Direct current and extra-high voltage transformer substation air-space-ground collaborative inspection decision-making method and system

Abstract

The invention relates to a space-air-ground collaborative inspection decision-making method and system for a direct-current and extra-high-voltage transformer substation, and relates to the technical field of transformer substation inspection. Performing inspection on the transformer substation in response to the daily inspection task to generate inspection initial data and data source equipment; inputting the inspection initial data and the corresponding data source equipment into an edge calculation module for analysis to determine an abnormal confidence coefficient; judging whether the abnormal confidence is not less than a confidence threshold; if not, continuing to respond to the daily inspection task to inspect the transformer substation so as to generate inspection initial data; if yes, acquiring an abnormal detection device corresponding to the abnormal confidence coefficient, and determining a recheck detection device according to the abnormal detection device; controlling the re-check detection device to assist the anomaly detection device to perform cooperative re-check so as to generate re-check detection data; and analyzing the re-check detection data to determine an equipment early warning signal. The method and the device have the effect of improving the timeliness of inspection response.

Description

Technical Field

[0001] This application relates to the technical field of substation inspection, and in particular to a decision-making method and system for air-ground-space collaborative inspection of DC and UHV substations. Background Technology

[0002] DC and UHV substations are the core hubs of modern smart grids, undertaking the critical task of long-distance, large-capacity, and high-efficiency power transmission. With the construction of new power systems, the requirements for safe, reliable, intelligent, and efficient operation and maintenance of these substations are increasing.

[0003] Among related technologies, "air-ground-space" collaborative inspection is an important technical support for the intelligent operation and maintenance of DC and UHV substations. It uses equipment such as drones, quadruped robot dogs, and fixed high-definition eagle eyes to inspect substations, and uploads the inspection data to a data integration platform for multi-source data fusion. The equipment health model identifies the characteristics of multi-source data to determine whether the equipment has defects, and determines and recommends treatment plans based on knowledge graphs and equipment defects.

[0004] Regarding the technologies mentioned above, current "air-ground-space" collaborative inspection intelligence achieves data collaboration, but lacks behavioral collaboration between inspection devices. For example, when a fixed high-definition eagle eye detects an anomaly, it can only send the abnormal data to the data integration platform for data analysis. Drones and robot dogs cannot perform timely verification, resulting in a delayed inspection response. There is still room for improvement. Summary of the Invention

[0005] To improve the timeliness of inspection response, this application provides a decision-making method and system for air-ground-space collaborative inspection of DC and UHV substations.

[0006] Firstly, this application provides a decision-making method for air-ground-space coordinated inspection of DC and UHV substations, adopting the following technical solution: Decision-making methods for integrated air-ground-space inspection of DC and UHV substations include: Collect daily inspection tasks; Pre-set drones, pre-set robot dogs, and pre-set fixed equipment respond to daily inspection tasks to inspect pre-set substations, generating initial inspection data and corresponding data source equipment; The initial inspection data and the corresponding data source devices are input into a preset edge computing module for analysis to determine the anomaly confidence level. Determine whether the anomaly confidence level is not less than the preset confidence level threshold; If not, the drones, robot dogs, and fixed equipment continue to respond to the daily inspection tasks to inspect the substation and generate initial inspection data. If so, collect the anomaly detection equipment corresponding to the anomaly confidence level, and determine the verification detection equipment based on the anomaly detection equipment; The control verification and testing equipment assists the anomaly detection equipment in performing collaborative verification to generate verification and testing data; The verification and testing data are analyzed to determine the equipment's early warning signals.

[0007] Optionally, the steps of inputting the initial inspection data and the corresponding data source device into a preset edge computing module for analysis to determine the anomaly confidence level include: Based on the initial inspection data and the corresponding data source equipment, the corresponding equipment anomaly feature weights are found in the preset equipment anomaly feature weight relationship; Based on the initial inspection data, find the corresponding maximum and minimum data values ​​in the preset data threshold relationship; Determine whether the initial inspection data meets the requirement of the minimum data value; If it does not meet the requirements, the initial inspection data will be discarded. If the conditions are met, the initial inspection data, the weights of equipment anomaly characteristics, the maximum and minimum data values ​​are analyzed to determine the anomaly confidence level.

[0008] Optionally, the steps to analyze the initial inspection data, the weights of equipment anomaly features, the maximum and minimum data values ​​to determine the anomaly confidence level include: The initial inspection data is normalized based on the maximum and minimum data values ​​to generate the degree of deviation of abnormal features; The degree of deviation of abnormal features is weighted and summed according to the weight of the abnormal features of the equipment to generate an initial confidence level; Based on the data source equipment, the corresponding equipment reliability coefficient is found in the preset equipment reliability coefficient relationship; The initial confidence level is corrected based on the equipment reliability coefficient to generate anomaly confidence levels.

[0009] Optionally, the steps of controlling the verification and testing equipment to assist the anomaly detection equipment in performing collaborative verification to generate verification and testing data include: Collect the location of abnormal equipment and verify the location of the verification equipment for the testing equipment; Based on the initial inspection data and the location of abnormal equipment, the corresponding data type and collaborative review location are found in the preset data review correspondence; Control the movement of the verification and testing equipment from the verification equipment location to the collaborative verification location; Based on the type of verification data, the verification and detection equipment is controlled to detect the collaborative verification location in order to generate collaborative verification data; Link initial inspection data and collaborative review data to generate review and inspection data.

[0010] Optionally, the step of controlling the verification and testing equipment to move from the verification equipment location to the collaborative verification location includes: The location of the verification equipment and the location of the collaborative verification are analyzed based on the preset path planning algorithm to determine the basic movement path; Collect a dynamic raster map of the basic movement path; Extract the location of dynamic obstacles from the preset substation grid map based on the dynamic grid map; The locations of dynamic obstacles are filtered based on the basic movement path and the preset obstacle influence radius to generate the locations of influencing obstacles; Analyze the locations of obstacles and the basic movement path to determine the optimal movement path; The optimized movement path control moves the verification and testing equipment from the verification equipment location to the collaborative verification location.

