Electric power line fault positioning and repairing system based on artificial intelligence

Abstract

The invention belongs to the technical field of power line fault positioning and repairing, and discloses a power line fault positioning and repairing system based on artificial intelligence. The system comprises a reference data acquisition module, a fluctuation data acquisition module, a data processing module, a fault prediction value generation module, a dynamic resource decision module, a safety problem tracing module and a data communication module, and is characterized in that a reference data set and a fluctuation data set are acquired, the reference data set and the fluctuation data set are preprocessed to obtain a feature data set, and the feature data set is sent to the data communication module; analyzing the reference data set and the feature data set to obtain a fault prediction value, processing the fault prediction value to obtain a fault decision feedback report, and analyzing the fault decision feedback report to obtain a safety responsibility report. The method has the remarkable advantages of high fault prediction accuracy, good fault auxiliary decision-making effect and high maintenance process loss responsibility tracing capability.

Description

Technical Field

[0001] The present invention relates to the technical field of power line fault location and repair, and more specifically, to a power line fault location and repair system based on artificial intelligence. Background Art

[0002] Power lines refer to the lines used to transmit electrical energy between power plants, substations, and power users. Power lines are an important part of the power supply system and are responsible for the transmission and distribution of electrical energy. Thanks to the rapid development of my country's economy in recent years, electricity, as the main energy source in today's society, has been widely popularized in my country. Due to my country's vast territory, the transmission of electricity requires the laying of power lines over a large area. The uneven quality of the laying environment will directly affect the safe use of power lines. Traditional power line fault location mostly relies on manual inspection and maintenance or reporting of problems caused by power line failures. This not only greatly increases the workload of staff, but also seriously affects the normal use of electricity. Relying on the development of artificial intelligence, current power line fault location and repair rely on various sensor data to automatically detect and locate power line faults. However, there are also problems such as poor reliability of prediction data, lack of corresponding auxiliary decision-making, and unclear division of responsibility for losses during the maintenance process. Therefore, how to effectively solve these problems has become a major issue that the current power industry needs to face.

[0003] The patent application with publication number CN206863161U discloses a power line fault location system based on the Internet of Things and DSP. By timely storing a large amount of power line fault information, the fault information is uniformly managed, shared and analyzed and can be formed into a big data sharing system. At the same time, it can be provided to multiple remote control terminals for review, sharing and timely dispatch of monitoring personnel to repair power lines, thereby realizing intelligent power line fault location and repair.

[0004] However, although the above-mentioned power line fault location system based on the Internet of Things and DSP has achieved intelligent power line fault location and repair to a certain extent by analyzing the fault information related to the power line, there are still problems in the use of the power line fault location and repair system, such as low accuracy of fault prediction data, lack of corresponding auxiliary decision-making and unclear division of responsibility for losses in the maintenance process, which results in waste of maintenance resources, increased maintenance scheduling response time and shirking of responsibility for losses in the maintenance process, thereby causing loss of benefits to power companies.

[0005] In view of this, the present invention proposes an artificial intelligence-based power line fault location and repair system to solve the above problems. Summary of the Invention

[0006] In order to overcome the above-mentioned defects of the prior art and achieve the above-mentioned objectives, the present invention provides the following technical solutions, including:

[0007] A benchmark data collection module is used to collect benchmark data sets, including line length data, line usage data, line material data, and environmental risk data;

[0008] Fluctuation data acquisition module, used to collect fluctuation data sets, which include real-time load data and line temperature data;

[0009] A data processing module is used to preprocess the benchmark data set and the fluctuation data set to obtain a feature data set;

[0010] Furthermore, the preprocessing methods for the benchmark dataset and the fluctuation dataset include:

[0011] A1. By quantifying the aging risk of line usage data, the specific calculation formula for aging risk quantification is:

[0012] ;

[0013] Get aging characteristic data ,in, Use data for the line, is the maximum permissible service life;

[0014] A2 calculates the average load characteristics based on real-time load data. The specific calculation formula for the average load characteristics is:

[0015] ;

[0016] Get average load characteristic data ,in, is the sample size, For the Real-time load data;

[0017] A3, by calculating the temperature change characteristics of the line temperature data, the specific calculation formula of the temperature change characteristics is:

[0018] ;

[0019] Get temperature change characteristic data ,in, Current time Line temperature data , For the previous time point Line temperature data , is the time interval;

[0020] A4. Risk characteristics are calculated based on the average load characteristic data and the temperature change characteristic data. The specific calculation formula for risk characteristics is:

[0021] ;

[0022] Get risk profile data ,in, is the maximum allowable current value, is the maximum allowable line temperature value;

[0023] A5, packing aging characteristic data, average load characteristic data, temperature change characteristic data and risk characteristic data to obtain a characteristic data set;

