Multi-unmanned aerial vehicle electric power inspection task allocation method and system

Abstract

The invention discloses a multi-unmanned aerial vehicle electric power inspection task allocation method and system, and the method comprises the steps: building an optimization objective function according to the constraints of a total flight distance, an average flight height, and a terrain threat; by improving a random step size search strategy based on distance weight, the predation stage of the black wing plinualism algorithm is optimized, and the global search ability of the algorithm is enhanced; an optimal individual selection strategy based on fitness and distance is introduced, and the global search efficiency and optimization precision of the black-wing plinary algorithm in the migration stage are enhanced; a cooperative predation mechanism of the Arctic Sea-Sharrow algorithm is introduced, positions are updated through individual cooperation, and the ability of the algorithm to jump out of local optimum is effectively improved. The method has remarkable advantages in the aspects of task allocation efficiency and precision in complex mountain environment simulation, collision-free paths between generated task points are shorter, and an efficient solution is provided for unmanned aerial vehicle task allocation and path planning under complex terrains.

Description

Technical Field

[0001] This invention relates to the field of multi-UAV power line inspection task allocation technology, specifically to a multi-UAV power line inspection task allocation method and system based on the Hybrid Black-winged kite and Arctic puffin Optimizer (HBAO). Background Technology

[0002] In the field of power transmission line inspection, UAV inspection technology, with its advantages in aerial photography and image recognition, has become an important technical means, demonstrating high efficiency, low cost, and safety in power infrastructure inspection. To ensure the efficient execution and optimal utilization of resources for UAV inspection tasks, factors such as the complexity of the regional environment, the shortest obstacle avoidance path, and the number of task points need to be considered to achieve reasonable task allocation and effective scheduling. Numerous existing optimization algorithms exist. For example, Wang Shupeng et al. proposed a genetic algorithm that adaptively adjusts the crossover and mutation rates to overcome the shortcomings of standard genetic algorithms, which are prone to getting trapped in local optima; Wang Gui et al. used an improved genetic algorithm incorporating simulated annealing, combined with the Metropolis criterion to optimize the offspring population, avoiding local optima and achieving fast convergence; Dong et al. proposed a clustering-based consensus-based bundling strategy to achieve distance-based UAV grouping; Gao et al. designed a multi-layer coded genetic algorithm suitable for heterogeneous multi-UAV task allocation problems; Zhang et al. combined the gravity search algorithm with the genetic algorithm to construct a GSA-GA hybrid algorithm model; Peng Xuanhao et al. proposed a chaotic backward learning wolf pack optimization algorithm based on an auction mechanism under Levy's Fight optimization to enhance the algorithm's optimization ability. However, these optimization methods typically solve task allocation and path planning separately, failing to fully consider their coupling relationship. Furthermore, in complex scenarios, limitations in population size and computational resources may lead to insufficient search space coverage, affecting global optimization accuracy and algorithm convergence performance. In 2024, inspired by the migration and predation behavior of the black-winged kite, Wang et al. proposed the Black Kite Algorithm (BKA), integrating Cauchy mutation and leader strategies to enhance the algorithm's global search capability. However, the convergence speed and accuracy of the single algorithm still have room for improvement. Summary of the Invention

[0003] The purpose of this invention is to provide a method and system for allocating power line inspection tasks using multiple unmanned aerial vehicles (UAVs) to overcome the problems existing in the prior art. This invention introduces the cooperative predation mechanism of the Arctic Puffin Algorithm (APO) and combines the relative positional relationship between individuals with multi-directional random perturbation to achieve a dynamic balance between global search and local search, further improving the convergence of the optimization algorithm. It can also be applied to the allocation of power line inspection tasks in large-scale mountainous environments.

[0004] To achieve the above objectives, the present invention adopts the following technical solution: The method for allocating multiple drone power line inspection tasks includes the following steps: Step 1: Establish the optimization objective function for the UAV power line inspection task; Step 2: Obtain the Black-winged Kite search algorithm; Step 3: Based on the requirements of the UAV power line inspection mission, the black-winged kite search algorithm obtained in Step 2 is improved to obtain the black-winged kite-argus hybrid optimizer. Step 4: Use the Black-winged Kite-Arctic Puffin hybrid optimizer obtained in Step 3 to optimize the objective function of the multi-UAV power line inspection task, and obtain the multi-UAV power line inspection task allocation result.

