Unmanned aerial vehicle energy optimization method, system and equipment for power grid inspection

Abstract

The invention belongs to the technical field of unmanned aerial vehicle routing inspection, and particularly relates to an unmanned aerial vehicle energy optimization method, system and equipment for power grid routing inspection, and the method comprises the steps: firstly obtaining power grid routing inspection data and power grid environment data of a power grid in a historical time period; determining power grid abnormal data, an abnormal solution and scheme feedback information of each cell according to the power grid inspection data; based on the power grid abnormal data and the power grid environment data, obtaining a plurality of cell comprehensive areas and inspection key positions; updating the exception solution based on the scheme feedback information to obtain a standard exception scheme; and performing power grid abnormity inspection on the inspection key position, and if a power grid abnormity exists in a power grid inspection result, determining a corresponding real-time solution according to the standard abnormity scheme. The method can directly focus a key area, greatly improves the inspection efficiency, can provide a standard abnormality scheme to reduce the loss caused by the abnormality of the power grid, and improves the operation reliability and stability of the power grid.

Description

Technical Field

[0001] This invention belongs to the field of unmanned aerial vehicle (UAV) inspection technology, specifically relating to an energy optimization method, system, and equipment for UAVs used in power grid inspection. Background Technology

[0002] Power grid mesh is a spatialized, hierarchical, and collaborative power grid management unit under the new power system. Its core idea is to divide the complex power grid system into several interconnected grid units according to dimensions such as geographical distribution, equipment type, and voltage level. Each unit integrates transmission, substation, distribution equipment and environmental parameters to form an integrated inspection and maintenance system of "transmission-transformation-distribution-loop".

[0003] Currently, power grid inspections are still primarily conducted manually on a regular basis. The core process involves maintenance personnel conducting on-site inspections of transmission, transformation, and distribution equipment within the grid at fixed intervals (e.g., monthly), recording equipment status, environmental parameters, and potential defects. While manual inspections offer some reliability in simple scenarios, their limitations become increasingly apparent as the power grid expands in scale and complexity. To overcome the bottlenecks of manual inspections, the industry is accelerating its transformation towards intelligent, drone-based grid inspections. These drones are equipped with multi-modal sensors such as infrared thermal imagers, lidar, visible light cameras, and partial discharge detectors, enabling them to simultaneously collect over 10 types of parameters, including equipment temperature, deformation, discharge, and contamination, covering more than 90% of common defect types and significantly improving the efficiency of power grid inspections.

[0004] While drone inspections have improved the efficiency of power grid inspections, each drone has a fixed flight energy limit. If the flight energy exceeds the design threshold, there is a risk of insufficient battery life, mission interruption, or even equipment damage. This problem is particularly prominent in complex power grid scenarios (such as mountainous areas, cross-river lines, and high-altitude regions), becoming a core bottleneck restricting the large-scale application of drones. Therefore, improving the energy utilization rate of drones has become an urgent issue to be addressed. Summary of the Invention

[0005] The purpose of this invention is to address the aforementioned problems in the existing technology by providing a method, system, and device for UAV energy optimization for power grid grid inspection. This device can not only directly focus on key areas, avoiding wasting time and electricity in irrelevant areas and greatly improving inspection efficiency, but also provide standard anomaly solutions to reduce losses caused by power grid anomalies and improve the reliability and stability of power grid operation.

[0006] To achieve the above objectives, the technical solution of the present invention is as follows:

[0007] In a first aspect, the present invention provides an energy optimization method for unmanned aerial vehicles (UAVs) used for power grid grid inspection, the UAV energy optimization method comprising:

[0008] Acquire power grid inspection data and power grid environment data within a historical time period;

[0009] Based on the power grid inspection data, determine the abnormal power grid data of each cell in the power grid at different time points, as well as the abnormal solutions and solution feedback information when resolving the abnormal power grid data;

[0010] By comprehensively analyzing the power grid anomaly data and the power grid environment data, multiple integrated cell regions and the corresponding key inspection locations for each integrated cell region are obtained.

[0011] The anomaly solution is updated based on the feedback information from the aforementioned solution to obtain a standard anomaly solution;

[0012] Based on the key inspection locations, a power grid inspection command is generated, controlling the inspection drone to fly to the key inspection locations to conduct power grid anomaly inspections, and receiving the power grid inspection results transmitted by the inspection drone.

[0013] Determine whether there is a power grid anomaly in the power grid inspection results. If so, determine a real-time solution corresponding to the power grid inspection results based on the standard anomaly scheme.

[0014] The comprehensive analysis of the power grid anomaly data and the power grid environment data yields multiple integrated cell regions and corresponding key inspection locations for each integrated cell region, including:

[0015] Based on the power grid environment data, determine the target environment data corresponding to the power grid anomaly data;

[0016] By analyzing the variation patterns of the power grid anomaly data and the target environment data, the anomaly step pattern of the power grid mesh under different environments is obtained;

[0017] Collect environmental change data of the power grid grid within a preset time period, and determine the estimated abnormal data of the power grid grid within the preset time period based on the environmental change data and the abnormal step pattern;

[0018] The predicted abnormal data is analyzed by regionalization and key location selection to obtain multiple cell comprehensive areas and the corresponding key inspection locations for each cell comprehensive area.

[0019] The step of updating the anomaly solution based on the feedback information of the solution to obtain a standard anomaly solution includes:

[0020] Based on the feedback information of the solution, determine the grid operation status of the grid after abnormal maintenance, and determine whether the grid operation status meets the preset operation status. If it does not meet the preset operation status, determine the position of each grid cell that does not meet the preset operation status, and determine the initial inspection sequence and grid inspection parameters of the inspection drone based on the abnormal solution.