[0024] A fault prediction value generation module is used to analyze the benchmark data set and the feature data set to obtain a fault prediction value;

[0025] Furthermore, the fault prediction value generation module also includes a historical data reading module, a system model construction module, a data support module and a prediction value transmission module, wherein:

[0026] The historical data reading module is used to support the system to retrieve the benchmark data set and feature data set stored in the database;

[0027] The system model building module is used to support the system in building the model required for calculation;

[0028] The data support module is used to support manual input of actual fault values, preset cycles and detection thresholds, and support system retrieval;

[0029] The predicted value transmission module is used to support the system in transmitting the fault prediction value;

[0030] Furthermore, the steps of analyzing the benchmark dataset and the feature dataset include:

[0031] Step 1: Based on the historical data reading module, a set of historical benchmark data sets and historical feature data sets stored in the database are extracted and grouped and labeled accordingly according to the order of timestamps from recent to far. The labeling results are L1, L2, L3, ..., Ln, and the labeling results are used as the sample set;

[0032] Step 2: Based on the system model construction module, the sample set is divided into an 80% training set and a 20% training set to establish a risk prediction model;

[0033] Step 3: Substitute into the calculation formula:

[0034] ;

[0035] Get the fault prediction value ,in, is the dynamic weight factor, is the activation function, is the number of feature data combinations, For the The feature weight factor of the feature data combination, For the The characteristic unit of the characteristic data combination, is the logarithm, 、 and is the physical weight factor;

[0036] Step 4: Based on the fault prediction value in step 3 and the actual fault value in the data support module, calculate the error of the fault prediction value. The specific calculation formula of the error is:

[0037] ;

[0038] Get the error value ,in, is the number of training set samples, For the Actual fault value , For the Failure prediction value, is the maximum allowable value of the fault prediction value;

[0039] According to the error threshold in the data support module, when the error value is greater than or equal to the error threshold, the dynamic feature weight factor is adjusted, and the process returns to step 3 to substitute the adjusted dynamic feature weight factor into the fault prediction value calculation formula. When the error value is less than the error threshold, the fault prediction value is output to step 6.

[0040] Step 5: Based on the feature weight factor in step 3 and according to the preset period in the data support module, the feature weight factor is updated. The specific calculation formula for the update is:

[0041] ;

[0042] Get the new feature weight factor ,in, is the normalization function, is the temperature parameter, For the The mean square error of the feature data combination;

[0043] The updated new feature weight factor Substitute into the fault prediction value calculation formula in step 3;

[0044] Step 6: Based on the fault prediction value in step 4, and according to the detection threshold and validation set in the data support module, the fault prediction value is tested. The specific expression formula of the test is:

[0045]

[0046] in, is the number of samples in the validation set, For samples The failure prediction value of The previous risk prediction model for the sample The calculated failure prediction value, is the detection threshold;

[0047] when When satisfied, output the fault prediction value to step seven. In case of rejection, return to step 1 to retrain the risk prediction model;

[0048] Step 7: Based on the prediction value transmission module, the fault prediction value is output to the dynamic resource decision module;

[0049] Furthermore, the feature unit The specific expression formula is:

[0050] ;

[0051] in, is the hyperbolic tangent function, is the line length data, is the base, 、 、 and For the Dynamic feature weight factors;

[0052] Furthermore, the specific calculation formula for adjusting the parameters of the dynamic feature weight factor is:

[0053] ;

[0054] Get the new dynamic feature weight factor ,in, For the Dynamic feature weight factor , Including dynamic feature weight factors 、 、 and , is the learning rate, For the The partial derivative of the characteristic unit, For the Dynamic feature weight factor The partial derivative of

[0055] Dynamic resource decision module, used to process the fault prediction value and obtain the fault decision feedback report;

[0056] Furthermore, the fault prediction value is processed in the following manner:

[0057] When the fault prediction value is less than 30, a preventive inspection report is generated; when the fault prediction value is greater than or equal to 30 and less than 70, a scheduling maintenance report is generated; when the fault prediction value is greater than or equal to 70, an emergency repair report is generated;

[0058] The preventive inspection report includes a statement that the predicted failure level is low and asks staff to conduct preventive inspections according to the maintenance cycle;

[0059] The dispatch maintenance report includes a description of the predicted fault extent, and the staff is asked to notify the nearby maintenance team to go to the faulty line section for inspection and maintenance;

[0060] The emergency repair report includes a statement indicating that the fault is predicted to be severe and that staff are requested to immediately cut off the power supply to the faulty line section and notify nearby maintenance teams to inspect and repair the faulty line section;

[0061] Package preventive inspection reports, scheduled maintenance reports, and emergency repair reports to obtain fault decision feedback reports;

[0062] The safety issue tracing module is used to analyze the fault decision feedback report and obtain the safety responsibility report;

[0063] Furthermore, the methods for analyzing the fault decision feedback report include:

[0064] B1. When a fault decision feedback report includes a scheduled maintenance report or an emergency repair report, determine whether the fault event actually occurred. If so, retrieve the full-link log. If not, generate a non-responsible party responsibility report.