[0011] Optionally, the steps of analyzing the locations of influencing obstacles and the basic movement path to determine the optimal movement path include: The location of the affected path is extracted from the basic movement path based on the location of the affected obstacle. The locations of the influencing obstacles and the influencing paths are calculated based on a pre-defined influence direction model to generate an influence direction vector. The location of the obstacle and the location of the path of influence are calculated based on the preset influence distance model to generate the obstacle influence distance; Data acquisition equipment support coefficient; The influence direction vector and obstacle influence distance are corrected based on the equipment support coefficient to generate a position optimization vector; The location of the influence path in the basic movement path is optimized based on the location optimization vector to generate an optimized movement path.

[0012] Optionally, the steps of analyzing the verification test data to determine equipment warning signals include: The verification test data were analyzed to determine the verification confidence level; The corresponding equipment warning level is found in the preset confidence level relationship based on the review confidence level; Based on the review confidence level and review detection data, the corresponding treatment strategy is found in the preset treatment knowledge graph; Associate equipment warning levels and handling strategies to generate equipment warning signals.

[0013] Optionally, the steps of analyzing the verification test data to determine the verification confidence level include: Based on the verification and testing data, determine the verification data for drones, robot dogs, and fixed equipment; The verification data of drones, robot dogs, and fixed equipment were analyzed separately to determine the confidence levels of drones, robot dogs, and fixed equipment. The reliability coefficients of drones, robot dogs, and fixed equipment are collected. The confidence scores of the drone, robot dog, and fixed equipment are weighted and voted on based on the reliability coefficients of the drone, robot dog, and fixed equipment to generate a verification confidence score.

[0014] Secondly, this application provides a space-air-ground collaborative inspection and decision-making system for DC and UHV substations, which adopts the following technical solution: A decision-making system for air-ground-space collaborative inspection of DC and UHV substations includes: The data acquisition module is used to collect data on daily inspection tasks and anomaly detection equipment. A memory for storing the program of the air-ground-space collaborative inspection decision-making method for DC and UHV substations as described in any of the above items; The processor and the program in the memory can be loaded and executed by the processor to implement the air-ground-space collaborative inspection decision method for DC and UHV substations as described in any of the above.

[0015] In summary, this application includes at least one of the following beneficial technical effects: 1. By using drones, robotic dogs, and fixed equipment to respond to daily inspection tasks, substations are inspected to obtain initial inspection data and corresponding data source equipment. The edge computing module analyzes the data to determine the anomaly confidence level. When the anomaly confidence level is not less than the confidence level threshold, the system controls the compliance detection equipment to assist the anomaly detection equipment in collaborative verification, ensuring the coordination of inspection behavior. When an anomaly occurs, the remaining inspection equipment can promptly conduct collaborative inspections, thereby improving the timeliness of inspection response. 2. By finding the corresponding verification data type and collaborative verification location in the data verification correspondence based on the initial inspection data and the location of abnormal equipment, the verification and detection equipment is moved to the collaborative verification location. The equipment is then tested according to the verification data type to obtain collaborative verification data. This allows the verification and detection equipment to detect data related to the initial inspection data within its own detection range, thereby assisting in verifying the authenticity of the initial inspection data and ensuring the accuracy of the verification. 3. By collecting a dynamic grid map of the basic movement path through fixed equipment, dynamic obstacle locations are extracted from the substation grid map based on the dynamic grid map, and then the locations of influencing obstacles are filtered out. The basic movement path is then optimized based on the locations of influencing obstacles to obtain an optimized movement path, enabling mutual support between inspection equipment and ensuring the efficiency of inspection equipment movement. Attached Figure Description

[0016] Figure 1 This is a flowchart of the air-ground-space collaborative inspection decision-making method for DC and UHV substations in the embodiments of this application.

[0017] Figure 2 This is a flowchart of the steps in this application embodiment to input the initial inspection data and the corresponding data source device into a preset edge computing module for analysis in order to determine the anomaly confidence level.

[0018] Figure 3 This is a flowchart of the steps in this application embodiment to analyze the initial inspection data, the weight of equipment abnormality features, the maximum value of the data, and the minimum value of the data to determine the confidence level of the abnormality.

[0019] Figure 4 This is a flowchart of the steps in this application embodiment to control the verification and detection equipment to assist the anomaly detection equipment in performing collaborative verification in order to generate verification and detection data.

[0020] Figure 5 This is a flowchart of the steps in this application embodiment to control the verification and testing equipment to move from the verification equipment position to the collaborative verification position.

[0021] Figure 6 This is a flowchart illustrating the steps in this application embodiment to analyze the location of influencing obstacles and the basic movement path to determine the optimized movement path.

[0022] Figure 7 This is a flowchart of the steps in this application embodiment to analyze the verification test data to determine the equipment warning signal.

[0023] Figure 8 This is a flowchart of the steps in this application embodiment to analyze the verification test data to determine the verification confidence level. Detailed Implementation

[0024] To make the purpose, technical solution, and advantages of this application clearer, the following description is provided in conjunction with the appendix. Figures 1 to 8 The present application will be further described in detail below with reference to embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the application.

[0025] Reference Figure 1 This application discloses a method for joint air-ground-space inspection decision-making for DC and UHV substations, including the following steps: Step S100: Collect daily inspection tasks.