[0005] Furthermore, the objective function for the UAV power line inspection task is established in step 1, as follows: The initial path points are generated using the following formula:

[0006] In the formula, x i , y i , z i This represents the coordinates of the currently generated path point in three-dimensional space; D This represents the Euclidean distance between the current task point and the next task point; a x , a y , a z These are the distance coefficients in three directions; R This represents the distance threshold between the generated path point and the next task point; rand(1) represents a random number between 0 and 1; Terrain threat constraints are defined as follows:

[0007] In the formula L H ( i , j ) is the waypoint ( i , j The corresponding terrain threat level; F H , F H` Terrain threat coefficient; h ( i , j () represents the flight altitude of the waypoint; H ( i ,j () represents the vertical elevation of the terrain corresponding to the track point; The constraints for the length and altitude of the flight path segment are defined as follows:

[0008]

[0009]

[0010] in, L length Indicates the constraint on the length of the flight path segment; L height Indicates the altitude constraint of the flight path segment; N Indicates the number of drones; P jk and d Pjk Indicates task point j arrive k The track segment and track segment length; N pjk Indicates track segment P jk The number of track points in the data; z l , z l+1 These represent the current waypoint and the next waypoint, respectively. z Axis coordinates; It is a binary decision variable used to represent the task point. j and k Is the path between them determined by the first... i The drone was used for the operation. By weighting and combining terrain threats and flight path constraints, an optimization objective function for the UAV power line inspection task is constructed, as shown in the following formula:

[0011] in, F The objective function for optimizing UAV power line inspection tasks; w 1, w 2, w 3 represents the weighting coefficients for each constraint.

[0012] Furthermore, the black-winged kite search algorithm in step 2 includes the following steps: Step 2.1: Randomly initialize the black-winged kite population; Step 2.2: Update the predation phase location based on the black-winged kite population obtained in Step 2.1; Step 2.3: Update the migration phase location based on the black-winged kite population location obtained in Step 2.2.

[0013] Furthermore, in step 2.2, the black-winged kite adjusts its wings and tail angle according to wind speed during flight, hovers in the air to observe prey, and then attacks swiftly. The mathematical model is as follows:

[0014] in, and They represent the first i Only the black-winged kite in the first t and t+ The first iteration in step 1 j Dimensional position; rand It is a random number with a value in the range [0, 1]. p It is a constant; t Indicates the current iteration number; T Indicates the total number of iterations; In step 2.3, if the fitness value of the current population is worse than that of the random population, the current leader relinquishes its leadership position and joins the migrating population, indicating that it is unsuitable to lead the population migration. Conversely, if the fitness value of the current population is better than that of the random population, the leader will continue to guide the population until it reaches the target location. The mathematical model is as follows:

[0015]

[0016] in, Indicates the number to date t In the next iteration, the Black-winged Kite leader was... j The position of the dimension; m Indicates the migration coefficient; F i Indicates the first i Only the black-winged kite in the first t Fitness value in the next iteration; F r Indicates the first t In each iteration, the fitness value of a black-winged kite individual at a random location; C (0,1) represents the Cauchy variation.

[0017] Furthermore, step 3 improves the black-winged kite search algorithm to obtain a black-winged kite-argus hybrid optimizer, specifically including the following steps: Step 3.1: Initialize the black-winged kite population and construct initial fitness values; Step 3.2: Optimize the predation phase of the Black-winged Kite algorithm by adopting a distance-weighted random step-size search strategy and updating the predation phase position of the initial population obtained in Step 3.1. Step 3.3: Introduce a selection strategy based on fitness and distance into the migration stage, update the population position obtained in Step 3.2 for the migration stage, and select the optimal leader; Step 3.4: Introduce the cooperative predation mechanism of the Arctic auk algorithm to update the population position obtained in Step 3.3; Step 3.5: Calculate the objective function value, and take the task sequence with the minimum fitness and the corresponding path points as the current optimal solution; Step 3.6: Determine if the maximum number of iterations has been reached. If yes, output the result; otherwise, return to step 3.2.

[0018] Furthermore, in step 3.1, the black-winged kite population is initialized using the following formula:

[0019] in, pop Indicates population size; dim For a given problem, the dimension; X ij Indicates the first i Only the black-winged kite in the first j The position of each black-winged kite is generated uniformly in the solution space, expressed as:

[0020] in, rand It is a random number within the range [0,1]. B ub and B lb These are the solutions to the problems. j The upper and lower boundaries of the dimension; During the initialization phase, the Black-winged Kite algorithm determines the leader position by selecting the individual with the best fitness value. X L It is considered the globally optimal solution for the population. X L The selection is based on minimizing the fitness value, mathematically expressed as:

[0021]

[0022] in, f Indicates the fitness value. f best This represents the optimal fitness value. find ( f best ==f ( X iThe function finds the individual Blackwing Kites corresponding to the optimal fitness value.