[0021] The first inspection position and the second inspection position are determined according to the location of the power grid cell and the initial inspection sequence. The first inspection position is the power grid inspection position where the power grid operation status first fails to meet the preset operation status, and the second inspection position is the power grid inspection position where the power grid operation status last fails to meet the preset operation status.

[0022] The target inspection parameters of the inspection drone between the first inspection position and the second inspection position are determined based on the power grid inspection parameters.

[0023] Based on the power grid operation status, determine the power grid operation changes between the first inspection position and the second inspection position, and perform simulated inspection of the power grid grid based on the power grid operation changes and the target inspection parameters to obtain the exclusive inspection parameters corresponding to each operation status in the power grid operation changes.

[0024] The specific inspection parameters are updated according to the initial inspection sequence to update the target inspection parameters, and the updated target inspection parameters are imported into the anomaly solution to obtain a standard anomaly solution.

[0025] The analysis of the variation patterns of the abnormal power grid data and the target environment data yields the abnormal stepwise patterns of the power grid mesh under different environments, including:

[0026] Construct a power grid anomaly coordinate system, wherein the X-axis of the power grid anomaly coordinate system represents different time nodes, the Y-axis of the power grid anomaly coordinate system represents anomaly data of different anomaly conditions of the power grid grid, and the Z-axis of the power grid anomaly coordinate system represents different environmental rule parameters;

[0027] The abnormal power grid data and the target environment data are imported into the abnormal power grid coordinate system according to the time development nodes to obtain the abnormal change curve corresponding to each cell in the power grid grid;

[0028] The abnormal change curve is calculated to obtain the abnormal curve variability of each cell under different power grid environments.

[0029] The variability of the abnormal curve is matrix-normalized based on the arrangement of each cell in the power grid to obtain the abnormal step pattern of the power grid under different environments.

[0030] The calculation of the curve volatility value of the abnormal change curve to obtain the abnormal curve volatility of each cell under different power grid environments includes:

[0031] Determine whether each abnormal change curve has fluctuations. If it does, define the starting point after the curve fluctuation as the first node, the peak point after the curve fluctuation as the second node, and the ending point after the curve fluctuation as the third node.

[0032] Determine the first difference data corresponding to the first node and the second node, and the second difference data corresponding to the second node and the third node, and calculate the ratio of the curve change period between the first node and the second node to the first difference data to obtain the first curve undulation;

[0033] The second curve variability is obtained by calculating the ratio of the curve change time period between the second node and the third node to the second difference data.

[0034] The first curve variability and the second curve variability are summarized according to the time ratio to obtain the abnormal curve variability of each cell under different power grid environments.

[0035] The analysis of the predicted anomaly data through regionalization and key location selection yields multiple integrated cell regions and corresponding key inspection locations for each integrated cell region, including:

[0036] The estimated abnormal data is compared with the preset abnormal data range to determine whether the estimated abnormal data meets the preset abnormal data range. If it does, the power grid cell is the abnormal cell, and the abnormal location of the abnormal cell is recorded. Based on the concentrated distribution characteristics of the abnormal location, the power grid is regionalized to obtain a comprehensive area of ​​multiple cells.

[0037] For each cell in the comprehensive area, the estimated abnormal data corresponding to different power grid cells are compared pairwise to determine the extreme cell with the largest extreme value in the estimated abnormal data. The position corresponding to the extreme cell is marked to obtain the key inspection position.

[0038] The controlled inspection drone flies to the key inspection location to conduct power grid anomaly inspections, including:

[0039] Obtain the current real-time location of each inspection drone, and simulate the flight arrival of the current real-time location with the inspection key location to obtain the target drone that arrives at the inspection key location in the shortest time.

[0040] Determine whether the target drone has an inspection mission. If it does, calculate the inspection mission duration and sum the inspection mission duration with the flight time of the target drone to the inspection key location to obtain the total inspection duration.

[0041] The inspection time of the second inspection drone flying to the key inspection location is determined, and the inspection time is compared with the total inspection time. If the inspection time is less than the total inspection time, the second inspection drone is designated as the inspection drone performing the power grid inspection task. The second inspection drone is an idle drone and the inspection task time is second only to the target drone.

[0042] Secondly, the present invention provides an unmanned aerial vehicle (UAV) energy optimization system for power grid inspection, the UAV energy optimization system comprising:

[0043] The data acquisition module is used to acquire power grid inspection data and power grid environment data within a historical time period.

[0044] The solution determination module is used to determine, based on the power grid inspection data, the abnormal power grid data of each cell in the power grid at different time nodes, as well as the abnormal solutions and solution feedback information when resolving the abnormal power grid data;

[0045] The power grid analysis module is used to comprehensively analyze the power grid anomaly data and the power grid environment data to obtain multiple cell comprehensive areas and the corresponding key inspection locations for each cell comprehensive area;

[0046] The solution update module is used to update the abnormal solution based on the solution feedback information to obtain a standard abnormal solution;

[0047] The power grid inspection module is used to generate power grid inspection commands based on the key inspection locations, control the inspection drone to fly to the key inspection locations to conduct power grid anomaly inspections, and receive the power grid inspection results transmitted by the inspection drone.