[0026] Routine inspection tasks refer to the daily inspection tasks within DC and UHV substations, including inspection routes and objects for drones, robotic cameras, and fixed equipment. For example, drones mainly collect normal and infrared images of insulator strings, lightning rod tops, and high-altitude transmission lines. Robotic cameras collect normal images, vibration, temperature, humidity, and other parameters of cable trenches, equipment bottoms, and instruments. Fixed equipment uses high-definition eagle-eye cameras and infrared thermal imagers, installed within the substation for comprehensive data collection. Specific routine inspection tasks are planned by operators based on the type and location of equipment within the substation, and adjusted according to the specific faults and usage conditions of the equipment.

[0027] Step S101: Preset drones, preset robot dogs, and preset fixed equipment respond to daily inspection tasks to inspect preset substations to generate initial inspection data and corresponding data source equipment.

[0028] After determining the daily inspection tasks, the processing terminal controls drones, robot dogs, and fixed equipment to inspect the substation according to the corresponding paths and inspection objects in the daily inspection tasks, and generates initial inspection data for each, thereby providing data analysis objects for subsequent determination of whether the inspected equipment has anomalies.

[0029] Initial inspection data refers to the data detected by drones, robotic dogs, and fixed equipment during substation inspections, such as temperature data.

[0030] The data source device refers to the device from which the initial inspection data originates. For example, if infrared temperature data is collected by a fixed device, then the source device for the temperature data is the fixed device. By determining the data source device, the confidence level of the initial inspection data can be determined through data such as the reliability of that device.

[0031] Step S102: Input the initial inspection data and the corresponding data source device into the preset edge computing module for analysis to determine the anomaly confidence level.

[0032] After determining the initial inspection data and the corresponding data source devices, the corresponding edge computing module analyzes the initial inspection data and the data source devices to obtain the anomaly confidence level of the data. The specific method is described in [reference needed]. Figure 2 The steps involved provide a score indicating whether the device is malfunctioning.

[0033] An edge computing module refers to a module installed in drones, robotic dogs, and fixed equipment to analyze whether inspection data is abnormal. The analysis logic of the edge computing module in this embodiment refers to... Figure 2 The steps involve calculating the confidence level of the initial inspection data to determine whether the data anomalies are correct.

[0034] Anomaly confidence level refers to the confidence level that the initial inspection data is anomalous. The confidence level ranges from 0 to 1. The higher the anomaly confidence level, the higher the probability that the initial inspection data is anomalous. It is determined by the edge computing module after analyzing the initial inspection data and the corresponding data source device. For specific methods, please refer to [link / reference]. Figure 2 The steps.

[0035] Step S103: Determine whether the anomaly confidence level is not less than the preset confidence level threshold.

[0036] The confidence threshold refers to the boundary value for distinguishing whether the initial inspection data is abnormal. In this embodiment, 0.7 is used as an example. If the confidence is greater than 0.7, the data is determined to be abnormal. If the confidence is less than 0.7, the data is determined to be normal.

[0037] By processing the terminal to determine whether the anomaly confidence level is not less than the confidence level threshold, it is possible to determine whether the initial inspection data is abnormal, and further determine whether other inspection equipment is needed for support and verification, thereby ensuring the coordination of behavior between inspection equipment and improving the timeliness of inspection response.

[0038] Step S1031: If not, the drone, robot dog, and fixed equipment continue to respond to the daily inspection task to inspect the substation and generate initial inspection data.

[0039] If the processing terminal determines that the anomaly confidence level is less than the confidence level threshold, it indicates that the initial inspection data is normal data and the corresponding equipment being inspected is in a normal state. Therefore, the drone, robot dog, and fixed equipment continue to be controlled to inspect the substation according to the task path and the inspection object corresponding to the daily inspection task, and the initial inspection data continues to be collected for analysis, so as to continuously monitor the changes in the status of equipment in the substation.

[0040] Step S1032: If yes, collect the anomaly detection device corresponding to the anomaly confidence level, and determine the verification detection device based on the anomaly detection device.

[0041] If the processing terminal determines that the confidence level of the anomaly is not less than the confidence level threshold, it indicates that the initial inspection data is abnormal data and the corresponding device being inspected is in an abnormal state. Therefore, the anomaly detection device corresponding to the anomaly confidence level is called, and the verification detection device is determined based on the anomaly detection device to provide data support for subsequent verification.

[0042] Anomaly detection equipment refers to equipment that detects anomalies within a substation. When the processing terminal determines that the anomaly confidence level of the initial inspection data is not less than the confidence level threshold, the data source equipment corresponding to the initial inspection data is defined as anomaly detection equipment.

[0043] Verification and testing equipment refers to inspection equipment used to assist in verifying whether equipment in a substation is abnormal. The processing terminal determines the type of abnormality detection equipment based on the abnormality detection equipment. For example, if the abnormality detection equipment is a fixed device, then the type of verification and testing equipment is a robot dog and a drone. After determining the type of verification and testing equipment, the location of the abnormality detection equipment is confirmed, and then the drone and robot dog closest to the abnormality detection equipment are selected as the verification and testing equipment.

[0044] Step S104: Control the verification and testing equipment to assist the anomaly detection equipment in performing collaborative verification to generate verification and testing data.

[0045] After determining the verification and testing equipment, the processing terminal controls the verification and testing equipment to assist the anomaly detection equipment in collaborative verification, thereby obtaining verification and testing data. The specific method is described in [reference needed]. Figure 4 The steps involve achieving behavioral coordination among inspection equipment, using the overall detection of the inspection equipment to determine whether there are any abnormalities in the equipment within the substation, thereby improving the timeliness and accuracy of inspection response.

[0046] Verification and testing data refers to the verification data of equipment in the substation by three types of inspection equipment. For example, the abnormality detection equipment is a fixed device, and the data detected is the temperature value and the rate of temperature change. The verification and testing equipment is a drone and a robot dog. The data detected by the drone is infrared temperature, hot spot area and temperature gradient, while the data detected by the robot dog is contact temperature and vibration frequency.