[0023] Furthermore, step 3.2 employs a distance-weighted random step-size search strategy to optimize the predation phase of the Black-winged Kite algorithm. The specific steps are as follows: By introducing the distance between the leader's position and the current individual's position, the step size of each iteration is dynamically adjusted, enabling individuals to quickly approach the optimal individual. Simultaneously, by introducing positive and negative random perturbation factors, individuals can explore in multiple directions. The mathematical model is as follows:

[0024]

[0025]

[0026] in, R For positive and negative random perturbation factors, sign For symbolic functions, L t Indicates the first t The Black-Winged Kite's leadership position in the next iteration. r It is a random number with a value in the range [0, 1]. p It is a constant.

[0027] Furthermore, step 3.3 introduces a selection strategy based on fitness and distance into the migration phase, and selects the optimal leader. The specific steps are as follows: Step 3.3.1: Obtain the individual position corresponding to the optimal fitness as the optimal solution. P best ; Step 3.3.2: Calculate candidate solutions P and the optimal solution P best Euclidean distance between Dis ; Step 3.3.3: Based on the results of Step 3.3.2, obtain the normalized vectors of the optimal fitness and Euclidean distance. Fit n and Dis n ; Step 3.3.4: Calculate the composite score vector FDB = 0.6* Fit n +0.4* Dis n ; Step 3.3.5: Based on the results of Step 3.3.4, select the individual with the smallest composite score vector as the optimal individual and obtain the optimal leader position index; Step 3.4 introduces the cooperative predation mechanism of the Arctic auk algorithm. The specific steps are as follows: The cooperative hunting behavior of the Arctic auk is incorporated into the black-winged kite algorithm to simulate the multi-directional cooperative hunting behavior of the Arctic auk. The cooperative hunting position equation of the Arctic auk is as follows:

[0028] in, F As a cooperative factor, it is used to adjust the predation behavior of Arctic puffins. L Represents Lévy's flight, variable r 1. r 2. r 3 is a random integer, ranging from 1 to... pop -1, while candidate solutions It is an individual randomly selected from the current population that meets the following conditions. and .

[0029] Furthermore, in step 4, the obtained Black-winged Kite-Arctic Puffin hybrid optimizer is used to optimize the allocation of multi-UAV power line inspection tasks. The specific method is as follows: Each path is a line connecting multiple trajectory points. A single black-winged kite in the population is defined as a path, and the black-winged kite with the lowest cost represents the optimal path. When optimizing the path allocation for multi-UAV power line inspection tasks using a black-winged kite-argus hybrid optimizer, the specific steps include: Step 4.1: Obtain the initial black-winged kite population using the method in Step 3.1; Step 4.2: Calculate the fitness of each black-winged kite in the initial black-winged kite population using the optimization objective function of the UAV power inspection task obtained in Step 1. Sort the black-winged kite population according to the fitness and select the best black-winged kite. Step 4.3: Update the position of the black-winged kite population using the methods in steps 3.2 to 3.3, update the position of the black-winged kite population using the cooperative predation mechanism of the Arctic auk algorithm in step 3.4, calculate the fitness of each black-winged kite, sort the black-winged kite population according to the fitness, and take the task sequence with the lowest fitness and the corresponding path point as the current optimal solution. Step 4.4: Determine if the maximum number of iterations has been reached. If yes, output the optimal black-winged kite, i.e., the optimal power inspection task allocation result. Otherwise, return to step 4.3.

[0030] A multi-UAV power line inspection task allocation system, including: Objective function establishment module: used to establish the optimization objective function for UAV power line inspection tasks; Algorithm Acquisition Module: Used to acquire the black-winged kite search algorithm; Algorithm Improvement Module: This module is used to improve the obtained Black-winged Kite search algorithm according to the requirements of UAV power line inspection tasks, resulting in a Black-winged Kite-Arctic Puffin hybrid optimizer. Solution module: The objective function of the multi-UAV power line inspection task is optimized using the Black-winged Kite-Arctic Puffin hybrid optimizer to obtain the task allocation results of the multi-UAV power line inspection task.