[0048] The anomaly resolution module is used to determine whether there is a power grid anomaly in the power grid inspection results. If so, it determines a real-time solution corresponding to the power grid inspection results based on the standard anomaly solution.

[0049] Thirdly, the present invention provides an unmanned aerial vehicle (UAV) energy optimization device for power grid grid inspection, the UAV energy optimization device including a memory and a processor; the memory is used to store computer program code and transmit the computer program code to the processor; the processor is used to execute the aforementioned UAV energy optimization method according to the instructions in the computer program code.

[0050] Fourthly, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the aforementioned UAV energy optimization method.

[0051] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0052] The UAV energy optimization method for power grid inspection described in this invention first acquires power grid inspection data and power grid environmental data within a historical time period, providing comprehensive and fundamental information support for subsequent analysis. Then, based on the power grid inspection data, it determines the power grid anomaly data for each cell at different time points, accurately locating historical problem occurrence points. Simultaneously, it records anomaly solutions and feedback information, providing a basis for solution optimization. Through comprehensive analysis of power grid anomaly data and environmental data, and by correlating historical anomalies with the environment, it identifies problem-prone areas, thereby obtaining comprehensive areas for multiple cells and corresponding key inspection locations for each comprehensive area. This makes inspections more targeted, avoids blind inspections, increases the probability of discovering potential anomalies, and improves inspection efficiency and accuracy. After acquiring anomaly solutions and feedback information, the feedback information reflects the actual effectiveness and shortcomings of historical anomaly solutions. Based on this feedback, the anomaly solutions are updated, removing unreasonable or ineffective parts while retaining effective measures. Furthermore, by combining feedback from different cases, more universal and efficient solution strategies can be integrated and optimized. Through continuous updates and improvements, a standard anomaly solution is finally obtained. When facing similar power grid anomalies, standard anomaly solutions provide more scientific and standardized guidance, reducing the time and cost of resolving anomalies and improving the success rate and quality of problem-solving. Subsequently, based on the analyzed key inspection locations, power grid inspection commands are generated, controlling inspection drones to fly to these key locations for anomaly inspection. Since the key inspection locations are areas prone to anomalies identified through comprehensive analysis of historical data, the drones can directly focus on these key areas, avoiding wasting time and energy in irrelevant areas, greatly improving inspection efficiency. Simultaneously, receiving the power grid inspection results transmitted by the inspection drones allows for timely acquisition of on-site information. This targeted inspection method not only quickly detects power grid anomalies but also ensures the comprehensiveness and accuracy of the inspection, providing a reliable basis for timely anomaly handling and effectively guaranteeing the safe and stable operation of the power grid. After receiving the power grid inspection results from the inspection drone and determining that there is a power grid anomaly, the system can quickly find a real-time solution corresponding to the current power grid inspection results from among many solutions based on the standard anomaly solution. This eliminates the need to develop a solution on-site, saving a lot of time. It enables timely handling of power grid anomalies, prevents the anomaly from worsening, reduces losses caused by power grid anomalies, ensures the power grid quickly returns to normal operation, and improves the reliability and stability of power grid operation. Attached Figure Description

[0053] Figure 1 This is a flowchart of the UAV energy optimization method described in this invention.

[0054] Figure 2 This is a schematic diagram of the structure of the UAV energy optimization system described in this invention.

[0055] Figure 3 This is a schematic diagram of the structure of the UAV energy optimization device described in this invention. Detailed Implementation

[0056] The present invention will now be described in further detail with reference to specific embodiments and accompanying drawings.

[0057] Example 1:

[0058] See Figure 1 An energy optimization method for unmanned aerial vehicles (UAVs) used for power grid inspection is performed in the following steps:

[0059] Step S10: Obtain power grid inspection data and power grid environment data within the historical time period.

[0060] Specifically, power grid inspection data refers to various data and information recorded during the inspection of the power grid. This data indicates the operating status of power grid equipment and whether any abnormalities exist. Examples include parameters such as voltage, current, and temperature, as well as information on whether equipment has malfunctioned and the type of malfunction. Power grid environmental data refers to various data related to the environment in which the power grid operates, indicating the condition of the power grid's operating environment. This includes meteorological data (such as temperature, humidity, wind speed, and rainfall).

[0061] Various sensors, such as voltage sensors, current sensors, and temperature sensors, are installed on power grid equipment to collect real-time operating parameters as power grid inspection data. Simultaneously, meteorological monitoring stations are set up at suitable locations around the power grid to acquire meteorological data. The data collected by the sensors and monitoring equipment is preliminarily processed and stored through a data acquisition terminal. Relevant power grid inspection data and power grid environmental data are extracted from the data acquisition terminal according to a set historical time period.

[0062] Step S11: Based on the power grid inspection data, determine the abnormal power grid data of each cell in the power grid at different time points, as well as the abnormal solutions and solution feedback information when resolving abnormal power grid data.

[0063] Specifically, a power grid grid is a series of interconnected sub-regions into which the entire power grid area is divided according to factors such as geographical scope, equipment distribution, or management needs. It serves as the basic unit for refined management and analysis of the power grid. A cell is a smaller regional unit further subdivided within the power grid grid. Each cell contains a certain number of power grid devices and is set up for more precise location and analysis of power grid problems. A time node is a specific moment or time period on the timeline, used to represent the specific time when data records or events occur, such as hourly, daily, or weekly intervals. Power grid anomaly data refers to data in power grid inspection data that exceeds the normal range or does not meet the normal operating standards of equipment. It represents data related to abnormal situations such as power grid equipment failures and performance degradation. Anomaly solutions refer to specific solutions and methods developed for the anomalies represented by power grid anomaly data. These are action plans that eliminate power grid anomalies and restore normal equipment operation through technical means and operational procedures. Solution feedback information refers to information collected after implementing solutions for power grid anomalies, regarding the effectiveness of the solutions, whether the problems were solved, and whether new problems arose. It represents the evaluation and feedback of the anomaly solutions.

