A power grid inspection method and system based on power private network data processing
By establishing a fault classifier and a power sensor system, rapid and accurate identification and repair of power grid faults were achieved, solving the tool carrying problem caused by the uncertainty of fault repair in existing technologies and improving the efficiency of power grid fault repair.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HUBEI ELECTRIC POWER CO JINGZHOU POWER SUPPLY CO
- Filing Date
- 2022-12-13
- Publication Date
- 2026-06-12
AI Technical Summary
The existing power grid has a wide variety of faults, and it is difficult to determine the repair methods and tools required for different faults. This results in inspection personnel carrying a large number of tools or prolonging the fault repair time, and may even cause economic losses.
A fault classifier is established, multiple fault databases are set up, data is collected in real time through power sensors, scale normalization processing and feature algorithm calculation are performed to identify fault categories, and fault areas are determined based on sensor numbers to provide corresponding repair methods.
It enables rapid and accurate fault identification and repair, reduces inspection intensity, improves fault repair efficiency, and reduces delays caused by improper tool carrying.
Smart Images
Figure CN116032003B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of power grid detection technology, specifically, it relates to a power grid inspection method and system based on power grid data processing. Background Technology
[0002] The power system is a crucial infrastructure of modern society, and its safe and reliable operation is essential for the normal functioning of people's social and economic lives. Electricity travels through stages such as generation, transmission, and distribution from production to use. In the transmission stage, transmission lines act as the "main arteries" of the power system, carrying electrical energy.
[0003] With the increasing frequency of extreme weather events, large-scale power outages caused by extreme weather are becoming more frequent, severely impacting people's lives and production. When extreme weather causes multiple faults in the power distribution system, the fault indicators can initially determine the fault location. However, to further pinpoint the exact location of the fault and implement load restoration operations, it is necessary to dispatch inspection personnel to inspect the fault area. Since the speed of load restoration is largely affected by the efficiency of the inspection personnel, and multiple fault areas often exist in the system under extreme weather conditions, it is crucial to optimize the scheduling of inspection personnel to obtain the inspection personnel scheduling scheme with the shortest inspection time for all lines.
[0004] Therefore, to ensure the safe operation of the power grid, inspection is an indispensable task. Currently, power grid inspection mainly relies on manual inspections conducted by inspection personnel on a regular and scheduled basis. This has the following problems: due to constraints from various factors such as weather conditions, environmental factors, personnel quality, and sense of responsibility, the quality of inspections cannot be guaranteed; information reflecting the operating status and equipment defects cannot be fed back in a timely manner; and when a power grid fault occurs, it needs to be repaired. However, power grid faults are diverse, and different faults require different repair methods and tools. When inspection personnel go to repair, it is difficult to determine which type of fault it is, which requires carrying a large number of repair tools, or going to the fault site to determine the fault and then notifying other personnel to bring the corresponding repair tools, which prolongs the repair time of the power grid fault and may even cause serious economic losses. Summary of the Invention
[0005] The problem to be solved
[0006] Given the diverse nature of existing power grid faults and the different repair methods and tools required for each, it is difficult for inspection personnel to determine the specific fault when they go to the site for repairs. This necessitates carrying a large number of repair tools or going to the fault site to identify the fault before notifying other personnel to bring the corresponding repair tools, which prolongs the repair time of power grid faults and may even cause serious economic losses. This invention provides a power grid inspection method and system based on power dedicated network data processing.
[0007] Technical solution
[0008] To solve the above problems, the present invention adopts the following technical solution.
[0009] A power grid inspection method based on power private network data processing includes the following steps:
[0010] Step 1: Establish a fault classifier. The fault classifier contains multiple fault databases, and each fault database corresponds to a fault signal class.
[0011] Step 2: Obtain historical operation data of the power grid, filter the historical data, extract fault assessment indicators from the historical data, determine the type of fault signal according to the fault assessment indicators, and store the historical fault signals in the corresponding fault database.
[0012] Step 3: Install power sensors on the dedicated power grid lines, number the power sensors according to the installation area and line, and collect power signal data of the dedicated power grid in real time under stable operating conditions.
[0013] Step 4: When the power sensor sends abnormal power signal data, calculate the raw data of the power data characteristics of the abnormal power signal data.
