Power line patrol unmanned aerial vehicle control method and system

Abstract

The invention relates to the technical field of unmanned aerial vehicles, in particular to a power line patrol unmanned aerial vehicle control method and system. Firstly, an unmanned aerial vehicle group is deployed; acquiring real-time power data, real-time flight data, real-time environment data and real-time unmanned aerial vehicle state data; analyzing the real-time power data and the real-time flight data to obtain a power line model; analyzing the power line model to obtain a first power inspection path group; further analyzing the real-time environment and the real-time unmanned aerial vehicle state data to obtain a safety state value and a next flight state of the unmanned aerial vehicle; when the next flight state of the unmanned aerial vehicle is stopped, determining the inspection point safety state value of the next to-be-inspected point; and finally, monitoring each unmanned aerial vehicle in the unmanned aerial vehicle group, and when the safety state of a certain unmanned aerial vehicle is identified to be abnormal, analyzing the real-time unmanned aerial vehicle state data, the real-time electric power data and the real-time flight data to obtain a second electric power inspection path group.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a control method and system for an electric power line inspection unmanned aerial vehicle. Background Art

[0002] Power line inspection is an important task of regular inspection and maintenance of power lines and their ancillary equipment. Its purpose is to timely discover and deal with hidden dangers in power lines and ensure the safe and stable operation of the power system. Power lines are exposed to the external environment for a long time and are easily affected by factors such as natural disasters and man-made damage. Regular line inspection can effectively prevent faults and ensure the reliability and continuity of power supply.

[0003] With the advancement of science and technology, drone technology has been increasingly widely used in the field of power line inspection. Drones have the advantages of flexibility, high flight speed and wide coverage, and can quickly reach high altitudes and complex terrain areas that are difficult to reach for inspection. Compared with traditional ground inspection methods, drone line inspection can greatly improve inspection efficiency, reduce labor costs and inspection time. However, traditional drone line inspection technology mainly relies on remote control, which is complicated to operate and easily leads to fluctuations in inspection accuracy and efficiency;

[0004] In the prior art, there are still some defects in the research on the control of drones used for power line inspection. For example, a drone line inspection trajectory planning method and system proposed in the Chinese patent with publication number CN113359864B first divides the target line inspection area for the drone; then, obtains the map of the area, and queries the location coordinates of all power facilities and drone power swap stations on the map. Next, a first trajectory point and a second trajectory point including all power facilities in the target line inspection area are generated; at the same time, a supply point covering the drone power swap station in the area is generated; based on the first trajectory point, a preset path covering these points is formed; finally, according to the power supply demand of the drone, the preset path is adjusted to ensure that the drone can complete the line inspection task efficiently and accurately. The Chinese patent with publication number CN113011780B proposes a task allocation method for multi-drone collaborative power inspection. First, the parking point location and the set of tasks and drones are obtained. Then, the actual distance between each task tower and the parking point is calculated. Next, a task allocation model is constructed. The optimization goal of the model is to minimize the time it takes the longest drone to complete its assigned task. At the same time, the assumptions and constraints of the model are set based on the actual needs of power inspection. Finally, the task allocation model is solved by using a genetic algorithm to obtain the overall optimal solution that satisfies all constraints, thereby achieving the optimization goal.

[0005] Existing drone control technology for power inspection mainly relies on remote control and preset flight paths for inspection, and has the problem of insufficient adaptability to dynamic environments.

[0006] Therefore, a control method and system for a power line inspection UAV are proposed. Summary of the invention

[0007] The purpose of the present invention is to provide a method and system for controlling a power line inspection drone. First, deploy a drone group; obtain real-time power data, real-time flight data, real-time environment data, and real-time drone status data; then, analyze the real-time power data and real-time flight data to obtain a power line model; analyze the power line model to obtain a first power inspection path group; further, analyze the real-time environment and real-time drone status data to obtain a safety status value and the next flight status of the drone; when the next flight status of the drone is an abnormal state, determine the safety status value of the next inspection point to be inspected; finally, monitor each drone in the drone group separately, and when it is identified that the flight status of a drone is abnormal, analyze the real-time drone status data, real-time power data, and real-time flight data of the drones in the drone group to obtain a second power inspection path group.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A control method for a power line inspection drone, comprising:

[0010] Deploy a drone group; the drone group includes multiple drones; each drone is equipped with a high-definition camera, GPS and environmental sensors;

[0011] Obtain real-time power data, real-time flight data, real-time environmental data, and real-time drone status data;

[0012] Using a graph theory algorithm to analyze the real-time power data and the flight data to obtain a power line model; applying a multi-objective optimization dynamic path planning algorithm to analyze the power line model to obtain a first power inspection path group;

[0013] By analyzing the real-time environmental data and the real-time drone status data, a safety status value is obtained; the safety status value is used to determine the next flight status of the drone; the flight status includes stopping, adjusting the flight strategy and maintaining flight; the flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed;

[0014] If the next flight state of the drone is to stop, the safety state of the drone is abnormal, the position of the drone and the next inspection point is analyzed to obtain the safety state value of the inspection point; and whether to reallocate the inspection point for inspection is determined according to the safety state value of the inspection point;

[0015] Each of the drones in the drone group is monitored separately. When an abnormal state is identified in a drone, a real-time dynamic task allocation algorithm among multiple drones is used to obtain a second power inspection path group based on the real-time drone status data, the real-time power data and the real-time flight data of each drone in the drone group.

