Power grid equipment fault diagnosis system and method based on image recognition

Abstract

The invention relates to the technical field of power grid fault recognition, in particular to a power grid equipment fault diagnosis system and method based on image recognition, and the system comprises an image collection module which is used for collecting the power grid equipment image data of each power grid equipment; the data preprocessing module is used for preprocessing the image data of the power grid equipment; the data diagnosis module is used for performing data diagnosis on the thunderstorm weather characteristic information according to a data preprocessing result; the fault identification module is used for judging the power grid equipment fault type of each power grid equipment according to the data diagnosis result and the power grid equipment importance grade standard type; and the fault maintenance module is used for judging the maintenance means of the power grid equipment corresponding to the fault type of the power grid equipment according to the fault type of the power grid equipment and the actual fault positioning information of each power grid equipment. According to the invention, the power grid equipment fault can be accurately positioned, and the maintenance efficiency of the power grid equipment fault is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of power grid fault identification, and particularly to a power grid equipment fault diagnosis system and method based on image recognition. Background Art

[0002] When lightning weather causes damage to power grid equipment and triggers a power outage accident, it is necessary to quickly and accurately identify the faults of power grid equipment and promptly notify maintenance personnel for repair to ensure that residents can still use electrical equipment normally during lightning weather. After lightning weather, it is necessary to inspect a large number of power grid equipment to determine whether any equipment has been damaged by lightning. However, the traditional manual inspection method is inefficient and difficult to detect hidden faults of power grid equipment.

[0003] Chinese Patent Publication No.: CN111726578A discloses a power plant fault maintenance and dispatching system based on artificial intelligence image recognition, including an integrated monitoring system. The integrated monitoring system includes an image management service platform, a regulation cloud power system, and a power D5000 system. A plurality of power equipment monitoring devices installed in the power plant are connected to the image management service system, an employee positioning management system is connected to the integrated monitoring system, an artificial intelligence fault diagnosis system is connected to the integrated monitoring system, and a plurality of intelligent safety helmets are connected to the employee positioning management system. However, this solution is difficult to quickly provide the location and fault type judgment of the faulty equipment after a fault occurs in power grid equipment during thunderstorm weather. Summary of the Invention

[0004] Therefore, the present invention provides a power grid equipment fault diagnosis system and method based on image recognition to overcome the problem in the prior art that the efficiency of repairing power grid equipment faults is low because it is difficult to quickly provide the location and fault type judgment of the faulty equipment after a fault occurs in power grid equipment during thunderstorm weather.

[0005] To achieve the above object, on the one hand, the present invention provides a power grid equipment fault diagnosis system based on image recognition, the system includes: An image acquisition module for acquiring power grid equipment image data of each power grid equipment; A data preprocessing module for preprocessing the power grid equipment image data according to a multi-mode data processing scheme to obtain a data preprocessing result, and further for optimizing the data preprocessing result according to power grid equipment image data processing rules; A data diagnosis module for performing data diagnosis on thunderstorm weather feature information according to the optimized data preprocessing result to obtain a data diagnosis result, and further for adjusting the acquisition process of the power grid equipment image data according to the data diagnosis result; A fault identification module, which is used to judge the grid equipment fault types of each grid equipment according to the data diagnosis result and the grid equipment importance level standard type, and is also used to output the actual fault location information of each grid equipment according to the grid equipment fault type and the fault location device; A fault repair module, which is used to judge the repair means of the grid equipment corresponding to the grid equipment fault type according to the grid equipment fault type and the actual fault location information of each grid equipment, and is also used to optimize the repair means of the grid equipment corresponding to the grid equipment fault type according to the safety repair coefficient.

[0006] Further, the data preprocessing module matches and analyzes the grid equipment image data with a multi-mode data processing scheme according to a data matching method to obtain an actual matching degree FP, compares the actual matching degree FP with a preset matching degree FP0, makes an output judgment on the data processing scheme type standard according to the comparison result, and outputs the data processing scheme type standard according to the judgment result, where: When FP < FP0, the data preprocessing module determines not to output the data processing scheme type standard; When FP ≥ FP0, the data preprocessing module determines to output the data processing scheme type standard, compares the grid equipment image data with the data processing scheme type standard, judges the data processing scheme type according to the comparison result, and preprocesses the grid equipment image data according to the judgment result to obtain a data preprocessing result.

[0007] Further, the data preprocessing module obtains the minimum value Xmin and the maximum value Xmax of the data preprocessing result according to the grid equipment image data processing rule, and calculates the grid equipment image standard optimization data Xnew according to the data preprocessing result X, the minimum value Xmin of the data preprocessing result, and the maximum value Xmax of the data preprocessing result, and sets and compares the grid equipment image standard optimization data Xnew with a preset grid equipment image standard optimization data X0new, makes an optimization judgment on the data preprocessing result according to the comparison result, and optimizes the data preprocessing result according to the optimization judgment result to obtain an optimized data preprocessing result.

[0008] Further, the data diagnosis module trains the convolutional neural network model according to the preset standard thunderstorm weather characteristic information dataset, outputs the convolutional neural network model that meets the preset accuracy rate as the thunderstorm weather characteristic information recognition model, and performs data diagnosis on the power grid equipment image data according to the thunderstorm weather characteristic information recognition model to obtain the thunderstorm weather characteristic information corresponding to the power grid equipment image data of each power grid equipment, and uses it as the data diagnosis result.

[0009] Further, the data diagnosis module evaluates the data diagnosis result according to the data comprehensive evaluation method to obtain the actual comprehensive evaluation value A1, compares the actual comprehensive evaluation value A1 with the preset comprehensive evaluation value A0, makes an adjustment judgment on the acquisition process of the power grid equipment image data according to the comparison result, and adjusts the acquisition process of the power grid equipment image data according to the judgment result.

[0010] Further, the fault identification module establishes a decision tree model according to the data diagnosis result and the power grid equipment importance level standard type, where: The fault identification module sets whether each power grid equipment has a power grid equipment fault as the first root node of the decision tree model; The fault identification module compares the key node information of each power grid equipment in the power grid topology structure with the preset key node information, outputs the power grid equipment importance level standard type according to the comparison result, and sets the power grid equipment importance level standard type as the second root node of the decision tree model; The fault identification module compares the thunderstorm weather characteristic information in the data diagnosis result with the preset thunderstorm fault characteristic information of the power grid equipment, outputs the power grid equipment fault timeliness type according to the comparison result, and sets the power grid equipment fault timeliness type as the third root node of the decision tree model; The fault identification module inputs the power grid equipment image data into the decision tree model to judge the power grid equipment fault type of each power grid equipment, where: When the power grid equipment image data reaches the leaf node marked as an instantaneous fault of an important equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment fault type of the power grid equipment is an instantaneous fault of an important equipment caused by lightning; When the power grid equipment image data reaches the leaf node marked as a permanent fault of an important equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment fault type of the power grid equipment is a permanent fault of an important equipment caused by lightning; When the power grid equipment image data reaches the leaf node marked as the instantaneous fault of unimportant equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the power grid equipment is the instantaneous fault of unimportant equipment caused by lightning; When the power grid equipment image data reaches the leaf node marked as the permanent fault of unimportant equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the power grid equipment is the permanent fault of unimportant equipment caused by lightning; When the power grid equipment image data reaches the leaf node marked as the fault of important equipment not caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the power grid equipment is the fault of important equipment not caused by lightning; When the power grid equipment image data reaches the leaf node marked as the fault of unimportant equipment not caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the power grid equipment is the fault of unimportant equipment not caused by lightning; When the power grid equipment image data reaches the leaf node marked as normal in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment is normal.