[0014] Step 5: Perform scale normalization on the raw data of the extracted power data features to normalize the raw data of the power data features to data with a mean of 0 and a variance of 1.
[0015] Step 6: Use feature algorithms to calculate the scale-normalized power feature data to obtain the abnormal power signal data feature data;
[0016] Step 7: Calculate the similarity between the power signal data feature data and the features of each fault database. Based on the calculated similarity results, determine the category to which the abnormal power signal data fault belongs, and complete the identification of abnormal power signal data faults.
[0017] Step 8: Staff determine the area of the fault based on the power sensor number that sent the abnormal data, and take corresponding repair measures based on the fault identification results of the abnormal power signal data.
[0018] Preferably, the nearest neighbor prototype classification algorithm of orthogonal quantum particle swarm optimization is used to extract the measurement fault assessment index from historical data. This algorithm employs a multi-collapse-orthogonal crossover quantum particle swarm optimization to iteratively optimize and select the best particles and effective prototypes. The distance from the test data to the selected prototypes is calculated to classify the test data. The training data is then used to learn, and the fitness value of each particle is obtained until the termination condition is met. The classification accuracy of the population is then calculated to obtain the final result. Furthermore, the particle swarm optimization algorithm formula is as follows:
[0019] Vi=Vi+C1×rand()×(pbesti-Xi)+C2×rand()×(gbesti-Xi)
[0020] Xi = Xi + Vi
[0021] Where i = 1, 2, 3...N, N is the total number of particles, Vi is the velocity of the particles, rand() is a random number between (0, 1), Xi is the current position of the particles, C1 and C2 are learning factors, in general C1 = C2 = 2, the maximum value of Vi is Vmax, if Vi is greater than Vmax, then Vi = Vmax.
[0022] Preferably, the scale normalization formula for the scale normalization process is as follows:
[0023]
[0024]
[0025] Where z represents the scale-normalized data, and x represents the original data. δ is the mean of the original data, and δ is the variance of the original data.
[0026] Furthermore, the calculation of the scale-normalized power characteristic data using the feature algorithm is based on the assumption that the abnormal power signal data matrix is X. m×n Where m is the data signal quantity and n is the number of features, the mean of matrix X is first centered to determine the principal direction w with the largest projection variance, as shown in the following formula:
[0027]
[0028] ||w|| 2 =1
[0029] Extract the first k principal directions to form the load matrix, W = (w1,...w k The result after feature extraction of abnormal power signal data is X. new =XW.
[0030] Preferably, the similarity calculation between the abnormal power signal data and historical fault signals in the fault database is performed using the correlation coefficient similarity, and the calculation formula is as follows:
[0031]
[0032]
[0033] Where X represents abnormal power signal data and Y represents historical fault signals.
[0034] Preferably, the fault database also includes a diagnostic and repair database, which stores the cause, path, and solution corresponding to the category to which the fault signal belongs. After the abnormal power signal data fault is identified, the corresponding cause, path, and solution will be displayed to the staff. The diagnostic and repair database will regularly modify and delete old data and update and reorganize the data in the diagnostic and repair database.
[0035] Preferably, when abnormal power signal data cannot be identified in the fault class and fault database within the fault classifier, a professional will conduct an assessment and diagnosis. At the same time, based on the assessment and diagnosis results, a new fault signal class and fault database will be established in the fault classifier, and the assessment and diagnosis results will be saved to the fault database.
[0036] Furthermore, after the abnormal power signal data fault identification is completed, it is stored in the corresponding fault database according to the fault signal category. The historical fault signals in the fault database are updated. Professionals will not only conduct assessments and diagnoses to find new fault causes, paths and solutions, but also regularly modify and delete old data in the fault database and reorganize the data in the fault database.
[0037] A power grid inspection system based on private power grid data processing includes:
[0038] The fault classification module is used to establish multiple fault databases and store historical fault data into the corresponding fault databases according to the category to which the fault signal belongs.
[0039] The fault assessment module is used to filter historical data of the power grid, extract fault assessment indicators from the historical data, and determine the category of the fault signal according to the fault assessment indicators.