[0064] Step S12: Perform comprehensive analysis on power grid anomaly data and power grid environment data to obtain multiple cell comprehensive areas and the corresponding key inspection locations for each cell comprehensive area.

[0065] Specifically, based on power grid environmental data, target environmental data corresponding to power grid anomaly data is determined. The variation patterns of both the power grid anomaly data and the target environmental data are analyzed to obtain the anomaly gradient pattern of the power grid grid under different environments. Environmental change data of the power grid grid is collected within a preset time period, and based on the environmental change data and the anomaly gradient pattern, the estimated anomaly data of the power grid grid within the preset time period is determined. The estimated anomaly data is then analyzed by regionalization and key location selection to obtain multiple integrated cell areas and the corresponding key inspection locations for each integrated cell area.

[0066] Target environmental data refers to power grid environmental data that is correlated with specific power grid anomaly data. It refers to the data corresponding to specific environmental factors that affect power grid anomalies after screening and analysis. For example, when equipment malfunctions due to overheating, the corresponding target environmental data is the high temperature weather data at that time. The anomaly ladder pattern refers to the ladder-like variation of anomaly data in the power grid under different environmental conditions. It indicates that as environmental factors change to varying degrees, the frequency and severity of power grid anomalies also exhibit hierarchical and gradual changes. Environmental change data refers to the dynamic changes of various parameters of the power grid's environment within a preset time period. This data, obtained through continuous monitoring and recording, reflects how environmental factors change during that period. Environmental factors include temperature (real-time temperature, daily maximum / minimum temperature, temperature change rate), humidity (relative humidity, dew point temperature), precipitation (rainfall, snowfall, precipitation intensity), and wind speed / direction (instantaneous wind speed, average wind speed, maximum wind speed), etc. Environmental data processing utilizes sensors deployed on power grid equipment / towers (such as temperature and humidity sensors, wind speed sensors), weather stations, drone inspections, and satellite remote sensing (large-scale vegetation / Environmental factor data is acquired (based on terrain), then cleaned and standardized to remove outliers and fill in missing data, ensuring accuracy and consistency. The processed environmental factor data is then used for feature extraction based on trend changes, and sliding window mean values ​​(e.g., the sliding mean of environmental parameters within 1 minute, reflecting short-term environmental factor trends) are collected. These sliding window mean values ​​are used to determine the changes in environmental factors over time. Predicted anomaly data refers to data that predicts potential anomalies in the power grid within a preset time period based on environmental change data and anomaly gradient patterns. This prediction is derived through analysis and inference and is used to anticipate potential anomalies in the power grid so that appropriate measures can be taken. A cell-level integrated area refers to an area formed by grouping multiple cells with similar or related anomaly characteristics after regional analysis of the predicted anomaly data. This area is designed for more effective inspection and management. Key inspection locations refer to locations within each cell-level integrated area that significantly impact the operating status of power grid equipment, are prone to anomalies, or whose anomalies would significantly affect the entire area. These are the areas that require focused attention and inspection during inspections.

[0067] The target environmental data and the anomaly gradient pattern are used as training samples to train a neural network model, resulting in a power grid anomaly model. Then, environmental change data is input into the power grid anomaly model for identification, yielding estimated anomalies that the power grid may experience within a preset time period under the background of environmental change data. The preset time period refers to the time between the most recent fault occurrence and the current time, and is not a fixed time period; it changes in real time as the fault occurs and time progresses.

[0068] Specifically, the abnormal step pattern is obtained according to the following steps: Construct a power grid anomaly coordinate system. The X-axis of the power grid anomaly coordinate system represents different time nodes, the Y-axis represents abnormal data under different anomaly conditions of the power grid grid, and the Z-axis represents different environmental rule parameters. Import the power grid anomaly data and target environment data into the power grid anomaly coordinate system according to time development nodes to obtain the anomaly change curve corresponding to each cell in the power grid grid. Calculate the curve volatility value of the anomaly change curve to obtain the anomaly curve variability of each cell under different power grid environments. Perform matrix normalization on the anomaly curve variability according to the arrangement relationship of each cell in the power grid grid to obtain the anomaly step pattern of the power grid grid under different environments.

[0069] The power grid anomaly coordinate system can be constructed using data analysis software (such as MATLAB, Python's relevant scientific computing libraries, etc.). By defining the variables and ranges of the X, Y, and Z axes in the data analysis software, the X-axis can be set with different time scales according to actual needs, the Y-axis is set according to the type and range of power grid anomaly data, and the Z-axis is determined according to the type and value range of environmental rule parameters. The power grid anomaly data and target environment data are organized into a specific data format, such as a table, containing information such as time, anomaly data, and environmental rule parameters. Using the software's import function, the data is accurately imported into the power grid anomaly coordinate system according to time development nodes. Using the software's plotting function, anomaly change curves are plotted for each cell. Then, program code is written to calculate the curve volatility values, obtaining the anomaly curve volatility of each cell under different power grid environments. Based on the cell layout information in the power grid grid, a matrix is ​​created in the software, and the anomaly curve volatility is filled into the matrix according to the cell layout relationship, completing matrix regularization, thereby obtaining the anomaly step pattern of the power grid grid under different environments.