[0016] The real-time power data includes the number, location, inspection status and safety status of multiple inspection points; the real-time flight data includes obstacle location, safety area and flight restricted area; the real-time environmental data includes visibility, temperature, humidity, wind speed and wind direction; the real-time drone status data includes number, power, load data, location, safety status and task execution progress; the safety status includes normal status and abnormal status; the abnormal status includes the drone whose safety status value is lower than the safety threshold 1, the power is 0 and the position remains unchanged for a long time; the inspection status includes waiting for inspection and inspected.

[0017] The multi-objective optimization dynamic path planning algorithm includes a potential field method and an RRT* algorithm; the calculation formula of the first power inspection path group is:

[0018]

[0019] Among them, {P opt} is the first power inspection path group, including the optimal path P for each drone i , L(P i ),C(P i ),T(P i ) are the path length, energy consumption and completion time of UAV i, n is the number of UAVs, argmin() is used to calculate the minimized path group, α1, α2, α3 are L(P i ),C(P i ),T(P i )’s weight.

[0020] The calculation formula of the safety status value is:

[0021]

[0022] Among them, T1 is the safety state value, w k is the weight of the kth environmental parameter, E k is the real-time data of the kth environmental parameter, m is the number of environmental parameters, and f k is the nonlinear transformation function for the kth parameter, and σ is the sigmoid activation function;

[0023] The environmental parameters include visibility, temperature, humidity, wind speed and wind direction;

[0024] The calculation formula of the safety status value of the inspection point is:

[0025] T2 = γ·T1 + δ·h(D);

[0026] Among them, T2 is the safety status value of the inspection point, γ, δ are weight coefficients, and h(D) is the distance risk function calculated based on the distance D from the drone to the inspection point;

[0027] The calculation formula of the distance D from the drone to the inspection point is:

[0028]

[0029] Among them, (X1, Y1, Z1) is the current position of the drone, and (X2, Y2, Z2) is the position of the next inspection point;

[0030] The calculation formula of the distance risk function is:

[0031]

[0032] Among them, v is the shape parameter, and D0 is the distance between the UAV and the midpoint of the next inspection point.

[0033] The adjusted flight strategy is obtained by analyzing the real-time environmental data and the real-time UAV status data using an improved particle swarm optimization algorithm and an A* search algorithm.

[0034] The calculation formula of the second power inspection path group is:

[0035]

[0036] Among them, {P new} is the second power inspection path group, including the second optimal path P of each drone j , L(P j ),C(P j ),T(P j ) are the path length, energy consumption and completion time of UAV j, e is the number of UAVs with normal safety status, argmin() is used to calculate the minimized path group, λ1,λ2,λ3 are L(P i ),C(P i ),T(P i )’s weight.

[0037] A power line inspection drone control system, comprising:

[0038] A deployment unit, used to deploy a drone group; the drone group includes a plurality of drones; each of the drones is equipped with a high-definition camera, GPS and environmental sensors;

[0039] A data acquisition unit, used to acquire real-time power data, real-time flight data, real-time environmental data, and real-time drone status data;

[0040] A first power inspection path group acquisition unit is used to acquire a first power inspection path group; the acquisition process of the first power inspection path group includes: analyzing the real-time power data and the real-time flight data by using a graph theory algorithm to acquire a power line model; applying a multi-objective optimization dynamic path planning algorithm to analyze the power line model to acquire the first power inspection path group;

[0041] The UAV flight status analysis unit is used to obtain a safety status value by analyzing the real-time environmental data and the real-time UAV status data; and to determine the next flight status of the UAV by analyzing the safety status value; the flight status includes stopping, adjusting the flight strategy and maintaining flight; and the flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed;

[0042] A patrol point safety status analysis unit is used to obtain a patrol point safety status value of the next patrol point of the drone whose next flight state is stopped; by analyzing the patrol point safety status value, it is determined whether to reallocate the patrol point for inspection;

[0043] An abnormal status updating unit is used to obtain a second power inspection path group; the process of obtaining the second power inspection path group includes: monitoring each of the drones in the drone group separately, and when an abnormal state is identified in a drone, using a real-time dynamic task allocation algorithm among multiple drones, and obtaining the second power inspection path group based on the real-time drone status data, the real-time power data and the real-time flight data of each drone in the drone group.

[0044] The real-time power data includes the number, location, inspection status and safety status of multiple inspection points; the real-time flight data includes obstacle location, safety area and flight restricted area; the real-time environmental data includes visibility, temperature, humidity, wind speed and wind direction; the real-time drone status data includes number, power, load data, location, safety status and task execution progress; the safety status includes normal status and abnormal status; the abnormal status includes the drone whose safety status value is lower than the safety threshold 1, the power is 0 and the position remains unchanged for a long time; the inspection status includes waiting for inspection and inspected; the safety status includes safe and dangerous, which is determined by the inspection point safety status value of the inspection point.

[0045] The multi-objective optimization dynamic path planning algorithm includes a potential field method and an RRT* algorithm; the calculation formula of the first power inspection path group is:

[0046]

[0047] Among them, {P opt} is the optimal path group, including the optimal path P of each drone i , L(P i ),C(P i ),T(P i ) are the path length, energy consumption and completion time of UAV i, n is the number of UAVs, argmin() is used to calculate the minimized path group, α1, α2, α3 are L(P i ),C(P i ),T(P i )’s weight.

[0048] The calculation formula of the safety status value is:

[0049]

[0050] Among them, T1 is the safety state value, w k is the weight of the kth environmental parameter, E k is the real-time data of the kth environmental parameter, m is the number of environmental parameters, and f k is the nonlinear transformation function for the kth parameter, and σ is the sigmoid activation function;

[0051] The environmental parameters include visibility, temperature, humidity, wind speed and wind direction.

