Distribution network inspection auxiliary system based on artificial intelligence

Abstract

The invention discloses a distribution network inspection auxiliary system based on artificial intelligence, and relates to the technical field of distribution network inspection, the distribution network inspection auxiliary system comprises a distribution network inspection platform, the distribution network inspection platform is in communication connection with a data acquisition module, an RFID identification module, a GIS map module, a data analysis module, an anomaly detection module and a report generation module, all the modules are connected through electric signals. According to the distribution network inspection auxiliary system based on artificial intelligence, the GPS track data of the inspection personnel and the state data of the equipment are received and processed in real time, the preset inspection route and the equipment parameter standard are automatically compared, and whether the inspection personnel deviates from the standard route and whether the equipment is in an abnormal state or not is recognized; the real-time feedback mechanism reduces the time and errors of manual inspection, so that the inspection work is more efficient and accurate, and meanwhile, the time sequence analysis is utilized to predict equipment faults in advance, thereby realizing preventive maintenance and avoiding the influence of sudden faults on the operation of a power grid.

Description

Technical Field

[0001] The present invention relates to the technical field of distribution network inspection, and in particular to a distribution network inspection auxiliary system based on artificial intelligence. Background Art

[0003] As the power equipment of the State Grid becomes increasingly complex, inspection work is crucial to ensuring the safe operation of the State Grid. The main responsibility of the inspectors is to monitor and check the health of transmission lines, substations, distribution equipment and other related infrastructure to ensure that these facilities can operate safely and stably. By comparing the scheduled inspection route with the actual walking track of the inspectors, it can be confirmed whether the inspectors strictly follow the prescribed route for inspection without missing or bypassing important checkpoints. Track comparison helps to detect the behavior of not performing tasks according to the established route as early as possible and take corrective measures.

[0004] However, traditional inspection and supervision rely on manual records and judgments, which are easily affected by human factors and are inefficient, resulting in accuracy being easily affected. With the continuous advancement of GPS positioning technology and data analysis methods, how to collect data from key sites through GPS and RFID technology, and use the geographic information system to compare the collected actual trajectory with the preset standard route to correct the behavior of not performing the task according to the established route is the problem we need to solve. To this end, an artificial intelligence-based distribution network inspection auxiliary system is proposed. Summary of the invention

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an artificial intelligence-based distribution network inspection auxiliary system, including a distribution network inspection platform, wherein the distribution network inspection platform is communicatively connected with a data acquisition module, an RFID identification module, a GIS map module, a data analysis module, an anomaly detection module, and a report generation module, wherein the modules are electrically connected with each other; The data acquisition module is used to collect key data during the inspection process, using GPS and RFID technology to automatically collect the location data of the inspection personnel and the identification information of key sites, and obtain equipment status data to ensure the real-time and accuracy of the data, providing a reliable basis for subsequent analysis and comparison; The RFID identification module uses RFID technology to quickly identify key sites encountered during the inspection process and record relevant information, thereby improving the efficiency and accuracy of equipment identification and reducing the possibility of human identification errors; The GIS map module is used to store and process geographic data, including preset inspection routes and actual walking tracks of inspection personnel, and visualize the map information of the inspection area, the actual walking tracks of inspection personnel, and the preset inspection routes, provide intuitive map views, and perform spatial data analysis, so that managers can intuitively understand the inspection progress and route conditions, and facilitate remote monitoring and command; The data analysis module is used to conduct in-depth analysis of the collected data, including comparison of the actual walking trajectory with the preset inspection route, analysis of equipment status data, analysis of potential problems and anomalies, such as failure to perform tasks according to the established route, equipment failure, etc., to provide support for subsequent decision-making; The detection alarm module automatically triggers an alarm mechanism according to the degree of abnormality when it finds that the inspection personnel deviate from the standard route or the equipment status is abnormal, and promptly notifies relevant personnel to take corrective measures to prevent the problem from escalating; The report generation module is used to generate an inspection report based on the abnormal detection and scheduling results, including the walking trajectory of the inspection personnel, equipment status, and abnormal situation information, providing comprehensive inspection feedback to the management to facilitate subsequent summary and improvement.

[0006] Preferably, in the data collection module, the process of collecting key data during the inspection process includes: The inspectors carry an inspection terminal with an integrated GPS module. The inspection terminal receives GPS satellite signals in real time, calculates and determines the inspectors’ current location (latitude and longitude information), and uploads the location data to the data center of the distribution network inspection platform in real time through wireless networks (such as 4G / 5G, Wi-Fi, etc.); RFID tags are deployed at key sites on the inspection route. Each RFID tag contains a unique electronic code to identify the site information, including site number, equipment type, and inspection requirements. Key sites include equipment to be inspected and preset inspection points. The inspection terminal carried by the inspection personnel has a built-in RFID reader / writer to read the information in the RFID tag and upload it to the data center of the distribution network inspection platform in real time through the wireless network; Various sensors (such as temperature sensors, pressure sensors, vibration sensors, etc.) are installed on the equipment to be inspected to monitor the operating status of the equipment in real time. The sensors convert the collected equipment operating status data (such as temperature value, pressure value, vibration frequency, etc.) into digital signals and transmit them to the data center of the distribution network inspection platform via wireless methods (such as Zigbee, LoRa, NB-IoT, etc.).