[0040] The data acquisition module is used to install power sensors on the dedicated power grid lines to collect power signal data of the dedicated power grid under stable operating conditions in real time.
[0041] The data calculation module is used to calculate the raw data of the power data characteristics of abnormal power signal data.
[0042] The scale normalization module is used to perform scale normalization processing on the raw data of the calculated and extracted power data features;
[0043] The feature calculation module is used to calculate the power feature data after scale normalization to obtain the feature data of abnormal power signal data.
[0044] The fault diagnosis module is used to calculate the similarity between the power signal data feature data and the features of each fault database, and to determine the type of fault to which the abnormal power signal data belongs.
[0045] The results display module is used to display the corresponding fault area and repair methods based on the fault type of the abnormal power signal data.
[0046] A power grid inspection method and system based on power grid data processing involves establishing a fault classifier with multiple fault databases, each corresponding to a fault signal category. Historical operational data from the power grid is acquired, filtered, and fault assessment indicators are extracted. The historical fault data is then classified according to these indicators and stored in the corresponding fault database. Power sensors are installed on the power grid lines, numbered according to their installation area and line. These sensors continuously collect power signal data under stable operating conditions. When an abnormal power signal is detected, calculations are performed on the abnormal power signal data. The raw data for calculating the power data characteristics of the abnormal power signal is then processed. The raw power data features are scale-normalized to a mean of 0 and a variance of 1. A feature algorithm is then used to calculate the abnormal power signal data features. The similarity between these features and the features in each fault database is calculated. Based on the similarity results, the fault category of the abnormal power signal data is determined, thus completing the abnormal power signal data fault identification. Workers determine the fault location based on the power sensor number that sent the abnormal data and take corresponding repair measures based on the fault identification results. This accurate fault diagnosis reduces the workload of workers during inspections, allowing them to carry the necessary repair tools promptly and improving the efficiency of fault repair.
[0047] Beneficial effects
[0048] Compared with the prior art, the beneficial effects of the present invention are as follows:
[0049] (1) Before calculating the raw data of abnormal power signal data, the present invention performs scale normalization processing to eliminate the dimensional influence between various characteristic indicators of the raw data of abnormal power signal data. After the raw data of abnormal power signal data is processed by data standardization, each indicator is at the same order of magnitude, which is suitable for comprehensive comparison and evaluation.
[0050] (2) The present invention uses the nearest neighbor prototype classification algorithm of orthogonal quantum particle swarm to extract the measurement fault assessment index from historical data. This classification algorithm can accurately obtain the fault assessment index, making the fault identification signal and subsequent calculation results more accurate, and obtaining better assessment and diagnosis results.
[0051] (3) The present invention stores the abnormal power signal data that has completed fault identification into the corresponding fault database according to the fault category. Then, the abnormal power signal data can be used as historical fault signals to participate in the calculation and fault identification of the next abnormal power data. As the number of times it is used increases, the standard fault library in the fault classifier and the historical fault data in the fault database gradually increase, and the fault classification results of abnormal power data become more and more accurate. Attached Figure Description
[0052] To more clearly illustrate the technical solutions in the embodiments or examples of this application, the accompanying drawings used in the embodiments or examples will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other drawings can be obtained according to these drawings without creative effort.
[0053] Figure 1 This is a schematic diagram of the steps of the present invention;
[0054] Figure 2 This is a schematic diagram of the process of the present invention;
[0055] Figure 3 This is a schematic diagram of the system structure of the present invention. Detailed Implementation
[0056] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some embodiments of this application, but not all embodiments. Generally, the components of the embodiments of this application described and shown in the accompanying drawings can be arranged and designed in various different configurations.
[0057] Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of the application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.
[0058] Example 1
[0059] like Figure 1 and Figure 2 As shown, a power grid inspection method based on power grid data processing has the following main steps:
[0060] A fault classifier is established, which contains multiple fault databases. Each fault database corresponds to a fault signal category. Historical operating data of the power grid is obtained, the historical data is filtered, fault assessment indicators are extracted from the historical data, the historical fault data are classified according to the fault assessment indicators, and the historical fault signals are stored in the corresponding fault databases.
