Power equipment fault diagnosis method and system based on GIS partial discharge map

Abstract

The present invention discloses a method and system for diagnosing power equipment faults based on a GIS partial discharge map, and relates to the technical field of fault detection. The present invention extracts data of partial discharge maps in different areas to construct feature vectors and calculates global statistics, and uses the global statistics to standardize the feature vectors of different areas; determines a divergence value to trigger sensor calibration, and performs a preliminary judgment on the power equipment fault through the calibrated feature data; analyzes the fault through a physical model to determine the type of power equipment fault; calculates the real-time insulation degradation index of the power equipment, and uploads the real-time insulation degradation index to the cloud; aggregates the feature vector statistics in all areas, and updates the global statistics to form a closed loop; the cloud formulates a dynamic threshold adjustment rule based on the received real-time insulation degradation index according to the insulation degradation index, and uses the dynamic threshold adjustment rule to update the insulation degradation index threshold.

Description

Technical Field

[0001] The present invention relates to the technical field of fault detection, and in particular to a method and system for diagnosing power equipment faults based on GIS partial discharge maps. Background Art

[0002] Power equipment faces the dual challenges of safe operation and optimized transformation. As the main substation equipment of the power grid, the safe operation of gas-insulated switchgear (GIS) is an important basis for ensuring the reliable operation of the power system. GIS has the advantages of small footprint, high operational reliability, long maintenance cycle, and easy transportation and installation. However, once a GIS fails, it may cause a large-scale power outage in the system, resulting in significant economic losses. With the development of sensor technology, various signals generated by partial discharge, such as ultra-high frequency (UHF) signals, can be measured more accurately. By processing and analyzing these signals, partial discharge spectra, such as phase-resolved partial discharge spectra (PRPD), are drawn. Researchers began to try to identify different types of partial discharge faults by observing and analyzing the characteristics of these spectra, such as distinguishing between tip discharge, particle discharge, suspended discharge, and air gap discharge.

[0003] In today's partial discharge detection and diagnosis of power equipment, due to the large working area of ​​the power equipment, data from multiple areas cannot be shared, resulting in the diagnostic model in each location being based only on local data and unable to utilize information from other areas, thus affecting the accuracy and generalization ability of the model. Summary of the Invention

[0004] The purpose of the present invention is to provide a method and system for diagnosing power equipment faults based on GIS partial discharge maps to solve the problems raised in the prior art.

[0005] To achieve the above object, the present invention provides the following technical solutions:

[0006] A method for diagnosing faults of power equipment based on GIS partial discharge maps, the method comprising the following steps:

[0007] S100, collecting statistics of all operating areas of the power equipment, collecting partial discharge spectra in different areas, extracting data of the partial discharge spectra in different areas to construct feature vectors and calculate global statistics, and using the global statistics to standardize the feature vectors of different areas;

[0008] Furthermore, the specific steps for standardizing the feature data of different regions using the global mean are as follows:

[0009] S101, count all areas where the power equipment is working, collect partial discharge spectra in different areas, and select data related to discharge physical quantities in the partial discharge spectra in different areas, including the discharge pulse voltage Vp , pulse density per unit time D p , pulse rise time T r and the average discharge charge Q avg ; Use four types of data to construct feature vectors F=[V p , D p , T r ,Q avg ], calculate the local eigenvector statistics in each region, including the mean and standard deviation; aggregate the statistics of different regions to obtain the global eigenvector statistics;

[0010] S102, using global eigenvector statistics to standardize the extracted feature data of different regions, the formula is:

[0011]

[0012] In the formula, F i k ' represents the i-th eigenvector in the k-th region after standardization, F i k represents the i-th eigenvector in the k-th region, u k represents the average value of the eigenvector in the kth region, σ k represents the standard deviation of the eigenvector of the kth region, σ global represents the global standard deviation, u global represents the global average.

[0013] By collecting statistics from all operating areas of power equipment and collecting partial discharge patterns within different areas, comprehensive information on the equipment's operating status can be obtained. Extracting pattern data to construct feature vectors and calculating global statistics helps transform complex discharge pattern information into quantifiable and analyzable characteristic data, providing a foundation for subsequent analysis and judgment. Using global statistics to standardize feature vectors eliminates the influence of data dimensions and scales in different areas, making data from different areas comparable and improving the accuracy of subsequent analysis and calculations.