[0052] Compared with the prior art, the present invention has the following beneficial effects:

[0053] 1. The present invention can effectively obtain the first power inspection path group by using graph theory algorithms to analyze real-time power data and real-time flight data, combined with a dynamic path planning algorithm for multi-objective optimization. This method takes into account path length, energy consumption, and completion time, and optimizes the design of the inspection path, so that the drone can cover all necessary inspection points in the shortest time with the lowest energy consumption when performing inspection tasks. This efficient path planning significantly improves the efficiency and economy of power inspections and reduces operating costs.

[0054] 2. The present invention calculates the safety status value by real-time analysis of environmental data and drone status data. This value is used to determine the flight status of the drone in real time and adjust the flight strategy. In addition, the inspection points are classified as normal or abnormal according to their actual conditions. For inspection points in normal status, the present invention ensures that these points can be reallocated to other drones for inspection when a drone fails. For inspection points in abnormal status, inspections are temporarily not carried out until environmental conditions meet safety requirements. This dynamic adjustment mechanism enhances the safety and flexibility of the inspection process and ensures the smooth completion of the inspection task.

[0055] 3. The present invention can dynamically reallocate tasks of the remaining drones and recalculate the inspection path when an abnormal state is identified in a drone group. This process is achieved through a real-time dynamic task allocation algorithm to ensure that all normal inspection points are re-covered in the second power inspection path group, so as not to affect the overall progress and effect of the inspection task. This timely task reallocation capability significantly improves the adaptability and robustness of the drone inspection system, enabling the system to effectively respond to emergencies and maintain the continuity of inspection tasks. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 A flow chart of a control method for a power line inspection drone provided by an embodiment of the present invention;

[0057] Figure 2 A structural diagram of a power line inspection drone control system provided by an embodiment of the present invention;

[0058] Figure 3 A schematic diagram of a portion of a first power inspection path group provided in Embodiment 1 of the present invention;

[0059] Figure 4 A schematic diagram of a portion of a second power inspection path group provided in Embodiment 1 of the present invention;

[0060] Figure 5 A schematic diagram of part of the first power inspection path group provided in Embodiment 2 of the present invention. DETAILED DESCRIPTION

[0061] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0062] There are many studies on the control of UAVs for power line inspection in the prior art, but there are some defects. In order to solve the defects in the prior art, the purpose of the present invention is to provide a control method for a power line inspection UAV. This method is applied to a power line inspection UAV control system. For specific method flow chart and system structure diagram, see Figure 1 and Figure 2 .

[0063] Embodiment 1

[0064] This embodiment analyzes the control of a drone used for power line inspection A, and the specific analysis ideas are as follows:

[0065] Deploy drone groups to conduct inspections on power lines; obtain real-time power data, real-time flight data, real-time environmental data, and real-time drone status data to determine inspection routes and analyze the status of drones and inspection points during the inspection process;

[0066] By analyzing the real-time power data and the real-time flight data, a power line model is obtained; by analyzing the power line model, a first power inspection path group is obtained; the drones in the drone group inspect the inspection points according to the inspection paths corresponding to the drones in the first power inspection path group;

[0067] During the inspection process, the safety status of each drone in the drone group is monitored separately. When the safety status of a drone is identified as abnormal, the real-time drone status data, real-time power data and real-time flight data of the drones in the drone group are further analyzed to obtain the second power inspection path group.

[0068] Among them, abnormal states include the safety state value being lower than the safety threshold 1, the power being 0, and the position being unchanged for a long time;

[0069] By monitoring the power of the drone and the location information through the GPS equipped by the drone, the safety status of the drone with a power of 0 and a long-term unchanged position is updated as an abnormal state;

[0070] The real-time environmental data of the drone is obtained by using the environmental sensors, cameras and GPS equipped on the drone, and the safety status value of the drone is obtained by analyzing the real-time environmental data;

[0071] Determine the next flight state of the drone according to the safety state value of the drone; the flight state includes stopping, adjusting the flight strategy and maintaining flight; adjusting the flight strategy includes changing the flight direction and changing the flight angle;

[0072] When the next flight state of the drone is to maintain flight, the first power inspection path group continues to be executed; when the next flight state of the drone is to adjust the flight strategy, the improved particle swarm optimization algorithm and A* search algorithm are used to analyze the real-time environmental data and the real-time drone status data to obtain the adjusted flight strategy; the flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed; when the next flight state of the drone is to stop, the safety state of the drone is updated to an abnormal state, and the inspection point safety state value of the next inspection point of the drone is determined;

[0073] When the next flight state of the drone is to stop, the safety state of the next point to be inspected of the drone is updated according to the safety state value of the inspection point; the real-time drone state data, real-time power data and real-time flight data of the drones in the drone group are further analyzed to obtain the second power inspection path group. Among them, the safety state of the next point to be inspected includes normal state and abnormal state; when the safety state of the inspection point is normal state and the inspection state is to be inspected, when obtaining the second power inspection path group, the inspection point belongs to the inspection point to be considered; when the safety state of the inspection point is abnormal state, the inspection point is not considered when determining the second power inspection path group at that time; when the power inspection path group is re-determined next time, it is added to the inspection point to be inspected, and the drone is considered to be assigned to the inspection; in the process of determining the second power inspection path group, only drones with a normal safety state and inspection points with a normal safety state and an inspection state to be inspected are considered.