[0070] When calculating the curve volatility, it is determined whether each abnormal curve exhibits volatility. If so, the starting point after the volatility is defined as the first node, the peak point as the second node, and the ending point as the third node. The first difference data between the first and second nodes, and the second difference data between the second and third nodes are determined. The ratio of the curve change period between the first and second nodes to the first difference data is calculated to obtain the first curve volatility. The ratio of the curve change period between the second and third nodes to the second difference data is also calculated to obtain the second curve volatility. The first and second curve volatility are then summarized according to time proportions to obtain the abnormal curve volatility of each cell under different power grid environments.

[0071] When dividing the grid into comprehensive regions and selecting key inspection locations, the estimated anomaly data is compared with the preset anomaly data range to determine if the estimated anomaly data matches the preset range. If it does, the grid cell is designated as an anomaly cell, and its location is recorded. Based on the concentrated distribution of anomaly locations, the grid is regionalized to obtain multiple comprehensive regions. For each comprehensive region, the estimated anomaly data corresponding to different grid cells is compared pairwise to identify the extreme cell with the largest extreme value. The location of this extreme cell is then marked, thus determining the key inspection locations.

[0072] The comparison between the estimated anomaly data and the preset anomaly data range can be achieved using a Geographic Information System (GIS) platform. The geographic information of the power grid grid is imported into the GIS platform, and the estimated anomaly data is associated with the power grid cells. The spatial analysis function of the GIS platform is used to compare the estimated anomaly data with the preset anomaly data range. When the estimated anomaly data matches the preset anomaly range, the corresponding power grid cell is marked as an anomaly cell on the GIS platform, and its location coordinates are recorded. Using the clustering analysis tool of the GIS platform, the power grid grid is regionalized based on the concentrated distribution characteristics of anomaly locations, generating multiple cell composite regions. Within each cell composite region, by writing scripts or using the built-in calculation function of the GIS platform, the estimated anomaly data corresponding to different power grid cells is compared pairwise to find the extreme cell with the largest extreme value. Its location is specially marked on the GIS platform, such as using icons of different colors, to obtain the key inspection locations. The preset anomaly range is a data interval pre-set based on the normal operation standards of the power grid, historical data, and relevant technical specifications, used to determine whether the estimated anomaly data exceeds the normal range, thereby determining whether an anomaly has occurred in the power grid.

[0073] Step S13: Update the abnormal solution based on the solution feedback information to obtain the standard abnormal solution.

[0074] Specifically, based on the feedback information from the solution, the grid operation status after abnormal maintenance is determined, and it is judged whether the grid operation status conforms to the preset operation status. If not, the location of each grid cell that does not conform to the preset operation status is determined, and the initial inspection sequence and grid inspection parameters of the inspection drone are determined according to the abnormal solution. Based on the grid cell location and the initial inspection sequence, the first and second inspection positions are determined. The first inspection position is the grid inspection position where the first instance of non-conformity with the preset operation status occurs, and the second inspection position is the grid inspection position where the last instance of non-conformity occurs occurs. The target inspection parameters for the inspection drone between the first and second inspection positions are determined based on the grid inspection parameters. The changes in grid operation between the first and second inspection positions are determined based on the grid operation status, and a simulated inspection of the grid grid is performed based on the changes in grid operation and the target inspection parameters. The inspection parameters are adjusted during the simulated inspection until the adjusted inspection parameters can ensure that the grid operation status after abnormal maintenance conforms to the preset operation status. The inspection parameters at this point are used as the specific inspection parameters corresponding to the current changes in grid operation. The dedicated inspection parameters are updated to the target inspection parameters according to the initial inspection sequence, and the updated target inspection parameters are imported into the anomaly solution to obtain the standard anomaly solution.

[0075] Step S14: Generate power grid inspection instructions based on key inspection locations, control the inspection drone to fly to the key inspection locations to conduct power grid anomaly inspections, and receive the power grid inspection results transmitted by the inspection drone.

[0076] Specifically, before controlling the drone inspection, the current real-time position of each inspection drone is obtained, and a flight arrival simulation is performed between the current real-time position and the key inspection position to obtain the target drone that reaches the key inspection position in the shortest time. It is then determined whether the target drone has an inspection task. If so, the inspection task duration is calculated, and this duration is summed with the flight time of the target drone to the key inspection position to obtain the total inspection time. The inspection time of the second inspection drone to the key inspection position is then determined, and compared with the total inspection time. If the inspection time is less than the total inspection time, the second inspection drone is designated as the inspection drone performing the power grid inspection task. The second inspection drone is an idle drone whose inspection task duration is second only to the target drone.

[0077] Step S15: Determine whether there is a power grid anomaly in the power grid inspection results. If so, determine the real-time solution corresponding to the power grid inspection results based on the standard anomaly solution.