[0061] The fault database also includes a diagnostic and repair database, which stores the causes, paths, and solutions corresponding to the category of the fault signal. After the abnormal power signal data fault is identified, the corresponding causes, paths, and solutions will be displayed to the staff. The diagnostic and repair database will regularly modify and delete old data and update and reorganize the data in the diagnostic and repair database.
[0062] The fault assessment index is extracted from historical data using a nearest-neighbor prototype classification algorithm based on orthogonal quantum particle swarm optimization. This algorithm employs multiple collapse-orthogonal crossover quantum particle swarm optimization to select the best particles and effective prototypes. The distance from the test data to the selected prototypes is calculated to classify the test data. The training data is then used to learn, and the fitness value of each particle is obtained until a termination condition is met. The classification accuracy of the population is then calculated to obtain the final result. The particle swarm optimization algorithm formula is as follows:
[0063] Vi=Vi+C1×rand()×(pbesti-Xi)+C2×rand()×(gbesti-Xi)
[0064] Xi = Xi + Vi
[0065] Where i = 1, 2, 3...N, N is the total number of particles, Vi is the velocity of the particles, rand() is a random number between (0, 1), Xi is the current position of the particles, C1 and C2 are learning factors, in general C1 = C2 = 2, the maximum value of Vi is Vmax, if Vi is greater than Vmax, then Vi = Vmax.
[0066] Power sensors are installed on dedicated power grid lines and numbered according to their installation area and line. The power sensors collect power signal data of the dedicated power grid in real time under stable operating conditions. When the power sensors send abnormal power signal data, the abnormal power signal data is calculated to obtain the raw data of the power data characteristics of the abnormal power signal data.
[0067] The raw data of the extracted power data features are subjected to scale normalization processing to normalize the raw data of the power data features to data with a mean of 0 and a variance of 1. The scale normalization formula for the scale normalization processing is as follows:
[0068]
[0069]
[0070] Where z represents the scale-normalized data, and x represents the original data. δ is the mean of the original data, and δ is the variance of the original data.
[0071] Feature algorithms are used to calculate the scale-normalized power characteristic data to obtain the characteristic data of abnormal power signals. The calculation of the abnormal power signal data matrix using feature algorithms is performed with the feature algorithm denoted as X. m×n Where m is the data signal quantity and n is the number of features, the mean of matrix X is first centered to determine the principal direction w with the largest projection variance, as shown in the following formula:
[0072]
[0073] ||w|| 2 =1
[0074] Extract the first k principal directions to form the load matrix, W = (w1,...w k The result after feature extraction of abnormal power signal data is X. new =XW.
[0075] The similarity between the power signal data features and the features of each fault database is calculated. Based on the calculated similarity results, the fault category of the abnormal power signal data is determined, thus completing the identification of abnormal power signal data faults. The similarity between the abnormal power signal data and historical fault signals in the fault database is calculated using the correlation coefficient similarity, and the calculation formula is as follows:
[0076]
[0077]
[0078] Where X represents abnormal power signal data and Y represents historical fault signals.
[0079] Staff members determine the area of the fault based on the power sensor number that sends abnormal data, and take corresponding repair measures based on the fault identification results of the abnormal power signal data.
[0080] When abnormal power signal data cannot be identified into the corresponding fault signal category and fault database within the fault classifier, professional personnel conduct an assessment and diagnosis. Based on the assessment and diagnosis results, a new fault signal category and fault database are established within the fault classifier, and the assessment and diagnosis results are saved to this fault database. After the abnormal power signal data is identified as a fault, it is stored in the corresponding fault database according to the fault signal category. The historical fault signals in the fault database are updated. Professional personnel not only conduct assessments and diagnoses to find new fault causes, paths, and solutions, but also regularly modify and delete old data in the fault database and reorganize the data in the fault database.
