result, obtain the adjusted monitoring frequency, and use the adjusted monitoring frequency as the preset monitoring frequency, then execute S1; when the comprehensive discharge analysis result is abnormal, obtain maintenance data, and send the comprehensive discharge analysis result and the maintenance data, where the maintenance data at least includes: discharge location, current discharge category, and the risk index corresponding to each discharge category in the current discharge category.

[0115] Specifically, the existing technology is adopted to obtain the discharge position. For example, the time difference positioning method or the array signal processing method can be used to obtain the discharge position of the ultra-high frequency electromagnetic wave signal; the triangulation method or the acoustic imaging method can be used to obtain the discharge position of the ultrasonic signal; the traveling wave method or the current transformer method can be used to obtain the discharge position of the high-frequency current pulse signal.

[0116] The risk index corresponding to each discharge category in the current discharge category is: the risk level corresponding to the risk level range to which the target discharge evaluation value of this discharge category belongs.

[0117] Further, the sub-steps for obtaining the adjusted monitoring frequency are as follows:

[0118] S41: Obtain historical monitoring data, and obtain the frequency index value according to the historical monitoring data and the target discharge evaluation value. Among them, the historical monitoring data includes: the current total number of monitors and the current total number of anomalies.

[0119]

[0120] Among them, Psyz is the frequency index value; Gjc is the current total number of anomalies; Zjc is the current total number of monitors; is the hazard degree weight in the target weight data packet corresponding to the target discharge category with the analysis serial number i; Spg i is the target discharge evaluation value corresponding to the target discharge category with the analysis serial number i, i ∈ [1, I], I is the total number of target discharge evaluation values, and I is a natural number.

[0121] S42: Traverse the monitoring frequency data table according to the frequency index value, and determine that the adjustment frequency value corresponding to the frequency index range to which the frequency index value belongs is the target adjustment frequency value.

[0122] S43: Obtain the adjusted monitoring frequency according to the target adjustment frequency value and the preset monitoring frequency. The adjusted monitoring frequency = the preset monitoring frequency + the target adjustment frequency value.

[0123] Specifically, the preset monitoring frequency is the number of times to obtain comprehensive monitoring data per unit time. The target adjustment frequency value is the number of times to increase the acquisition of comprehensive monitoring data per unit time.

[0124] When the comprehensive discharge analysis result is critical, adjusting the monitoring frequency in a timely manner according to the historical abnormal situation and the hazard degree of the target discharge evaluation value can further improve the real-time performance and accuracy of the monitoring.

[0125] Further, when the comprehensive discharge analysis result is critical, it also includes: obtaining and sending predicted discharge data, where the predicted discharge data includes: various predicted discharge indicators, such as: discharge amount, discharge intensity, discharge frequency, discharge phase, and discharge energy.

[0126] Furthermore, the expression for predicting the discharge index is as follows:

[0127]

[0128] where Yfd k is the k-th predicted discharge index; β 0 , β e , β r , β v are all regression coefficients; Xsb e is the e-th data in the device historical data, e ∈ [1, E], and E is the total type of data in the device historical data involved in calculating the k-th predicted discharge index; Xhj r is the r-th index in the environmental factor data, r ∈ [1, R], and R is the total type of indices in the environmental factor data involved in calculating the k-th predicted discharge index; Xlb v is the discharge characteristic parameter of the corresponding discharge category, v ∈ [1, V], and V is the total amount of discharge characteristic parameters of the discharge category involved in calculating the k-th predicted discharge index; τ is the error term.

[0129] Specifically, the regression coefficients and the error term are set according to the actual situation. The predicted discharge data can assist the user terminal to know in advance the time and location where the device may fail, facilitating the advance arrangement of maintenance plans, thus avoiding serious consequences such as production interruption and safety accidents caused by sudden device failures; it can timely maintain and repair the device before serious damage occurs, thereby reducing the wear and aging of the device caused by over-discharge or abnormal discharge and extending the service life of the device.

[0130] Among them, the device historical data includes at least: device voltage historical data, device current historical data, device power historical data, device discharge amount historical data, device discharge frequency historical data, device discharge pulse width historical data, device temperature historical data, device insulation resistance historical data, and device operation time historical data.

[0131] The indices in the environmental factor data include at least: temperature, humidity, air pressure, pollution degree, altitude, and electromagnetic interference intensity.

[0132] The discharge characteristic parameters include at least: discharge pulse characteristic parameters (such as: pulse amplitude), discharge statistical characteristic parameters (such as: discharge repetition rate), and spectral characteristic parameters (such as: frequency band width).

[0133] The beneficial effects achieved by this application are as follows:

[0134] (1) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application can monitor and analyze various discharge data, and the analysis results are comprehensive and accurate.

[0135] (2) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application fully consider the influence of different discharge categories, equipment data, and environmental data on the analysis results of partial discharge conditions, effectively improving the comprehensiveness and accuracy of the analysis.

[0136] (3) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application can adjust the monitoring frequency in a timely manner according to real-time conditions, ensuring the real-time and accurate monitoring.