[0047] Step S105: Analyze the verification test data to determine the equipment warning signal.

[0048] After determining the verification test data, the processing terminal performs unified data analysis on the verification test data to determine the equipment warning signal. The specific method is as follows: Figure 7 These steps ensure the accuracy of the review.

[0049] Equipment early warning signals are signals that indicate the level of equipment abnormality and the handling method. They are obtained by the processing terminal after analyzing the verification and testing data. Through equipment early warning signals, maintenance personnel can understand the status of equipment in the substation in a timely manner, thereby improving the maintenance efficiency of equipment in the substation.

[0050] Reference Figure 2 The steps for analyzing initial inspection data and corresponding data source devices into a preset edge computing module to determine anomaly confidence levels include: Step S200: Based on the initial inspection data and the corresponding data source equipment, find the corresponding equipment anomaly feature weight in the preset equipment anomaly feature weight relationship.

[0051] Among them, the equipment abnormality feature weight relationship refers to the correspondence between different feature data detected by different types of inspection equipment and the feature data weight. For example, the feature data weight of temperature value detected by fixed equipment is 0.6, while the feature data weight of temperature change rate is 0.4. The operator determines the weight according to the importance of the feature data in the status and forms a mapping table that corresponds one-to-one with the type of inspection equipment, feature data and feature data weight.

[0052] Equipment anomaly feature weights refer to the weights of different feature data in the initial inspection data. The processing terminal finds the weights in the mapping table corresponding to the equipment anomaly feature weights based on the initial inspection data and the corresponding data source equipment. By determining the equipment anomaly feature weights, data support is provided for the subsequent fusion of different feature data in the initial inspection data to form data describing the overall status of the inspected equipment.

[0053] Step S201: Based on the initial inspection data, find the corresponding maximum and minimum data values ​​in the preset data threshold relationship.

[0054] Among them, the data threshold relationship refers to the correspondence between different characteristic data and data thresholds. For example, the minimum value of equipment temperature is 25 degrees Celsius and the maximum value is 80 degrees Celsius. The operator forms a mapping table by matching the initial inspection data with the data thresholds one by one.

[0055] The maximum value of data refers to the abnormal maximum value of different characteristic data in the initial inspection data, and the minimum value of data refers to the normal minimum value of different characteristic data in the initial inspection data. The data is obtained by the processing terminal by looking up the mapping table corresponding to the data threshold relationship based on the initial inspection data.

[0056] Step S202: Determine whether the initial inspection data meets the requirement of minimum data value.

[0057] The requirement for the minimum data value refers to the error range that exceeds the minimum data value. The specific error range depends on the type of feature data. For example, for temperature, the error range is within 5 degrees Celsius.

[0058] By processing the terminal, it is determined whether the initial inspection data exceeds the error range of the minimum data value, thereby determining whether the data is abnormal and providing data support for whether the confidence level of the initial inspection data needs to be calculated subsequently.

[0059] Step S2021: If it does not meet the requirements, the initial inspection data will be removed.

[0060] If the processing terminal determines that the initial inspection data is within the error range of the minimum data value, it indicates that the initial inspection data is normal. At this time, the equipment in the substation is normal, the anomaly confidence level is 0, and therefore the initial inspection data is removed.

[0061] Step S2022: If the conditions are met, analyze the initial inspection data, the weight of equipment anomaly characteristics, the maximum value of the data, and the minimum value of the data to determine the anomaly confidence level.

[0062] If the processing terminal determines that the initial inspection data exceeds the error range of the minimum data value, it indicates that the equipment in the substation may be in an abnormal state. Therefore, the initial inspection data, the weight of the equipment abnormality characteristics, the maximum data value, and the minimum data value are analyzed to determine the anomaly confidence level. The specific method is described in [reference needed]. Figure 3 The steps.

[0063] Reference Figure 3 The steps for analyzing initial inspection data, equipment anomaly characteristic weights, maximum and minimum data values ​​to determine the anomaly confidence level include: Step S300: Normalize the initial inspection data based on the maximum and minimum data values ​​to generate the degree of deviation of abnormal features.

[0064] Among them, the degree of abnormal feature deviation refers to the degree to which the feature data in the initial inspection data deviates from the normal data. The value of the degree of abnormal feature deviation ranges from 0 to 1. The larger the value, the greater the deviation. The processing terminal uses the difference between the maximum and minimum data values ​​as the denominator and the difference between the corresponding feature data and the minimum data value in the initial inspection data as the numerator to normalize the initial inspection data and obtain the degree of abnormal feature deviation, thereby quantifying the degree of abnormality of the feature data in the initial inspection data.

[0065] Step S301: The degree of deviation of abnormal features is weighted and summed according to the weight of the abnormal features of the equipment to generate an initial confidence level.

[0066] The initial confidence level refers to the confidence level of equipment anomalies within the substation determined after integrating all feature data from the initial inspection data. It is obtained by the processing terminal through a weighted summation of the anomaly feature offsets of the corresponding feature data based on the equipment anomaly feature weights. By determining the initial confidence level, the offsets of all feature data are aggregated to describe the equipment status, ensuring the accuracy of the confidence level.

[0067] Step S302: Find the corresponding equipment reliability coefficient in the preset equipment reliability coefficient relationship based on the data source equipment.

[0068] Among them, the equipment reliability coefficient relationship refers to the correspondence between different inspection equipment and equipment reliability coefficient. The operator records the number of times the equipment is inspected and the number of times it is correct within a fixed period of time, and then calculates the quotient of the two to obtain the baseline reliability coefficient. Then, the current reliability coefficient is obtained by weighted summation based on the baseline reliability coefficient and the preset minimum reliability coefficient, thus forming a mapping table that maps the equipment to the current reliability coefficient.