[0081] As described above, in this example, a fault classifier is established, containing multiple fault databases. Each fault database corresponds to a fault signal category. Historical operating data of the power grid is acquired, filtered, and fault assessment indicators are extracted. The historical fault data is then classified according to these indicators, and the fault signals are stored in their corresponding fault databases. Power sensors are installed on the power grid lines, numbered according to their installation area and line. These sensors collect power signal data in real time under stable operating conditions. When a power sensor sends abnormal power signal data, the abnormal power signal data is calculated to determine the anomaly. The raw power data features of the power signal data are obtained through a scale normalization process. This normalizes the raw power data features to a mean of 0 and a variance of 1. A feature algorithm is then used to calculate the abnormal power signal data features. The similarity between these features and the features in each fault database is calculated. Based on the similarity results, the fault category of the abnormal power signal data is determined, thus completing the abnormal power signal data fault identification. Workers then determine the fault location based on the power sensor number that sent the abnormal data and implement corresponding repair measures according to the fault identification results.
[0082] Example 2
[0083] like Figure 3 As shown, a power grid inspection system based on power grid data processing includes:
[0084] The fault classification module is used to establish multiple fault databases and store historical fault data into the corresponding fault databases according to the category to which the fault signal belongs.
[0085] The fault assessment module is used to filter historical data of the power grid, extract fault assessment indicators from the historical data, and determine the category of the fault signal according to the fault assessment indicators.
[0086] The data acquisition module is used to install power sensors on the dedicated power grid lines to collect power signal data of the dedicated power grid under stable operating conditions in real time.
[0087] The data calculation module is used to calculate the raw data of the power data characteristics of abnormal power signal data.
[0088] The scale normalization module is used to perform scale normalization processing on the raw data of the calculated and extracted power data features;
[0089] The feature calculation module is used to calculate the power feature data after scale normalization to obtain the feature data of abnormal power signal data.
[0090] The fault diagnosis module is used to calculate the similarity between the power signal data feature data and the features of each fault database, and to determine the type of fault to which the abnormal power signal data belongs.
[0091] The results display module is used to display the corresponding fault area and repair methods based on the fault type of the abnormal power signal data.
[0092] As described above, in this example, multiple fault databases are established through the fault classification module, and historical fault data is stored in the corresponding fault database according to the fault signal category. The fault assessment module filters the historical data of the power grid and determines the category of the fault signal according to the fault assessment indicators. The data acquisition module collects power signal data of the power grid in real time under stable operating conditions. The data calculation module calculates the raw data of the power data characteristics of abnormal power signal data. The scale normalization module performs scale normalization processing. The feature calculation module calculates the power feature data to obtain the feature data of abnormal power signal data. The fault judgment module calculates the similarity between the power signal data feature data and the features of each fault database to determine the fault category of the abnormal power signal data. The result display module displays the corresponding fault area and repair method according to the fault category of the abnormal power signal data.
[0093] The embodiments described above are merely preferred embodiments of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications, improvements, and substitutions without departing from the concept of the present invention, and these all fall within the protection scope of the present invention.
Claims
1. A power grid inspection method based on power private network data processing, characterized in that, The following steps are adopted: Step 1: Establish a fault classifier. The fault classifier contains multiple fault databases, and each fault database corresponds to a fault signal class. Step 2: Obtain historical operation data of the power grid, filter the historical data, extract fault assessment indicators from the historical data, determine the type of fault signal according to the fault assessment indicators, and store the historical fault signals in the corresponding fault database. Step 3: Install power sensors on the dedicated power grid lines, number the power sensors according to the installation area and line, and collect power signal data of the dedicated power grid in real time under stable operating conditions. Step 4: When the power sensor sends abnormal power signal data, calculate the raw data of the power data characteristics of the abnormal power signal data. Step 5: Perform scale normalization on the raw data of the extracted power data features to normalize the raw data of the power data features to data with a mean of 0 and a variance of 1. Step 6: Use feature algorithms to calculate the scale-normalized power feature data to obtain the abnormal power signal data feature data; Step 7: Calculate the similarity between the power signal data feature data and the features of each fault database. Based on the calculated similarity results, determine the category to which the abnormal power signal data fault belongs, and complete the identification of abnormal power signal data faults. Step 8: Staff determine the area of the fault based on the power sensor number that sent the abnormal data, and take corresponding repair measures based on the fault identification results of the abnormal power signal data.