[0014] S200, calculating the divergence values ​​of the normalized phase-resolved partial discharge (PRPD) distributions in different regions and globally, determining the divergence values ​​to trigger sensor calibration, and performing a preliminary fault determination of the power equipment based on the calibrated characteristic data;

[0015] Furthermore, the specific steps for performing preliminary judgment on power equipment faults using the calibrated characteristic data are as follows:

[0016] S201. Determine regional faults based on real-time data collected over a time window. Calculate regional probability distributions based on phase intervals for local PRPD distributions in different regions. The formula is: regional probability distribution = number of pulses in the phase interval / total number of pulses in the region. Weight the probability distributions for each region in the cloud and average them to generate a global probability distribution. The weights are manually set. Calculate the divergence values ​​of the regional and global PRPD distributions using the regional and global probability distributions. The formula is:

[0017]

[0018] In the formula, D JS k represents the divergence value of the PRPD distribution in region k and the global PRPD distribution, P k represents the probability distribution of region k, P global represents the global probability distribution; M represents the average probability distribution of region k and the global average probability distribution, D KL Indicates the extraction of the DL divergence in the brackets; the calculation formula of M is:

[0019]

[0020] S202, when D JS k <0.1, the PRPD distribution of region k is aligned with the global PRPD distribution, and the sensor does not need to be calibrated. JS k When ≥0.1, it is judged that the PRPD distribution of region k and the global one are not aligned, and sensor calibration is started;

[0021] Calculating the divergence of the normalized, regional and global phase-resolved partial discharge (PRPD) patterns can measure the degree of difference between regional PD patterns and the global pattern. By determining the divergence and triggering sensor calibration, the accuracy and reliability of the collected data can be ensured, providing a more accurate basis for initial diagnosis of power equipment faults.

[0022] S203, extracting the mean and standard deviation of the discharge pulse voltage under the current global normal working state from the standardized eigenvector statistics for testing, and calculating the regional outlier value, the formula is:

[0023]

[0024] In the formula, Z represents the regional outlier value, n represents the number of pulse sampling points in the current time window, represents the normalized mean value of the discharge pulse voltage in the current time window, σVnormal represents the standard deviation of the discharge pulse voltage under global normal working conditions, and u V normal Indicates the average value of the discharge pulse voltage under global normal working conditions;

[0025] S203 , using normal distribution to set a regional abnormality threshold Zy, when |Z|>Zy, determine that a fault exists in the region and mark it, perform the same judgment on all regions, and output a fault mark.

[0026] The output results directly drive the subsequent fault type identification (S300) and global optimization (S400-S500), forming a closed-loop diagnosis process.

[0027] Initial fault assessment based on calibrated characteristic data can promptly detect potential fault hazards before obvious faults occur in power equipment, provide early warning, and help take maintenance measures in advance to prevent further development of the fault, leading to equipment damage or power outages.

[0028] S300: After initially determining that the power equipment has failed, analyze the failure using a physical model to determine the type of failure; calculate the real-time insulation degradation index of the power equipment, and upload the real-time insulation degradation index to the cloud;

[0029] Furthermore, the specific steps for calculating the real-time insulation degradation index of the power equipment are as follows:

[0030] S301. Construct a physical model of the power equipment as follows:

[0031]

[0032] In the formula, V p represents the discharge pulse voltage, E represents the electric field strength at the fault, h represents the material constant, τ represents the ionization time constant, t r represents the pulse rise time, and d represents the size of the fault;

[0033] S302. Extracting the discharge pulse voltage, material constant, ionization time constant, pulse rise time, and anomaly size within the determined fault region and inputting these into a physical model to calculate the electric field strength at the fault region. Determining the characteristics of different partial discharge types based on professional indices in the field of power equipment. Comparing the calculated real-time fault electric field strength with the characteristics of different partial discharge types to determine the fault type of the real-time fault region.

[0034] After initially determining the fault of power equipment, analyzing the fault through physical models can deeply explore the cause and mechanism of the fault, accurately determine the fault type of power equipment, and provide strong support for formulating targeted maintenance plans.

[0035] S303. Construct an insulation degradation index model as follows:

[0036]

[0037] In the model, S(t) represents the insulation degradation index, V p j represents the jth discharge pulse voltage, Q j represents the charge of the jth discharge, E a represents the activation energy of the insulating material, k B represents the Boltzmann constant, T(t) represents the real-time temperature of the device, and N represents the number of discharges; the fault levels are set to S th1 and S th2 ; Use the fault level to judge the real-time fault degree, when S(t) th1 When S th1 ≤S(t) th2 When S(t)≥S th2 When the fault occurs, it is judged as a serious fault; the comprehensive fault type is obtained by combining the fault type and fault severity, and the real-time insulation degradation index is uploaded to the cloud.

[0038] Calculating the real-time insulation degradation index of power equipment and uploading it to the cloud provides real-time visibility into changes in the equipment's insulation performance. This index allows operators to promptly understand the insulation status of equipment, providing a valuable reference for maintenance and repair, ensuring safe and reliable operation.