[0011] Further, the fault identification module triggers the corresponding fault location device to collect the real-time operation data of each power grid equipment according to the fault type of the power grid equipment, compares the real-time operation data with the preset fault operation conditions of the power grid equipment, identifies and judges the occurrence of the power grid equipment fault according to the comparison result, and outputs the actual fault location information of the power grid equipment according to the judgment result, where: When the real-time operation data is inconsistent with the preset fault operation conditions of the power grid equipment, the fault identification module determines that no power grid equipment fault has occurred, and the fault identification module does not output the actual fault location information of the power grid equipment; When the real-time operation data is consistent with the preset fault operation conditions of the power grid equipment, the fault identification module determines that a power grid equipment fault has occurred, locates the fault according to the real-time operation data and the power grid topology structure of the power grid equipment, obtains the actual fault location information of the power grid equipment, and outputs the actual fault location information of the power grid equipment.

[0012] Further, the fault repair module determines the repair means for the power grid equipment corresponding to the power grid equipment fault type according to the power grid equipment fault type of each power grid equipment and the actual fault location information corresponding to the power grid equipment, where: When the fault identification module determines that the power grid equipment fault type of the power grid equipment is an instantaneous fault of an important equipment caused by lightning, the fault repair module determines that the repair means for the power grid equipment corresponding to the power grid equipment fault type is the first fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a permanent fault of an important equipment caused by lightning, the fault repair module determines that the repair means for the power grid equipment corresponding to the power grid equipment fault type is the second fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is an instantaneous fault of a non-important equipment caused by lightning, the fault repair module determines that the repair means for the power grid equipment corresponding to the power grid equipment fault type is the first fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a permanent fault of a non-important equipment caused by lightning, the fault repair module determines that the repair means for the power grid equipment corresponding to the power grid equipment fault type is the third fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a fault of an important equipment not caused by lightning, the fault repair module determines that the repair means for the power grid equipment corresponding to the power grid equipment fault type is the second fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a fault of a non-important equipment not caused by lightning, the fault repair module determines that the repair means for the power grid equipment corresponding to the power grid equipment fault type is the second fault repair means.

[0013] Further, the fault repair module calculates the safety repair coefficient SFT according to the lightning activity intensity factor SF1, the power grid equipment fault type location factor SF2, the repair personnel protection equipment and skill factor SF3, and the repair site environment factor SF4, and sets SFT = V1×SF1 + V2×SF2 + V3×SF3 + V4×SF4, where V1 represents the weight of the lightning activity intensity factor SF1, V2 represents the weight of the power grid equipment fault type location factor SF2, V3 represents the weight of the repair personnel protection equipment and skill factor SF3, V4 represents the weight of the repair site environment factor SF4, V1 + V2 + V3 + V4 = 1, and 0 ≤ SFT ≤ 1; The fault repair module determines the necessity of repairing the power grid equipment corresponding to the power grid equipment fault type according to the value range of the safety repair factor SFT, and optimizes the repair means of the power grid equipment corresponding to the power grid equipment fault type, where: When 0.6 ≤ SFT ≤ 1, the fault repair module determines that the necessity of repairing the power grid equipment corresponding to the power grid equipment fault type is allowed for repair, and optimizes the repair means of the power grid equipment corresponding to the power grid equipment fault type into repair optimization means; When 0.3 < SFT < 0.6, the fault repair module determines that the necessity of repairing the power grid equipment corresponding to the power grid equipment fault type is allowed for repair, compares the actual repair situation with the preset emergency repair situation, judges the urgency degree of the repair situation according to the comparison result, and optimizes the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the judgment result; When 0 ≤ SFT ≤ 0.3, the fault repair module determines that the necessity of repairing the power grid equipment corresponding to the power grid equipment fault type is not allowed for repair, and the fault repair module does not optimize the repair means of the power grid equipment corresponding to the power grid equipment fault type.

[0014] On the other hand, the present invention also provides a method for diagnosing power grid equipment faults based on image recognition, and the method includes: Step S1, collecting power grid equipment image data of each power grid equipment; Step S2, preprocessing the power grid equipment image data according to a multi-mode data processing scheme to obtain a data preprocessing result, and optimizing the data preprocessing result according to the power grid equipment image data processing rule to obtain an optimized data preprocessing result; Step S3, performing data diagnosis on the thunderstorm weather characteristic information according to the optimized data preprocessing result to obtain a data diagnosis result, and adjusting the acquisition process of the power grid equipment image data according to the data diagnosis result; Step S4, judging the power grid equipment fault type of each power grid equipment according to the data diagnosis result and the power grid equipment important level standard type, and outputting the actual fault location information of each power grid equipment according to the power grid equipment fault type and the fault location device; Step S5, judging the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the power grid equipment fault type and the actual fault location information of each power grid equipment, and optimizing the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the safety repair factor.

[0015] Compared with the prior art, the beneficial effects of the present invention are as follows. Through the image acquisition module, the system can efficiently acquire the image data of grid equipment, ensuring the timeliness and accuracy of data acquisition. Through the data preprocessing module, the system can perform standardization processing on the image data of grid equipment to ensure data accuracy. Through the data diagnosis module, the system can accurately diagnose the characteristic information of thunderstorm weather using the image data of grid equipment, and adjust the acquisition process of the image data of grid equipment according to the diagnosis results to ensure data accuracy and integrity. Through the fault identification module, the system can accurately identify the fault types of grid equipment for each grid equipment according to the characteristic information of thunderstorm weather and the standard type of the importance level of grid equipment, and identify the actual fault location information of each grid equipment according to the fault types of grid equipment and the fault location device for each grid equipment, improving the accuracy of fault location information identification. Through the fault repair module, the system can analyze the repair means according to the fault types of grid equipment and the actual fault location information for each grid equipment, and in combination with the safety repair coefficient, improving the repair efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a schematic structural diagram of the grid equipment fault diagnosis system based on image recognition in this embodiment; Figure 2 It is a schematic flow diagram of the grid equipment fault diagnosis method based on image recognition in this embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] In order to make the objectives and advantages of the present invention clearer, the present invention will be further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0018] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are only used to explain the technical principles of the present invention and do not limit the protection scope of the present invention.