[0065] B2: Based on the full-link log, locate the error link and determine whether the error value in error link 1 is greater than the error threshold. If so, generate a responsibility report for the AI algorithm team. If not, proceed to error determination link 2.

[0066] B3, judging whether there is an execution problem in step 2. If yes, a maintenance team responsibility report is generated. If not, the process proceeds to step 3;

[0067] B4, Error Judgment Step 3: Is there a data problem? If yes, generate a data provision team responsibility report. If no, proceed to Error Judgment Step 4;

[0068] B5: Determine if there is a line material problem in error step 4. If yes, generate a line installation team responsibility report; if no, generate a non-responsible party responsibility report;

[0069] B6: Package the AI algorithm team responsibility report, the maintenance team responsibility report, the data provider team responsibility report, the line installation team responsibility report, and the non-responsible party responsibility report to obtain a safety responsibility report.

[0070] Data communication module, used to process fault decision feedback reports and transmit them in conjunction with safety responsibility reports;

[0071] Furthermore, the method for processing the fault decision feedback report and transmitting it in combination with the safety responsibility report includes:

[0072] When the fault decision feedback report is a preventive inspection report, the power line location data is integrated and sent to the staff receiving end in a non-real-time manner;

[0073] When the fault decision feedback report is a dispatch maintenance report or an emergency repair report, it integrates the power line location data, nearby maintenance team data and estimated repair time, and sends it to the staff and maintenance team receiving end in real time;

[0074] Combine the fault decision feedback report with the safety responsibility report and send it to the management personnel in a non-real-time manner;

[0075] Further, S1: collecting a benchmark data set, the benchmark data set includes line length data, line usage data, line material data and environmental risk data;

[0076] S2: Collecting a fluctuation data set, which includes real-time load data and line temperature data;

[0077] S3: Preprocess the benchmark dataset and the fluctuation dataset to obtain the feature dataset;

[0078] S4: Analyze the benchmark data set and the feature data set to obtain the fault prediction value;

[0079] S5: Process the fault prediction value to obtain a fault decision feedback report;

[0080] S6: Analyze the fault decision feedback report to obtain a safety responsibility report;

[0081] S7: Process the fault decision feedback report and transmit it in conjunction with the safety responsibility report.

[0082] The technical effects and advantages of the artificial intelligence-based power line fault location and repair system of the present invention are as follows:

[0083] The present invention collects a benchmark data set, which includes line length data, line usage data, line material data, and environmental risk data; collects a fluctuation data set, which includes real-time load data and line temperature data; pre-processes the benchmark data set and the fluctuation data set to obtain a feature data set; analyzes the benchmark data set and the feature data set to obtain a fault prediction value; processes the fault prediction value to obtain a fault decision feedback report; analyzes the fault decision feedback report to obtain a safety responsibility report; processes the fault decision feedback report, and transmits it in conjunction with the safety responsibility report. This enables the system to maintain high-precision fault prediction capabilities in complex usage scenarios, thereby effectively improving the practicality and applicability of the system. In addition, the present invention can quickly generate corresponding auxiliary decision reports through analysis of the fault prediction values, greatly reducing the decision-making time required for staff to respond to sudden faults, thereby effectively ensuring the economic benefits of the power company. At the same time, through analysis of the fault decision feedback report, the system's responsibility tracing capabilities can be effectively improved, providing a data basis for the division of responsibility standards for power companies when they suffer losses during the maintenance process. Overall, the present invention has the significant advantages of high fault prediction accuracy, good fault auxiliary decision-making effects, and strong responsibility tracing capabilities for losses during the maintenance process. BRIEF DESCRIPTION OF THE DRAWINGS

[0084] Figure 1 Schematic diagram of the power line fault location and repair system based on artificial intelligence of the present invention;

[0085] Figure 2 Schematic diagram of the artificial intelligence-based power line fault location and repair method of the present invention. DETAILED DESCRIPTION

[0086] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0087] The terms used in the embodiments of the present invention are for the purpose of describing specific embodiments only and are not intended to limit the present invention. The singular forms "a," "an," "the," and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, and unless the context clearly indicates otherwise, "a plurality" generally includes at least two.

[0088] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting," depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to the determination" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)," depending on the context.

[0089] In addition, the step sequence in the following method embodiments is only an example and not a strict limitation.