[0039] S400, aggregate the eigenvector statistics in all regions and update the global statistics to form a closed loop;

[0040] Furthermore, the specific steps for updating the global statistics to form a closed loop are:

[0041] S401. Extract the eigenvector statistics in different regions and aggregate them to update the global statistics. The formula for the global eigenvector mean is:

[0042]

[0043] In the formula, u global represents the global eigenvector mean, A k represents the number of samples in region k, u k represents the mean of the eigenvectors of region k, where K represents the total number of working regions of the power equipment;

[0044] S402: Calculate the intra-group variance within a single region and the inter-group variance between different regions, and combine the two variances to obtain the global eigenvector variance. The formula is:

[0045]

[0046] (σ global ) 2 =(σ in )​​2 +(σ be ) 2 ;

[0047] In the formula, (σ in ) 2 represents the within-group variance within a single region, (σ be ) 2 represents the between-group variance between different regions, (σ global ) 2 represents the global eigenvector variance; u k represents the average value of the eigenvector in the kth region; calculating the global eigenvector statistics according to the time series, and assigning and updating the global eigenvector statistics in S102 to form a closed loop;

[0048] S403: Extract the normal discharge records in all areas, extract all data related to the physical model in the records, calculate the material constants in each record, and calculate the average of all recorded material constants, and use the average as the new material constant h. new Update the material constants of the physical model for all regions.

[0049] Aggregating the eigenvector statistics across all regions and updating the global statistics form a closed loop, enabling the system to continuously adjust and optimize itself based on newly collected data. This allows for timely reflection of changes in the operating status of power equipment, improving the system's ability to monitor and diagnose equipment faults, and forming a dynamic, adaptive monitoring system.

[0050] S500. After receiving the real-time insulation degradation index, the cloud formulates a dynamic threshold adjustment rule according to the insulation degradation index, and updates the insulation degradation index threshold using the dynamic threshold adjustment rule.

[0051] Furthermore, the specific steps for updating the insulation degradation index threshold using the dynamic threshold adjustment rule are as follows:

[0052] S501, after receiving the insulation degradation index in different areas, the cloud uses the cumulative distribution function to calculate the global insulation degradation index G global , the fault level is calculated using the global insulation degradation index, and the formula is:

[0053]

[0054] In the formula, S th1 global and S th2 global Respectively represent the global fault level;

[0055] S502. Localize the calculated global fault level for each region using the following formula:

[0056]

[0057] In the formula, S th1 k and S th2 k represents the localized fault level of the kth region; α represents the adjustment coefficient, which is set manually; the localized fault level is used to assign and update the fault levels in different regions in S303 to form a closed loop.

[0058] The cloud establishes dynamic threshold adjustment rules based on the real-time insulation degradation index it receives, updating the threshold. This allows for flexible adjustment of the fault diagnosis threshold based on the equipment's actual operating conditions and changes in insulation performance. This avoids the potential for misjudgments or missed diagnosis caused by fixed thresholds, improving the accuracy and reliability of fault diagnosis.

[0059] The power equipment fault diagnosis system based on GIS partial discharge map includes a data acquisition module, a data preprocessing module, a fault initial judgment module, a fault type judgment module, a global data update module and a dynamic threshold update module;

[0060] The data acquisition module is used to collect partial discharge spectra in different areas, select data related to discharge physical quantities in the partial discharge spectra in different areas, and construct feature vectors;

[0061] The data preprocessing module is used to calculate the eigenvectors to obtain the eigenvector statistics of different regions and perform standardization;

[0062] The fault preliminary judgment module is used to calculate the divergence value of the phase-resolved partial discharge map PRPD distribution in different regions and the global phase after standardization, judge the divergence value to trigger sensor calibration, and perform preliminary judgment on the power equipment fault based on the calibrated characteristic data;

[0063] The fault type judgment module is used to analyze the fault through a physical model after initially judging the fault of the power equipment to determine the type of fault of the power equipment; calculate the real-time insulation degradation index of the power equipment and upload the real-time insulation degradation index to the cloud;

[0064] The global data update module is used to aggregate the feature vector statistics in all regions and update the global statistics to form a closed loop;

[0065] The dynamic threshold updating module is used in the cloud to formulate a dynamic threshold adjustment rule according to the insulation degradation index after receiving the real-time insulation degradation index, and update the insulation degradation index threshold using the dynamic threshold adjustment rule.

[0066] The fault initial judgment module includes a PRPD distribution alignment unit and a fault marking unit;

[0067] The PRPD distribution alignment unit is used to calculate the divergence value of the regional and global PRPD distributions using the regional probability distribution and the global probability distribution to determine whether the sensor needs calibration;

[0068] The fault marking unit is used to extract the mean and standard deviation of the discharge pulse voltage under the current global normal working state from the standardized eigenvector statistics for inspection, calculate regional outliers, determine whether there is a fault in the region and mark it.

[0069] The fault type judgment module includes a fault type unit and a fault degree unit;

[0070] The fault type unit is used to calculate the electric field strength at the fault location using a physical model, obtain the characteristics of different partial discharge types based on professional indexes in the field of power equipment, and determine the real-time fault type based on the characteristics of different partial discharge types;

[0071] The fault degree unit is used to calculate the insulation degradation index of the fault location and use the fault level to determine the fault degree of the real-time fault location.