[0019] Please refer to Figure 1 as shown, which is a schematic structural diagram of the grid equipment fault diagnosis system based on image recognition in this embodiment. The system includes: An image acquisition module for acquiring the image data of grid equipment for each grid equipment; A data preprocessing module for preprocessing the image data of grid equipment according to a multi-mode data processing scheme to obtain a data preprocessing result, and further optimizing the data preprocessing result according to the image data processing rules of grid equipment to obtain an optimized data preprocessing result. The data preprocessing module is connected to the image acquisition module; A data diagnosis module is used to perform data diagnosis on the thunderstorm weather characteristic information based on the optimized data preprocessing result to obtain a data diagnosis result, and is also used to adjust the acquisition process of the power grid equipment image data according to the data diagnosis result. The data diagnosis module is connected to the image acquisition module and the data preprocessing module; A fault identification module is used to judge the power grid equipment fault types of each power grid equipment according to the data diagnosis result and the power grid equipment importance level standard type, and is also used to output the actual fault location information of each power grid equipment according to the power grid equipment fault type and the fault location device. The fault identification module is connected to the data diagnosis module; A fault repair module is used to judge the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the power grid equipment fault type and the actual fault location information of each power grid equipment, and is also used to optimize the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the safety repair coefficient. The fault repair module is connected to the fault identification module.

[0020] Specifically, the system is set in the power grid equipment fault control terminal. The system can efficiently collect the power grid equipment image data through the image acquisition module to ensure the timeliness and accuracy of data acquisition. The system can perform standardized processing on the power grid equipment image data through the data preprocessing module to ensure the accuracy of data. The system can accurately diagnose the thunderstorm weather characteristic information by using the power grid equipment image data through the data diagnosis module, and adjust the acquisition process of the power grid equipment image data according to the diagnosis result to ensure the accuracy and integrity of data. The system can accurately identify the power grid equipment fault types of each power grid equipment according to the thunderstorm weather characteristic information and the power grid equipment importance level standard type through the fault identification module, and identify the actual fault location information of each power grid equipment according to the power grid equipment fault type and the fault location device of each power grid equipment to improve the accuracy of fault location information identification. The system can analyze the repair means according to the power grid equipment fault type and the actual fault location information of each power grid equipment and in combination with the safety repair coefficient through the fault repair module to improve the repair efficiency.

[0021] Specifically, the power grid equipment refers to various equipment that constitutes a power system. The power grid equipment image data refers to the set of image data collected and visualized through a power grid equipment image acquisition device, including electrical image data, non-electrical image data, and power grid equipment operation status image data. The electrical image data refers to the set of electrical data used to collect real-time electrical performance parameter charts of power grid equipment, including voltage data and current data. The non-electrical image data refers to the set of non-electrical data used to collect real-time non-electrical performance parameter charts of power grid equipment, including power grid equipment vibration data and power grid equipment appearance image data. The power grid equipment vibration data refers to the real-time data of vibration signals generated by power grid equipment during operation. The power grid equipment appearance image data refers to the image information used to present the appearance of power grid equipment. The power grid equipment operation status image data refers to the set of power grid equipment operation status data used to collect real-time power grid equipment operation status charts of power grid equipment, including power grid equipment operation mode data, power grid equipment operation record data, and protection device action data. The power grid equipment operation mode data refers to the information used to record the operation mode of power grid equipment. The power grid equipment operation record data refers to the information used to record the operations performed on power grid equipment. The protection device action data refers to the information of the automatic actions of protection devices in the power grid when a fault is detected. The power grid equipment image acquisition device refers to the device used to collect and visualize the power grid equipment image data, including electrical image data acquisition equipment, non-electrical image data acquisition equipment, and power grid equipment operation status image data acquisition equipment. The electrical image data acquisition equipment refers to the equipment used to collect the electrical performance parameters of power grid equipment and convert the electrical performance parameters into real-time electrical performance parameter charts. The non-electrical image data acquisition equipment refers to the equipment used to collect the non-electrical performance parameters of power grid equipment and convert the non-electrical performance parameters into real-time non-electrical performance parameter charts. The power grid equipment operation status image data acquisition equipment refers to the equipment used to collect the power grid equipment operation status of power grid equipment and convert the power grid equipment operation status into real-time power grid equipment operation status charts.

[0022] Specifically, the data preprocessing module performs matching analysis on the power grid equipment image data and the multi-mode data processing scheme according to the data matching method to obtain the actual matching degree FP, compares the actual matching degree FP with the preset matching degree FP0, makes an output judgment on the data processing scheme type standard according to the comparison result, and outputs the data processing scheme type standard according to the judgment result, where: When FP < FP0, the data preprocessing module determines not to output the data processing scheme type standard; When FP ≥ FP0, the data preprocessing module determines to output the data processing scheme type standard, compares the power grid equipment image data with the data processing scheme type standard, determines the data processing scheme type according to the comparison result, and preprocesses the power grid equipment image data according to the determination result to obtain a data preprocessing result, where: If the power grid equipment image data is consistent with the electrical image data processing standard in the data processing scheme type standard, the data preprocessing module determines that the data processing scheme type is a preset electrical image data processing scheme, preprocesses the electrical image data according to the preset electrical image data processing scheme to obtain a first data preprocessing result, and outputs the first data preprocessing result as the data preprocessing result; If the power grid equipment image data is consistent with the non-electrical image data processing standard in the data processing scheme type standard, the data preprocessing module determines that the data processing scheme type is a preset non-electrical image data processing scheme, preprocesses the non-electrical image data according to the preset non-electrical image data processing scheme to obtain a second data preprocessing result, and outputs the second data preprocessing result as the data preprocessing result; If the power grid equipment image data is consistent with the power grid equipment operation state image data processing standard in the data processing scheme type standard, the data preprocessing module determines that the data processing scheme type is a preset power grid equipment operation state image data processing scheme, preprocesses the power grid equipment operation state image data according to the preset power grid equipment operation state image data processing scheme to obtain a third data preprocessing result, and outputs the third data preprocessing result as the data preprocessing result; If the power grid equipment image data is inconsistent with the data processing scheme type standard, the data preprocessing module determines not to preprocess the power grid equipment image data.