[0090] In practice, the server-side device deployed in the AI-based power line fault location and repair system may consist of one or more devices. The AI-based power line fault location and repair system can be implemented as a service instance, a virtual machine, or a hardware device. For example, the AI-based power line fault location and repair system can be implemented as a service instance deployed on one or more devices in a cloud node. Simply put, the AI-based power line fault location and repair system can be understood as software deployed on a cloud node, providing the AI-based power line fault location and repair system to each user. Alternatively, the AI-based power line fault location and repair system can be implemented as a virtual machine deployed on one or more devices in a cloud node. This virtual machine contains application software for managing each user. Alternatively, the AI-based power line fault location and repair system can be implemented as a server-side device composed of multiple hardware devices of the same or different types, with one or more hardware devices configured to provide the AI-based power line fault location and repair system to each user.

[0091] In terms of implementation, the AI-based power line fault location and repair system and the user end are mutually compatible. Specifically, if the AI-based power line fault location and repair system is an application installed on a cloud service platform, the user end serves as a client that establishes a communication connection with the application. Alternatively, if the AI-based power line fault location and repair system is implemented as a website, the user end serves as a webpage. Alternatively, if the AI-based power line fault location and repair system is implemented as a cloud service platform, the user end serves as a mini-program within an instant messaging application.

[0092] like Figure 1 , which is a system architecture diagram of an artificial intelligence-based power line fault location and repair system provided by one embodiment of the present invention.

[0093] The power line fault location and repair system based on artificial intelligence of the present invention can be set in a cloud server. In terms of implementation, it can be used as one or more service devices, or as an application installed on the cloud (such as a mobile service operator's server, server cluster, etc.), or it can also be developed as a website. According to the functions implemented, the power line fault location and repair system based on artificial intelligence may include a baseline data acquisition module, a fluctuation data acquisition module, a data processing module, a fault prediction value generation module, a dynamic resource decision module, a safety problem tracing module and a data communication module. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0094] In an embodiment of the present invention, in the power line fault location and repair system based on artificial intelligence, each of the above modules can be implemented independently and called with other modules. The call here can be understood as that a certain module can connect to multiple modules of another type and provide corresponding services to the multiple modules connected to it. For example, the sharing evaluation module can call the same information acquisition module to obtain the information collected by the information acquisition module. Based on the above characteristics, in the power line fault location and repair system based on artificial intelligence provided by the embodiment of the present invention, the scope of application of the power line fault location and repair system architecture based on artificial intelligence can be adjusted by adding modules and directly calling them without modifying the program code, thereby realizing cluster-type horizontal expansion, so as to achieve the purpose of quickly and flexibly expanding the power line fault location and repair system based on artificial intelligence. In actual applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in cloud servers.

[0095] Example 1

[0096] See also Figure 1 As shown, the power line fault location and repair system based on artificial intelligence in this embodiment includes:

[0097] The benchmark data acquisition module is used to acquire a benchmark data set, which includes line length data, line usage data, line material data and environmental risk data;

[0098] It should be explained that the length values of power lines in a specified area are collected through the geographic information system to obtain line length data; the commissioning date records of power lines in a specified area are collected through the asset management system, and the commissioning date is subtracted from the current time to obtain line usage data; the material data of power lines in a specified area are collected through the equipment ledger database and assigned values to obtain line material data. The specific expression formula for the assignment is: Through the geographic information system, the environmental classification in the designated area is collected and assigned values to obtain environmental risk data. The specific expression formula for the assignment is: ;

[0099] The fluctuation data acquisition module is used to collect a fluctuation data set, which includes real-time load data and line temperature data;

[0100] It should be explained that the instantaneous current value of the power line in the specified area is collected through the current transformer to obtain the real-time load data; the surface temperature value of the power line in the specified area is collected through the optical fiber temperature sensor to obtain the line temperature data;

[0101] The data processing module is used to pre-process the reference data set and the fluctuation data set to obtain a feature data set;

[0102] Furthermore, the preprocessing methods for the benchmark dataset and the fluctuation dataset include:

[0103] A1. By quantifying the aging risk of line usage data, the specific calculation formula for aging risk quantification is:

[0104] ;

[0105] Get aging characteristic data ,in, Use data for the line, is the maximum permissible service life;

[0106] A2 calculates the average load characteristics based on real-time load data. The specific calculation formula for the average load characteristics is:

[0107] ;

[0108] Get average load characteristic data ,in, is the sample size, For the Real-time load data;

[0109] A3, by calculating the temperature change characteristics of the line temperature data, the specific calculation formula of the temperature change characteristics is:

[0110] ;

[0111] Get temperature change characteristic data ,in, Current time Line temperature data , For the previous time point Line temperature data , is the time interval;

[0112] A4. Risk characteristics are calculated based on the average load characteristic data and the temperature change characteristic data. The specific calculation formula for risk characteristics is:

[0113] ;

[0114] Get risk profile data ,in, is the maximum allowable current value, is the maximum allowable line temperature value;