[0031] Compared with the prior art, the present invention has the following beneficial technical effects: This invention proposes a hybrid black-winged kite-argus optimizer for multi-UAV power line inspection tasks in large-scale complex mountainous environments (such as the Qinling Mountains). It jointly optimizes task allocation and path planning, significantly improving the applicability of the model in inspection task allocation scenarios.

[0032] This invention modifies the predation phase step size of BKA to a random step size based on distance weights, effectively preventing the algorithm from getting trapped in local optima. By applying the optimal individual selection strategy of FDB, the optimization efficiency and global search capability of the algorithm are further improved. To address the problem of insufficient optimization capability in the later stages of the BKA algorithm, the cooperative predation mechanism of the APO algorithm is introduced. By combining the relative positional relationship between individuals and multi-directional random perturbation, a dynamic balance between global search and local search is achieved, which significantly enhances the optimization performance of the algorithm in complex terrain.

[0033] Compared with other advanced methods, this invention has significant advantages in terms of task allocation efficiency and accuracy in the simulation of complex mountainous environments in the Qinling digital elevation model, and generates shorter collision-free paths between task points, providing an efficient solution for UAV task allocation and path planning in complex terrain. Attached Figure Description

[0034] The accompanying drawings are provided to further understand the invention and constitute a part of this invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0035] Figure 1 This is a flowchart illustrating the present invention; Figure 2 This is a flowchart of the FDB selection strategy algorithm involved in this invention; Figure 3 This is a flowchart of the HBAO algorithm involved in this invention; Figure 4 This is a comparison chart of the algorithm involved in this invention and other algorithms (P=20). Figure 5 This is a comparison chart of the algorithm involved in this invention and other algorithms (P=30). Figure 6This is a comparison chart of the solution results of the algorithm designed in this invention with various algorithms under different numbers of task points. Detailed Implementation

[0036] The present invention will now be described in further detail with reference to the accompanying drawings: Example 1 See Figure 1 The flowchart of this invention shows that the basic black-winged kite search algorithm has been optimized and improved, proposing a black-winged kite-argus hybrid optimizer. Addressing the insufficient optimization capability in the later stages of BKA (Black-winged Kite) migration, the aurora algorithm is introduced. First, a UAV power line inspection task model and optimization objective function are established. Second, the HBAO hybrid optimizer is designed, improving the distance-weighted stochastic step-size search strategy, optimizing the black-winged kite search algorithm, and introducing a fitness- and distance-based optimal individual selection (FDB) strategy. This enhances the global search efficiency and optimization accuracy of the black-winged kite algorithm during the migration phase. Figure 2 By introducing the cooperative predation mechanism of the Arctic auk algorithm, the diversity and efficiency of the global search are ensured, such as... Figure 3 Finally, a digital elevation model of the Qinling Mountains was established and compared with various algorithms. Figure 4 , Figure 5 and Figure 6 It can be seen that, when there are many inspection points, the solution obtained by the proposed multi-UAV inspection task allocation method for power transmission lines based on the hybrid optimization of black-winged kite and Arctic puffin has better overall performance than the four comparative algorithms, and the global cost is significantly reduced.

[0037] The specific steps are as follows: Step 1: Generate initial path points using the following formula:

[0038] In the formula, x i , y i , z i This represents the coordinates of the currently generated path point in three-dimensional space; D This represents the Euclidean distance between the current task point and the next task point; a x , a y , a z These are the distance coefficients in three directions; R This represents the distance threshold between the generated path point and the next task point; rand(1) represents a random number between 0 and 1.

[0039] Terrain threat constraints are defined as follows:

[0040] In the formula L H ( i , j ) is the waypoint ( i , j The corresponding terrain threat level; F H , F H` Terrain threat coefficient; h ( i , j () represents the flight altitude of the waypoint; H ( i , j ) represents the vertical elevation of the terrain corresponding to the track point.

[0041] The constraints for the length and altitude of the flight path segment are defined as follows:

[0042]

[0043]

[0044] in, L length Indicates the constraint on the length of the flight path segment; L height Indicates the altitude constraint of the flight path segment; N Indicates the number of drones; P jk and d Pjk Indicates task point j arrive k The track segment and track segment length; N pjk Indicates track segment P jk The number of track points in the data; z l , z l+1 These represent the current waypoint and the next waypoint, respectively. z Axis coordinates; It is a binary decision variable used to represent the task point. j and k Is the path between them determined by the first... i The operation was carried out using drones.