[0007] Preferably, in the RFID identification module, the process of identifying key sites in the inspection process includes: When the inspection personnel bring the inspection terminal close to the RFID tag, the built-in RFID reader / writer of the inspection terminal sends a radio frequency signal of a preset frequency. After receiving the radio frequency signal, the RFID tag obtains energy based on the coil induction current on the RFID tag, activating the control circuit and radio frequency circuit in the tag; After activation, the RFID tag converts the information stored in the chip (such as site number, equipment type, inspection requirements, etc.) into a response signal and sends it back to the RFID reader through the antenna; After the RFID reader receives the response signal sent by the RFID tag, it decodes and processes it to restore the original information. The restored original information is uploaded to the data center of the distribution network inspection platform in real time through wireless networks (such as 4G / 5G, Wi-Fi, etc.).

[0008] Preferably, in the GIS map module, the process of visually displaying the map information of the inspection area, the actual walking track of the inspection personnel, and the preset inspection route includes: Collect geographical data of the inspection area, including map information, terrain, buildings, roads, basic information of equipment location, and plan preset inspection routes based on the geographical data of the inspection area; Use GPS technology to track the location of inspectors in real time, record their walking tracks, store the actual walking tracks of inspectors and the preset inspection route data in the GIS system, and pre-process the stored data, including data cleaning, format conversion, coordinate unification, etc., to ensure the accuracy and consistency of the data; Visually display the map information of the inspection area, the actual walking tracks of the inspectors, and the preset inspection routes on the GIS map, analyze the differences between the actual walking tracks of the inspectors and the preset inspection routes, and identify deviations and omissions; Dynamically mark key sites, equipment locations, and fault point information during the inspection process on the map, and associate the contextual information of the inspection process; Utilizing the spatial analysis function of GIS, we optimize and analyze the walking paths of patrol personnel, provide suggestions for the shortest path, and spatially divide the patrol areas. We analyze the patrol difficulty and equipment density characteristics of different areas to provide a basis for the formulation of patrol plans.

[0009] Preferably, in the data analysis module, the process of analyzing potential problems and anomalies includes: Extract all relevant raw data from the inspection process from the data center of the distribution network inspection platform, including the actual walking tracks of the inspectors, the data read by the RFID tags, and the equipment status data collected by the sensors, and integrate the data from different sources to ensure the consistency and integrity of the data; Preprocess all relevant raw data extracted during the inspection process and unify the coordinate system. Improve data quality through preprocessing steps such as data cleaning, denoising, filling missing values, and data conversion (such as normalization and standardization), laying a good foundation for subsequent analysis. The location and time information of each track point is parsed from the actual walking track data of the inspectors, and the preset inspection route in the system is loaded. The actual walking track is compared with the preset inspection route by using the Hausdorff distance calculation method to obtain the track evaluation index and evaluate the difference between the two. According to the comparison results, the situations where the tasks are not performed according to the established route, such as deviation from the route and missed inspection areas, are identified and marked as abnormal; The equipment status data is classified and aggregated according to the dimensions of equipment type and time period for more in-depth analysis. The standard threshold of each equipment parameter is set. By comparing the actual data with the threshold, it is determined whether the equipment is in a normal state. The time series analysis method is used to analyze the change trend of the equipment status parameters over time, obtain the equipment anomaly assessment index, predict the failure trend, identify the parameter values ​​with abnormal trends, and mark them as potential equipment failure problems.

[0010] Preferably, the expression of the trajectory evaluation index is:

[0011] Among them, TEI is the trajectory evaluation index, which is used to quantify the difference between the actual walking trajectory and the preset inspection route, N is the number of points in the actual walking trajectory, p_i is the i-th point in the actual walking trajectory, q_h is the h-th point in the preset inspection route, d(p_i, q_h) is the Euclidean distance between points p_i and q_h, and the smaller the TEI value, the closer the actual walking trajectory is to the preset inspection route; The expression of the equipment abnormality evaluation index is: ; Among them, DEI is the equipment abnormality evaluation index, which is used to evaluate the changing trend of equipment status parameters over time and predict potential failures. T is the total number of time points, x_t is the ideal state value of the equipment at the tth moment, y_t is the actual measurement value of the equipment at the tth moment, and θ is the adjustment index, which is used to adjust the sensitivity of the difference between the ideal value and the actual value. The larger the value of DEI, the more the equipment state deviates from the ideal state and the higher the potential failure risk.