[0072] The global data update module includes a global mean unit and a global variance unit;

[0073] The global mean unit is used to extract the eigenvector statistics in different regions and aggregate them to update the global statistics;

[0074] The global variance unit is used to calculate the intra-group variance in a single region and the inter-group variance between different regions, and combine the two variances to obtain the global eigenvector variance.

[0075] Compared with the prior art, the present invention has the following beneficial effects:

[0076] 1. This invention solves the data silo problem by aligning statistical distributions and aggregating global parameters, allowing each region to diagnose data based on a unified benchmark. Furthermore, global parameter aggregation eliminates the need for private data, such as device data, to be shared between regions, thus protecting data security while resolving data silos.

[0077] 2. The dynamic threshold in the present invention is adjusted according to the global aging index distribution, and combined with the regional mutual verification mechanism to filter out false alarms, which greatly reduces the probability of false alarms of power equipment failures.

[0078] 3. When adjusting the local fault level, the present invention calculates the global fault level and uses the global fault level as a basis to locally adjust the local fault levels of different regions. This not only increases the breadth of the fault level in each region and ensures that there are no unique anomalies in the region, but also makes adjustments based on the characteristics of different regions, making the judgment more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 This is a module distribution diagram of the power equipment fault diagnosis system based on GIS partial discharge map of the present invention;

[0080] Figure 2 The figure is a schematic diagram of the steps of the power equipment fault diagnosis method based on GIS partial discharge map of the present invention. DETAILED DESCRIPTION

[0081] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0082] Example: Figure 1-Figure 2 As shown, the present invention provides a technical solution.

[0083] A method for diagnosing faults of power equipment based on GIS partial discharge maps, the method comprising the following steps:

[0084] S100, collecting statistics of all operating areas of the power equipment, collecting partial discharge spectra in different areas, extracting data of the partial discharge spectra in different areas to construct feature vectors and calculate global statistics, and using the global statistics to standardize the feature vectors of different areas;

[0085] The specific steps for standardizing the feature data of different regions using the global mean are:

[0086] S101, count all areas where the power equipment is working, collect partial discharge spectra in different areas, and select data related to discharge physical quantities in the partial discharge spectra in different areas, including the discharge pulse voltage V p , pulse density per unit time D p , pulse rise time T r and the average discharge charge Q avg ; Use four types of data to construct feature vectors F=[V p , D p , T r ,Q avg], calculate the local eigenvector statistics in each region, including the mean and standard deviation; aggregate the statistics of different regions to obtain the global eigenvector statistics;

[0087] S102, using global eigenvector statistics to standardize the extracted feature data of different regions, the formula is:

[0088]

[0089] In the formula, F i k ' represents the i-th eigenvector in the k-th region after standardization, F i k represents the i-th eigenvector in the k-th region, u k represents the average value of the eigenvector in the kth region, σ k represents the standard deviation of the eigenvector of the kth region, σ global represents the global standard deviation, u global represents the global average.

[0090] By collecting statistics from all operating areas of power equipment and collecting partial discharge patterns within different areas, comprehensive information on the equipment's operating status can be obtained. Extracting pattern data to construct feature vectors and calculating global statistics helps transform complex discharge pattern information into quantifiable and analyzable characteristic data, providing a foundation for subsequent analysis and judgment. Using global statistics to standardize feature vectors eliminates the influence of data dimensions and scales in different areas, making data from different areas comparable and improving the accuracy of subsequent analysis and calculations.

[0091] S200, calculating the divergence values ​​of the normalized phase-resolved partial discharge (PRPD) distributions in different regions and globally, determining the divergence values ​​to trigger sensor calibration, and performing a preliminary fault determination of the power equipment based on the calibrated characteristic data;

[0092] The specific steps for preliminary judgment of power equipment faults using calibrated characteristic data are as follows:

[0093] S201. Determine regional faults based on real-time data collected over a time window. Calculate regional probability distributions based on phase intervals for local PRPD distributions in different regions. The formula is: regional probability distribution = number of pulses in the phase interval / total number of pulses in the region. Weight the probability distributions for each region in the cloud and average them to generate a global probability distribution. The weights are manually set. Calculate the divergence values ​​of the regional and global PRPD distributions using the regional and global probability distributions. The formula is:

[0094]

[0095] In the formula, DJS k represents the divergence value of the PRPD distribution in region k and the global PRPD distribution, P k represents the probability distribution of region k, P global represents the global probability distribution; M represents the average probability distribution of region k and the global average probability distribution, D KL Indicates the extraction of the DL divergence in the brackets; the calculation formula of M is:

[0096]

[0097] S202, when D JS k <0.1, the PRPD distribution of region k is aligned with the global PRPD distribution, and the sensor does not need to be calibrated. JS k When ≥0.1, it is judged that the PRPD distribution of region k and the global one are not aligned, and sensor calibration is started;

[0098] Calculating the divergence of the normalized, regional and global phase-resolved partial discharge (PRPD) patterns can measure the degree of difference between regional PD patterns and the global pattern. By determining the divergence and triggering sensor calibration, the accuracy and reliability of the collected data can be ensured, providing a more accurate basis for initial diagnosis of power equipment faults.