[0074] In this embodiment, by deploying a drone group, and acquiring and analyzing real-time power data, real-time environmental data, and real-time drone status data, efficient inspection of power lines is achieved. First, by analyzing the real-time power data, a power line model is constructed, and the first power inspection path group is further obtained by analyzing the model. Then, the real-time environment and real-time drone status data are analyzed, the safety state value is calculated, and the next flight state of the drone is determined. When the next flight state of a drone is stopped, the system will evaluate the safety state value of the next inspection point to be inspected to ensure the safety and rationality of the inspection process. Finally, by continuously monitoring the status of each drone in the drone group, when the safety state of a drone is identified as abnormal, the status data and power data of the drone group are analyzed in real time, and the second power inspection path group is dynamically obtained to ensure the continuity and efficiency of the inspection task. The design of this system significantly improves the flexibility and adaptability of drone inspections, ensuring the safe and efficient execution of inspection work.

[0075] Reference Figure 1S10 in which S10 is applied to a deployment unit and a data acquisition unit of a power line patrol drone control system; the data acquisition unit is used to deploy a drone group; the drone group includes a plurality of drones; each of the drones is equipped with a high-definition camera, a GPS and an environmental sensor;

[0076] The data acquisition unit is used to acquire real-time power data, real-time flight data, real-time environmental data and real-time drone status data;

[0077] Furthermore, the real-time power data includes the number, location, inspection status and safety status of multiple inspection points; the real-time flight data includes obstacle location, safety area and flight restricted area; the real-time environmental data includes visibility, temperature, humidity, wind speed and wind direction; the real-time drone status data includes number, power, load data, location, safety status and task execution progress; the safety status includes normal status and abnormal status; the abnormal status includes the drone with a safety status value lower than the safety threshold 1, a power of 0 and a long-term unchanged position; the inspection status includes waiting for inspection and inspected. Table 1 is a data example table of some real-time power data;

[0078] Table 1 Data example of some real-time power data

[0079] serial number Location Inspection status Security Status 1 (34.0522,-118.2437) Waiting for inspection Normal state 2 (34.0522,-118.2440) Waiting for inspection Normal state 3 (34.0525,-118.2437) Waiting for inspection Normal state ... ... ... ...

[0080] In this embodiment, the drone group includes three drones; they are numbered 01, 02 and 03 respectively; the visibility in the real-time environmental data is obtained by analyzing the pictures taken by the camera equipped by the drone; the temperature, humidity, wind speed and wind direction are obtained by the environmental sensors equipped by the drone;

[0081] In this embodiment, the data acquisition unit can obtain real-time power data, real-time environmental data and real-time drone status data, providing comprehensive information support for power inspection. Real-time power data covers the location, inspection status and safety status of the inspection point, real-time environmental data includes visibility, temperature, humidity, wind speed and wind direction, and real-time drone status data includes number, power, load data, location, safety status and task execution progress. This comprehensive data collection capability enables drones to respond accurately to various inspection situations, optimize inspection paths and adjust flight strategies in real time, thereby significantly improving inspection efficiency and safety, ensuring efficient execution of inspection work and safe operation of drones.

[0082] Reference Figure 1S20 in the embodiment, S20 is applied to a first power inspection path group acquisition unit of a power line inspection drone control system, and is used to acquire a first power inspection path group; the acquisition process of the first power inspection path group includes: acquiring a power line model by analyzing the real-time power data and the real-time flight data using a graph theory algorithm; analyzing the power line model by applying a multi-objective optimization dynamic path planning algorithm to acquire the first power inspection path group;

[0083] Furthermore, by using a graph theory algorithm to analyze the real-time power data, the specific process of obtaining the power line model includes:

[0084] Constructing a graph model: defining nodes and edges; the nodes are key points for inspection in the power system. In this embodiment, the key points include transmission towers, substations, and connection boxes; the edges represent power lines and connect corresponding nodes; the attributes of the nodes include the location, inspection status, and safety status of the inspection points; the weights of the edges are determined by the line length, transmission capacity, and urgency of maintenance;

[0085] Create a graph: construct a graph G = (V, E); where V is the node set and E is the edge set;

[0086] Expanded graph: Mark the locations of all obstacles and restricted flight zones in the graph. The nodes corresponding to these locations are specially marked to prevent path planning from crossing these areas.

[0087] Safety area processing: Mark all safety areas. When planning a route, you can give priority to routes that pass through or are close to these safety areas to increase flight safety.

[0088] Connectivity analysis: Analyze the connectivity of the graph to confirm that all nodes are reachable and there are no isolated parts.

[0089] Furthermore, the multi-objective optimization dynamic path planning algorithm includes a potential field method and an RRT* algorithm;

[0090] The potential field method is a popular navigation and obstacle avoidance algorithm that guides the movement of a drone by creating a virtual force field around it. These force fields include attractive forces and repulsive forces; the attractive forces are forces directed toward the target, pushing the drone toward the target; the repulsive forces are forces from obstacles, pushing the drone away from the obstacles;

[0091] The attraction F attr The expression formula is:

[0092] F attr =-k attr (xx-xx goal );

[0093] Among them, kattr is the attraction coefficient, xx,xx goal They are the real-time position and target position of the UAV respectively;

[0094] In this embodiment, the target position is the position of the next point to be inspected;

[0095] The repulsive force F rep The expression formula is:

[0096]

[0097] Among them, xx obs is the obstacle position, d is the distance from the drone to the obstacle, and d max is the maximum distance at which the repulsive force acts, k rep is the repulsion coefficient;

[0098] RRT* is a sampling-based path planning algorithm that can effectively find a path from the starting point to the end point in a complex environment. Compared with the original RRT algorithm, RRT* adds support for path optimization while maintaining the rapid exploration of the path. The implementation process includes:

[0099] Initialization: Starting from the starting point, initialize the tree.