[0012] Preferably, in the detection alarm module, the process of automatically triggering the alarm mechanism according to the degree of abnormality includes: Receive GPS trajectory data uploaded by patrol personnel in real time, use the preset patrol route and Hausdorff distance calculation method to compare the actual walking trajectory of patrol personnel with the preset patrol route, obtain the trajectory evaluation index, and set the trajectory deviation threshold. When the trajectory evaluation index exceeds the trajectory deviation threshold, it is determined that the patrol personnel deviates from the standard route; Receive the equipment status data uploaded by various sensors in real time, compare and analyze the received data according to the preset equipment parameter standard threshold, apply the time series analysis method, predict the change trend of equipment status parameters, calculate the equipment abnormality evaluation index, and set the abnormality evaluation threshold. When the equipment abnormality evaluation index exceeds the threshold, it is judged that the equipment status is abnormal; According to the severity of the inspection personnel's trajectory deviation and the potential risk of equipment abnormality, different weights are assigned to the trajectory evaluation index and the equipment abnormality evaluation index. The trajectory evaluation index and the equipment abnormality evaluation index are weighted and calculated according to the assigned weights to obtain the abnormality warning coefficient. Different alarm levels are set according to the abnormal warning coefficient, namely, the first alarm level, the second alarm level and the third alarm level. The alarm level increases gradually from the first to the third abnormality level. In combination with the urgency of the inspection task and the severity of the consequences of the failure, the alarm threshold of the abnormal warning coefficient is set to match the alarm level. Combined with the calculation results of abnormality detection and abnormality warning coefficient, the current alarm level is determined, and the abnormality degree of the current inspection process is analyzed to generate corresponding alarm information, where the alarm information includes the abnormality type (inspection personnel deviate from the route or equipment status is abnormal), abnormal location (specific coordinates or equipment number), alarm level (quantitatively expressed by warning coefficient) and recommended corrective measures; Through multiple channels such as SMS, email, and instant messaging software, the alarm information is sent to relevant personnel (such as inspection team leaders, operation and maintenance personnel, equipment managers, etc.) in a timely manner to ensure that relevant personnel can quickly receive the alarm information and understand the specific details and response measures of the abnormal situation.

[0013] Preferably, the expression of the abnormal warning coefficient is: ; Among them, AWC is the abnormal warning coefficient, α and β are weight factors used to adjust the relative importance of trajectory evaluation index and equipment abnormal evaluation index in the total warning coefficient, TEI_k is the trajectory evaluation index at the kth time point, (TEI) ̅ is the average of the trajectory evaluation index at all time points, n is the number of data points of the trajectory evaluation index, DEI_j is the state abnormal evaluation index of the jth device, m is the number of data points of the equipment abnormal evaluation index, and the higher the AWC, the more serious the abnormal situation.

[0014] Preferably, the plurality of alarm levels correspond to the plurality of alarm thresholds, the alarm thresholds include an upper threshold and a lower threshold, and the correspondence between the plurality of alarm levels and the plurality of alarm thresholds is: First level alarm level AWC≤AWC_M; Secondary alarm level AWC_M <AWC≤AWC_H; The third level alarm level AWC>AWC_H; Among them, AWC is the abnormal warning coefficient, EC_yg is the lower threshold corresponding to the second-level alarm level and the upper threshold corresponding to the first-level alarm level, and AWC_H is the lower threshold corresponding to the third-level alarm level and the upper threshold corresponding to the second-level alarm level.

[0015] Preferably, in the report generation module, the generation process of the inspection report includes: Restoring and comparing the walking paths of the inspectors to evaluate whether the inspection is completed according to the preset route, marking the specific location and time of deviation from the route, evaluating the health status and potential risks of the equipment based on the equipment status data and the preset thresholds, and determining the alarm level of the detected abnormal situation based on the abnormal detection and scheduling results; Based on the determined alarm level and the corresponding alarm information, an inspection report is generated, including but not limited to: Inspection overview: basic inspection information, such as inspection time, inspection personnel, inspection area, etc. Walking trajectory analysis: walking trajectory map of inspection personnel, comparison results with preset inspection routes, detailed description of deviations, equipment status report: summary information of equipment status, including a list of equipment in normal and abnormal states, trend analysis of key parameters, abnormal situation summary: list of abnormal events, location, time, type, possible causes and recommended corrective measures, inspection efficiency analysis: completion of inspection tasks, coverage of inspection routes, efficiency evaluation of inspection personnel; The contents of the inspection reports are standardized and distributed to relevant personnel, and the reports are archived for easy query and analysis.

[0016] The present invention provides a distribution network inspection auxiliary system based on artificial intelligence. It has the following beneficial effects: 1. The distribution network inspection auxiliary system based on artificial intelligence receives and processes the GPS trajectory data of the inspectors and the status data of the equipment in real time, automatically compares the preset inspection routes and equipment parameter standards, and identifies whether the inspectors deviate from the standard routes and whether the equipment is in an abnormal state. The instant feedback mechanism reduces the time and error of manual inspection, making the inspection work more efficient and accurate. At the same time, it uses time series analysis to predict equipment failures in advance, realize preventive maintenance, and avoid the impact of sudden failures on power grid operation.