[0099] S203, extracting the mean and standard deviation of the discharge pulse voltage under the current global normal working state from the standardized eigenvector statistics for testing, and calculating the regional outlier value, the formula is:

[0100]

[0101] In the formula, Z represents the regional outlier value, n represents the number of pulse sampling points in the current time window, Represents the normalized mean discharge pulse voltage in the current time window, σV normal Indicates the standard deviation of the discharge pulse voltage under global normal working conditions, u V normal Indicates the average value of the discharge pulse voltage under global normal working conditions;

[0102] S203 , using normal distribution to set a regional abnormality threshold Zy, when |Z|>Zy, determine that a fault exists in the region and mark it, perform the same judgment on all regions, and output a fault mark.

[0103] The output results directly drive the subsequent fault type identification (S300) and global optimization (S400-S500), forming a closed-loop diagnosis process.

[0104] Initial fault assessment based on calibrated characteristic data can promptly detect potential fault hazards before obvious faults occur in power equipment, provide early warning, and help take maintenance measures in advance to prevent further development of the fault, leading to equipment damage or power outages.

[0105] S300: After initially determining that the power equipment has failed, analyze the failure using a physical model to determine the type of failure; calculate the real-time insulation degradation index of the power equipment, and upload the real-time insulation degradation index to the cloud;

[0106] The specific steps for calculating the real-time insulation degradation index of power equipment are as follows:

[0107] S301. Construct a physical model of the power equipment as follows:

[0108]

[0109] In the formula, V p represents the discharge pulse voltage, E represents the electric field strength at the fault, h represents the material constant, τ represents the ionization time constant, t r represents the pulse rise time, and d represents the size of the fault;

[0110] S302. Extracting the discharge pulse voltage, material constant, ionization time constant, pulse rise time, and anomaly size within the determined fault region and inputting these into a physical model to calculate the electric field strength at the fault region. Determining the characteristics of different partial discharge types based on professional indices in the field of power equipment. Comparing the calculated real-time fault electric field strength with the characteristics of different partial discharge types to determine the fault type of the real-time fault region.

[0111] After initially determining the fault of power equipment, analyzing the fault through physical models can deeply explore the cause and mechanism of the fault, accurately determine the fault type of power equipment, and provide strong support for formulating targeted maintenance plans.

[0112] S303. Construct an insulation degradation index model as follows:

[0113]

[0114] In the model, S(t) represents the insulation degradation index, V p j represents the jth discharge pulse voltage, Q j represents the charge of the jth discharge, E a represents the activation energy of the insulating material, k B represents the Boltzmann constant, T(t) represents the real-time temperature of the device, and N represents the number of discharges; the fault levels are set to S th1 and S th2; Use the fault level to judge the real-time fault degree, when S(t) th1 When S th1 ≤S(t) th2 When S(t)≥S th2 When the fault occurs, it is judged as a serious fault; the comprehensive fault type is obtained by combining the fault type and fault severity, and the real-time insulation degradation index is uploaded to the cloud.

[0115] Calculating the real-time insulation degradation index of power equipment and uploading it to the cloud provides real-time visibility into changes in the equipment's insulation performance. This index allows operators to promptly understand the insulation status of equipment, providing a valuable reference for maintenance and repair, ensuring safe and reliable operation.

[0116] S400, aggregate the eigenvector statistics in all regions and update the global statistics to form a closed loop;

[0117] The specific steps to update the global statistics to form a closed loop are:

[0118] S401. Extract the eigenvector statistics in different regions and aggregate them to update the global statistics. The formula for the global eigenvector mean is:

[0119]

[0120] In the formula, u global represents the global eigenvector mean, A k represents the number of samples in region k, u k represents the mean of the eigenvectors of region k, where K represents the total number of working regions of the power equipment;

[0121] S402: Calculate the intra-group variance within a single region and the inter-group variance between different regions, and combine the two variances to obtain the global eigenvector variance. The formula is:

[0122]

[0123]

[0124] (σ global ) 2 =(σ in ) 2 +(σ be ) 2 ;

[0125] In the formula, (σ in ) 2 represents the within-group variance within a single region, (σ be ) 2 represents the between-group variance between different regions, (σ​​global ) 2 represents the global eigenvector variance; u k represents the average value of the eigenvector in the kth region; calculating the global eigenvector statistics according to the time series, and assigning and updating the global eigenvector statistics in S102 to form a closed loop;

[0126] S403: Extract the normal discharge records in all areas, extract all data related to the physical model in the records, calculate the material constants in each record, and calculate the average of all recorded material constants, and use the average as the new material constant h. new Update the material constants of the physical model for all regions.

[0127] Aggregating the eigenvector statistics across all regions and updating the global statistics form a closed loop, enabling the system to continuously adjust and optimize itself based on newly collected data. This allows for timely reflection of changes in the operating status of power equipment, improving the system's ability to monitor and diagnose equipment faults, and forming a dynamic, adaptive monitoring system.