[0137] (4) The comprehensive monitoring and analysis method and system for partial discharge data of power equipment in this application can obtain predicted discharge data when the comprehensive discharge analysis result is critical. By using the predicted discharge data, it can assist the user terminal to know in advance the time and location where the equipment may fail, facilitating the advance arrangement of maintenance plans, thereby avoiding serious consequences such as production interruption and safety accidents caused by sudden equipment failures; and it can promptly perform maintenance and repair when the equipment has not yet been severely damaged, thereby reducing the wear and aging of the equipment caused by excessive or abnormal discharge and extending the service life of the equipment.

[0138] Although the preferred embodiments of this application have been described, those skilled in the art can make additional changes and modifications once they know the basic creative concepts. Therefore, the protection scope of this application is intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of this application. Obviously, those skilled in the art can make various changes and variations to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application belong to the scope of the protection of this application and its equivalent technologies, this application also intends to include these modifications and variations.

Claims

1. A comprehensive monitoring and analysis method for partial discharge data of power equipment, characterized in that: The steps include: S1: Obtaining comprehensive monitoring data according to a preset monitoring frequency, wherein the comprehensive monitoring data at least includes: original signals, equipment data and environmental data; S2: classifying and processing the original signals in the comprehensive monitoring data to determine the current discharge category, wherein the current discharge category includes at least one or more of internal discharge, corona discharge, surface discharge, floating potential discharge and poor contact discharge; S3: Analyze the discharge situation according to the current discharge category, equipment data and environmental data to obtain a comprehensive discharge analysis result, wherein the comprehensive discharge analysis result is: normal, abnormal or critical; S4: When the comprehensive discharge analysis result is normal, send the comprehensive discharge analysis result; when the comprehensive discharge analysis result is critical, send the comprehensive discharge analysis result, and obtain the adjusted monitoring frequency, and use the adjusted monitoring frequency as the preset monitoring frequency, and execute S1; when the comprehensive discharge analysis result is abnormal, obtain maintenance data, and send the comprehensive discharge analysis result and maintenance data, wherein the maintenance data at least includes: discharge position, current discharge category, and risk index corresponding to each discharge category in the current discharge category.

2. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 1 is characterized in that: The sub-steps of classifying the original signal in the comprehensive monitoring data and determining the current discharge category are as follows: S21: classifying and processing the ultra-high frequency electromagnetic wave signal in the original signal to determine the first discharge category; S22: classifying and processing the ultrasonic signal in the original signal to determine the second discharge category; S23: classifying the high-frequency current pulse signal in the original signal to determine a third discharge category; S24: The first discharge category, the second discharge category, and the third discharge category are used as current discharge categories.

3. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 2 is characterized in that: The sub-steps of classifying and processing the ultra-high frequency electromagnetic wave signal in the original signal and determining the first discharge category are as follows: S211: performing a first preprocessing on the ultra-high frequency electromagnetic wave signal in the original signal to obtain a preprocessed ultra-high frequency electromagnetic wave signal, wherein the first preprocessing at least includes: noise reduction processing and filtering processing; S212: extracting features from the preprocessed ultra-high frequency electromagnetic wave signal to obtain a first partial discharge feature, wherein the first partial discharge feature includes: a first time domain feature, a first frequency domain feature, a first time-frequency domain feature, and a statistical feature; S213: inputting the first partial discharge feature into a pre-trained first classification model, and using the first classification model to analyze the first partial discharge feature to determine a first discharge category, wherein the first discharge category is: internal discharge, corona discharge or creeping discharge.

4. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 2 is characterized in that: The sub-steps of classifying and processing the ultrasonic signal in the original signal and determining the second discharge category are as follows: S221: performing a second preprocessing on the ultrasonic signal in the original signal to obtain a preprocessed ultrasonic signal, wherein the second preprocessing at least includes: denoising processing and normalization processing; S222: extracting features from the preprocessed ultrasonic signal to obtain a second partial discharge feature, wherein the second partial discharge feature includes: a second time domain feature, a second frequency domain feature, and a second time-frequency domain feature; S223: Inputting the second partial discharge feature into a pre-trained second classification model, and using the second classification model to analyze the second partial discharge feature to determine a second discharge category, wherein the second discharge category is: internal discharge, surface discharge or corona discharge.

5. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 2 is characterized in that: The sub-steps of classifying the high-frequency current pulse signal in the original signal and determining the third discharge category are as follows: S231: performing a third preprocessing on the high-frequency current pulse signal in the original signal to obtain a preprocessed high-frequency current pulse signal, wherein the third preprocessing at least includes: filtering processing and normalization processing; S232: extracting features from the preprocessed high-frequency current pulse signal to obtain a third partial discharge feature, wherein the third partial discharge feature includes: a third time domain feature, a third frequency domain feature, a third time-frequency domain feature, and a second statistical feature; S233: inputting the third partial discharge feature into a pre-trained third classification model, and the third classification model analyzes the third partial discharge feature to determine a third discharge category, wherein the third discharge category is: internal discharge, floating potential discharge, or poor contact discharge.

6. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 1 is characterized in that: The sub-steps for analyzing the discharge situation based on the current discharge category, equipment data, and environmental data to obtain the comprehensive discharge analysis results are as follows: S31: reading the current discharge category, and randomly generating an analysis sequence number for each discharge category in the current discharge category, the analysis sequence numbers are increased in a random order; S32: taking the discharge category corresponding to the minimum value in the analysis sequence number as the target discharge category, traversing the impact weight database according to the target discharge category, and determining the impact weight data packet whose standard discharge category is consistent with the target discharge category as the target weight data packet, wherein the target weight data packet at least includes: the discharge category hazard value, the hazard degree weight, the equipment factor weight, the weight of each equipment indicator, the environmental factor weight and the weight of each environmental indicator; S33: Calculating a target discharge evaluation value according to the target weight data packet, the device data, and the environmental data; S34: judging the analysis sequence number of the target discharge category according to the total number of discharge categories in the current discharge category; if the total number of discharge categories in the current discharge category is greater than the analysis sequence number of the target discharge category, the analysis sequence number of the target discharge category is eliminated, and S32 is executed; if the total number of discharge categories in the current discharge category is equal to the analysis sequence number of the target discharge category, the analysis sequence number of the target discharge category is eliminated, and S35 is executed; S35: Analyze all target discharge evaluation values ​​to obtain a comprehensive discharge analysis result.

7. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 6 is characterized in that: The expression of the target discharge evaluation value is: Among them, Spg i The target discharge evaluation value corresponding to the target discharge category with the analysis number i; To analyze the hazard weight in the target weight data packet corresponding to the target discharge category with sequence number i; Fys i To analyze the discharge category hazard value in the target weight data packet corresponding to the target discharge category with sequence number i; To analyze the equipment factor weight in the target weight data packet corresponding to the target discharge category with sequence number i; Szb n is the normalized value of the nth device indicator in the device data; η n To analyze the weight of the nth device indicator in the target weight data packet corresponding to the target discharge category with serial number i, n∈[1,N], N is a natural number, representing the total number of device indicators in the device data; To analyze the environmental factor weight in the target weight data packet corresponding to the target discharge category with sequence number i; Hzb m is the normalized value of the mth environmental indicator in the environmental data; m To analyze the weight of the mth environmental indicator in the target weight data packet corresponding to the target discharge category with sequence number i, m∈[1,M], M is a natural number, representing the total number of environmental indicators in the environmental data.

8. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 6 is characterized in that: The sub-steps for analyzing all target discharge evaluation values ​​and obtaining comprehensive discharge analysis results are as follows: S351: traversing the evaluation threshold database according to the target discharge category corresponding to each target discharge evaluation value, and determining a standard threshold table whose standard discharge category is consistent with the target discharge category as the target threshold table; S352: Analyze the corresponding target discharge evaluation value using the target threshold table to generate a sub-discharge result. If the target discharge evaluation value is within the normal threshold range of the target threshold table, the generated sub-discharge result is normal. If the target discharge evaluation value is within the critical threshold range of the target threshold table, the generated sub-discharge result is critical; if the target discharge evaluation value is within the abnormal threshold range of the target threshold table, the generated sub-discharge result is abnormal; S353: generating a comprehensive discharge analysis result according to all sub-discharge results; if all sub-discharge results are normal, the generated comprehensive discharge analysis result is normal; If there are one or more anomalies in all sub-discharge results, the generated comprehensive discharge analysis result is abnormal; If there are one or more criticalities in all sub-discharge results and no abnormalities, the comprehensive discharge analysis result is critical.

9. The comprehensive monitoring and analysis method for partial discharge data of power equipment according to claim 1, characterized in that: The sub-steps to obtain the adjusted monitoring frequency are as follows: S41: Obtain historical monitoring data, and obtain a frequency index value according to the historical monitoring data and the target discharge evaluation value, wherein the historical monitoring data includes: the total number of current monitoring times and the total number of current abnormal times; S42: Traversing the monitoring frequency data table according to the frequency index value, determining that the adjustment frequency value corresponding to the frequency index range to which the frequency index value belongs is the target adjustment frequency value; S43: Obtaining the adjusted monitoring frequency according to the target adjustment frequency value and the preset monitoring frequency, where the adjusted monitoring frequency = the preset monitoring frequency + the target adjustment frequency value.

10. A comprehensive monitoring and analysis system for partial discharge data of power equipment, characterized in that: include: Multiple user terminals and an integrated monitoring and analysis center; Among them, the user terminal is used to send comprehensive monitoring data to the comprehensive monitoring and analysis center according to the preset monitoring frequency; receive comprehensive discharge analysis results and maintenance data; Comprehensive monitoring and analysis center: used to execute the comprehensive monitoring and analysis method of partial discharge data of power equipment as described in any one of claims 1-9.

Images