[0017] 2. The distribution network inspection auxiliary system based on artificial intelligence ensures that the inspectors perform their tasks according to the established inspection routes by comparing their walking trajectories with the preset inspection routes in real time. It uses the Hausdorff distance calculation method to accurately identify deviations in the inspection paths, thereby reducing missed inspections and false inspections. In addition, by identifying key sites and equipment through RFID technology, it reduces recognition errors caused by human factors, improves the accuracy and reliability of inspection work, and ensures the stable operation of power equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a block diagram of the distribution network inspection auxiliary system based on artificial intelligence of the present invention; Figure 2 Flowchart for analyzing potential problems and anomalies for the present invention; Figure 3 This is a flow chart of the present invention which automatically triggers an alarm mechanism according to the degree of abnormality. DETAILED DESCRIPTION

[0019] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. The embodiments of the present invention are provided for the purpose of illustration and description, and are not intended to be exhaustive or to limit the present invention to the disclosed forms. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiments are selected and described in order to better illustrate the principles and practical applications of the present invention, and to enable those of ordinary skill in the art to understand the present invention and thereby design various embodiments with various modifications suitable for specific uses.

[0020] The first embodiment, as Figure 1 , Figure 2 As shown, the present invention provides a technical solution: a distribution network inspection auxiliary system based on artificial intelligence, including a distribution network inspection platform, the distribution network inspection platform is communicatively connected with a data acquisition module, an RFID identification module, a GIS map module, a data analysis module, an abnormality detection module and a report generation module, wherein the electrical signals between the modules are connected; The data acquisition module is used to collect key data during the inspection process. It uses GPS and RFID technology to automatically collect the location data of the inspectors and the identification information of key sites, and obtains the equipment status data to ensure the real-time and accuracy of the data, providing a reliable basis for subsequent analysis and comparison. The inspectors carry an inspection terminal with an integrated GPS module. The inspection terminal receives GPS satellite signals in real time, calculates and determines the current location of the inspectors (latitude and longitude information), and uploads the location data to the data center of the distribution network inspection platform in real time through wireless networks (such as 4G / 5G, Wi-Fi, etc.). RFID tags are deployed at key sites on the inspection route. Each RFID tag contains a unique electronic code for The information identifying the site includes the site number, equipment type, and inspection requirements. The key sites include the equipment to be inspected and the preset inspection points. The inspection terminal carried by the inspectors has a built-in RFID reader / writer to read the information in the RFID tag and upload it to the data center of the distribution network inspection platform in real time through the wireless network. Various sensors (such as temperature sensors, pressure sensors, vibration sensors, etc.) are installed on the equipment to be inspected to monitor the operating status of the equipment in real time. The sensors convert the collected equipment operating status data (such as temperature value, pressure value, vibration frequency, etc.) into digital signals and transmit them to the data center of the distribution network inspection platform through wireless methods (such as Zigbee, LoRa, NB-IoT, etc.); The RFID identification module uses RFID technology to quickly identify key sites encountered during the inspection process and record relevant information to improve the efficiency and accuracy of equipment identification and reduce the possibility of human identification errors. When the inspector carries the inspection terminal close to the RFID tag, the built-in RFID reader in the inspection terminal sends a radio frequency signal of a preset frequency. After receiving the radio frequency signal, the RFID tag obtains energy based on the coil induction current on the RFID tag to activate the control circuit and radio frequency circuit in the tag. After activation, the RFID tag converts the information stored in the chip (such as site number, equipment type, inspection requirements, etc.) into a response signal and sends it back to the RFID reader through the antenna. After receiving the response signal sent by the RFID tag, the RFID reader decodes and processes it to restore the original information. The restored original information is uploaded to the data center of the distribution network inspection platform in real time through a wireless network (such as 4G / 5G, Wi-Fi, etc.); The GIS map module is used to store and process geographic data, including preset inspection routes and the actual walking tracks of inspection personnel. It also visualizes the map information of the inspection area, the actual walking tracks of inspection personnel, and the preset inspection routes, provides intuitive map views, and conducts spatial data analysis, so that managers can intuitively understand the inspection progress and route conditions, and facilitate remote monitoring and command. It collects geographic data of the inspection area, including map information, terrain, buildings, roads, and basic information on equipment locations. It also plans preset inspection routes based on the geographic data of the inspection area, uses GPS technology to track the location of inspection personnel in real time, records their walking tracks, and stores the actual walking tracks of inspection personnel and preset inspection route data in the GIS system. The system is preprocessed for the stored data, including data cleaning, format conversion, coordinate unification, etc., to ensure the accuracy and consistency of the data. The map information of the inspection area, the actual walking trajectory of the inspection personnel and the preset inspection route are visually displayed on the GIS map. The differences between the actual walking trajectory of the inspection personnel and the preset inspection route are analyzed to identify deviations and omissions. The key sites, equipment locations, and fault point information in the inspection process are dynamically marked on the map, and the context information of the inspection process is associated. The spatial analysis function of GIS is used to optimize the walking path of the inspection personnel, provide the shortest path suggestion, and spatially divide the inspection area. The inspection difficulty and equipment density characteristics of different areas are analyzed to provide a basis for the formulation of the inspection plan. The data analysis module is used to conduct in-depth analysis of the collected data, including the comparison of the actual walking trajectory with the preset inspection route, the analysis of the equipment status data, and the analysis of potential problems and anomalies, such as failure to perform tasks according to the established route, equipment failure, etc., to provide support for subsequent decision-making. All relevant raw data in the inspection process are extracted from the data center of the distribution network inspection platform, including the actual walking trajectory of the inspectors, the data read by the RFID tags, and the equipment status data collected by the sensors. The data from different sources are integrated to ensure the consistency and integrity of the data. All relevant raw data in the inspection process are pre-processed and the coordinate system is unified. Through pre-processing steps such as data cleaning, denoising, filling missing values, and data conversion (such as normalization and standardization), the data quality is improved, laying a good foundation for subsequent analysis. The location and time information of each trajectory point is parsed from the walking trajectory data, and the preset inspection route in the system is loaded. The actual walking trajectory is compared with the preset inspection route by using the Hausdorff distance calculation method to obtain the trajectory evaluation index, evaluate the difference between the two, and identify the situation where the task is not performed according to the established route, such as deviation from the route and missed inspection area, and mark it as abnormal. The equipment status data is classified and aggregated according to the dimensions of equipment type and time period for more in-depth analysis, and the standard threshold of each equipment parameter is set. By comparing the actual data with the threshold, it is determined whether the equipment is in a normal state. The time series analysis method is used to analyze the change trend of the equipment status parameters over time, obtain the equipment abnormality evaluation index, predict the failure trend, identify the parameter value with abnormal trend, and mark it as a potential equipment failure problem; Furthermore, the expression of trajectory evaluation index is:

[0021] Among them, TEI is the trajectory evaluation index, which is used to quantify the difference between the actual walking trajectory and the preset inspection route. N is the number of points in the actual walking trajectory, p_i is the i-th point in the actual walking trajectory, q_h is the h-th point in the preset inspection route, and d(p_i, q_h) is the Euclidean distance between points p_i and q_h. The smaller the TEI value, the closer the actual walking trajectory is to the preset inspection route. The maximum distance from each actual trajectory point to the nearest preset trajectory point is calculated, and these maximum distances are squared and averaged, and finally the square root is taken to ensure that even if only one point deviates seriously, it can significantly affect the overall evaluation; The expression of the equipment abnormality evaluation index is: ; Where DEI is the equipment abnormality evaluation index, which is used to evaluate the changing trend of equipment status parameters over time and predict potential failures. T is the total number of time points. x_t is the ideal state value of the equipment at the tth moment. y_t is the actual measured value of the equipment at the tth moment. θ is the adjustment index, which is used to adjust the sensitivity of the difference between the ideal value and the actual value. The larger the value of DEI, the more the equipment state deviates from the ideal state and the higher the potential failure risk. The logic function log(1+e^θ(x_t-y_t)) is used to calculate the abnormality of the equipment state at each time point. As time increases, these values ​​are averaged to observe long-term trends. The detection alarm module automatically triggers the alarm mechanism according to the degree of abnormality when it is found that the inspection personnel deviate from the standard route or the equipment status is abnormal, and promptly notifies the relevant personnel to take corrective measures to prevent the problem from escalating; The report generation module is used to generate inspection reports based on the abnormal detection and scheduling results, including the walking trajectory of the inspection personnel, equipment status, and abnormal situation information, providing comprehensive inspection feedback to the management to facilitate subsequent summary and improvement.