[0115] A5, packing aging characteristic data, average load characteristic data, temperature change characteristic data and risk characteristic data to obtain a characteristic data set;

[0116] The fault prediction value generation module is used to analyze the reference data set and the feature data set to obtain a fault prediction value;

[0117] Furthermore, the fault prediction value generation module also includes a historical data reading module, a system model building module, a data support module and a prediction value transmission module, wherein:

[0118] The historical data reading module is used to support the system to retrieve the benchmark data set and feature data set stored in the database;

[0119] The system model building module is used to support the system in building the model required for calculation;

[0120] The data support module is used to support manual input of actual fault values, preset cycles and detection thresholds, and support system retrieval;

[0121] The predicted value transmission module is used to support the system in transmitting the fault prediction value;

[0122] Furthermore, the steps of analyzing the benchmark dataset and the feature dataset include:

[0123] Step 1: Based on the historical data reading module, a set of historical benchmark data sets and historical feature data sets stored in the database are extracted and grouped and labeled accordingly according to the order of timestamps from recent to far. The labeling results are L1, L2, L3, ..., Ln, and the labeling results are used as the sample set;

[0124] Step 2: Based on the system model construction module, the sample set is divided into an 80% training set and a 20% training set to establish a risk prediction model;

[0125] Step 3: Substitute into the calculation formula:

[0126] ;

[0127] Get the fault prediction value ,in, is the dynamic weight factor, is the activation function, is the number of feature data combinations, For the The feature weight factor of the feature data combination, For the The characteristic unit of the characteristic data combination, is the logarithm, 、 and is the physical weight factor;

[0128] It needs to be explained that the activation function is used to convert The calculation result is constrained to the range of 0 to 1;

[0129] Step 4: Based on the fault prediction value in step 3 and the actual fault value in the data support module, calculate the error of the fault prediction value. The specific calculation formula of the error is:

[0130] ;

[0131] Get the error value ,in, is the number of training set samples, For the Actual fault value , For the Failure prediction value, is the maximum allowable value of the fault prediction value;

[0132] According to the error threshold in the data support module, when the error value is greater than or equal to the error threshold, the dynamic feature weight factor is adjusted, and the process returns to step 3 to substitute the adjusted dynamic feature weight factor into the fault prediction value calculation formula. When the error value is less than the error threshold, the fault prediction value is output to step 6.

[0133] Step 5: Based on the feature weight factor in step 3 and according to the preset period in the data support module, the feature weight factor is updated. The specific calculation formula for the update is:

[0134] ;

[0135] Get the new feature weight factor ,in, is the normalization function, is the temperature parameter, For the The mean square error of the feature data combination;

[0136] It should be explained that the normalization function is used to distribute weights; the temperature parameter is used to control the concentration of weight distribution, and the temperature parameter is greater than 0;

[0137] The updated new feature weight factor Substitute into the fault prediction value calculation formula in step 3;

[0138] Step 6: Based on the fault prediction value in step 4, and according to the detection threshold and validation set in the data support module, the fault prediction value is tested. The specific expression formula of the test is:

[0139]

[0140] in, is the number of samples in the validation set, For samples The failure prediction value of The previous risk prediction model for the sample The calculated failure prediction value, is the detection threshold;

[0141] when When satisfied, output the fault prediction value to step seven. In case of rejection, return to step 1 to retrain the risk prediction model;

[0142] Step 7: Based on the prediction value transmission module, the fault prediction value is output to the dynamic resource decision module;

[0143] Furthermore, the feature unit The specific expression formula is:

[0144] ;

[0145] in, is the hyperbolic tangent function, is the line length data, is the base, 、 、 and For the Dynamic feature weight factors;

[0146] It needs to be explained that The calculation result is constrained to be within the range of (-1, 1);

[0147] Furthermore, the specific calculation formula for adjusting the parameters of the dynamic feature weight factor is:

[0148] ;

[0149] Get the new dynamic feature weight factor ,in, For the Dynamic feature weight factor , Including dynamic feature weight factors 、 、 and , is the learning rate, For the The partial derivative of the characteristic unit, For the Dynamic feature weight factor The partial derivative of

[0150] The dynamic resource decision module is used to process the fault prediction value and obtain a fault decision feedback report;

[0151] Furthermore, the method for processing the fault prediction value includes:

[0152] When the fault prediction value is less than 30, a preventive inspection report is generated; when the fault prediction value is greater than or equal to 30 and less than 70, a scheduling maintenance report is generated; when the fault prediction value is greater than or equal to 70, an emergency repair report is generated;

[0153] The preventive inspection report includes a statement that the predicted failure level is low and asks staff to conduct preventive inspections according to the maintenance cycle;