[0128] S500. After receiving the real-time insulation degradation index, the cloud formulates a dynamic threshold adjustment rule according to the insulation degradation index, and updates the insulation degradation index threshold using the dynamic threshold adjustment rule.

[0129] The specific steps for updating the insulation degradation index threshold using the dynamic threshold adjustment rule are as follows:

[0130] S501, after receiving the insulation degradation index in different areas, the cloud uses the cumulative distribution function to calculate the global insulation degradation index G global , the fault level is calculated using the global insulation degradation index, and the formula is:

[0131]

[0132] In the formula, S th1 global and S th2 global Respectively represent the global fault level;

[0133] S502. Localize the calculated global fault level for each region using the following formula:

[0134]

[0135] In the formula, S th1 k and S th2 krepresents the localized fault level of the kth region; α represents the manually set adjustment coefficient. The localized fault level is used to assign and update the fault levels of the different regions in S303, forming a closed loop. The cloud establishes dynamic threshold adjustment rules based on the received real-time insulation degradation index and updates the insulation degradation index threshold. This allows for flexible adjustment of the fault diagnosis threshold based on the actual operating conditions of the equipment and changes in insulation performance. This avoids the potential for misjudgment or missed diagnosis caused by fixed thresholds, improving the accuracy and reliability of fault diagnosis.

[0136] The power equipment fault diagnosis system based on GIS partial discharge map includes a data acquisition module, a data preprocessing module, a fault initial judgment module, a fault type judgment module, a global data update module and a dynamic threshold update module;

[0137] The data acquisition module is used to collect partial discharge spectra in different areas, select data related to discharge physical quantities in the partial discharge spectra in different areas, and construct feature vectors;

[0138] The data preprocessing module is used to calculate the eigenvectors to obtain the eigenvector statistics of different regions and perform standardization;

[0139] The fault preliminary judgment module is used to calculate the divergence value of the phase-resolved partial discharge map PRPD distribution in different regions and the global phase after standardization, judge the divergence value to trigger sensor calibration, and perform preliminary judgment on the power equipment fault based on the calibrated characteristic data;

[0140] The fault type judgment module is used to analyze the fault through a physical model after initially judging the fault of the power equipment to determine the type of fault of the power equipment; calculate the real-time insulation degradation index of the power equipment and upload the real-time insulation degradation index to the cloud;

[0141] The global data update module is used to aggregate the feature vector statistics in all regions and update the global statistics to form a closed loop;

[0142] The dynamic threshold updating module is used in the cloud to formulate a dynamic threshold adjustment rule according to the insulation degradation index after receiving the real-time insulation degradation index, and update the insulation degradation index threshold using the dynamic threshold adjustment rule.

[0143] The fault initial judgment module includes a PRPD distribution alignment unit and a fault marking unit;

[0144] The PRPD distribution alignment unit is used to calculate the divergence value of the regional and global PRPD distributions using the regional probability distribution and the global probability distribution to determine whether the sensor needs calibration;

[0145] The fault marking unit is used to extract the mean and standard deviation of the discharge pulse voltage under the current global normal working state from the standardized eigenvector statistics for inspection, calculate regional outliers, determine whether there is a fault in the region and mark it.

[0146] The fault type judgment module includes a fault type unit and a fault degree unit;

[0147] The fault type unit is used to calculate the electric field strength at the fault location using a physical model, obtain the characteristics of different partial discharge types based on professional indexes in the field of power equipment, and determine the real-time fault type based on the characteristics of different partial discharge types;

[0148] The fault degree unit is used to calculate the insulation degradation index of the fault location and use the fault level to determine the fault degree of the real-time fault location.

[0149] The global data update module includes a global mean unit and a global variance unit;

[0150] The global mean unit is used to extract the eigenvector statistics in different regions and aggregate them to update the global statistics;

[0151] The global variance unit is used to calculate the intra-group variance in a single region and the inter-group variance between different regions, and combine the two variances to obtain the global eigenvector variance.

[0152] Example 1: When making a preliminary fault assessment for partial discharge in power equipment, the characteristics of different partial discharge types are as follows:

[0153]

[0154] Assume that the data for inputting the physical model and insulation degradation model are: V p =70mv, t r =1.5μS, T(t)=313K, h=1.2, τ=0.2μS, d=0.05mm, Q=50; E=12kV / mm calculated in the physical model, and the calculated insulation degradation index S(t)=1500;

[0155] The fault type was initially determined to be surface discharge;

[0156] Assume that the fault levels are S th1 =1000, S th2 =2500; the fault degree is judged to be a moderate fault; and the final fault type is "surface discharge-moderate fault".