[0023] Specifically, the data matching method refers to a method for performing matching analysis on the power grid equipment image data and a multi-mode data processing solution. In this embodiment, the specific implementation of the data matching method is not limited. For example, it can be set to perform matching analysis on the power grid equipment image data and the multi-mode data processing solution through the cosine similarity method. The multi-mode data processing solution refers to adopting a corresponding data processing solution for the power grid equipment image data. The actual matching degree FP refers to the value obtained by performing matching analysis on the power grid equipment image data and the multi-mode data processing solution according to the data matching method. The preset matching degree FP0 refers to a preset value for comparison with the actual matching degree FP, such as 0.95. The data processing scheme type standard refers to the standard of a preset processing scheme for power grid equipment image data. The data processing scheme type refers to the specific processing scheme type determined according to the data processing scheme type standard. The data preprocessing result refers to the result obtained by preprocessing the power grid equipment image data based on the comparison result between the power grid equipment image data and the data processing scheme type standard. The preset electrical image data processing scheme refers to the specific processing scheme formulated according to the electrical image data processing standard. In this embodiment, the implementation manner of the preset electrical image data processing scheme is not limited. For example, it can be set to perform anomaly analysis on the electrical image data according to the electrical image data processing standard and process the abnormal values of the electrical image data based on the normal discharge mode library. The first data preprocessing result refers to the result obtained by preprocessing the electrical image data according to the preset electrical image data processing scheme. The preset non-electrical image data processing scheme refers to the specific processing scheme formulated according to the non-electrical image data processing standard. In this embodiment, the implementation manner of the preset non-electrical image data processing scheme is not limited. For example, it can be set to perform anomaly analysis on the non-electrical image data according to the non-electrical image data processing standard and delete the abnormal values of the non-electrical image data. The second data preprocessing result refers to the result obtained by preprocessing the non-electrical image data according to the preset non-electrical image data processing scheme. The preset power grid equipment operation state image data processing scheme refers to the specific processing scheme formulated according to the power grid equipment operation state image data processing standard. In this embodiment, the implementation manner of the preset power grid equipment operation state image data processing scheme is not limited. For example, it can be set to perform anomaly analysis on the power grid equipment operation state image data according to the power grid equipment operation state image data processing standard and process the abnormal values of the power grid equipment operation state image data based on the normal data of the electrical image data and the normal data of the non-electrical image data. The third data preprocessing result refers to the result obtained by preprocessing the power grid equipment operation state image data according to the preset power grid equipment operation state image data processing scheme. The electrical image data processing standard refers to the processing standards and requirements formulated for the electrical image data. The non-electrical image data processing standard refers to the processing standards and requirements formulated for the non-electrical image data. The power grid equipment operation state image data processing standard refers to the processing standards and requirements formulated for the power grid equipment operation state image data..

[0024] Specifically, through the data preprocessing module, accurate data standardization processing can be achieved, the corresponding data processing scheme type can be automatically selected, and the processing efficiency can be improved.

[0025] Specifically, the data preprocessing module obtains the minimum value Xmin of the data preprocessing result and the maximum value Xmax of the data preprocessing result according to the power grid equipment image data processing rules, and calculates the optimized data Xnew of the power grid equipment image standard based on the data preprocessing result X, the minimum value Xmin of the data preprocessing result, and the maximum value Xmax of the data preprocessing result, and sets , and compares the optimized data Xnew of the power grid equipment image standard with the preset optimized data X0new of the power grid equipment image, makes an optimization judgment on the data preprocessing result according to the comparison result, and optimizes the data preprocessing result according to the optimization judgment result to obtain the optimized data preprocessing result, where: When Xnew ≥ X0new, the data preprocessing module determines not to optimize the data preprocessing result; When Xnew < X0new, the data preprocessing module determines to optimize the data preprocessing result and sets the data preprocessing result optimization parameter to optimize the optimized data Xnew of the power grid equipment image standard, and sets , the optimized optimized data of the power grid equipment image standard is X1new, and sets , and outputs the optimized optimized data of the power grid equipment image standard as the optimized data preprocessing result.

[0026] Specifically, the power grid equipment image data processing rule refers to a set of standards for processing and optimizing power grid equipment image data. The minimum value Xmin of the data preprocessing result refers to the lowest value obtained after the power grid equipment image data is preprocessed. The maximum value Xmax of the data preprocessing result refers to the highest value obtained after the power grid equipment image data is preprocessed. The preset optimized data X0new of the power grid equipment image refers to a preset value compared with the optimized data Xnew of the power grid equipment image, such as 0.95. The data preprocessing result optimization parameter refers to the parameter for optimizing the data preprocessing result when Xnew < X0new. The optimized optimized data of the power grid equipment image standard refers to the result obtained by optimizing the optimized data Xnew of the power grid equipment image standard according to the data preprocessing result optimization parameter , and the optimized data preprocessing result refers to the result obtained by optimizing the data preprocessing result according to the power grid equipment image data processing rules.

[0027] Specifically, through the data preprocessing module, the optimization requirements of the power grid equipment image data can be accurately judged to ensure data quality.

[0028] Specifically, the data diagnosis module trains the convolutional neural network model according to the preset standard thunderstorm weather characteristic information data set, outputs the convolutional neural network model that meets the preset accuracy rate as the thunderstorm weather characteristic information recognition model, and performs data diagnosis on the grid equipment image data according to the thunderstorm weather characteristic information recognition model to obtain the thunderstorm weather characteristic information corresponding to the grid equipment image data of each grid equipment, which is used as the data diagnosis result.

[0029] Specifically, the thunderstorm weather characteristic information recognition model refers to a model that meets the preset accuracy rate obtained by training the convolutional neural network model according to the preset standard thunderstorm weather characteristic information data set. The preset standard thunderstorm weather characteristic information data set refers to a data set preset for training the convolutional neural network model in the storage form of grid equipment image data - thunderstorm weather characteristic information of each grid equipment. The convolutional neural network model refers to a machine learning model used to extract features from the grid equipment image data of each grid equipment and predict the thunderstorm weather characteristic information of the grid equipment image data of each grid equipment. In this embodiment, the training method of the convolutional neural network model is not limited. For example, it can be set to input the thunderstorm weather characteristic information training set into the convolutional neural network model for training, and input the thunderstorm weather characteristic information test set into the trained convolutional neural network model to optimize and iterate the parameters in the convolutional neural network model until the accuracy rate of the output result of the thunderstorm weather characteristic information test set of the convolutional neural network model reaches the preset accuracy rate, and then output the convolutional neural network model as the thunderstorm weather characteristic information recognition model. The preset accuracy rate refers to a preset value reflecting the training situation of the convolutional neural network model, such as 95%. The data diagnosis result refers to the thunderstorm weather characteristic information corresponding to the grid equipment image data of each grid equipment.

[0030] Specifically, through the thunderstorm weather characteristic information recognition model of the data diagnosis module, the accuracy and efficiency of the recognition of the grid equipment state under thunderstorm weather can be improved.

[0031] Specifically, the data diagnosis module evaluates the data diagnosis result according to the data comprehensive evaluation method to obtain the actual comprehensive evaluation value A1, compares the actual comprehensive evaluation value A1 with the preset comprehensive evaluation value A0, makes an adjustment judgment on the acquisition process of the grid equipment image data according to the comparison result, and adjusts the acquisition process of the grid equipment image data according to the judgment result, where: When A1≥A0, the data diagnosis module determines not to adjust the acquisition process of the grid equipment image data; When A1 < A0, the data diagnosis module determines to adjust the acquisition process of the power grid equipment image data, and adjusts the acquisition frequency of the power grid equipment image data of each power grid equipment in the image acquisition module to a preset increased acquisition frequency range.