[0045] To ensure the global optimality and flight safety of the UAV trajectory, terrain threats and trajectory constraints are weighted and integrated to construct the optimization objective function for the UAV power line inspection task, as shown in the following formula:

[0046] in, F The objective function for optimizing UAV power line inspection tasks; w 1, w 2, w 3 represents the weighting coefficients for each constraint.

[0047] Step 2: The basic black-winged kite search algorithm can be divided into the following steps: The black-winged kite population is randomly initialized using the following formula:

[0048] in, pop Indicates population size; dim For a given problem, the dimension; X ij Indicates the first i Only the black-winged kite in the first j The position of each black-winged kite is generated uniformly in the solution space, expressed as:

[0049] in, rand It is a random number within the range [0,1]. B ub and B lb These are the solutions to the problems. j The upper and lower boundaries of a dimension.

[0050] During the initialization phase, the Black-winged Kite algorithm determines the leader position by selecting the individual with the best fitness value. X L This is considered the globally optimal solution for the population. X L The selection is based on minimizing the fitness value, mathematically expressed as:

[0051]

[0052] in, f Indicates the fitness value. f best This represents the optimal fitness value. find ( f best ==f ( X i The function finds the individual Blackwing Kites corresponding to the optimal fitness value.

[0053] The black-winged kite adjusts its wings and tail angle according to wind speed during flight, hovering in the air to observe its prey before swiftly striking. The mathematical model is as follows:

[0054]

[0055] in, and They represent the first i Only the black-winged kite in the first t and t+ The first iteration in step 1 j Dimensional position; rand It is a random number with a value in the range [0, 1]; a constant. p The value is 0.9; t Indicates the current iteration number; T Indicates the total number of iterations; If the current leader's fitness value is worse than that of a random population, the current leader will relinquish its leadership position and join the migrating population, indicating that it is unsuitable to lead the migration. Conversely, if the current leader's fitness value is better than that of a random population, the leader will continue to guide the population until it reaches the target location. The mathematical model is as follows:

[0056]

[0057] Indicates the number to date t In the next iteration, the Black-winged Kite leader was... j The position of the dimension; m Indicates the migration coefficient; F i Indicates the first i Only the black-winged kite in the first t Fitness value in the next iteration; F r Indicates the first t In each iteration, the fitness value of a black-winged kite individual at a random location; C (0,1) represents the Cauchy variation.

[0058] Step 3: A distance-weighted stochastic step-size search strategy was designed to improve the predation behavior of BKA. The mathematical model of its predation phase is as follows:

[0059]

[0060]

[0061] in,R For positive and negative random perturbation factors, sign For symbolic functions, L t Indicates the first t The Black-Winged Kite's leadership position in the next iteration. r It is a random number with a value in the range [0, 1], a constant. p The value is 0.9.

[0062] The migration behavior of BKA was improved by introducing a fitness- and distance-based selection strategy, FDB. By weighting the fitness values ​​and distance vectors of all individuals in the population, the representative individual that contributes most to population optimization is selected for position updating in each iteration.

[0063] The Freedom-Based Search (FDB) strategy was introduced into the BKA migration behavior to replace random selection of individuals, thereby improving the effectiveness of location updates by selecting more representative individuals. Applying the FDB strategy helps the algorithm escape local optima traps and enhances its global search capabilities.

[0064] To address the shortcomings of the global search capability in the early stages of the BKA algorithm, an improved hybrid strategy is proposed, incorporating the cooperative predation behavior of the Arctic auk into the BKA algorithm. The cooperative predation position equation of the Arctic auk is as follows:

[0065] in, F As a cooperative factor, it is used to adjust the predation behavior of Arctic puffins. L This indicates that Levi is flying. Variable r 1. r 2. r 3 is a random integer, ranging from 1 to... pop -1, while candidate solutions It is an individual randomly selected from the current population that meets the following conditions. and Through the aforementioned improved strategies, the algorithm can not only quickly find the optimal solution, but also maintain diversity throughout the solution space, thus demonstrating superior performance in handling the task allocation problem of multiple UAV inspections of power transmission lines.