[0022] The second embodiment is based on the first embodiment. Figure 3 As shown, in the detection alarm module, the process of automatically triggering the alarm mechanism according to the degree of abnormality includes: Receive GPS trajectory data uploaded by patrol personnel in real time, use the preset patrol route and Hausdorff distance calculation method to compare the actual walking trajectory of patrol personnel with the preset patrol route, obtain the trajectory evaluation index, and set the trajectory deviation threshold. When the trajectory evaluation index exceeds the trajectory deviation threshold, it is determined that the patrol personnel deviates from the standard route. Receive equipment status data uploaded by various sensors in real time, compare and analyze the received data according to the preset equipment parameter standard threshold, apply time series analysis method to predict the change trend of equipment status parameters, calculate the equipment abnormality evaluation index, and set the abnormality evaluation threshold. When the equipment abnormality evaluation index exceeds the threshold, it is determined that the equipment status is abnormal. According to the severity of the patrol personnel's trajectory deviation and the potential risk of abnormal equipment status, different weights are assigned to the trajectory evaluation index and the equipment abnormality evaluation index. The trajectory evaluation index and the equipment abnormality evaluation index are weighted according to the assigned weights to obtain the abnormal warning coefficient. Different alarm levels are set according to the abnormal warning coefficient, namely the first alarm level, the second alarm level and the third alarm level. The alarm level increases gradually from the first to the third level. In combination with the urgency of the inspection task and the severity of the consequences of the failure, the alarm threshold of the abnormal warning coefficient is set to match the alarm level. The current alarm level is determined in combination with the calculation results of the abnormal detection and the abnormal warning coefficient. The abnormal degree of the current inspection process is analyzed to generate corresponding alarm information. The alarm information includes the abnormal type (the inspection personnel deviates from the route or the equipment status is abnormal), the abnormal location (specific coordinates or equipment number), the alarm level (quantitatively expressed by the warning coefficient) and the recommended corrective measures. The alarm information is sent to relevant personnel (such as the inspection team leader, operation and maintenance personnel, equipment administrator, etc.) in a timely manner through multiple channels such as SMS, email, and instant messaging software to ensure that relevant personnel can quickly receive the alarm information and understand the specific details of the abnormal situation and the response measures; Furthermore, the expression of the abnormal warning coefficient is: ; Among them, AWC is the abnormal warning coefficient, α and β are weight factors used to adjust the relative importance of trajectory evaluation index and equipment abnormal evaluation index in the total warning coefficient, TEI_k is the trajectory evaluation index at the kth time point, (TEI) ̅ is the average value of the trajectory evaluation index at all time points, n is the number of data points of the trajectory evaluation index, DEI_j is the state abnormal evaluation index of the jth device, m is the number of data points of the equipment abnormal evaluation index, the higher the AWC, the more serious the abnormal situation, and the increase of AWC with the increase of TEI and DEI means that the more serious the deviation and abnormal situation are, the higher the abnormal warning coefficient is; Furthermore, multiple alarm levels correspond to multiple alarm thresholds. The alarm thresholds include upper thresholds and lower thresholds. The correspondence between multiple alarm levels and multiple alarm thresholds is as follows: For the first-level alarm level, AWC ≤ AWC_M; (low-level early warning): Slight deviation or anomaly, which may not require immediate action but should be monitored and recorded. Notify the patrol personnel and maintenance team for preliminary inspection and data verification. For the second-level alarm level, AWC_M < AWC ≤ AWC_H; (medium-level early warning): Moderate deviation or anomaly, which requires further investigation and possible on-site inspection. Dispatch on-site technical support for detailed inspection and prepare for necessary maintenance. For the third-level alarm level, AWC > AWC_H; (high-level early warning): Severe deviation or anomaly, which may pose an immediate threat to safety or operation and requires an emergency response. Immediate measures should be taken, such as stopping the operation of relevant equipment for emergency repair, and at the same time notify all relevant stakeholders. Among them, AWC is the abnormal early warning coefficient, EC_yg is the lower threshold corresponding to the second-level alarm level minus the upper threshold corresponding to the first-level alarm level, and AWC_H is the lower threshold corresponding to the third-level alarm level minus the upper threshold corresponding to the second-level alarm level. In the report generation module, the generation process of the patrol report includes: Restore and compare the walking trajectory of the patrol personnel to evaluate whether the patrol is completed according to the preset route, mark the specific location and time of the deviation from the route, evaluate the health status and potential risks of the equipment based on the equipment status data in combination with the preset thresholds, and determine the alarm level of the detected abnormal situation according to the abnormal detection and scheduling results. Generate a patrol report based on the determined alarm level and corresponding alarm information, including but not limited to: Patrol overview: Basic information of the patrol, such as patrol time, patrol personnel, patrol area, etc.; Walking trajectory analysis: Walking trajectory map of the patrol personnel, comparison result with the preset patrol route, and detailed description of the deviation situation; Equipment status report: Summary information of the equipment status, including lists of normal and abnormal equipment, and trend analysis of key parameters; Abnormal situation summary: List of abnormal events, location, time, type, possible causes, and recommended corrective measures; Patrol efficiency analysis: Completion of the patrol task, coverage rate of the patrol route, and efficiency evaluation of the patrol personnel. Standardize the content of the patrol report, distribute it to relevant personnel, and archive the report for easy query and analysis.

[0023] Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field and related fields without creative work should fall within the scope of protection of the present invention. The structures, devices and operating methods not specifically described and explained in the present invention are implemented according to the conventional means in the field unless otherwise specified and limited.