[0032] Specifically, the data comprehensive evaluation method refers to the method for evaluating the data diagnosis result. In this embodiment, the specific implementation scheme of the data comprehensive evaluation method is not limited. For example, it can be set to evaluate the data diagnosis result through the principal component analysis method. The actual comprehensive evaluation value A1 refers to the result obtained by evaluating the data diagnosis result according to the data comprehensive evaluation method. The preset comprehensive evaluation value A0 refers to the preset value used for comparison with the actual comprehensive evaluation value A1, such as 90%. The acquisition process of the power grid equipment image data refers to the process of acquiring the power grid equipment image data according to the power grid equipment image acquisition device. The preset increased acquisition frequency range refers to the standard range for controlling the increased acquisition frequency preset when A1 < A0, such as increasing from once every 10 minutes to once every 5 minutes.

[0033] Specifically, through the data diagnosis module, anomalies can be quickly identified, and the acquisition frequency of the power grid equipment image data can be immediately adjusted to improve the accuracy of power grid equipment monitoring.

[0034] Specifically, the fault identification module establishes a decision tree model based on the data diagnosis result and the power grid equipment importance level standard type, where: The fault identification module sets whether each power grid equipment has a power grid equipment fault as the first root node of the decision tree model; The fault identification module compares the key node information of each power grid equipment in the power grid topology structure with the preset key node information, outputs the power grid equipment importance level standard type according to the comparison result, and sets the power grid equipment importance level standard type as the second root node of the decision tree model, where: When the key node information is consistent with the preset key node information, the fault identification module outputs the power grid equipment importance level standard type as important equipment and outputs the important equipment as the second root node of the decision tree model; When the key node information is inconsistent with the preset key node information, the fault identification module outputs the power grid equipment importance level standard type as non-important equipment and outputs the non-important equipment as the second root node of the decision tree model; The fault identification module compares the thunderstorm weather feature information in the data diagnosis result with the preset thunderstorm fault feature information of grid equipment, outputs the fault timeliness type of grid equipment according to the comparison result, and sets the fault timeliness type of grid equipment as the third root node of the decision tree model, where: When the thunderstorm weather feature information is consistent with the preset thunderstorm fault feature information of grid equipment, the fault identification module outputs the fault timeliness type of grid equipment as the instantaneous lightning interference feature, and outputs the instantaneous lightning interference feature as the third root node of the decision tree model; When the thunderstorm weather feature information is inconsistent with the preset thunderstorm fault feature information of grid equipment, the fault identification module outputs the fault timeliness type of grid equipment as the permanent fault feature, and outputs the permanent fault feature as the third root node of the decision tree model; The fault identification module inputs the grid equipment image data into the decision tree model to judge the fault type of each grid equipment, where: When the grid equipment image data reaches the leaf node marked as the instantaneous fault of important equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the grid equipment is the instantaneous fault of important equipment caused by lightning; When the grid equipment image data reaches the leaf node marked as the permanent fault of important equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the grid equipment is the permanent fault of important equipment caused by lightning; When the grid equipment image data reaches the leaf node marked as the instantaneous fault of unimportant equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the grid equipment is the instantaneous fault of unimportant equipment caused by lightning; When the grid equipment image data reaches the leaf node marked as the permanent fault of unimportant equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the grid equipment is the permanent fault of unimportant equipment caused by lightning; When the grid equipment image data reaches the leaf node marked as the fault of important equipment not caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the grid equipment is the fault of important equipment not caused by lightning; When the grid equipment image data reaches the leaf node marked as the fault of unimportant equipment not caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the fault type of the grid equipment is the fault of unimportant equipment not caused by lightning; When the power grid equipment image data reaches the leaf node marked as normal in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment is normal.

[0035] Specifically, the important level standard type of the power grid equipment refers to the standard for classifying the importance of equipment in the power grid. The decision tree model refers to an algorithm for judging the fault type of power grid equipment. The power grid equipment fault type refers to the type of abnormal state that occurs in the equipment in the power grid. The power grid topology structure refers to the structural information that describes the connection relationship between various equipment in the power grid and the power flow path. The key node information refers to the information of nodes with important influence in the power grid topology structure, such as substations. The preset key node information refers to the standard of key node information preset in the decision tree model, such as nodes with a line load rate of 80%. The power grid equipment fault time effect type refers to the time type of power grid equipment failure in thunderstorm weather. The preset power grid equipment thunderstorm fault characteristic information refers to the characteristic information preset in the decision tree model for judging whether the power grid equipment fails due to thunderstorm weather, such as the voltage fluctuation duration is 10 minutes. The instantaneous lightning interference characteristic refers to the abnormal state that the power grid equipment temporarily presents due to lightning interference. The permanent fault characteristic refers to the long-term irreparable fault state of the power grid equipment caused by lightning. The leaf node of the instantaneous fault of important equipment caused by lightning refers to the terminal node in the decision tree model that represents the instantaneous fault of important equipment caused by lightning. The instantaneous fault of important equipment caused by lightning refers to the fault caused by lightning interference, resulting in a temporary abnormal state of important equipment. The leaf node of the permanent fault of important equipment caused by lightning refers to the terminal node in the decision tree model that represents the permanent fault of important equipment caused by lightning. The permanent fault of important equipment caused by lightning refers to the fault caused by lightning, resulting in a long-term irreparable abnormal state of important equipment. The leaf node of the instantaneous fault of unimportant equipment caused by lightning refers to the terminal node in the decision tree model that represents the instantaneous fault of unimportant equipment caused by lightning. The instantaneous fault of unimportant equipment caused by lightning refers to the fault caused by lightning interference, resulting in a temporary abnormal state of unimportant equipment. The leaf node of the permanent fault of unimportant equipment caused by lightning refers to the terminal node in the decision tree model that represents the permanent fault of unimportant equipment caused by lightning. The permanent fault of unimportant equipment caused by lightning refers to the fault caused by lightning, resulting in a long-term irreparable abnormal state of unimportant equipment. The leaf node of the fault of important equipment not caused by lightning refers to the terminal node in the decision tree model that represents the fault of important equipment not caused by lightning. The fault of important equipment not caused by lightning refers to the fault caused by non-lightning reasons, resulting in an abnormal state of important equipment. The leaf node of the fault of unimportant equipment not caused by lightning refers to the terminal node in the decision tree model that represents the fault of unimportant equipment not caused by lightning. The fault of unimportant equipment not caused by lightning refers to the fault caused by non-lightning reasons, resulting in an abnormal state of unimportant equipment.

[0036] Specifically, by constructing a decision tree model through the fault identification module, the fault judgment of power grid equipment can be quickly carried out, improving the efficiency of fault identification.