[0078] Specifically, acquiring power grid inspection data and environmental data within historical timeframes provides comprehensive and fundamental information support for subsequent analysis. Then, based on the power grid inspection data, anomaly data for each cell at different time points is determined, enabling precise location of historical problem occurrences. Simultaneously, anomaly solutions and feedback information are recorded, providing a basis for solution optimization. Comprehensive analysis of power grid anomaly and environmental data, utilizing the correlation between historical anomalies and the environment, identifies problem-prone areas, resulting in multiple cell-wide integrated areas and corresponding key inspection locations for each area. This makes inspections more targeted, avoids blind inspections, increases the probability of discovering potential anomalies, and improves inspection efficiency and accuracy. After obtaining anomaly solutions and feedback information, the feedback reflects the actual effectiveness and shortcomings of historical anomaly solutions. Updating anomaly solutions based on this feedback removes unreasonable and ineffective parts, retaining effective measures. Furthermore, combining feedback from different cases allows for the integration and optimization of more universal and efficient solution strategies. Through continuous updates and improvements, a standard anomaly solution is finally obtained. When facing similar power grid anomalies, standardized anomaly solutions provide more scientific and standardized guidance, reducing time and costs in resolving anomalies and improving the success rate and quality of problem-solving. Power grid inspection commands are generated based on the analyzed key inspection locations, controlling inspection drones to fly to these locations for anomaly inspection. Since these key inspection locations are areas prone to anomalies identified through comprehensive analysis of historical data, the drones can directly focus on these key areas, avoiding wasting time and energy in irrelevant areas, greatly improving inspection efficiency. Simultaneously, receiving the power grid inspection results transmitted by the inspection drones allows for timely acquisition of on-site information. This targeted inspection method not only quickly detects power grid anomalies but also ensures the comprehensiveness and accuracy of the inspection, providing a reliable basis for timely anomaly handling and effectively guaranteeing the safe and stable operation of the power grid.

[0079] Upon receiving the power grid inspection results from the inspection drone and determining that a power grid anomaly exists, the system can quickly find a real-time solution corresponding to the current inspection results from numerous solutions based on standard anomaly protocols. This eliminates the need for on-site ad-hoc solution development, saving significant time, enabling timely handling of power grid anomalies, preventing further deterioration, reducing losses caused by anomalies, ensuring rapid restoration of normal power grid operation, and improving the reliability and stability of power grid operation.

[0080] Example 2:

[0081] See Figure 2An energy optimization system for unmanned aerial vehicles (UAVs) used for power grid inspection includes a data acquisition module, a scheme determination module, a power grid analysis module, a scheme update module, a power grid inspection module, and an anomaly resolution module.

[0082] The data acquisition module is used to acquire power grid inspection data and power grid environment data of the power grid grid within a historical time period;

[0083] The solution determination module is used to determine, based on the power grid inspection data, abnormal power grid data of each cell in the power grid at different time nodes, as well as abnormal solutions and solution feedback information when resolving the abnormal power grid data;

[0084] The power grid analysis module is used to comprehensively analyze the power grid anomaly data and the power grid environment data to obtain multiple integrated cell regions and the corresponding key inspection locations for each integrated cell region. Specifically, the power grid analysis module is used to determine the target environment data corresponding to the power grid anomaly data according to the following steps: analyzing the change patterns of the power grid anomaly data and the target environment data to obtain the anomaly ladder pattern of the power grid grid under different environments; collecting environmental change data of the power grid grid within a preset time period, and determining the estimated anomaly data of the power grid grid within the preset time period based on the environmental change data and the anomaly ladder pattern; performing regionalization and key location selection analysis on the estimated anomaly data to obtain multiple integrated cell regions and the corresponding key inspection locations for each integrated cell region. Specifically, the power grid analysis module is used to obtain the anomaly ladder pattern of the power grid grid under different environments according to the following steps: constructing a power grid anomaly coordinate system, where the X-axis of the power grid anomaly coordinate system represents different time nodes, the Y-axis of the power grid anomaly coordinate system represents anomaly data of different anomaly conditions of the power grid grid, and the Z-axis of the power grid anomaly coordinate system represents different environmental rule parameters; and guiding the power grid anomaly data and the target environment data according to the time development nodes. The abnormal curves corresponding to each cell in the power grid are obtained by inputting into the power grid anomaly coordinate system. The curve volatility values ​​of the abnormal curves are calculated to obtain the abnormal curve volatility of each cell under different power grid environments. The abnormal curve volatility is matrix-normalized according to the arrangement of each cell in the power grid to obtain the abnormal step pattern of the power grid under different environments. Specifically, the power grid analysis module obtains the abnormal curve volatility of each cell under different power grid environments according to the following steps: It is determined whether each abnormal curve exhibits volatility. If so, the starting point after the curve volatility is defined as the first node, the peak point after the curve volatility as the second node, and the ending point after the curve volatility as the third node. The first difference data corresponding to the first node and the second node, and the second difference data corresponding to the second node and the third node are determined. The ratio of the curve change period between the first node and the second node to the first difference data is calculated to obtain the first curve volatility. The ratio of the curve change period between the second node and the third node to the second difference data is calculated to obtain the second curve volatility. The first curve volatility and the second curve volatility are summarized according to the time ratio to obtain the abnormal curve volatility of each cell under different power grid environments.Specifically, the power grid analysis module is used to obtain multiple integrated cell regions and the corresponding key inspection locations for each integrated cell region according to the following steps: The estimated anomaly data is compared with a preset anomaly data range to determine whether the estimated anomaly data conforms to the preset anomaly data range. If it does, the power grid cell is designated as an anomaly cell, and the anomaly location of the anomaly cell is recorded. Based on the concentrated distribution characteristics of the anomaly locations, the power grid is regionalized to obtain multiple integrated cell regions. The estimated anomaly data corresponding to different power grid cells in each integrated cell region are compared pairwise to determine the extreme cell with the largest extreme value among the estimated anomaly data. The location corresponding to the extreme cell is marked to obtain the key inspection locations.