[0100] Iterative growth: In each iteration, a point is randomly sampled, the nearest node is found in the tree as the parent node of the point, and an attempt is made to expand a new node to the sampled point.

[0101] Path optimization: When a new node is introduced, check whether there is a shorter path connecting to the node. If so, reconnect the parent node of the node.

[0102] Collision detection: Collision detection is performed when adding each new node to ensure that the path does not cross obstacles.

[0103] Target Detection: Once a target is found that can be reached directly from any node in the tree without collision, the algorithm attempts to connect to the target.

[0104] In this embodiment, the implementation process of the multi-objective optimization dynamic path planning algorithm based on the potential field method and the RRT* algorithm includes:

[0105] Environmental modeling: Build an environmental model that includes all known obstacles and points to be inspected. This model is represented as a continuous or gridded space that includes obstacle locations, safety areas, and restricted flight zones.

[0106] Generate guidance fields: In the environment model, the potential field method is applied to generate guidance fields for each drone. This includes for each drone:

[0107] Attraction: Generate an attraction field pointing to the target for the inspection point.

[0108] Repulsion: Generates a repulsion field from obstacles and other drones, ensuring drones maintain a safe distance during flight.

[0109] Path sampling and growth: Path sampling and growth are performed using the RRT* algorithm under the guidance of the potential field. The potential field provides a force to push toward the goal, helping RRT* avoid obstacles and develop in a more efficient direction when exploring the path;

[0110] Sampling strategy: By adjusting the sampling strategy of RRT*, it is more inclined to sample in areas with strong attraction, while considering repulsion and avoiding areas with dense obstacles.

[0111] Node expansion: When performing node expansion of RRT*, consider whether the path segment from the current node to the new sampling point will increase the total path energy calculated by the potential field method. This includes path length, obstacle avoidance capability, and energy consumption.

[0112] Path optimization: After the RRT* tree is formed, the generated path is optimized to further reduce the path cost and improve flight efficiency. The path is corrected using the information of the potential field, the path curve is optimized, unnecessary turns are reduced, and the path smoothness and flight efficiency are improved.

[0113] Real-time Updates: During the flight of the UAV, the potential field is updated in real time and the RRT* path is recalculated based on changes in the environment, including new obstacles or changing weather conditions.

[0114] The calculation formula of the first power inspection path group is:

[0115]

[0116] Among them, {P opt} is the first power inspection path group, including the optimal path P for each drone i , L(P i ),C(P i ),T(P i ) are the path length, energy consumption and completion time of UAV i, n is the number of UAVs, argmin() is used to calculate the minimized path group, α1, α2, α3 are L(P i ),C(P i ),T(P i )’s weight.

[0117] In this embodiment, the obtained first power inspection path group includes three inspection paths, and the three inspection paths are respectively assigned to drones with corresponding numbers 01, 02 and 03; Figure 3Shown is a schematic diagram of part of the first power inspection path group;

[0118] By using graph theory algorithms combined with real-time power data to obtain a power line model, and applying multi-objective optimization dynamic path planning algorithms, such as the potential field method and the RRT* algorithm, the first power inspection path group is effectively obtained. This process ensures that the inspection path of each drone is both efficient and energy-saving, while taking into account the optimization of completion time. By accurately calculating the path length, energy consumption and completion time of the drone, and adjusting the weight parameters using a multi-objective optimization algorithm, the present invention achieves precise control and optimization of the drone inspection path. This method significantly improves the operational efficiency of power inspections, ensuring that drones can complete inspection tasks in the shortest time with the lowest energy consumption, thereby reducing operating costs and improving inspection quality.

[0119] Reference Figure 1 S30 in the embodiment, S30 is applied to a UAV flight state analysis unit and a patrol point safety state analysis unit of a power line patrol UAV control system; the UAV flight state analysis unit is used to obtain a safety state value by analyzing the real-time environmental data and the real-time UAV state data; by analyzing the safety state value, the next flight state of the UAV is determined; the flight state includes stopping, adjusting the flight strategy and maintaining flight; the flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed; the patrol point safety state analysis unit is used to obtain the patrol point safety state value of the next patrol point of the UAV whose next flight state is stopping; by analyzing the patrol point safety state value, it is determined whether to reallocate the patrol point for inspection;

[0120] Furthermore, the calculation formula of the safety status value is:

[0121]

[0122] Among them, T1 is the safety state value, w k is the weight of the kth environmental parameter, E k is the real-time data of the kth environmental parameter, m is the number of environmental parameters, and f k is the nonlinear transformation function for the kth parameter, and σ is the sigmoid activation function;

[0123] The environmental parameters include visibility, temperature, humidity, wind speed and wind direction;

[0124] When T1≤T min When T min <T1<T max When T1≥T maxWhen T, the next flight state of the UAV is to maintain flight; min ,T max It is a preset value based on the analysis of historical data; T min is the safety threshold 1, T max is the safety threshold 2;

[0125] The calculation formula of the safety status value of the inspection point is:

[0126] T2 = γ·T1 + δ·h(D);

[0127] Among them, T2 is the safety status value of the inspection point, γ, δ are weight coefficients, and h(D) is the distance risk function calculated based on the distance D from the drone to the inspection point;

[0128] When T2<T3, the safety status of the inspection point is abnormal, and when T2≥T3, the safety status of the inspection point is normal; T3 is a value preset based on the analysis of historical data;

[0129] The calculation formula of the distance D from the drone to the inspection point is:

[0130]

[0131] Among them, (X1, Y1, Z1) is the current position of the drone, and (X2, Y2, Z2) is the position of the next inspection point;

[0132] The calculation formula of the distance risk function is:

[0133]

[0134] Among them, v is the shape parameter, and D0 is the distance between the UAV and the midpoint of the next inspection point.