[0069] The equipment reliability coefficient refers to the credibility of the test data from the data source equipment. The equipment reliability coefficient ranges from 0 to 1. The larger the value, the higher the credibility. It is obtained by the processing terminal by looking up the equipment reliability coefficient relationship based on the data source equipment.

[0070] Step S303: Correct the initial confidence level based on the equipment reliability coefficient to generate anomaly confidence level.

[0071] In this step, the abnormal confidence level is the same as that in step S102. It is obtained by multiplying the reliability coefficient of the processing terminal computing device by the initial confidence level, thereby correcting the initial confidence level according to the credibility of the data source device and ensuring the accuracy of the confidence level.

[0072] Reference Figure 4 The steps for controlling the verification and testing equipment to assist the anomaly detection equipment in performing collaborative verification to generate verification and testing data include: Step S400: Collect the location of the abnormal device and the location of the verification device for the verification test.

[0073] The abnormal equipment location refers to the location of the equipment exhibiting abnormality within the substation, while the verification equipment location refers to the location of the movable equipment within the verification and testing equipment. The abnormal equipment location is obtained through fusion localization using LiDAR and UWB on the abnormal detection equipment, while the verification equipment location is obtained through fusion localization using LiDAR and UWB on the movable equipment within the verification and testing equipment. Determining the abnormal and verification equipment locations provides data support for subsequently determining the specific verification location of the verification and testing equipment.

[0074] Step S401: Based on the initial inspection data and the location of abnormal equipment, find the corresponding data type and collaborative review location in the preset data review correspondence.

[0075] Among them, the data verification correspondence refers to the correspondence between different types of data at different locations and the verification data at the verification location. For example, if a fixed device detects an abnormal temperature data at a certain device, the corresponding verification data are infrared temperature value, hot spot area, temperature gradient, contact temperature and vibration frequency. The location is the aerial inspection location and ground inspection location set for the device location. The operator will then form a mapping table by matching the device location and abnormal data with the verification location and verification data one by one.

[0076] The data type to be verified refers to the data type that the verification and testing equipment needs to verify. The coordinate verification location refers to the location of the verification and testing equipment when verifying the data. It is obtained by the processing terminal by looking up the mapping table corresponding to the data verification correspondence based on the initial inspection data and the location of the abnormal equipment.

[0077] Step S402: Control the verification and testing equipment to move from the verification equipment position to the collaborative verification position.

[0078] After determining the collaborative verification location, a movement path is planned based on the location of the verification equipment and the collaborative verification location. This allows the verification and testing equipment to move from the verification equipment location to the collaborative verification location according to the movement path. The specific method is described in [reference needed]. Figure 5 These steps provide basic support for subsequent verification and testing equipment to collect verification data.

[0079] Step S403: Control the verification detection equipment to detect the collaborative verification position according to the verification data type to generate collaborative verification data.

[0080] In this process, after the verification and testing equipment moves to the collaborative verification position, the corresponding sensor on the verification and testing equipment is controlled to detect the equipment at the collaborative verification position according to the verification data type, thereby obtaining collaborative verification data.

[0081] Collaborative verification data refers to data that is consistent with the data type detected by the verification testing equipment at the collaborative verification location. By determining the collaborative verification data, the overall data collection of substation equipment can be supplemented from another perspective, thereby more comprehensively reflecting the status of equipment in the substation.

[0082] Step S404: Link the initial inspection data and collaborative review data to generate review inspection data.

[0083] In this step, the verification and testing data is consistent with the verification and testing data in step S104. The processing terminal stores the initial inspection data and the collaborative verification data in the same data packet. By associating the data of the anomaly detection device and the verification and testing device, the substation equipment is detected as a whole, thereby comprehensively reflecting the true condition of the substation equipment.

[0084] Reference Figure 5 The steps for controlling the verification and testing equipment to move from the verification equipment location to the collaborative verification location include: Step S500: Analyze the location of the verification equipment and the location of the collaborative verification according to the preset path planning algorithm to determine the basic movement path.

[0085] Among them, the path planning algorithm refers to the algorithm used to plan the movement path between two locations within a substation. In the embodiments of this application, A is used as an example. * Take the algorithm as an example.

[0086] The basic movement path refers to the movement path from the verification equipment location to the collaborative verification location. The processing terminal maps the verification equipment location and the collaborative verification location onto a static substation map, and then incorporates factors such as distance, time, and accessibility as constraints into the path planning. In this way, the path planning algorithm is used to find the optimal path and ensure the basic availability of the basic movement path.

[0087] Step S501: Collect a dynamic raster map of the basic movement path.

[0088] Among them, the dynamic grid map refers to the dynamic map of the basic movement path. The processing terminal controls the fixed equipment near the basic movement path to collect images of the basic movement path, and uses a convolutional neural network to identify obstacles in the images. After the obstacles are identified, their positions are mapped onto the static map of the basic movement path, thereby generating a dynamic grid map. By determining the dynamic grid map, the changes in dynamic obstacles on the basic movement path can be understood in real time, so as to adjust the path in advance and ensure the smooth movement of the verification and inspection equipment.

[0089] Step S502: Extract the location of dynamic obstacles from the preset substation grid map based on the dynamic grid map.

[0090] Among them, the substation grid map refers to the static grid map of the substation, which is stored in the processing terminal by the operator. The grid map marks fixed obstacles in the substation, such as transformers and other equipment.

[0091] Dynamic obstacle location refers to the position of dynamic obstacles within a substation. The processing terminal compares the markers of the same grid cells in the dynamic grid map and the substation grid map; if they differ, the coordinates of that grid cell are defined as the dynamic obstacle location. Determining the dynamic obstacle location provides data support for subsequent dynamic obstacle avoidance. Step S503: Filter the dynamic obstacle locations based on the basic movement path and the preset obstacle influence radius to generate the influencing obstacle locations.