[0154] The dispatch maintenance report includes a description of the predicted fault extent, and the staff is asked to notify the nearby maintenance team to go to the faulty line section for inspection and maintenance;

[0155] The emergency repair report includes a statement indicating that the fault is predicted to be severe and that staff are requested to immediately cut off the power supply to the faulty line section and notify nearby maintenance teams to inspect and repair the faulty line section;

[0156] Package preventive inspection reports, scheduled maintenance reports, and emergency repair reports to obtain fault decision feedback reports;

[0157] The safety problem tracing module is used to analyze the fault decision feedback report to obtain a safety responsibility report;

[0158] Further, the methods for analyzing the fault decision feedback report include:

[0159] B1. When a fault decision feedback report includes a scheduled maintenance report or an emergency repair report, determine whether the fault event actually occurred. If so, retrieve the full-link log. If not, generate a non-responsible party responsibility report.

[0160] It should be explained that the full-link log includes error values, fault decision feedback reports, and actual results;

[0161] B2: Based on the full-link log, locate the error link and determine whether the error value in error link 1 is greater than the error threshold. If so, generate a responsibility report for the AI algorithm team. If not, proceed to error determination link 2.

[0162] B3, judging whether there is an execution problem in step 2. If yes, a maintenance team responsibility report is generated. If not, the process proceeds to step 3;

[0163] It should be explained that implementation problems include problems such as the maintenance team not arriving in time or causing damage to power lines during the maintenance process;

[0164] B4, Error Judgment Step 3: Is there a data problem? If yes, generate a data provision team responsibility report. If no, proceed to Error Judgment Step 4;

[0165] It needs to be explained that data problems include data errors in the collected data;

[0166] B5: Determine if there is a line material problem in error step 4. If yes, generate a line installation team responsibility report; if no, generate a non-responsible party responsibility report;

[0167] It should be explained that line material issues include line material data not meeting line material usage standards;

[0168] B6: Package the AI algorithm team responsibility report, the maintenance team responsibility report, the data provider team responsibility report, the line installation team responsibility report, and the non-responsible party responsibility report to obtain a safety responsibility report.

[0169] The data communication module is used to process the fault decision feedback report and transmit it in conjunction with the safety responsibility report;

[0170] Furthermore, the method for processing the fault decision feedback report and transmitting it in combination with the safety responsibility report includes:

[0171] When the fault decision feedback report is a preventive inspection report, the power line location data is integrated and sent to the staff receiving end in a non-real-time manner;

[0172] It needs to be explained that non-real-time methods such as email;

[0173] When the fault decision feedback report is a dispatch maintenance report or an emergency repair report, it integrates the power line location data, nearby maintenance team data and estimated repair time, and sends it to the staff and maintenance team receiving end in real time;

[0174] Combine the fault decision feedback report with the safety responsibility report and send it to the management personnel in a non-real-time manner;

[0175] The present embodiment has the following beneficial effects: By collecting a benchmark data set, which includes line length data, line usage data, line material data, and environmental risk data, collecting a fluctuation data set, which includes real-time load data and line temperature data, preprocessing the benchmark data set and the fluctuation data set to obtain a feature data set, analyzing the benchmark data set and the feature data set to obtain a fault prediction value, processing the fault prediction value to obtain a fault decision feedback report, analyzing the fault decision feedback report to obtain a safety responsibility report, processing the fault decision feedback report, and transmitting it in conjunction with the safety responsibility report, the system can maintain high-precision fault prediction capabilities in the face of complex usage scenarios, thereby effectively improving the practicality and applicability of the system. In addition, the present invention can quickly generate corresponding auxiliary decision reports through analysis of the fault prediction values, greatly reducing the decision-making time required for staff to respond to sudden faults, thereby effectively ensuring the economic benefits of the power company. At the same time, through analysis of the fault decision feedback report, the responsibility tracing capability of the system can be effectively improved, providing a data basis for the division of responsibility standards for power companies when they suffer losses during the maintenance process. Overall, the present invention has the significant advantages of high fault prediction accuracy, good fault auxiliary decision-making effect, and strong responsibility tracing capability for losses during the maintenance process.

[0176] Example 2

[0177] See also Figure 2 As shown, for parts not described in detail in this embodiment, please refer to the description of Example 1. A method for locating and repairing power line faults based on artificial intelligence is provided, the method comprising: S1: collecting a benchmark data set, the benchmark data set including line length data, line usage data, line material data, and environmental risk data;

[0178] S2: Collecting a fluctuation data set, which includes real-time load data and line temperature data;

[0179] S3: Preprocess the benchmark dataset and the fluctuation dataset to obtain the feature dataset;

[0180] S4: Analyze the benchmark data set and the feature data set to obtain the fault prediction value;

[0181] S5: Process the fault prediction value to obtain a fault decision feedback report;

[0182] S6: Analyze the fault decision feedback report to obtain a safety responsibility report;

[0183] S7: Process the fault decision feedback report and transmit it in conjunction with the safety responsibility report.