[0037] Specifically, the fault identification module triggers the corresponding fault location device according to the fault type of the power grid equipment to collect the real-time operation data of each power grid equipment, compares the real-time operation data with the preset fault operation conditions of the power grid equipment, identifies and judges the occurrence of the power grid equipment fault according to the comparison result, and outputs the actual fault location information of the power grid equipment according to the judgment result, where: When the real-time operation data is inconsistent with the preset fault operation conditions of the power grid equipment, the fault identification module determines that no power grid equipment fault has occurred in the power grid equipment, and the fault identification module does not output the actual fault location information of the power grid equipment; When the real-time operation data is consistent with the preset fault operation conditions of the power grid equipment, the fault identification module determines that a power grid equipment fault has occurred in the power grid equipment, locates the fault according to the real-time operation data and the power grid topology structure of the power grid equipment, obtains the actual fault location information of the power grid equipment, and outputs the actual fault location information of the power grid equipment.

[0038] Specifically, the fault location device refers to a device for detecting the fault location of power grid equipment, the real-time operation data refers to various data generated by each power grid equipment in the operating state, the preset fault operation conditions of the power grid equipment refer to the criteria preset for judging whether the power grid equipment has a fault. For example, the fault current of the transmission line is 1200 amperes. In this embodiment, the specific implementation manner of fault location is not limited. For example, it can be set to locate the power grid equipment with a power grid equipment fault by the traveling wave ranging method. The occurrence of the power grid equipment fault refers to the specific situation of the power grid equipment having a fault, and the actual fault location information refers to the information obtained by locating the fault according to the real-time operation data and the power grid topology structure of the power grid equipment.

[0039] Specifically, by comparing the fault location device and the real-time operation data through the fault identification module, the accuracy of fault identification can be improved, and the efficiency of power grid operation and maintenance can be improved.

[0040] Specifically, the fault repair module judges the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the power grid equipment fault type of each power grid equipment and the actual fault location information corresponding to the power grid equipment, where: When the fault identification module determines that the fault type of the grid equipment is an instantaneous fault of important equipment caused by lightning, the fault repair module determines that the repair means for the grid equipment corresponding to the grid equipment fault type is the first fault repair means; When the fault identification module determines that the fault type of the grid equipment is a permanent fault of important equipment caused by lightning, the fault repair module determines that the repair means for the grid equipment corresponding to the grid equipment fault type is the second fault repair means; When the fault identification module determines that the fault type of the grid equipment is an instantaneous fault of non-important equipment caused by lightning, the fault repair module determines that the repair means for the grid equipment corresponding to the grid equipment fault type is the first fault repair means; When the fault identification module determines that the fault type of the grid equipment is a permanent fault of non-important equipment caused by lightning, the fault repair module determines that the repair means for the grid equipment corresponding to the grid equipment fault type is the third fault repair means; When the fault identification module determines that the fault type of the grid equipment is a fault of important equipment not caused by lightning, the fault repair module determines that the repair means for the grid equipment corresponding to the grid equipment fault type is the second fault repair means; When the fault identification module determines that the fault type of the grid equipment is a fault of non-important equipment not caused by lightning, the fault repair module determines that the repair means for the grid equipment corresponding to the grid equipment fault type is the second fault repair means.

[0041] Specifically, the repair means for the grid equipment corresponding to the grid equipment fault type refers to the measures for repairing the grid equipment corresponding to the grid equipment fault type. The first fault repair means refers to the measures pre-established for the instantaneous faults of important equipment caused by lightning and the instantaneous faults of non-important equipment caused by lightning. In this embodiment, the specific implementation manner of the first fault repair means is not limited. For example, it can be set to repair through the self-recovery function of the grid. The second fault repair means refers to the measures pre-established for the permanent faults of important equipment caused by lightning, the faults of important equipment not caused by lightning, and the faults of non-important equipment not caused by lightning. In this embodiment, the specific implementation manner of the second fault repair means is not limited. For example, it can be set to repair through the automated monitoring remote platform of the grid. The third fault repair means refers to the measures pre-established for the permanent faults of non-important equipment caused by lightning. In this embodiment, the specific implementation manner of the third fault repair means is not limited. For example, it can be set to repair by adjusting the transformer tap of the adjacent substation.

[0042] Specifically, the fault repair module can quickly and accurately determine the repair means of the power grid equipment corresponding to the fault type of the power grid equipment, improving the reliability of power grid operation.

[0043] Specifically, the fault repair module calculates the safety repair factor SFT based on the lightning activity intensity factor SF1, the power grid equipment fault type location factor SF2, the maintenance personnel protection equipment and skill factor SF3, and the maintenance site environment factor SF4, and sets SFT = V1×SF1 + V2×SF2 + V3×SF3 + V4×SF4. V1 represents the weight of the lightning activity intensity factor SF1, V2 represents the weight of the power grid equipment fault type location factor SF2, V3 represents the weight of the maintenance personnel protection equipment and skill factor SF3, V4 represents the weight of the maintenance site environment factor SF4, V1 + V2 + V3 + V4 = 1, and 0 ≤ SFT ≤ 1; The fault repair module judges the necessity of repairing the power grid equipment corresponding to the fault type of the power grid equipment according to the value range of the safety repair factor SFT, and optimizes the repair means of the power grid equipment corresponding to the fault type of the power grid equipment according to the repair necessity, where: When 0.6 ≤ SFT ≤ 1, the fault repair module determines that the necessity of repairing the power grid equipment corresponding to the fault type of the power grid equipment is allowed for repair, and optimizes the repair means of the power grid equipment corresponding to the fault type of the power grid equipment into the repair optimization means; When 0.3 < SFT < 0.6, the fault repair module determines that the necessity of repairing the power grid equipment corresponding to the fault type of the power grid equipment is allowed for repair, and compares the actual repair situation with the preset emergency repair situation, judges the urgency degree of the repair situation according to the comparison result, and optimizes the repair means of the power grid equipment corresponding to the fault type of the power grid equipment according to the judgment result, where: If the actual repair situation is inconsistent with the preset emergency repair situation, the fault repair module determines that the urgency degree of the repair situation is a non-emergency repair situation, and the fault repair module does not optimize the repair means of the power grid equipment corresponding to the fault type of the power grid equipment; If the actual repair level is consistent with the preset emergency repair situation, the fault repair module determines that the urgency degree of the repair situation is an emergency repair situation, and optimizes the repair means of the power grid equipment corresponding to the fault type of the power grid equipment into the repair optimization means; When 0 ≤ SFT ≤ 0.3, the fault repair module determines that the necessity of repairing the power grid equipment corresponding to the fault type of the power grid equipment is not allowed for repair, and the fault repair module does not optimize the repair means of the power grid equipment corresponding to the fault type of the power grid equipment.