[0092] The obstacle influence radius refers to the radius that affects the movement of the verification and testing equipment. It is determined by the operator based on the size of the verification and testing equipment, and is usually taken as 1 to 1.5 meters.

[0093] The location of an influencing obstacle refers to a dynamic obstacle that affects the movement of the verification and testing equipment. The processing terminal calculates the shortest distance between the dynamic obstacle location and the basic movement path. If the shortest distance is less than the obstacle's influence radius, it indicates that the obstacle will affect the normal movement of the verification and testing equipment; therefore, this dynamic obstacle location is identified as an influencing obstacle location. By identifying influencing obstacle locations, locations that do not affect the movement of the verification and testing equipment are removed, significantly reducing the amount of data analyzed.

[0094] Step S504: Analyze the location of the influencing obstacle and the basic movement path to determine the optimal movement path.

[0095] The optimized movement path refers to the path taken after avoiding dynamic obstacles within the basic movement path. It is obtained by the processing terminal optimizing the basic movement path based on the location of the influencing obstacles. Specific methods are described in [reference needed]. Figure 6 The process involves determining an optimized movement path, enabling the verification and testing equipment to move smoothly to the target location without being affected by obstacles.

[0096] Step S505: Based on the optimized movement path, control the verification and inspection equipment to move from the verification equipment location to the collaborative verification location.

[0097] After determining the optimized movement path, the processing terminal controls the movable equipment in the verification and testing equipment to move from the verification equipment position to the collaborative verification position according to the optimized movement path. During the movement, the verification and testing equipment detects new obstacles in the path in real time and makes real-time position adjustments.

[0098] Reference Figure 6 The steps to analyze the location of obstacles and the basic movement path to determine the optimal movement path include: Step S600: Extract the location of the influencing path from the basic movement path based on the location of the influencing obstacle.

[0099] Among them, the affected path location refers to the location in the basic movement path affected by the location of the obstacle. The processing terminal draws a circle with the affected obstacle location as the center and the radius of the influence to identify the coordinates in the basic movement path that are located within the circle, which is the affected path location. By determining the affected path location, the coordinates in the basic movement path that need to be adjusted are identified, providing data support for subsequent path optimization.

[0100] Step S601: Calculate the location of the influencing obstacle and the location of the influencing path according to the preset influence direction model to generate the influence direction vector.

[0101] The influence direction model refers to the model that calculates the direction of influence of the location of the obstacle on the location of the path. Specifically, it calculates the influence direction vector by using the vector from the location of the obstacle to the location of the path as the numerator and the distance between the location of the obstacle and the location of the path as the denominator.

[0102] The influence direction vector refers to the direction in which the location of the obstacle affects the location of the path. It is calculated by the processing terminal by inputting the location of the obstacle and the location of the path into the influence direction model. By determining the influence direction vector, the adjustment direction of the basic movement path is determined.

[0103] Step S602: Calculate the location of the affected obstacle and the location of the affected path according to the preset influence distance model to generate the obstacle influence distance.

[0104] The influence distance model is a model for calculating the influence strength of the location of an obstacle on the location of the path. Specifically, it uses the negative of the square of the distance between the location of the obstacle and the location of the path as the influence strength, thus reflecting the rule that the closer the distance, the stronger the influence.

[0105] The obstacle influence distance refers to the distance by which the location of an obstacle affects the location of the path. It is calculated by the processing terminal by inputting the location of the obstacle and the location of the path into the influence distance model. By determining the obstacle influence distance, the adjustment distance of the basic movement path can be determined.

[0106] Step S603: Collect equipment support coefficients.

[0107] Among them, the equipment support coefficient refers to the reliability of fixed equipment in collecting obstacle information for different devices. It is determined by the operator based on the accuracy of the fixed equipment in collecting obstacles in the past. The equipment support coefficient ranges from 0 to 1. The larger the value, the higher the reliability.

[0108] Step S604: Correct the influence direction vector and obstacle influence distance according to the equipment support coefficient to generate a position optimization vector.

[0109] The location optimization vector refers to the optimized direction and distance that affect the path location. It is determined by multiplying the support coefficient of the processing terminal calculation device by the distance affected by the obstacle, thereby determining the specific adjustment distance. Then, the dot product of the adjustment distance and the direction vector is calculated to obtain the location optimization vector.

[0110] Step S605: Optimize the position of the influence path in the basic movement path according to the position optimization vector to generate an optimized movement path.

[0111] In this step, the optimized movement path is the same as that in step S504. The processing terminal adjusts the position of the influencing path in the basic movement path along the direction and distance corresponding to the position optimization vector. Then, cubic spline interpolation, Bézier curve fitting, or gradient descent smoothing are used to smooth the adjusted path to obtain the optimized movement path, ensuring that the verification and inspection equipment can move smoothly to the verification position.

[0112] Reference Figure 7 The steps for analyzing the verification and testing data to determine equipment warning signals include: Step S700: Analyze the verification test data to determine the verification confidence level.

[0113] Among them, the verification confidence level refers to the credibility of the verification data detected by the verification testing equipment. The verification confidence level ranges from 0 to 1. The higher the verification confidence level, the higher the credibility of the substation equipment malfunctioning. It is determined by the processing terminal after analyzing the verification testing data. The specific method is as follows: Figure 8 The steps.

[0114] Step S701: Based on the verification confidence level, find the corresponding equipment warning level in the preset confidence level relationship.

[0115] The confidence level relationship refers to the correspondence between different confidence levels and warning levels. For example, when the confidence level is 0 to 0.5, it can be determined that the confidence level is low and it is a false alarm of the inspection equipment, so the warning level is a false alarm level; when the confidence level is 0.5 to 0.7, the confidence level is moderate, and the warning level is a general warning; when the confidence level is 0.7 to 0.9, the confidence level is high, and the warning level is an important warning; when the confidence level is 0.9 to 1, the confidence level is the highest, and the warning level is an emergency warning. Operators will form a mapping table by matching the confidence level with the warning level one by one.