[0184] Example 3

[0185] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0186] Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims rather than the foregoing description, and all changes that come within the meaning and range of equivalents of the claims are intended to be embraced therein. Any reference to a figure in a claim should not be construed as limiting the claim to which it relates.

[0187] The embodiments of the present application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, techniques, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to achieve optimal results.

[0188] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. Terms such as "first" and "second" are used to indicate names and do not imply any particular order.

[0189] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. The power line fault location and repair system based on artificial intelligence is characterized by: The system includes: a reference data acquisition module, a fluctuation data acquisition module, a data processing module, a fault prediction value generation module, a dynamic resource decision module, a safety problem tracing module and a data communication module, wherein: The benchmark data acquisition module is used to acquire a benchmark data set, which includes line length data, line usage data, line material data and environmental risk data; The fluctuation data acquisition module is used to collect a fluctuation data set, which includes real-time load data and line temperature data; The data processing module is used to pre-process the reference data set and the fluctuation data set to obtain a feature data set; The fault prediction value generation module is used to analyze the reference data set and the feature data set to obtain a fault prediction value; The dynamic resource decision module is used to process the fault prediction value and obtain a fault decision feedback report; The safety problem tracing module is used to analyze the fault decision feedback report to obtain a safety responsibility report; The data communication module is used to process the fault decision feedback report and transmit it in combination with the safety responsibility report.

2. The artificial intelligence-based power line fault location and repair system according to claim 1, characterized in that: The preprocessing methods for the benchmark dataset and the fluctuation dataset include: A1. By quantifying the aging risk of line usage data, the specific calculation formula for aging risk quantification is: ； Get aging characteristic data ,in, Use data for the line, is the maximum permissible service life; A2 calculates the average load characteristics based on real-time load data. The specific calculation formula for the average load characteristics is: ； Get average load characteristic data ,in, is the sample size, For the Real-time load data; A3, by calculating the temperature change characteristics of the line temperature data, the specific calculation formula of the temperature change characteristics is: ； Get temperature change characteristic data ,in, Current time Line temperature data , For the previous time point Line temperature data , is the time interval; A4. Risk characteristics are calculated based on the average load characteristic data and the temperature change characteristic data. The specific calculation formula for risk characteristics is: ； Get risk profile data ,in, is the maximum allowable current value, is the maximum allowable line temperature value; A5, packages the aging characteristic data, average load characteristic data, temperature change characteristic data and risk characteristic data to obtain a characteristic data set.

3. The artificial intelligence-based power line fault location and repair system according to claim 1, characterized in that: The fault prediction value generation module also includes a historical data reading module, a system model construction module, a data support module and a prediction value transmission module, wherein: The historical data reading module is used to support the system to retrieve the benchmark data set and feature data set stored in the database; The system model building module is used to support the system in establishing the model required for calculation; The data support module is used to support manual input of actual fault values, preset cycles and detection thresholds, and support system retrieval; The predicted value transmission module is used to support the system in transmitting the fault prediction value.

4. The artificial intelligence-based power line fault location and repair system according to claim 3, characterized in that: The steps for analyzing the benchmark dataset and feature dataset include: Step 1: Based on the historical data reading module, a set of historical benchmark data sets and historical feature data sets stored in the database are extracted and grouped and labeled accordingly according to the order of timestamps from recent to far. The labeling results are L1, L2, L3, ..., Ln, and the labeling results are used as the sample set; Step 2: Based on the system model construction module, the sample set is divided into an 80% training set and a 20% training set to establish a risk prediction model; Step 3: Substitute into the calculation formula: ； Get the fault prediction value ,in, is the dynamic weight factor, is the activation function, is the number of feature data combinations, For the The feature weight factor of the feature data combination, For the The characteristic unit of the characteristic data combination, is the logarithm, 、 and is the physical weight factor; Step 4: Based on the fault prediction value in step 3 and the actual fault value in the data support module, calculate the error of the fault prediction value. The specific calculation formula of the error is: ； Get the error value ,in, is the number of training set samples, For the Actual fault value , For the Failure prediction value, is the maximum allowable value of the fault prediction value; According to the error threshold in the data support module, when the error value is greater than or equal to the error threshold, the dynamic feature weight factor is adjusted, and the process returns to step 3 to substitute the adjusted dynamic feature weight factor into the fault prediction value calculation formula. When the error value is less than the error threshold, the fault prediction value is output to step 6. Step 5: Based on the feature weight factor in step 3 and according to the preset period in the data support module, the feature weight factor is updated. The specific calculation formula for the update is: ； Get the new feature weight factor ,in, is the normalization function, is the temperature parameter, For the The mean square error of the feature data combination; The updated new feature weight factor Substitute into the fault prediction value calculation formula in step 3; Step 6: Based on the fault prediction value in step 4, and according to the detection threshold and validation set in the data support module, the fault prediction value is tested. The specific expression formula of the test is: ； in, is the number of samples in the validation set, For samples The failure prediction value of The previous risk prediction model for the sample The calculated failure prediction value, is the detection threshold; when When satisfied, output the fault prediction value to step seven. In case of rejection, return to step 1 to retrain the risk prediction model; Step 7: Based on the prediction value transmission module, the fault prediction value is output to the dynamic resource decision module.