[0135] By comprehensively analyzing the real-time environmental data and the status data of the UAV, the safety status value is obtained, and the next flight status of the UAV is judged accordingly, which effectively improves the accuracy and real-time performance of flight safety management. Through this analysis, the UAV can choose to stop flying, adjust the flight strategy, or continue to maintain the current flight status according to the evaluation results of the safety status value. The adjustment strategy includes changing the flight direction and flight angle to adapt to environmental changes. In addition, the inspection point safety status analysis unit further analyzes the safety status values ​​of the inspection points affected by the stop of the flight status, and determines whether these inspection points need to be reallocated to ensure the continuity and efficiency of the inspection work. This dynamic flight status management and inspection point reallocation mechanism not only improves the flexibility and response speed of the UAV inspection tasks, but also significantly enhances the safety and reliability of the entire inspection process.

[0136] Furthermore, the adjusted flight strategy is obtained by analyzing the real-time environmental data and the real-time UAV status data using an improved particle swarm optimization algorithm and an A* search algorithm.

[0137] By comprehensively analyzing the real-time environmental data and the status data of the UAV, the safety status value is obtained, and the next flight status of the UAV is judged accordingly, which effectively improves the accuracy and real-time performance of flight safety management. Through this analysis, the UAV can choose to stop flying, adjust the flight strategy, or continue to maintain the current flight status according to the evaluation results of the safety status value. The adjustment strategy includes changing the flight direction and flight angle to adapt to environmental changes. In addition, the inspection point safety status analysis unit further analyzes the safety status values ​​of the inspection points affected by the stop of the flight status, and determines whether these inspection points need to be reallocated to ensure the continuity and efficiency of the inspection work. This dynamic flight status management and inspection point reallocation mechanism not only improves the flexibility and response speed of the UAV inspection tasks, but also significantly enhances the safety and reliability of the entire inspection process.

[0138] In this embodiment, after inspecting inspection point 5, the drone numbered 02 flies to the inspection line of inspection point 8, and the safety quality detection value is less than T min The next flight state of the drone is stopped; the safety state of drone No. 02 is updated from normal state to abnormal state.

[0139] Further, the inspection point safety status value of inspection point 8 is calculated; in this embodiment, the inspection point safety status value of the inspection point is less than T3, and the safety status of the inspection point is abnormal; therefore, inspection point 8 is not considered when calculating the second power inspection path group in the next step; in this embodiment, the current real-time drone status data is shown in Table 2;

[0140] Table 2 Current real-time drone status data table

[0141]

[0142] Reference Figure 1 S40 in the embodiment, S40 is applied to an abnormal state update unit of a power line inspection UAV control system, and is used to obtain a second power inspection path group; the process of obtaining the second power inspection path group includes: monitoring each of the UAVs in the UAV group separately, and when an abnormal state is identified in a certain UAV, using a real-time dynamic task allocation algorithm among multiple UAVs, and obtaining the second power inspection path group according to the real-time UAV state data and the real-time power data of each UAV in the UAV group.

[0143] Furthermore, the calculation formula of the second power inspection path group is:

[0144]

[0145] Among them, {P new} is the second power inspection path group, including the second optimal path P of each drone j , L(P j ),C(P j ),T(P j ) are the path length, energy consumption and completion time of UAV j, e is the number of UAVs with normal safety status, argmin() is used to calculate the minimized path group, λ1,λ2,λ3 are L(P i ),C(P i ),T(P i )’s weight.

[0146] By implementing a real-time dynamic task allocation algorithm, the second power inspection path group is effectively obtained to deal with abnormal conditions in the drone group. When any drone is detected to have a fault or abnormality, the system can quickly reallocate inspection tasks and recalculate and optimize the inspection path based on the real-time status data and power data of each drone. This method not only ensures the continuity of the inspection task, but also reduces the task completion time, energy consumption and improves the overall inspection efficiency by dynamically optimizing the path. Furthermore, the calculation of the second power inspection path group adopts a multi-objective optimization strategy, weighing the path length, energy consumption and time, effectively improving the response speed and adaptability of the system, thereby optimizing the operating status of the drone and the execution of the inspection task while ensuring safety and efficiency.

[0147] In this embodiment, since the safety status of the drone numbered 02 is updated to an abnormal state and cannot continue to perform the inspection task, it is necessary to calculate the second power inspection path group to reallocate the inspection tasks to the drones numbered 01 and 02; in this embodiment, by using a real-time dynamic task allocation algorithm among multiple drones, the real-time drone status data and real-time power data of each drone in the drone group are analyzed, and the second power inspection path group for the drones numbered 01 and 02 to continue the inspection is obtained; a schematic diagram of part of the second power inspection path group is shown in FIG. Figure 4 As shown;

[0148] Embodiment 2

[0149] This embodiment is based on the description of a method and system for controlling a power line inspection drone in the first embodiment, and analyzes the control of a drone used for power line inspection B.