[0116] Equipment warning level refers to the warning level of equipment in a substation, including false alarm, general, important and emergency warnings. The warning level is obtained by the processing terminal by looking up the corresponding mapping table of confidence level relationship according to the verification confidence level. By determining the equipment warning level, the operation and maintenance personnel can be promptly reminded according to the credibility of the warning, which helps the operation and maintenance personnel understand the status of the equipment.

[0117] Step S702: Based on the verification confidence level and verification detection data, find the corresponding treatment strategy in the preset treatment knowledge graph.

[0118] Among them, the handling knowledge graph refers to the handling strategies corresponding to different confidence levels and abnormal data. For example, if the verified test data shows an abnormal temperature but the confidence level is low, the corresponding handling strategy is to record the data. When the confidence level is moderate or high, the corresponding handling strategy is to include the equipment in the key inspection scope and increase the frequency of inspection and verification. When the confidence level is the highest, the corresponding handling strategy is to immediately issue an alarm to remind maintenance personnel to carry out on-site maintenance.

[0119] The handling strategy refers to the method for handling substation equipment anomalies detected by inspection equipment. It is obtained by the processing terminal through matching in the handling knowledge graph based on the verification confidence level and the verification detection data.

[0120] Step S703: Associate the equipment warning level and handling strategy to generate an equipment warning signal.

[0121] In this step, the equipment warning signal is the same as that in step S105. The processing terminal stores the equipment warning level and the handling strategy in the same data packet. Through the equipment warning level, the operation and maintenance personnel can understand the abnormal situation of the equipment in the substation in a timely manner and determine the best handling method to ensure the good operating status of the equipment in the substation.

[0122] Reference Figure 8 The steps for analyzing the verification test data to determine the verification confidence level include: Step S800: Determine the verification data for the UAV, the robot dog, and the fixed equipment based on the verification and testing data.

[0123] Among them, UAV verification data refers to the data obtained after UAVs verify and inspect substation equipment, robot dog verification data refers to the data obtained after robot dogs verify and inspect substation equipment, and fixed equipment verification data refers to the data obtained after fixed equipment verifies and inspects substation equipment. These data are retrieved by the processing terminal from the verification and inspection data. By identifying UAV verification data, robot dog verification data, and fixed equipment verification data, data analysis objects are provided for subsequent analysis of the confidence levels of the three.

[0124] Step S801: Analyze the verification data of the UAV, the robot dog, and the fixed equipment respectively to determine the confidence level of the UAV, the robot dog, and the fixed equipment.

[0125] Among them, the confidence level of the drone refers to the credibility of the drone verification data, the confidence level of the robot dog refers to the credibility of the robot dog verification data, and the confidence level of the fixed equipment refers to the credibility of the fixed equipment verification data. These are obtained by the processing terminal after analyzing the drone verification data, robot dog verification data, and fixed equipment verification data respectively. The specific analysis process is as follows... Figure 3 The logic of the intermediate steps is consistent and will not be elaborated here.

[0126] Step S802: Collect the reliability coefficient of the drone, the reliability coefficient of the robot dog, and the reliability coefficient of the fixed equipment.

[0127] Among them, the reliability coefficient of the UAV refers to the reliability of the UAV when verifying data, the reliability coefficient of the robot dog refers to the reliability of the robot dog when verifying data, and the reliability coefficient of the fixed equipment refers to the reliability coefficient of the fixed equipment when verifying data. The operator determines the reliability coefficient based on the accuracy of the three when verifying data, and stores the reliability coefficient in the processing terminal for later retrieval.

[0128] Step S803: Based on the reliability coefficients of the UAV, the robot dog, and the fixed equipment, perform a weighted vote on the confidence scores of the UAV, the robot dog, and the fixed equipment to generate a verification confidence score.

[0129] In this step, the verification confidence level is consistent with that in step S700. The processing terminal calculates the sum of the reliability coefficients of the UAV, the robot dog, and the fixed equipment. Then, it calculates the quotients of the UAV reliability coefficient, the robot dog reliability coefficient, and the fixed equipment reliability coefficient with the sum of their respective coefficients to obtain the weights of the UAV, the robot dog, and the fixed equipment. The calculated weights are then used to perform a weighted summation of the UAV confidence level, the robot dog confidence level, and the fixed equipment confidence level, and the verification confidence level is obtained by fusing the confidence levels of the three.

[0130] Based on the same inventive concept, embodiments of this application provide a space-air-ground collaborative inspection and decision-making system for DC and UHV substations, including: The data acquisition module is used to collect data on daily inspection tasks, anomaly detection equipment, verification equipment locations, dynamic grid maps, equipment support coefficients, drone reliability coefficients, robot dog reliability coefficients, and fixed equipment reliability coefficients. The memory is used to store the program for the air-ground-space collaborative inspection decision-making method of DC and UHV substations; The processor and memory can load and execute programs to realize a collaborative air-ground-space inspection decision-making method for DC and UHV substations.

[0131] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0132] This application provides a computer-readable storage medium storing a computer program that can be loaded by a processor and executed as a decision-making method for air-ground-space coordinated inspection of DC and UHV substations.

[0133] Computer storage media include, for example, USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media that can store program code.

[0134] Based on the same inventive concept, embodiments of this application provide a smart terminal, including a memory and a processor, wherein the memory stores a computer program that can be loaded and executed by the processor to implement a decision-making method for air-ground-space collaborative inspection of DC and UHV substations.

[0135] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the above-described division of functional modules is used as an example. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device, and unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0136] The above are all preferred embodiments of this application and are not intended to limit the scope of protection of this application. Any feature disclosed in this specification (including the abstract and drawings) may be replaced by other equivalent or similar features unless specifically stated otherwise. That is, unless specifically stated otherwise, each feature is only one example of a series of equivalent or similar features.