[0066] Step 4: Optimize the allocation of multi-UAV power line inspection tasks using the Black-winged Kite-Arctic Puffin hybrid optimizer. The specific method is as follows: Each inspection path consists of several trajectory points, and the path can be viewed as an ordered connection line of these trajectory points. Each individual black-winged kite in the population represents a complete path, and the individual with the minimum cost is the current optimal path. When optimizing path allocation for multi-UAV power line inspection tasks, a black-winged kite-argus hybrid optimization algorithm is adopted. The specific optimization process includes the following steps: Step 4.1: Obtain the initial black-winged kite population using the method in Step 2; Step 4.2: Calculate the fitness of each black-winged kite in the initial black-winged kite population using the optimization objective function of the UAV power inspection task obtained in Step 1. Sort the black-winged kite population according to the fitness and select the best black-winged kite. Step 4.3: Update the position of the black-winged kite population using the method in Step 3, calculate the fitness of each black-winged kite, sort the black-winged kite population according to the fitness, and take the task sequence with the smallest fitness and the corresponding path point as the current optimal solution. Step 4.4: Determine if the maximum number of iterations has been reached. If yes, output the optimal black-winged kite, i.e., the optimal power inspection task allocation result. Otherwise, return to step 4.3.

[0067] Figure 4 The diagram shows the path planning performance of different algorithms when the number of task points is 20. Figure 5 The diagram shows the path planning results of different algorithms when the number of task points is 30. Figure 6 The graph shows the solution results of each algorithm for different numbers of task points.

[0068] Figure 4 (a) shows the path planning performance of the BKA algorithm when the number of task points is 20. Figure 4 (b) shows the path planning performance of the SSA algorithm when the number of task points is 20. Figure 4 (c) shows the path planning performance of the APO algorithm when the number of task points is 20. Figure 4 (d) shows the path planning performance of the HHO algorithm when the number of task points is 20. Figure 4 (e) shows the path planning effect of the HBAO algorithm when the number of task points is 20; Figure 5 (a) shows the path planning performance of the BKA algorithm when the number of task points is 30. Figure 5 (b) shows the path planning performance of the SSA algorithm when the number of task points is 30. Figure 5 (c) shows the path planning performance of the APO algorithm when the number of task points is 30. Figure 5 (d) shows the path planning performance of the HHO algorithm when the number of task points is 30. Figure 5 (e) shows the path planning effect of the HBAO algorithm when the number of task points is 30; Figure 6(a) is a graph showing the results of each algorithm when the number of task points is 20. Figure 6 (b) shows the results of each algorithm when the number of task points is 30. It can be seen that the paths generated by BKA and HHO are generally smooth, but their task allocation results are poor, with obvious path reversals. In contrast, the paths generated by APO and SSA have many jagged points, significantly increasing flight costs and resulting in poor performance. In HBAO, the paths between task points are not only smooth, but the task allocation results are also reasonable, demonstrating superior convergence and coverage. Furthermore, HBAO has a stronger ability to explore towards the optimal location, and the total path cost of the task allocation results is lower.

[0069] Example 2 This invention provides a multi-UAV power line inspection task allocation system, comprising: Objective function establishment module: used to establish the optimization objective function for UAV power line inspection tasks; Algorithm Acquisition Module: Used to acquire the black-winged kite search algorithm; Algorithm Improvement Module: This module is used to improve the obtained Black-winged Kite search algorithm according to the requirements of UAV power line inspection tasks, resulting in a Black-winged Kite-Arctic Puffin hybrid optimizer. Solution module: The objective function of the multi-UAV power line inspection task is optimized using the Black-winged Kite-Arctic Puffin hybrid optimizer to obtain the task allocation results of the multi-UAV power line inspection task.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit its scope of protection. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that after reading the present invention, they can still make various changes, modifications or equivalent substitutions to the specific implementation of the invention, but these changes, modifications or equivalent substitutions are all within the scope of protection of the pending claims of the invention.

Claims

1. A method for allocating multiple unmanned aerial vehicle (UAV) power line inspection tasks, characterized in that: Includes the following steps: Step 1: Establish the optimization objective function for the UAV power line inspection task; Step 2: Obtain the Black-winged Kite search algorithm; Step 3: Based on the requirements of the UAV power line inspection mission, the black-winged kite search algorithm obtained in Step 2 is improved to obtain the black-winged kite-argus hybrid optimizer. Step 4: Use the Black-winged Kite-Arctic Puffin hybrid optimizer obtained in Step 3 to optimize the objective function of the multi-UAV power line inspection task, and obtain the multi-UAV power line inspection task allocation result.