[0150] In this embodiment, the drone group includes 4 drones; they are numbered 01, 02, 03 and 04 respectively; the visibility in the real-time environmental data is obtained by analyzing the pictures taken by the camera equipped by the drone; the temperature, humidity, wind speed and wind direction are obtained by the environmental sensors equipped by the drone;

[0151] In this embodiment, the obtained first power inspection path group includes four inspection paths, and the four inspection paths are respectively assigned to drones with corresponding numbers 01, 02, 03 and 04; Figure 5 Shown is a schematic diagram of part of the first power inspection path group;

[0152] In this embodiment, after inspecting inspection point 3, the drone numbered 03 flies to the inspection line of inspection point 4, and the safety quality detection value is less than T max and greater than T min , the next flight state of the drone is to adjust the flight strategy;

[0153] Furthermore, the adjusted flight strategy is obtained by analyzing the real-time environment data and the real-time UAV status data using an improved particle swarm optimization algorithm and an A* search algorithm. The specific acquisition process includes:

[0154] Particle initialization: Each particle represents a possible flight strategy, including the settings of flight angle, height and speed.

[0155] Particle attributes: The attributes of each particle include angle θ, height hg and velocity vg, as well as the corresponding fitness score, which reflects the adaptability of the strategy to the current flight conditions.

[0156] The fitness score of each particle is calculated as:

[0157] fitness=fi1·Safety(θ,hg)+fi2·Energy(vg)+fi3·Time(vg);

[0158] Among them, fitness is the fitness score, Safety (θ, hg) is the safety margin based on the adjustment of flight angle and altitude to adapt to environmental obstacles and wind direction, Energy (vg) represents the energy efficiency function, which is used to evaluate the energy consumption at a given speed; Time (vg) is the time efficiency function, which is used to evaluate the time required to complete the task;

[0159] Update the particle's position and velocity;

[0160] The position update formula is:

[0161] x q (t+1)=x q (t)+vgq (t+1);

[0162] Among them, x q (t) is the position of particle q at time t; vg q (t+1) is the velocity of particle q at time t+1;

[0163] The speed update formula is:

[0164]

[0165] in, is the inertia weight, which is used to control the momentum of the particle; is the learning factor, controlling the particle to move to the best historical position p best and the global optimal position g best degree of attraction.

[0166] Based on the flight parameters determined by particle swarm optimization, the A* search algorithm calculates the optimal path from the current position to the target position, taking into account environmental obstacles and flight restricted areas.

[0167] The cost function f(n) of the A* search algorithm is:

[0168] f(n)=g(n)+h(n);

[0169] Where g(n) is the actual cost from the starting point to node n, and h(n) is the estimated cost from node n to the target point, calculated using Euclidean distance or Manhattan distance.

[0170] The path selection process includes:

[0171] Select the node with the lowest f(n) value from the open list as the current node;

[0172] Update the g(n) and h(n) values ​​of the nodes around the current node, and recalculate f(n);

[0173] The flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed;

[0174] The adjustment of the flight angle θ new The formula is:

[0175] θ new =θ curr +k θ ·(θ desi -θ curr )+δ wind ;

[0176] Among them, θ curr ,θ desiare the current flight angle and the ideal flight angle calculated based on the target path, k θ is the angle adjustment sensitivity coefficient, δ wind It is the wind direction deviation compensation calculated based on the real-time wind direction data to offset the crosswind effect;

[0177] The adjustment of the flight altitude h new The formula is:

[0178] h new =h curr +k h ·(h desi -h curr )+δ h ;

[0179] Among them, h curr ,h desi are the current flight altitude and the ideal flight altitude calculated based on the target path, k h is the height adjustment sensitivity coefficient, δ h It is the altitude adjustment compensation value calculated based on real-time environmental data;

[0180] The adjustment of the flight speed vg new The formula is:

[0181] vg new =vg curr +k vg ·(vg desi -vg curr )+δ vg ;

[0182] Among them, vg curr ,vg desi are the current flight speed and the ideal flight speed calculated based on the target path, k vg is the speed adjustment sensitivity coefficient, δ vg It is the speed adjustment compensation value calculated based on real-time environmental data;

[0183] Furthermore, the drone numbered 03 adjusts its flight strategy and continues to perform the unfinished inspection tasks in the first power inspection path group;

[0184] Further, each drone in the drone group continues to be monitored;

[0185] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A control method for a power line inspection drone, characterized in that: include: deploying drone groups; The drone group includes a plurality of drones; each of the drones is equipped with a high-definition camera, GPS and environmental sensors; Obtain real-time power data, real-time flight data, real-time environmental data, and real-time drone status data; Using a graph theory algorithm to analyze the real-time power data and the flight data to obtain a power line model; applying a multi-objective optimization dynamic path planning algorithm to analyze the power line model to obtain a first power inspection path group; By analyzing the real-time environmental data and the real-time drone status data, a safety status value is obtained; the safety status value is used to determine the next flight status of the drone; the flight status includes stopping, adjusting the flight strategy and maintaining flight; the flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed; If the next flight state of the drone is stopped, the safety state of the drone is abnormal, and the position of the drone and the next inspection point is analyzed to obtain the safety state value of the inspection point; Determining whether to reallocate the inspection point for inspection according to the safety status value of the inspection point; Each of the drones in the drone group is monitored separately. When an abnormal state is identified in a drone, a real-time dynamic task allocation algorithm among multiple drones is used to obtain a second power inspection path group based on the real-time drone status data, the real-time power data and the real-time flight data of each drone in the drone group.

2. A power line inspection drone control method according to claim 1, characterized in that: The real-time power data includes the number, location, inspection status and safety status of multiple inspection points; the real-time flight data includes obstacle location, safety area and flight restricted area; the real-time environmental data includes visibility, temperature, humidity, wind speed and wind direction; the real-time drone status data includes number, power, load data, location, safety status and task execution progress; the safety status includes normal status and abnormal status; the abnormal status includes the drone whose safety status value is lower than the safety threshold 1, the power is 0 and the position remains unchanged for a long time; the inspection status includes waiting for inspection and inspected.