[0085] The solution update module is used to update the anomaly solution based on the solution feedback information to obtain a standard anomaly solution. Specifically, the solution update module is used to update the anomaly solution according to the following steps to obtain a standard anomaly solution: determine the grid operation status of the grid after anomaly maintenance based on the solution feedback information, and determine whether the grid operation status conforms to the preset operation status. If it does not conform, determine the position of each grid cell that does not conform to the preset operation status; determine the initial inspection sequence and grid inspection parameters of the inspection drone according to the anomaly solution; determine the first inspection position and the second inspection position according to the grid cell position and the initial inspection sequence. The first inspection position is the first time the grid operation status does not conform to the preset operation status. The system identifies the power grid inspection locations based on the power grid conditions, with the second inspection location being the last location where the power grid operation condition does not conform to the preset operation condition. Based on the power grid inspection parameters, target inspection parameters for the inspection drone are determined between the first and second inspection locations. The system also determines the power grid operation changes between the first and second inspection locations based on the power grid operation conditions, and performs simulated inspections of the power grid grid based on these changes and the target inspection parameters to obtain specific inspection parameters for each operation condition. Finally, the specific inspection parameters are updated to the target inspection parameters according to the initial inspection sequence, and the updated target inspection parameters are imported into the anomaly solution to obtain a standard anomaly solution.

[0086] The power grid inspection module is used to generate power grid inspection commands based on the key inspection locations, control the inspection drones to fly to the key inspection locations to perform power grid anomaly inspections, and receive the power grid inspection results transmitted by the inspection drones. Specifically, the power grid inspection module is used to control the inspection drones to fly to the key inspection locations to perform power grid anomaly inspections according to the following steps: obtaining the current real-time location of each inspection drone, and simulating the flight arrival of the current real-time location with the key inspection location to obtain the target drone that reaches the key inspection location in the shortest time; determining whether the target drone is undergoing inspection. If a task exists, the inspection task duration is calculated, and the inspection task duration is summed with the flight time of the target drone to the inspection key location to obtain the total inspection duration. The inspection time of the second inspection drone to the inspection key location is determined, and the inspection time is compared with the total inspection time. If the inspection time is less than the total inspection time, the second inspection drone is designated as the inspection drone performing the power grid inspection task. The second inspection drone is an idle drone and its inspection task duration is second only to that of the target drone.

[0087] The anomaly resolution module is used to determine whether there is a power grid anomaly in the power grid inspection results. If there is, it determines a real-time solution corresponding to the power grid inspection results based on the standard anomaly solution.

[0088] Example 3:

[0089] See Figure 3 A drone energy optimization device for power grid grid inspection includes a memory and a processor; the memory is used to store computer program code and transmit the computer program code to the processor; the processor is used to execute the drone energy optimization method described in Embodiment 1 according to the instructions in the computer program code.

[0090] Example 4:

[0091] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the UAV energy optimization method described in Embodiment 1.

Claims

1. An energy optimization method for unmanned aerial vehicles (UAVs) used for power grid grid inspection, characterized in that: The UAV energy optimization method includes: Acquire power grid inspection data and power grid environment data within a historical time period; Based on the power grid inspection data, determine the abnormal power grid data of each cell in the power grid at different time points, as well as the abnormal solutions and solution feedback information when resolving the abnormal power grid data; By comprehensively analyzing the power grid anomaly data and the power grid environment data, multiple integrated cell regions and the corresponding key inspection locations for each integrated cell region are obtained. The anomaly solution is updated based on the feedback information from the aforementioned solution to obtain a standard anomaly solution; Based on the key inspection locations, a power grid inspection command is generated, controlling the inspection drone to fly to the key inspection locations to conduct power grid anomaly inspections, and receiving the power grid inspection results transmitted by the inspection drone. Determine whether there is a power grid anomaly in the power grid inspection results. If so, determine a real-time solution corresponding to the power grid inspection results based on the standard anomaly scheme.

2. The UAV energy optimization method for power grid grid inspection according to claim 1, characterized in that: The comprehensive analysis of the power grid anomaly data and the power grid environment data yields multiple integrated cell regions and corresponding key inspection locations for each integrated cell region, including: Based on the power grid environment data, determine the target environment data corresponding to the power grid anomaly data; By analyzing the variation patterns of the power grid anomaly data and the target environment data, the anomaly step pattern of the power grid mesh under different environments is obtained; Collect environmental change data of the power grid grid within a preset time period, and determine the estimated abnormal data of the power grid grid within the preset time period based on the environmental change data and the abnormal step pattern; The predicted abnormal data is analyzed by regionalization and key location selection to obtain multiple cell comprehensive areas and the corresponding key inspection locations for each cell comprehensive area.

3. The UAV energy optimization method for power grid grid inspection according to claim 1, characterized in that: The step of updating the anomaly solution based on the feedback information of the solution to obtain a standard anomaly solution includes: Based on the feedback information of the solution, determine the grid operation status of the grid after abnormal maintenance, and determine whether the grid operation status meets the preset operation status. If it does not meet the preset operation status, determine the position of each grid cell that does not meet the preset operation status, and determine the initial inspection sequence and grid inspection parameters of the inspection drone based on the abnormal solution. The first inspection position and the second inspection position are determined according to the location of the power grid cell and the initial inspection sequence. The first inspection position is the power grid inspection position where the power grid operation status first fails to meet the preset operation status, and the second inspection position is the power grid inspection position where the power grid operation status last fails to meet the preset operation status. The target inspection parameters of the inspection drone between the first inspection position and the second inspection position are determined based on the power grid inspection parameters. Based on the power grid operation status, determine the power grid operation changes between the first inspection position and the second inspection position, and perform simulated inspection of the power grid grid based on the power grid operation changes and the target inspection parameters to obtain the exclusive inspection parameters corresponding to each operation status in the power grid operation changes. The specific inspection parameters are updated according to the initial inspection sequence to update the target inspection parameters, and the updated target inspection parameters are imported into the anomaly solution to obtain a standard anomaly solution.