2. The multi-UAV power line inspection task allocation method according to claim 1, characterized in that, Step 1 establishes the optimization objective function for the UAV power line inspection task, as follows: The initial path points are generated using the following formula: In the formula, x i , y i , z i This represents the coordinates of the currently generated path point in three-dimensional space; D This represents the Euclidean distance between the current task point and the next task point; a x , a y , a z These are the distance coefficients in the three directions; R This represents the distance threshold between the generated path point and the next task point; rand(1) represents a random number between 0 and 1; Terrain threat constraints are defined as follows: In the formula L H ( i , j ) is the waypoint ( i , j The corresponding terrain threat level; F H , F H` Terrain threat coefficient; h ( i , j () represents the flight altitude of the waypoint; H ( i , j () represents the vertical elevation of the terrain corresponding to the track point; The constraints for the length and altitude of the flight path segment are defined as follows: in, L length Indicates the constraint on the length of the flight path segment; L height Indicates the altitude constraint of the flight path segment; N Indicates the number of drones; P jk and d Pjk Indicates task point j arrive k The track segment and track segment length; N pjk Indicates track segment P jk The number of track points in the data; z l , z l+1 These represent the current waypoint and the next waypoint, respectively. z Axis coordinates; It is a binary decision variable used to represent the task point. j and k Is the path between them determined by the first... i The drone was used for the operation. By weighting and combining terrain threats and flight path constraints, an optimization objective function for the UAV power line inspection task is constructed, as shown in the following formula: in, F The objective function for optimizing UAV power line inspection tasks; w 1, w 2, w 3 represents the weighting coefficients for each constraint.

3. The multi-UAV power line inspection task allocation method according to claim 1, characterized in that, Step 2 of the black-winged kite search algorithm includes the following steps: Step 2.1: Randomly initialize the black-winged kite population; Step 2.2: Update the predation phase location based on the black-winged kite population obtained in Step 2.1; Step 2.3: Update the migration phase location based on the black-winged kite population location obtained in Step 2.

2.

4. The multi-UAV power line inspection task allocation method according to claim 3, characterized in that, In step 2.2, the black-winged kite adjusts its wings and tail angle according to wind speed during flight, hovers in the air to observe its prey, and then attacks swiftly. The mathematical model is as follows: in, and They represent the first i Only the black-winged kite in the first t and t+ The first iteration in step 1 j Dimensional position; rand It is a random number with a value in the range [0, 1]. p It is a constant; t Indicates the current iteration number; T Indicates the total number of iterations; In step 2.3, if the fitness value of the current population is worse than that of the random population, the current leader relinquishes its leadership position and joins the migrating population, indicating that it is unsuitable to lead the population migration. Conversely, if the fitness value of the current population is better than that of the random population, the leader will continue to guide the population until it reaches the target location. The mathematical model is as follows: in, Indicates the number to date t In the next iteration, the Black-winged Kite leader was... j The position of the dimension; m Indicates the migration coefficient; F i Indicates the first i Only the black-winged kite in the first t Fitness value in the next iteration; F r Indicates the first t In each iteration, the fitness value of a black-winged kite individual at a random location; C (0,1) represents the Cauchy variation.

5. The multi-UAV power line inspection task allocation method according to claim 3, characterized in that, Step 3 improves the black-winged kite search algorithm to obtain a black-winged kite-argus hybrid optimizer, which specifically includes the following steps: Step 3.1: Initialize the black-winged kite population and construct initial fitness values; Step 3.2: Optimize the predation phase of the Black-winged Kite algorithm by adopting a distance-weighted random step-size search strategy and updating the predation phase position of the initial population obtained in Step 3.

1. Step 3.3: Introduce a selection strategy based on fitness and distance into the migration stage, update the population position obtained in Step 3.2 for the migration stage, and select the optimal leader; Step 3.4: Introduce the cooperative predation mechanism of the Arctic auk algorithm to update the population position obtained in Step 3.3; Step 3.5: Calculate the objective function value, and take the task sequence with the minimum fitness and the corresponding path points as the current optimal solution; Step 3.6: Determine if the maximum number of iterations has been reached. If yes, output the result; otherwise, return to step 3.

2.

6. The multi-UAV power line inspection task allocation method according to claim 5, characterized in that, Step 3.1 initializes the black-winged kite population using the following formula: in, pop Indicates population size; dim For a given problem, the dimension; X ij Indicates the first i Only the black-winged kite in the first j The position of each black-winged kite is generated uniformly in the solution space, expressed as: in, rand It is a random number within the range [0,1]. B ub and B lb These are the solutions to the problems. j The upper and lower boundaries of the dimension; During the initialization phase, the Black-winged Kite algorithm determines the leader position by selecting the individual with the best fitness value. X L It is considered the globally optimal solution for the population. X L The selection is based on minimizing the fitness value, mathematically expressed as: in, f Indicates the fitness value. f best This represents the optimal fitness value. find ( f best ==f ( X i The function finds the individual Blackwing Kites corresponding to the optimal fitness value.