[0157] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

Claims

1. A method for diagnosing power equipment faults based on GIS partial discharge maps, characterized by: The method comprises the following steps: S100, collecting statistics of all operating areas of the power equipment, collecting partial discharge spectra in different areas, extracting data of the partial discharge spectra in different areas to construct feature vectors and calculate global statistics, and using the global statistics to standardize the feature vectors of different areas; S200, calculating the divergence values ​​of the normalized phase-resolved partial discharge (PRPD) distributions in different regions and globally, determining the divergence values ​​to trigger sensor calibration, and performing a preliminary fault determination of the power equipment based on the calibrated characteristic data; The specific steps for preliminary judgment of power equipment faults using calibrated characteristic data are as follows: S201. Determine regional faults based on real-time data collected over a time window. Calculate regional probability distributions based on phase intervals for local PRPD distributions in different regions. The formula is: regional probability distribution = number of pulses in the phase interval / total number of pulses in the region. Weight the probability distributions for each region in the cloud and average them to generate a global probability distribution. The weights are manually set. Calculate the divergence values ​​of regional and global PRPD distributions using the regional and global probability distributions. The formula is: ； In the formula, D JS k represents the divergence value of the PRPD distribution in region k and the global PRPD distribution, P k represents the probability distribution of region k, P global represents the global probability distribution; M represents the average probability distribution of region k and the global average probability distribution, D KL Indicates the extraction of the DL divergence in the brackets; the calculation formula of M is: ; S202, when D JS k <0.1, the PRPD distribution of region k is aligned with the global PRPD distribution, and the sensor does not need to be calibrated. JS k When ≥0.1, it is judged that the PRPD distribution of region k and the global one are not aligned, and sensor calibration is started; S203, extracting the mean and standard deviation of the discharge pulse voltage under the current global normal working state from the standardized eigenvector statistics for testing, and calculating the regional outlier value, the formula is: ； In the formula, Z represents the regional outlier value, n represents the number of pulse sampling points in the current time window, and `V p represents the normalized mean discharge pulse voltage in the current time window, σ V normal Indicates the standard deviation of the discharge pulse voltage under global normal working conditions, u V normal Indicates the average value of the discharge pulse voltage under global normal working conditions; S203. Set a regional abnormality threshold Zy using normal distribution. When |Z|>Zy, determine that a fault exists in the region and mark it. Perform the same judgment on all regions and output a fault mark. S300: After initially determining that the power equipment has failed, analyze the failure using a physical model to determine the type of failure; calculate the real-time insulation degradation index of the power equipment, and upload the real-time insulation degradation index to the cloud; The specific steps for calculating the real-time insulation degradation index of power equipment are as follows: S301. Construct a physical model of the power equipment as follows: ； In the formula, V p represents the discharge pulse voltage, E represents the electric field strength at the fault, h represents the material constant, τ represents the ionization time constant, t r represents the pulse rise time, and d represents the size of the fault; S302. Extracting the discharge pulse voltage, material constant, ionization time constant, pulse rise time, and anomaly size within the determined fault region and inputting these into a physical model to calculate the electric field strength at the fault region. Determining the characteristics of different partial discharge types based on professional indices in the field of power equipment. Comparing the calculated real-time fault electric field strength with the characteristics of different partial discharge types to determine the fault type of the real-time fault region. S303. Construct an insulation degradation index model as follows: ； In the model, S(t) represents the insulation degradation index, V p j represents the jth discharge pulse voltage, Q j represents the charge of the jth discharge, E a represents the activation energy of the insulating material, k B represents the Boltzmann constant, T(t) represents the real-time temperature of the device, and N represents the number of discharges; the fault levels are set to S th1 and S th2 ; Use the fault level to judge the real-time fault degree, when S(t) th1 When S th1 ≤S(t) th2 When S(t)≥S th2 When the fault is detected, it is judged as a serious fault; the comprehensive fault type is obtained by combining the fault type and fault severity, and the real-time insulation degradation index is uploaded to the cloud;​​ S400, aggregate the eigenvector statistics in all regions and update the global statistics to form a closed loop; S500. After receiving the real-time insulation degradation index, the cloud formulates a dynamic threshold adjustment rule according to the insulation degradation index, and updates the insulation degradation index threshold using the dynamic threshold adjustment rule.

2. The method for diagnosing power equipment faults based on GIS partial discharge maps according to claim 1, characterized in that: The specific steps of using the global mean to standardize the feature data of different regions in S100 are: S101, count all areas where the power equipment is working, collect partial discharge spectra in different areas, and select data related to discharge physical quantities in the partial discharge spectra in different areas, including the discharge pulse voltage V p , pulse density per unit time D p , pulse rise time T r and the average discharge charge Q avg ; Use four types of data to construct feature vectors F=[V p , D p , T r ,Q avg ], calculate the local feature vector statistics in each area, the statistics include the mean value and standard deviation; use the statistics of different areas to aggregate to obtain the global feature vector statistics; S102, use the global feature vector statistics to standardize the feature data extracted from different areas, the formula is: ； In the formula, F i k ' represents the i-th eigenvector in the k-th region after standardization, F i k represents the i-th eigenvector in the k-th region, u k represents the average value of the eigenvector in the kth region, σ k represents the standard deviation of the eigenvector of the kth region, σ global represents the global standard deviation, u global represents the global average.