5. The artificial intelligence-based power line fault location and repair system according to claim 4 is characterized in that: Feature Unit The specific expression formula is: ； in, is the hyperbolic tangent function, is the line length data, is the base, 、 、 and For the Dynamic feature weighting factors.

6. The artificial intelligence-based power line fault location and repair system according to claim 4, characterized in that: The specific calculation formula for adjusting the parameters of the dynamic feature weight factor is: ； Get the new dynamic feature weight factor ,in, For the Dynamic feature weight factor , Including dynamic feature weight factors 、 、 and , is the learning rate, For the The partial derivative of the characteristic unit, For the Dynamic feature weight factor The partial derivative of .

7. The power line fault location and repair system based on artificial intelligence according to claim 1 is characterized in that: Methods for processing the fault prediction value include: When the fault prediction value is less than 30, a preventive inspection report is generated; when the fault prediction value is greater than or equal to 30 and less than 70, a scheduling maintenance report is generated; when the fault prediction value is greater than or equal to 70, an emergency repair report is generated; The preventive inspection report includes a statement that the predicted failure level is low and asks staff to conduct preventive inspections according to the maintenance cycle; The dispatch maintenance report includes a description of the predicted fault extent, and the staff is asked to notify the nearby maintenance team to go to the faulty line section for inspection and maintenance; The emergency repair report includes a statement indicating that the fault is predicted to be severe and that staff are requested to immediately cut off the power supply to the faulty line section and notify nearby maintenance teams to inspect and repair the faulty line section; Package preventive inspection reports, scheduled maintenance reports and emergency repair reports to obtain failure decision feedback reports.

8. The power line fault location and repair system based on artificial intelligence according to claim 1 is characterized in that: Methods for analyzing the fault decision feedback report include: B1. When a fault decision feedback report includes a scheduled maintenance report or an emergency repair report, determine whether the fault event actually occurred. If so, retrieve the full-link log. If not, generate a non-responsible party responsibility report. B2: Based on the full-link log, locate the error link and determine whether the error value in error link 1 is greater than the error threshold. If so, generate a responsibility report for the AI algorithm team. If not, proceed to error determination link 2. B3, judging whether there is an execution problem in step 2. If yes, a maintenance team responsibility report is generated. If not, the process proceeds to step 3; B4, Error Judgment Step 3: Is there a data problem? If yes, generate a data provision team responsibility report. If no, proceed to Error Judgment Step 4; B5: Determine if there is a line material problem in error step 4. If yes, generate a line installation team responsibility report; if no, generate a non-responsible party responsibility report; B6, packages the artificial intelligence algorithm team responsibility report, maintenance team responsibility report, data provision team responsibility report, line installation team responsibility report and non-responsible party responsibility report to obtain the safety responsibility report.

9. The artificial intelligence-based power line fault location and repair system according to claim 1, characterized in that: The methods for processing the fault decision feedback report and transmitting it in conjunction with the safety responsibility report include: When the fault decision feedback report is a preventive inspection report, the power line location data is integrated and sent to the staff receiving end in a non-real-time manner; When the fault decision feedback report is a dispatch maintenance report or an emergency repair report, it integrates the power line location data, nearby maintenance team data and estimated repair time, and sends it to the staff and maintenance team receiving end in real time; The fault decision feedback report is combined with the safety responsibility report and sent to the management receiving end in a non-real-time manner.

10. A method for locating and repairing power line faults based on artificial intelligence, implemented by the power line fault locating and repairing system based on artificial intelligence according to any one of claims 1 to 9, characterized in that: The following steps are included: S1: Collect benchmark data sets, including line length data, line usage data, line material data, and environmental risk data; S2: Collecting a fluctuation data set, which includes real-time load data and line temperature data; S3: Preprocess the benchmark dataset and the fluctuation dataset to obtain the feature dataset; S4: Analyze the benchmark data set and the feature data set to obtain the fault prediction value; S5: Process the fault prediction value to obtain a fault decision feedback report; S6: Analyze the fault decision feedback report to obtain a safety responsibility report; S7: Process the fault decision feedback report and transmit it in conjunction with the safety responsibility report.

Images