4. The UAV energy optimization method for power grid grid inspection according to claim 2, characterized in that: The analysis of the variation patterns of the abnormal power grid data and the target environment data yields the abnormal stepwise patterns of the power grid mesh under different environments, including: Construct a power grid anomaly coordinate system, wherein the X-axis of the power grid anomaly coordinate system represents different time nodes, the Y-axis of the power grid anomaly coordinate system represents anomaly data of different anomaly conditions of the power grid grid, and the Z-axis of the power grid anomaly coordinate system represents different environmental rule parameters; The abnormal power grid data and the target environment data are imported into the abnormal power grid coordinate system according to the time development nodes to obtain the abnormal change curve corresponding to each cell in the power grid grid; The abnormal change curve is calculated to obtain the abnormal curve variability of each cell under different power grid environments. The variability of the abnormal curve is matrix-normalized based on the arrangement of each cell in the power grid to obtain the abnormal step pattern of the power grid under different environments.

5. The UAV energy optimization method for power grid grid inspection according to claim 4, characterized in that: The calculation of the curve volatility value of the abnormal change curve to obtain the abnormal curve volatility of each cell under different power grid environments includes: Determine whether each abnormal change curve has fluctuations. If it does, define the starting point after the curve fluctuation as the first node, the peak point after the curve fluctuation as the second node, and the ending point after the curve fluctuation as the third node. Determine the first difference data corresponding to the first node and the second node, and the second difference data corresponding to the second node and the third node, and calculate the ratio of the curve change period between the first node and the second node to the first difference data to obtain the first curve undulation; The second curve variability is obtained by calculating the ratio of the curve change time period between the second node and the third node to the second difference data. The first curve variability and the second curve variability are summarized according to the time ratio to obtain the abnormal curve variability of each cell under different power grid environments.

6. The UAV energy optimization method for power grid grid inspection according to claim 2, characterized in that: The analysis of the predicted anomaly data through regionalization and key location selection yields multiple integrated cell regions and corresponding key inspection locations for each integrated cell region, including: The estimated abnormal data is compared with the preset abnormal data range to determine whether the estimated abnormal data meets the preset abnormal data range. If it does, the power grid cell is the abnormal cell, and the abnormal location of the abnormal cell is recorded. Based on the concentrated distribution characteristics of the abnormal location, the power grid is regionalized to obtain a comprehensive area of ​​multiple cells. For each cell in the comprehensive area, the estimated abnormal data corresponding to different power grid cells are compared pairwise to determine the extreme cell with the largest extreme value in the estimated abnormal data. The position corresponding to the extreme cell is marked to obtain the key inspection position.

7. The UAV energy optimization method for power grid grid inspection according to claim 1, characterized in that: The controlled inspection drone flies to the key inspection location to conduct power grid anomaly inspections, including: Obtain the current real-time location of each inspection drone, and simulate the flight arrival of the current real-time location with the inspection key location to obtain the target drone that arrives at the inspection key location in the shortest time. Determine whether the target drone has an inspection mission. If it does, calculate the inspection mission duration and sum the inspection mission duration with the flight time of the target drone to the inspection key location to obtain the total inspection duration. The inspection time of the second inspection drone flying to the key inspection location is determined, and the inspection time is compared with the total inspection time. If the inspection time is less than the total inspection time, the second inspection drone is designated as the inspection drone performing the power grid inspection task. The second inspection drone is an idle drone and the inspection task time is second only to the target drone.

8. An unmanned aerial vehicle (UAV) energy optimization system for power grid inspection, characterized in that: The UAV energy optimization system includes: The data acquisition module is used to acquire power grid inspection data and power grid environment data within a historical time period. The solution determination module is used to determine, based on the power grid inspection data, the abnormal power grid data of each cell in the power grid at different time nodes, as well as the abnormal solutions and solution feedback information when resolving the abnormal power grid data; The power grid analysis module is used to comprehensively analyze the power grid anomaly data and the power grid environment data to obtain multiple cell comprehensive areas and the corresponding key inspection locations for each cell comprehensive area; The solution update module is used to update the abnormal solution based on the solution feedback information to obtain a standard abnormal solution; The power grid inspection module is used to generate power grid inspection commands based on the key inspection locations, control the inspection drone to fly to the key inspection locations to conduct power grid anomaly inspections, and receive the power grid inspection results transmitted by the inspection drone. The anomaly resolution module is used to determine whether there is a power grid anomaly in the power grid inspection results. If so, it determines a real-time solution corresponding to the power grid inspection results based on the standard anomaly solution.

9. An energy optimization device for unmanned aerial vehicles (UAVs) used for power grid inspection, characterized in that: The UAV energy optimization device includes a memory and a processor; the memory is used to store computer program code and transmit the computer program code to the processor; the processor is used to execute the UAV energy optimization method as described in claim 1 according to the instructions in the computer program code.

10. A computer-readable storage medium storing a computer program that, when executed by a processor, implements the UAV energy optimization method as described in claim 1.

Images