[0044] Specifically, the lightning activity intensity factor SF1 refers to the influencing factor of the lightning activity intensity on the maintenance work. The power grid equipment fault type location factor SF2 refers to the influencing factor of the location corresponding to the power grid equipment fault type on the maintenance work. The maintenance personnel's protective equipment and skills factor SF3 refers to the influencing factor of the maintenance personnel's protective equipment and skills on the maintenance work. The maintenance site environment factor SF4 refers to the influencing factor of the maintenance site environment on the maintenance work. The value range of the safety maintenance coefficient SFT refers to the standard used to evaluate the maintenance work. The necessity of maintenance refers to judging the necessity of the maintenance work according to the value range of the safety maintenance coefficient SFT. The maintenance optimization means refers to the measures for the maintenance work. In this embodiment, the specific implementation manner of the maintenance optimization means is not limited. For example, it can be set to perform maintenance through the maintenance means combined with safety measures. The safety measures refer to the measures taken to ensure the safety of maintenance personnel and the stable operation of power grid equipment, such as wearing protective equipment. The actual maintenance situation refers to the situation actually encountered during the maintenance process. The preset emergency maintenance situation refers to the maintenance situation that is preset for emergency handling according to historical experience and equipment characteristics, such as a power outage in the power supply area caused by a high-voltage line fault. The urgency degree of the maintenance situation refers to the degree of urgency required for the maintenance work.

[0045] Specifically, through the fault maintenance module, the safety and urgency of the maintenance task can be more accurately evaluated, and the safety and efficiency of the power grid equipment fault maintenance can be improved.

[0046] Please refer to Figure 2 as shown, which is a schematic flowchart of the power grid equipment fault diagnosis method based on image recognition in this embodiment. The method includes: Step S1, collect the power grid equipment image data of each power grid equipment; Step S2, preprocess the power grid equipment image data according to the multi-mode data processing scheme to obtain the data preprocessing result, and optimize the data preprocessing result according to the power grid equipment image data processing rule to obtain the optimized data preprocessing result; Step S3, perform data diagnosis on the thunderstorm weather characteristic information according to the optimized data preprocessing result to obtain the data diagnosis result, and adjust the acquisition process of the power grid equipment image data according to the data diagnosis result; Step S4, judge the power grid equipment fault type of each power grid equipment according to the data diagnosis result and the power grid equipment important level standard type, and output the actual fault location information of each power grid equipment according to the power grid equipment fault type and the fault location device; Step S5: Determine the maintenance means for the power grid equipment corresponding to the power grid equipment failure type according to the power grid equipment failure type and the actual fault location information of each power grid equipment, and optimize the maintenance means for the power grid equipment corresponding to the power grid equipment failure type according to the safety maintenance factor.

[0047] So far, the technical solution of the present invention has been described in conjunction with the preferred embodiments shown in the drawings. However, it is easy for those skilled in the art to understand that the protection scope of the present invention is obviously not limited to these specific embodiments. Without departing from the principle of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will fall within the protection scope of the present invention.

Claims

1. A power grid equipment fault diagnosis system based on image recognition, characterized in that: The system comprises: An image acquisition module is used to collect power grid equipment image data of each power grid equipment; A data preprocessing module, used to preprocess the power grid equipment image data according to a multi-mode data processing scheme to obtain a data preprocessing result, and also used to optimize the data preprocessing result according to a power grid equipment image data processing rule; A data diagnosis module, used to perform data diagnosis on the thunderstorm weather characteristic information according to the optimized data preprocessing result to obtain a data diagnosis result, and also used to adjust the collection process of the power grid equipment image data according to the data diagnosis result; A fault identification module, used to judge the type of power grid equipment fault of each power grid equipment according to the data diagnosis result and the type of power grid equipment importance level standard, and also used to output the actual fault location information of each power grid equipment according to the power grid equipment fault type and the fault location device; The fault repair module is used to judge the maintenance means of the power grid equipment corresponding to the power grid equipment fault type according to the power grid equipment fault type and the actual fault location information of each power grid equipment, and is also used to optimize the maintenance means of the power grid equipment corresponding to the power grid equipment fault type according to the safety maintenance factor.

2. The power grid equipment fault diagnosis system based on image recognition according to claim 1 is characterized in that: The data preprocessing module matches and analyzes the power grid equipment image data with the multi-mode data processing scheme according to the data matching method to obtain an actual matching degree FP, and compares the actual matching degree FP with the preset matching degree FP0, and outputs a judgment on the data processing scheme type standard according to the comparison result, and outputs the data processing scheme type standard according to the judgment result, wherein: When FP<FP0, the data preprocessing module determines not to output the data processing scheme type standard; When FP≥FP0, the data preprocessing module determines to output the data processing scheme type standard, and compares the power grid equipment image data with the data processing scheme type standard, judges the data processing scheme type according to the comparison result, and preprocesses the power grid equipment image data according to the judgment result to obtain the data preprocessing result.

3. The power grid equipment fault diagnosis system based on image recognition according to claim 2 is characterized in that: The data preprocessing module obtains the minimum value Xmin and the maximum value Xmax of the data preprocessing result according to the power grid equipment image data processing rule, and calculates the power grid equipment image standard optimization data Xnew according to the data preprocessing result X, the minimum value Xmin and the maximum value Xmax of the data preprocessing result, and sets The power grid equipment image standard optimization data Xnew is compared with the preset power grid equipment image standard optimization data X0new, the data preprocessing result is optimized according to the comparison result, and the data preprocessing result is optimized according to the optimization judgment result to obtain the optimized data preprocessing result.

4. The power grid equipment fault diagnosis system based on image recognition according to claim 1, characterized in that: The data diagnosis module trains the convolutional neural network model according to a preset standard thunderstorm weather characteristic information data set, outputs the convolutional neural network model that meets the preset accuracy as a thunderstorm weather characteristic information recognition model, and performs data diagnosis on the power grid equipment image data according to the thunderstorm weather characteristic information recognition model to obtain the thunderstorm weather characteristic information corresponding to the power grid equipment image data of each power grid equipment, and uses it as the data diagnosis result.

5. The power grid equipment fault diagnosis system based on image recognition according to claim 4 is characterized in that: The data diagnosis module evaluates the data diagnosis result according to the data comprehensive evaluation method to obtain an actual comprehensive evaluation value A1, and compares the actual comprehensive evaluation value A1 with the preset comprehensive evaluation value A0, and adjusts and judges the acquisition process of the power grid equipment image data according to the comparison result, and adjusts the acquisition process of the power grid equipment image data according to the judgment result.