3. The power equipment fault diagnosis method based on GIS partial discharge spectrum according to claim 2 is characterized in that: The specific steps of updating the global statistics to form a closed loop in S400 are: S401. Extract the eigenvector statistics in different regions and aggregate them to update the global statistics. The formula for the global eigenvector mean is: ； In the formula, u global represents the global eigenvector mean, A k represents the number of samples in region k, u k represents the mean of the eigenvectors of region k, where K represents the total number of working regions of the power equipment; S402: Calculate the intra-group variance within a single region and the inter-group variance between different regions, and combine the two variances to obtain the global eigenvector variance. The formula is: ； ； ； In the formula, (σ in ) 2 represents the within-group variance within a single region, (σ be ) 2 represents the between-group variance between different regions, (σ global ) 2 represents the global eigenvector variance; u k represents the average value of the eigenvector in the kth region; calculating the global eigenvector statistics according to the time series, and assigning and updating the global eigenvector statistics in S102 to form a closed loop; S403: Extract the normal discharge records in all areas, extract all data related to the physical model in the records, calculate the material constants in each record, and calculate the average of all recorded material constants, and use the average as the new material constant h. new Update the material constants of the physical model for all regions.

4. The method for diagnosing power equipment faults based on GIS partial discharge maps according to claim 3 is characterized in that: The specific steps of updating the insulation degradation index threshold using the dynamic threshold adjustment rule in S500 are: S501, after receiving the insulation degradation index in different areas, the cloud uses the cumulative distribution function to calculate the global insulation degradation index G global , the fault level is calculated using the global insulation degradation index, and the formula is: ； In the formula, S th1 global and S th2 global Respectively represent the global fault level; S502. Localize the calculated global fault level for each region using the following formula: ; In the formula, S th1 k and S th2 k represents the localized fault level of the kth region; α represents the adjustment coefficient, which is set manually; the localized fault level is used to assign and update the fault levels in different regions in S303 to form a closed loop.

5. A power equipment fault diagnosis system based on GIS partial discharge maps, using the power equipment fault diagnosis method based on GIS partial discharge maps according to any one of claims 1 to 4, characterized in that: The power equipment fault diagnosis system includes a data acquisition module, a data preprocessing module, a fault initial judgment module, a fault type judgment module, a global data update module and a dynamic threshold update module; The data acquisition module is used to collect partial discharge spectra in different areas, select data related to discharge physical quantities in the partial discharge spectra in different areas, and construct feature vectors; The data preprocessing module is used to calculate the eigenvectors to obtain the eigenvector statistics of different regions and perform standardization; The fault preliminary judgment module is used to calculate the divergence value of the phase-resolved partial discharge map PRPD distribution in different regions and the global phase after standardization, judge the divergence value to trigger sensor calibration, and perform preliminary judgment on the power equipment fault based on the calibrated characteristic data; The fault type judgment module is used to analyze the fault through a physical model after initially judging the fault of the power equipment to determine the type of fault of the power equipment; calculate the real-time insulation degradation index of the power equipment and upload the real-time insulation degradation index to the cloud; The global data update module is used to aggregate the feature vector statistics in all regions and update the global statistics to form a closed loop; The dynamic threshold updating module is used in the cloud to formulate a dynamic threshold adjustment rule according to the insulation degradation index after receiving the real-time insulation degradation index, and update the insulation degradation index threshold using the dynamic threshold adjustment rule.

6. The power equipment fault diagnosis system based on GIS partial discharge map according to claim 5 is characterized in that: The fault initial judgment module includes a PRPD distribution alignment unit and a fault marking unit; The PRPD distribution alignment unit is used to calculate the divergence value of the regional and global PRPD distributions using the regional probability distribution and the global probability distribution to determine whether the sensor needs calibration; The fault marking unit is used to extract the mean and standard deviation of the discharge pulse voltage under the current global normal working state from the standardized eigenvector statistics for inspection, calculate regional outliers, determine whether there is a fault in the region and mark it.

7. The power equipment fault diagnosis system based on GIS partial discharge map according to claim 5 is characterized in that: The fault type judgment module includes a fault type unit and a fault degree unit; The fault type unit is used to calculate the electric field strength at the fault location using a physical model, obtain the characteristics of different partial discharge types based on professional indexes in the field of power equipment, and determine the real-time fault type based on the characteristics of different partial discharge types; The fault degree unit is used to calculate the insulation degradation index of the fault location and use the fault level to determine the fault degree of the real-time fault location.

8. The power equipment fault diagnosis system based on GIS partial discharge map according to claim 5 is characterized by: The global data update module includes a global mean unit and a global variance unit; The global mean unit is used to extract the eigenvector statistics in different regions and aggregate them to update the global statistics; The global variance unit is used to calculate the intra-group variance in a single region and the inter-group variance between different regions, and combine the two variances to obtain the global eigenvector variance.

Images