2. The power grid inspection method based on power private network data processing according to claim 1, characterized in that: The fault assessment index is extracted from historical data and a nearest neighbor prototype classification algorithm of orthogonal quantum particle swarm optimization is adopted. The particle swarm optimization algorithm of multiple collapse-orthogonal cross quantum is used for iterative optimization to select the best particles and effective prototypes. The distance from the test data to the selected prototype is calculated to classify the test data. The training data is continued to learn and obtain the fitness value of each particle until the termination condition is met. The classification accuracy of the population is calculated to obtain the final result.
3. The power grid inspection method based on power private network data processing according to claim 2, characterized in that: The particle swarm optimization algorithm formula is as follows: Vi=Vi+C1×rand()×(pbesti-Xi)+C2×rand()×(gbesti-Xi) Xi = Xi + Vi Where i = 1, 2, 3...N, N is the total number of particles, Vi is the velocity of the particles, rand() is a random number between (0, 1), Xi is the current position of the particles, C1 and C2 are learning factors, in general C1 = C2 = 2, the maximum value of Vi is Vmax, if Vi is greater than Vmax, then Vi = Vmax.
4. The power grid inspection method based on power private network data processing according to claim 1, characterized in that: The scale normalization formula for the scale normalization process is as follows: Where z represents the scale-normalized data, and x represents the original data. δ is the mean of the original data, and δ is the variance of the original data.
5. A power grid inspection method based on power private network data processing according to claim 4, characterized in that: The calculation of scale-normalized power characteristic data using feature algorithms is based on the assumption that the abnormal power signal data matrix is X. m×n Where m is the data signal quantity and n is the number of features, the mean of matrix X is first centered to determine the principal direction w with the largest projection variance, as shown in the following formula: ||w|| 2 ＝1 Extract the first k principal directions to form the load matrix, W = (w1,...w k The result after feature extraction of abnormal power signal data is X. new =XW.
6. The power grid inspection method based on power private network data processing according to claim 1, characterized in that: The similarity between the abnormal power signal data and historical fault signals in the fault database is calculated using the correlation coefficient similarity, and the calculation formula is as follows: Where X represents abnormal power signal data and Y represents historical fault signals.
7. The power grid inspection method based on power private network data processing according to claim 1, characterized in that: The fault database also includes a diagnostic and repair database, which stores the causes, paths, and solutions corresponding to the category to which the fault signal belongs. After the abnormal power signal data fault is identified, the corresponding causes, paths, and solutions will be displayed to the staff. The diagnostic and repair database will regularly modify and delete old data and update and reorganize the data in the diagnostic and repair database.
8. A power grid inspection method based on power private network data processing according to claim 1, characterized in that: When abnormal power signal data cannot be identified in the fault classifier and fault database, professional personnel will conduct an assessment and diagnosis. At the same time, based on the assessment and diagnosis results, a new fault signal class and fault database will be established in the fault classifier, and the assessment and diagnosis results will be saved to the fault database.
9. A power grid inspection method based on dedicated power grid data processing according to claim 8, characterized in that: After the abnormal power signal data fault identification is completed, it is stored in the corresponding fault database according to the fault signal category. The historical fault signals in the fault database are updated. Professionals will not only conduct assessments and diagnoses to find new fault causes, paths and solutions, but also regularly modify and delete old data in the fault database and reorganize the data in the fault database.
10. A power grid inspection system based on dedicated power grid data processing, characterized in that, include: The fault classification module is used to establish multiple fault databases and store historical fault data into the corresponding fault databases according to the category to which the fault signal belongs. The fault assessment module is used to filter historical data of the power grid, extract fault assessment indicators from the historical data, and determine the category of the fault signal according to the fault assessment indicators. The data acquisition module is used to install power sensors on the dedicated power grid lines to collect power signal data of the dedicated power grid under stable operating conditions in real time. The data calculation module is used to calculate the raw data of the power data characteristics of abnormal power signal data. The scale normalization module is used to perform scale normalization processing on the raw data of the calculated and extracted power data features; The feature calculation module is used to calculate the power feature data after scale normalization to obtain the feature data of abnormal power signal data. The fault diagnosis module is used to calculate the similarity between the power signal data feature data and the features of each fault database, and to determine the type of fault to which the abnormal power signal data belongs. The results display module is used to display the corresponding fault area and repair methods based on the fault type of the abnormal power signal data.