Claims

1. A decision-making method for air-ground-space coordinated inspection of DC and UHV substations, characterized in that, include: Collect daily inspection tasks; Pre-set drones, pre-set robot dogs, and pre-set fixed equipment respond to daily inspection tasks to inspect pre-set substations, generating initial inspection data and corresponding data source equipment; The initial inspection data and the corresponding data source devices are input into a preset edge computing module for analysis to determine the anomaly confidence level. Determine whether the anomaly confidence level is not less than the preset confidence level threshold; If not, the drones, robot dogs, and fixed equipment continue to respond to the daily inspection tasks to inspect the substation and generate initial inspection data. If so, collect the anomaly detection equipment corresponding to the anomaly confidence level, and determine the verification detection equipment based on the anomaly detection equipment; The control verification and testing equipment assists the anomaly detection equipment in performing collaborative verification to generate verification and testing data; The verification and testing data are analyzed to determine the equipment's early warning signals.

2. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 1, characterized in that, The steps involved in inputting the initial inspection data and the corresponding data source devices into a preset edge computing module for analysis to determine the anomaly confidence level include: Based on the initial inspection data and the corresponding data source equipment, the corresponding equipment anomaly feature weights are found in the preset equipment anomaly feature weight relationship; Based on the initial inspection data, find the corresponding maximum and minimum data values ​​in the preset data threshold relationship; Determine whether the initial inspection data meets the requirement of the minimum data value; If it does not meet the requirements, the initial inspection data will be discarded. If the conditions are met, the initial inspection data, the weights of equipment anomaly characteristics, the maximum and minimum data values ​​are analyzed to determine the anomaly confidence level.

3. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 2, characterized in that, The steps for analyzing initial inspection data, equipment anomaly characteristic weights, maximum and minimum data values ​​to determine the anomaly confidence level include: The initial inspection data is normalized based on the maximum and minimum data values ​​to generate the degree of deviation of abnormal features; The degree of deviation of abnormal features is weighted and summed according to the weight of the abnormal features of the equipment to generate an initial confidence level; Based on the data source equipment, the corresponding equipment reliability coefficient is found in the preset equipment reliability coefficient relationship; The initial confidence level is corrected based on the equipment reliability coefficient to generate anomaly confidence levels.

4. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 1, characterized in that, The steps for controlling the verification and testing equipment to assist the anomaly detection equipment in performing collaborative verification to generate verification and testing data include: Collect the location of abnormal equipment and verify the location of the verification equipment for the testing equipment; Based on the initial inspection data and the location of abnormal equipment, the corresponding data type and collaborative review location are found in the preset data review correspondence; Control the movement of the verification and testing equipment from the verification equipment location to the collaborative verification location; Based on the type of verification data, the verification and detection equipment is controlled to detect the collaborative verification location in order to generate collaborative verification data; Link initial inspection data and collaborative review data to generate review and inspection data.

5. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 4, characterized in that, The steps for controlling the movement of the verification and testing equipment from the verification equipment location to the collaborative verification location include: The location of the verification equipment and the location of the collaborative verification are analyzed based on the preset path planning algorithm to determine the basic movement path; Collect a dynamic raster map of the basic movement path; Extract the location of dynamic obstacles from the preset substation grid map based on the dynamic grid map; The locations of dynamic obstacles are filtered based on the basic movement path and the preset obstacle influence radius to generate the locations of influencing obstacles; Analyze the locations of obstacles and the basic movement path to determine the optimal movement path; The optimized movement path control moves the verification and testing equipment from the verification equipment location to the collaborative verification location.

6. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 5, characterized in that, The steps to analyze the location of obstacles and the basic movement path to determine the optimal movement path include: The location of the affected path is extracted from the basic movement path based on the location of the affected obstacle. The locations of the influencing obstacles and the influencing paths are calculated based on a pre-defined influence direction model to generate an influence direction vector. The location of the obstacle and the location of the path of influence are calculated based on the preset influence distance model to generate the obstacle influence distance; Data acquisition equipment support coefficient; The influence direction vector and obstacle influence distance are corrected based on the equipment support coefficient to generate a position optimization vector; The location of the influence path in the basic movement path is optimized based on the location optimization vector to generate an optimized movement path.

7. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 1, characterized in that, The steps for analyzing the verification and testing data to determine equipment warning signals include: The verification test data were analyzed to determine the verification confidence level; The corresponding equipment warning level is found in the preset confidence level relationship based on the review confidence level; Based on the review confidence level and review detection data, the corresponding treatment strategy is found in the preset treatment knowledge graph; Associate equipment warning levels and handling strategies to generate equipment warning signals.

8. The air-ground-space coordinated inspection decision-making method for DC and UHV substations according to claim 7, characterized in that, The steps for analyzing the verification test data to determine the verification confidence level include: Based on the verification and testing data, determine the verification data for drones, robot dogs, and fixed equipment; The verification data of drones, robot dogs, and fixed equipment were analyzed separately to determine the confidence levels of drones, robot dogs, and fixed equipment. The reliability coefficients of drones, robot dogs, and fixed equipment are collected. The confidence scores of the drone, robot dog, and fixed equipment are weighted and voted on based on the reliability coefficients of the drone, robot dog, and fixed equipment to generate a verification confidence score.

9. A decision-making system for air-ground-space collaborative inspection of DC and UHV substations, characterized in that, include: The data acquisition module is used to collect data on daily inspection tasks and anomaly detection equipment. A memory for storing the program of the air-ground-space collaborative inspection decision method for DC and UHV substations as described in any one of claims 1 to 8; The processor and the program in the memory can be loaded and executed by the processor to implement the air-ground-space collaborative inspection decision method for DC and UHV substations as described in any one of claims 1 to 8.

Images