Claims

1. The distribution network inspection auxiliary system based on artificial intelligence includes a distribution network inspection platform, which is characterized by: The distribution network inspection platform is communicatively connected with a data acquisition module, an RFID identification module, a GIS map module, a data analysis module, an anomaly detection module and a report generation module, wherein electrical signals are connected between the modules; The data acquisition module is used to collect key data during the inspection process, automatically collect the location data of the inspection personnel and the identification information of key sites using GPS and RFID technology, and obtain the equipment status data; The RFID identification module uses RFID technology to quickly identify key sites encountered during the inspection process and record relevant information; The GIS map module is used to store and process geographic data, including preset inspection routes and actual walking tracks of inspection personnel, and to visualize map information of inspection areas, actual walking tracks of inspection personnel, and preset inspection routes; The data analysis module is used to conduct in-depth analysis of the collected data, including comparison of actual walking trajectory with preset inspection routes, analysis of equipment status data, and analysis of potential problems and anomalies; The detection alarm module automatically triggers an alarm mechanism according to the degree of abnormality when it finds that the inspection personnel deviate from the standard route or the equipment status is abnormal, and promptly notifies relevant personnel to take corrective measures; The report generation module is used to generate an inspection report according to the abnormality detection and scheduling results, including the walking track of the inspection personnel, the equipment status, and the abnormal situation information.

2. The distribution network inspection auxiliary system based on artificial intelligence according to claim 1 is characterized in that: In the data collection module, the process of collecting key data during the inspection process includes: The inspectors carry an inspection terminal with an integrated GPS module. The inspection terminal receives GPS satellite signals in real time, calculates and determines the inspectors' current location, and uploads the location data to the data center of the distribution network inspection platform in real time through the wireless network. RFID tags are deployed at key sites on the inspection route. Each RFID tag contains a unique electronic code to identify the site information, including site number, equipment type, and inspection requirements. Key sites include equipment to be inspected and preset inspection points. The inspection terminal carried by the inspection personnel has a built-in RFID reader / writer to read the information in the RFID tag and upload it to the data center of the distribution network inspection platform in real time through the wireless network; Various sensors are installed on the equipment to be inspected to monitor the operating status of the equipment in real time. The sensors convert the collected equipment operating status data into digital signals and transmit them wirelessly to the data center of the distribution network inspection platform.

3. The distribution network inspection auxiliary system based on artificial intelligence according to claim 2 is characterized in that: In the RFID identification module, the process of identifying key sites in the inspection process includes: When the inspection personnel bring the inspection terminal close to the RFID tag, the RFID reader / writer built into the inspection terminal sends a radio frequency signal of a preset frequency. After receiving the radio frequency signal, the RFID tag obtains energy based on the coil induction current on the RFID tag, activating the control circuit and radio frequency circuit in the tag. Once activated, the RFID tag converts the information stored in the chip into a response signal and sends it back to the RFID reader through the antenna; After the RFID reader receives the response signal sent by the RFID tag, it decodes and processes it to restore the original information. The restored original information is uploaded to the data center of the distribution network inspection platform in real time via the wireless network.

4. The distribution network inspection auxiliary system based on artificial intelligence according to claim 3 is characterized in that: In the GIS map module, the process of visually displaying the map information of the inspection area, the actual walking track of the inspection personnel, and the preset inspection route includes: Collect geographic data of the inspection area, including map information, terrain, buildings, roads, basic information of equipment location, and plan preset inspection routes based on the geographic data of the inspection area; Use GPS technology to track the location of patrol personnel in real time, record their walking tracks, store the actual walking tracks of patrol personnel and the preset patrol route data in the GIS system, and pre-process the stored data; Visually display the map information of the inspection area, the actual walking tracks of the inspectors, and the preset inspection routes on the GIS map, analyze the differences between the actual walking tracks of the inspectors and the preset inspection routes, and identify deviations and omissions; Dynamically mark key sites, equipment locations, and fault point information during the inspection process on the map, and associate the contextual information of the inspection process; Utilize the spatial analysis function of GIS to optimize the walking paths of patrol personnel, provide suggestions for the shortest path, divide the patrol areas into different areas, and analyze the patrol difficulty and equipment density characteristics of different areas.

5. The distribution network inspection auxiliary system based on artificial intelligence according to claim 4 is characterized in that: In the data analysis module, the process of analyzing potential problems and anomalies includes: Extract all relevant raw data from the inspection process from the data center of the distribution network inspection platform, including the actual walking tracks of the inspectors, the data read by the RFID tags, and the equipment status data collected by the sensors, and integrate the data from different sources; Pre-process all relevant raw data extracted during the inspection process and unify the coordinate system; The location and time information of each track point is parsed from the actual walking track data of the inspectors, and the preset inspection route in the system is loaded. The actual walking track is compared with the preset inspection route by using the Hausdorff distance calculation method to obtain the track evaluation index, evaluate the difference between the two, and identify the situation where the task is not performed according to the established route based on the comparison results; The equipment status data is classified and aggregated according to the dimensions of equipment type and time period for more in-depth analysis. The standard threshold of each equipment parameter is set. By comparing the actual data with the threshold, it is determined whether the equipment is in a normal state. The time series analysis method is used to analyze the change trend of the equipment status parameters over time, obtain the equipment anomaly assessment index, predict the failure trend, identify the parameter values ​​with abnormal trends, and mark them as potential equipment failure problems.