6. The power grid equipment fault diagnosis system based on image recognition according to claim 4, characterized in that: The fault identification module establishes a decision tree model according to the data diagnosis results and the type of importance level standard of the power grid equipment, wherein: The fault identification module sets whether a grid device fault occurs in each grid device as the first root node of the decision tree model; The fault identification module compares the key node information of each power grid device in the power grid topology with the preset key node information, outputs the importance level standard type of the power grid device according to the comparison result, and sets the importance level standard type of the power grid device as the second root node of the decision tree model; The fault identification module compares the thunderstorm weather characteristic information in the data diagnosis result with the preset thunderstorm fault characteristic information of the power grid equipment, outputs the power grid equipment fault time-effect type according to the comparison result, and sets the power grid equipment fault time-effect type as the third root node of the decision tree model; The fault identification module inputs the power grid equipment image data into a decision tree model to determine the power grid equipment fault type of each power grid equipment, wherein: When the power grid equipment image data reaches a leaf node marked as an important equipment instantaneous fault caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment fault type of the power grid equipment is an important equipment instantaneous fault caused by lightning; When the power grid equipment image data reaches a leaf node marked as a permanent failure of important equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment failure type of the power grid equipment is a permanent failure of important equipment caused by lightning; When the power grid equipment image data reaches a leaf node marked as a non-important equipment instantaneous fault caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment fault type of the power grid equipment is a non-important equipment instantaneous fault caused by lightning; When the power grid equipment image data reaches a leaf node marked as a permanent failure of non-important equipment caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment failure type of the power grid equipment is a permanent failure of non-important equipment caused by lightning; When the power grid equipment image data reaches a leaf node marked as an important equipment failure not caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment failure type of the power grid equipment is an important equipment failure not caused by lightning; When the power grid equipment image data reaches a leaf node marked as a non-important equipment fault not caused by lightning in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid equipment fault type of the power grid equipment is a non-important equipment fault not caused by lightning; When the power grid device image data reaches a leaf node marked as normal in the decision tree model along the path of the decision tree model, the fault identification module determines that the power grid device is normal.

7. The power grid equipment fault diagnosis system based on image recognition according to claim 6, characterized in that: The fault identification module triggers the corresponding fault location device according to the type of the power grid equipment fault to collect the real-time operation data of each power grid equipment, and compares the real-time operation data with the preset power grid equipment fault operation condition of the power grid equipment, identifies and judges the power grid equipment fault occurrence of the power grid equipment according to the comparison result, and outputs the actual fault location information of the power grid equipment according to the judgment result, wherein: When the real-time operation data is inconsistent with the preset power grid equipment fault operation condition, the fault identification module determines that the power grid equipment fault of the power grid equipment is that no power grid equipment fault has occurred, and the fault identification module does not output the actual fault location information of the power grid equipment; When the real-time operation data is consistent with the preset power grid equipment fault operation conditions, the fault identification module determines that the power grid equipment fault of the power grid equipment is that a power grid equipment fault has occurred, and locates the fault according to the real-time operation data of the power grid equipment and the power grid topology structure, obtains the actual fault location information of the power grid equipment, and outputs the actual fault location information of the power grid equipment.

8. The power grid equipment fault diagnosis system based on image recognition according to claim 7, characterized in that: The fault repair module determines the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the power grid equipment fault type of each power grid equipment and the actual fault location information corresponding to the power grid equipment, wherein: When the fault identification module determines that the power grid equipment fault type of the power grid equipment is an important equipment instantaneous fault caused by lightning, the fault repair module determines that the maintenance means of the power grid equipment corresponding to the power grid equipment fault type is a first fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a permanent fault of important equipment caused by lightning, the fault repair module determines that the repair means of the power grid equipment corresponding to the power grid equipment fault type is a second fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a non-important equipment instantaneous fault caused by lightning, the fault repair module determines that the maintenance means of the power grid equipment corresponding to the power grid equipment fault type is a first fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a permanent fault of non-important equipment caused by lightning, the fault repair module determines that the repair means of the power grid equipment corresponding to the power grid equipment fault type is a third fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is an important equipment fault not caused by lightning, the fault repair module determines that the maintenance means of the power grid equipment corresponding to the power grid equipment fault type is a second fault repair means; When the fault identification module determines that the power grid equipment fault type of the power grid equipment is a non-important equipment fault not caused by lightning, the fault repair module determines that the maintenance means of the power grid equipment corresponding to the power grid equipment fault type is a second fault repair means.

9. The power grid equipment fault diagnosis system based on image recognition according to claim 8, characterized in that: The fault maintenance module calculates the safety maintenance factor SFT according to the lightning activity intensity factor SF1, the power grid equipment fault type location factor SF2, the maintenance personnel protective equipment and skill factor SF3 and the maintenance site environment factor SF4, and sets SFT=V1×SF1+V2×SF2+V3×SF3+V4×SF4, where V1 represents the weight of the lightning activity intensity factor SF1, V2 represents the weight of the power grid equipment fault type location factor SF2, V3 represents the weight of the maintenance personnel protective equipment and skill factor SF3, and V4 represents the weight of the maintenance site environment factor SF4, V1+V2+V3+V4=1, 0≤SFT≤1; The fault repair module determines the necessity of repairing the power grid equipment corresponding to the power grid equipment fault type according to the value range of the safety repair factor SFT, and optimizes the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the necessity of repair, wherein: When 0.6≤SFT≤1, the fault repair module determines that the repair necessity of the power grid equipment corresponding to the power grid equipment fault type is that repair is allowed, and optimizes the repair means of the power grid equipment corresponding to the power grid equipment fault type to a repair optimization means; When 0.3<SFT<0.6, the fault repair module determines that the repair necessity of the power grid equipment corresponding to the power grid equipment fault type is allowed repair, and compares the actual repair situation with the preset emergency repair situation, judges the urgency of the repair situation according to the comparison result, and optimizes the repair means of the power grid equipment corresponding to the power grid equipment fault type according to the judgment result; When 0≤SFT≤0.3, the fault repair module determines that the maintenance necessity of the power grid equipment corresponding to the power grid equipment fault type is that maintenance is not allowed, and the fault repair module does not optimize the maintenance means of the power grid equipment corresponding to the power grid equipment fault type.

10. A method applied to a power grid equipment fault diagnosis system based on image recognition according to any one of claims 1 to 9, characterized in that: The method comprises: Step S1, collecting power grid equipment image data of each power grid equipment; Step S2, preprocessing the power grid equipment image data according to the multi-mode data processing scheme to obtain a data preprocessing result, and optimizing the data preprocessing result according to the power grid equipment image data processing rule to obtain an optimized data preprocessing result; Step S3, performing data diagnosis on the thunderstorm weather characteristic information according to the optimized data preprocessing result to obtain a data diagnosis result, and adjusting the acquisition process of the power grid equipment image data according to the data diagnosis result; Step S4, judging the power grid equipment fault type of each power grid equipment according to the data diagnosis result and the power grid equipment importance level standard type, and outputting the actual fault location information of each power grid equipment according to the power grid equipment fault type and the fault location device; Step S5, judging the maintenance means of the grid equipment corresponding to the grid equipment fault type according to the grid equipment fault type and the actual fault location information of each grid equipment, and optimizing the maintenance means of the grid equipment corresponding to the grid equipment fault type according to the safety maintenance factor.

Images