3. A power line inspection drone control method according to claim 1, characterized in that: The multi-objective optimization dynamic path planning algorithm includes a potential field method and an RRT* algorithm; the calculation formula of the first power inspection path group is: Among them, {P opt } is the first power inspection path group, including the optimal path P for each drone i , L(P i ),C(P i ),T(P i ) are the path length, energy consumption and completion time of UAV i, n is the number of UAVs, argmin() is used to calculate the minimized path group, α1, α2, α3 are L(P i ),C(P i ),T(P i )’s weight.

4. A power line inspection drone control method according to claim 1, characterized in that: The calculation formula of the safety status value is: Among them, T1 is the safety state value, w k is the weight of the kth environmental parameter, E k is the real-time data of the kth environmental parameter, m is the number of environmental parameters, and f k is the nonlinear transformation function for the kth parameter, and σ is the sigmoid activation function; The environmental parameters include visibility, temperature, humidity, wind speed and wind direction; The calculation formula of the safety status value of the inspection point is: T2 = γ·T1 + δ·h(D); Among them, T2 is the safety status value of the inspection point, γ, δ are weight coefficients, and h(D) is the distance risk function calculated based on the distance D from the drone to the inspection point; The calculation formula of the distance D from the drone to the inspection point is: Among them, (X1, Y1, Z1) is the current position of the drone, and (X2, Y2, Z2) is the position of the next inspection point; The calculation formula of the distance risk function is: Among them, v is the shape parameter, and D0 is the distance between the UAV and the midpoint of the next inspection point.

5. The control method of a power line inspection drone according to claim 1 is characterized in that: The adjusted flight strategy is obtained by analyzing the real-time environmental data and the real-time UAV status data using an improved particle swarm optimization algorithm and an A* search algorithm.

6. A power line inspection drone control method according to claim 1, characterized in that: The calculation formula of the second power inspection path group is: Among them, {P new } is the second power inspection path group, including the second optimal path P of each drone j , L(P j ),C(P j ),T(P j ) are the path length, energy consumption and completion time of UAV j, e is the number of UAVs with normal safety status, argmin() is used to calculate the minimized path group, λ1,λ2,λ3 are L(P i ),C(P i ),T(P i )’s weight.

7. A power line inspection drone control system, characterized in that: include: A deployment unit, used to deploy a drone group; the drone group includes a plurality of drones; each of the drones is equipped with a high-definition camera, GPS and environmental sensors; A data acquisition unit, used to acquire real-time power data, real-time flight data, real-time environmental data, and real-time drone status data; A first power inspection path group acquisition unit, used to acquire a first power inspection path group; The process of acquiring the first power inspection path group includes: analyzing the real-time power data and the real-time flight data by using a graph theory algorithm to acquire a power line model; applying a multi-objective optimization dynamic path planning algorithm to analyze the power line model to acquire the first power inspection path group; The UAV flight status analysis unit is used to obtain a safety status value by analyzing the real-time environmental data and the real-time UAV status data; and to determine the next flight status of the UAV by analyzing the safety status value; the flight status includes stopping, adjusting the flight strategy and maintaining flight; and the flight strategy adjustment includes adjusting the flight angle, adjusting the flight altitude and adjusting the flight speed; A patrol point safety status analysis unit is used to obtain a patrol point safety status value of the next patrol point of the drone whose next flight state is stopped; by analyzing the patrol point safety status value, it is determined whether to reallocate the patrol point for inspection; An abnormal status updating unit is used to obtain a second power inspection path group; the process of obtaining the second power inspection path group includes: monitoring each of the drones in the drone group separately, and when an abnormal state is identified in a drone, using a real-time dynamic task allocation algorithm among multiple drones, and obtaining the second power inspection path group based on the real-time drone status data, the real-time power data and the real-time flight data of each drone in the drone group.

8. The power line inspection drone control system according to claim 7 is characterized in that: The real-time power data includes the number, location, inspection status and safety status of multiple inspection points; the real-time flight data includes obstacle location, safety area and flight restricted area; the real-time environmental data includes visibility, temperature, humidity, wind speed and wind direction; the real-time drone status data includes number, power, load data, location, safety status and task execution progress; the safety status includes normal status and abnormal status; the abnormal status includes the drone whose safety status value is lower than the safety threshold 1, the power is 0 and the position remains unchanged for a long time; the inspection status includes waiting for inspection and inspected; the safety status includes safe and dangerous, which is determined by the inspection point safety status value of the inspection point.

9. The power line inspection drone control system according to claim 7 is characterized in that: The multi-objective optimization dynamic path planning algorithm includes a potential field method and an RRT* algorithm; the calculation formula of the first power inspection path group is: Among them, {P opt } is the optimal path group, including the optimal path P of each drone i , L(P i ),C(P i ),T(P i ) are the path length, energy consumption and completion time of UAV i, n is the number of UAVs, argmin() is used to calculate the minimized path group, α1, α2, α3 are L(P i ),C(P i ),T(P i )’s weight.

10. The power line inspection drone control system according to claim 7, characterized in that: The calculation formula of the safety status value is: Among them, T1 is the safety state value, w k is the weight of the kth environmental parameter, E k is the real-time data of the kth environmental parameter, m is the number of environmental parameters, and f k is the nonlinear transformation function for the kth parameter, and σ is the sigmoid activation function; The environmental parameters include visibility, temperature, humidity, wind speed and wind direction.

Images