6. The distribution network inspection auxiliary system based on artificial intelligence according to claim 5 is characterized in that: The expression of the trajectory evaluation index is: ； Among them, TEI is the trajectory evaluation index, which is used to quantify the difference between the actual walking trajectory and the preset inspection route, N is the number of points in the actual walking trajectory, p_i is the i-th point in the actual walking trajectory, q_h is the h-th point in the preset inspection route, d(p_i, q_h) is the Euclidean distance between points p_i and q_h, and the smaller the TEI value, the closer the actual walking trajectory is to the preset inspection route; The expression of the equipment abnormality evaluation index is: ; Among them, DEI is the equipment abnormality evaluation index, which is used to evaluate the changing trend of equipment status parameters over time and predict potential failures. T is the total number of time points, x_t is the ideal state value of the equipment at the tth moment, y_t is the actual measurement value of the equipment at the tth moment, and θ is the adjustment index, which is used to adjust the sensitivity of the difference between the ideal value and the actual value. The larger the value of DEI, the more the equipment state deviates from the ideal state and the higher the potential failure risk.

7. The artificial intelligence-based distribution network inspection auxiliary system according to claim 6 is characterized in that: In the detection alarm module, the process of automatically triggering the alarm mechanism according to the degree of abnormality includes: Receive GPS trajectory data uploaded by patrol personnel in real time, use the preset patrol route and Hausdorff distance calculation method to compare the actual walking trajectory of patrol personnel with the preset patrol route, obtain the trajectory evaluation index, and set the trajectory deviation threshold. When the trajectory evaluation index exceeds the trajectory deviation threshold, it is determined that the patrol personnel deviates from the standard route; Receive the equipment status data uploaded by various sensors in real time, compare and analyze the received data according to the preset equipment parameter standard threshold, apply the time series analysis method, predict the change trend of equipment status parameters, calculate the equipment abnormality evaluation index, and set the abnormality evaluation threshold. When the equipment abnormality evaluation index exceeds the threshold, it is judged that the equipment status is abnormal; According to the severity of the inspection personnel's trajectory deviation and the potential risk of equipment abnormality, different weights are assigned to the trajectory evaluation index and the equipment abnormality evaluation index. The trajectory evaluation index and the equipment abnormality evaluation index are weighted and calculated according to the assigned weights to obtain the abnormality warning coefficient. Different alarm levels are set according to the abnormal warning coefficient, namely, the first alarm level, the second alarm level and the third alarm level. The alarm level increases gradually from the first to the third abnormality level. In combination with the urgency of the inspection task and the severity of the consequences of the failure, the alarm threshold of the abnormal warning coefficient is set to match the alarm level. Combine the calculation results of abnormality detection and abnormality warning coefficient to determine the current alarm level, analyze the abnormality degree of the current inspection process, and generate corresponding alarm information, where the alarm information includes the abnormality type, abnormality location, alarm level and recommended corrective measures; The alarm information will be sent to relevant personnel in a timely manner through multiple channels such as SMS, email, and instant messaging software.

8. The artificial intelligence-based distribution network inspection auxiliary system according to claim 7 is characterized in that: The expression of the abnormal warning coefficient is: ； Among them, AWC is the abnormal warning coefficient, α and β are weight factors used to adjust the relative importance of trajectory evaluation index and equipment abnormal evaluation index in the total warning coefficient, TEI_k is the trajectory evaluation index at the kth time point, (TEI) ̅ is the average of the trajectory evaluation index at all time points, n is the number of data points of the trajectory evaluation index, DEI_j is the state abnormal evaluation index of the jth device, m is the number of data points of the equipment abnormal evaluation index, and the higher the AWC, the more serious the abnormal situation.

9. The artificial intelligence-based distribution network inspection auxiliary system according to claim 8, characterized in that: The multiple alarm levels correspond to the multiple alarm thresholds, the alarm thresholds include an upper threshold and a lower threshold, and the corresponding relationship between the multiple alarm levels and the multiple alarm thresholds is: First level alarm level AWC≤AWC_M; Secondary alarm level AWC_M <AWC≤AWC_H; The third level alarm level AWC>AWC_H; Among them, AWC is the abnormal warning coefficient, EC_yg is the lower threshold corresponding to the second-level alarm level and the upper threshold corresponding to the first-level alarm level, and AWC_H is the lower threshold corresponding to the third-level alarm level and the upper threshold corresponding to the second-level alarm level.

10. The distribution network inspection auxiliary system based on artificial intelligence according to claim 9 is characterized in that: In the report generation module, the generation process of the inspection report includes: Restoring and comparing the walking paths of the inspectors to evaluate whether the inspection is completed according to the preset route, marking the specific location and time of deviation from the route, evaluating the health status and potential risks of the equipment based on the equipment status data and the preset thresholds, and determining the alarm level of the detected abnormal situation based on the abnormal detection and scheduling results; Generate inspection reports based on the determined alarm levels and corresponding alarm information; Standardize the contents of the inspection reports, distribute them to relevant personnel, and archive the reports.

Images