7. The multi-UAV power line inspection task allocation method according to claim 5, characterized in that, Step 3.2 employs a distance-weighted random step-size search strategy to optimize the predation phase of the Black-winged Kite algorithm. The specific steps are as follows: By introducing the distance between the leader's position and the current individual's position, the step size of each iteration is dynamically adjusted, enabling individuals to quickly approach the optimal individual. Simultaneously, by introducing positive and negative random perturbation factors, individuals can explore in multiple directions. The mathematical model is as follows: in, R For positive and negative random perturbation factors, sign For symbolic functions, L t Indicates the first t The Black-Winged Kite's leadership position in the next iteration. r It is a random number with a value in the range [0, 1]. p It is a constant.

8. The multi-UAV power line inspection task allocation method according to claim 5, characterized in that, Step 3.3 introduces a selection strategy based on fitness and distance into the migration phase, and selects the optimal leader. The specific steps are as follows: Step 3.3.1: Obtain the individual position corresponding to the optimal fitness as the optimal solution. P best ; Step 3.3.2: Calculate candidate solutions P and the optimal solution P best Euclidean distance between Dis ; Step 3.3.3: Based on the results of Step 3.3.2, obtain the normalized vectors of the optimal fitness and Euclidean distance. Fit n and Dis n ; Step 3.3.4: Calculate the composite score vector FDB = 0.6* Fit n +0.4* Dis n ; Step 3.3.5: Based on the results of Step 3.3.4, select the individual with the smallest composite score vector as the optimal individual and obtain the optimal leader position index; Step 3.4 introduces the cooperative predation mechanism of the Arctic auk algorithm. The specific steps are as follows: The cooperative hunting behavior of the Arctic auk is incorporated into the black-winged kite algorithm to simulate the multi-directional cooperative hunting behavior of the Arctic auk. The cooperative hunting position equation of the Arctic auk is as follows: in, F As a cooperative factor, it is used to adjust the predation behavior of Arctic puffins. L Represents Lévy's flight, variable r 1. r 2. r 3 is a random integer, ranging from 1 to... pop -1, while candidate solutions It is an individual randomly selected from the current population that meets the following conditions. and .

9. The multi-UAV power line inspection task allocation method according to claim 5, characterized in that, In step 4, the obtained Black-winged Kite-Arctic Puffin hybrid optimizer is used to optimize the allocation of multi-UAV power line inspection tasks. The specific method is as follows: Each path is a line connecting multiple trajectory points. A single black-winged kite in the population is defined as a path, and the black-winged kite with the lowest cost represents the optimal path. When optimizing the path allocation for multi-UAV power line inspection tasks using a black-winged kite-argus hybrid optimizer, the specific steps include: Step 4.1: Obtain the initial black-winged kite population using the method in Step 3.1; Step 4.2: Calculate the fitness of each black-winged kite in the initial black-winged kite population using the optimization objective function of the UAV power inspection task obtained in Step 1. Sort the black-winged kite population according to the fitness and select the best black-winged kite. Step 4.3: Update the position of the black-winged kite population using the methods in steps 3.2 to 3.3, update the position of the black-winged kite population using the cooperative predation mechanism of the Arctic auk algorithm in step 3.4, calculate the fitness of each black-winged kite, sort the black-winged kite population according to the fitness, and take the task sequence with the lowest fitness and the corresponding path point as the current optimal solution. Step 4.4: Determine if the maximum number of iterations has been reached. If yes, output the optimal black-winged kite, i.e., the optimal power inspection task allocation result. Otherwise, return to step 4.

3.

10. A multi-UAV power line inspection task allocation system, characterized in that, include: Objective function establishment module: used to establish the optimization objective function for UAV power line inspection tasks; Algorithm Acquisition Module: Used to acquire the black-winged kite search algorithm; Algorithm Improvement Module: This module is used to improve the obtained Black-winged Kite search algorithm according to the requirements of UAV power line inspection tasks, resulting in a Black-winged Kite-Arctic Puffin hybrid optimizer. Solution module: The objective function of the multi-UAV power line inspection task is optimized using the Black-winged Kite-Arctic Puffin hybrid optimizer to obtain the task allocation results of the multi-UAV power line inspection task.

Images