Multi-unmanned aerial vehicle cooperative power grid detection method and device

Abstract

The invention discloses a multi-unmanned-aerial-vehicle cooperative power grid detection method and device, and the method comprises the steps: building a multi-unmanned-aerial-vehicle cooperative task planning model based on the dynamic cost and dynamic income required by unmanned aerial vehicle detection considering environmental factors; based on the multi-unmanned aerial vehicle cooperative task planning model and the multi-unmanned aerial vehicle cooperative decision-making mechanism, nonlinear multi-target problem iterative optimization is carried out, and optimal task distribution of multi-unmanned aerial vehicle cooperative power grid detection is determined; and based on the optimal task allocation, coordinating the multiple unmanned aerial vehicles to perform power grid detection. The unmanned aerial vehicle can be endowed with more intelligence, and the task allocation planning resolving efficiency of multi-unmanned aerial vehicle cooperative power grid detection is improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of power grid detection, and in particular relates to a multi-UAV collaborative power grid detection method and device. Background Art

[0002] In the maintenance of power systems, multi-UAV collaborative task grid inspection planning plays a vital role. This technology needs to comprehensively consider the terrain, communication signal strength and environmental factors of the UAV, and formulate the best allocation and task planning scheme for multiple UAVs to perform complex multi-grid inspection tasks. Especially in complex communication and large-scale power grid inspection environments, multi-UAV collaborative grid inspection planning technology has become a key technology to improve grid inspection efficiency and reduce maintenance costs. Since this problem is essentially a multi-constraint optimization combination problem, how to quickly and accurately carry out collaborative task planning in the face of the instability of communication signals and a large number of grid objects to be inspected has become a major challenge in the current power inspection field.

[0003] Although there are a small number of existing task allocation schemes for multiple UAVs, they still have the following defects: 1. The convergence speed and optimization ability of conventional optimization algorithms cannot be met; 2. The conventional optimization algorithm is directly applied, and the algorithm parameters are not adjusted during the optimization process or the adjustment accuracy of the algorithm parameters is not enough, resulting in relatively low task allocation accuracy in the corresponding scenarios; 3. Only the flight speed of the UAV and the task type are considered to determine the model constraints, while the communication quality, detection conditions, etc. are not considered, which cannot be well applied to the field of power grid detection. For example, invention patent application CN117170405A discloses a method for UAV task allocation based on multi-objective particle swarms. The method first constructs a heterogeneous UAV collaborative multi-task allocation model, and then adopts a constraint-based initialization strategy to ensure that the particles generated by the initial particle swarm are feasible solutions that meet the constraints; then, a two-stage optimal particle selection strategy is adopted to retain the searched individual optimal particles and population optimal particles. Next, a task-based small module particle update strategy is adopted to enable particles to quickly learn from the optimal particles, and at the same time, task-based mutation operations are performed with a certain probability, which can improve the overall performance of the algorithm; finally, the updated particles are corrected to ensure that they are feasible solutions that meet the constraints. Summary of the invention

[0004] In view of the defects existing in the above-mentioned prior art, the present invention provides a method and device for multi-UAV collaborative power grid detection, which can give UAVs more intelligence and improve the efficiency of task allocation planning and solution of multi-UAV collaborative power grid detection.

[0005] In a first aspect, the present invention provides a multi-UAV collaborative power grid detection method, comprising: Based on the dynamic cost and dynamic benefit required for drone detection considering environmental factors, a multi-drone collaborative task planning model is established; Based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, the nonlinear multi-objective problem is iteratively optimized to determine the optimal task allocation for multi-UAV collaborative power grid detection. Coordinate multiple UAVs for power grid inspection based on optimal task allocation.

[0006] Furthermore, based on the dynamic cost and dynamic benefit required for drone detection considering environmental factors, a multi-drone collaborative task planning model is established, including: Based on the detection cost and flight cost generated by the drone performing the detection task, determine the dynamic cost corresponding to the drone detection; Determine the dynamic benefits of the drone based on the drone’s detection accuracy and the detection value of the power grid target; According to the dynamic cost and dynamic benefit of UAVs, the objective function of multi-UAV collaborative mission planning is given; The constraints of UAVs taking environmental factors into account and the dynamic adjustment mechanism of the objective function are determined, and a multi-UAV collaborative mission planning model is established in combination with the objective function of multi-UAV collaborative mission planning.

[0007] Furthermore, the dynamic cost corresponding to drone detection is determined, including: Obtain the distance between the UAV and the power grid target and the minimum distance that meets the detection quality, and give the interference distance; Based on the interference of environmental factors on drone detection and the detection power of drones, combined with the interference distance, the detection cost of drones is determined; The real-time wind speed and terrain undulation parameters at the power grid target are obtained and normalized to give the terrain complexity coefficient at the power grid target; Obtain the maximum Euclidean distance of all UAVs relative to the power grid target, and combine the terrain complexity coefficient at the power grid target and the distance between the UAV and the power grid target to give the UAV's range cost; The detection cost and flight cost of the drone are superimposed to give the dynamic cost corresponding to the drone detection.

[0008] Furthermore, the dynamic cost corresponding to drone detection satisfies the following relationship:

[0009] In the formula, represents the dynamic cost of UAV i detecting power grid target j at time t; The detection cost generated by the drone performing the detection task satisfies the following relationship:

[0010] In the formula, represents the detection cost of UAV i detecting power grid target j at time t, α is the detection weight factor, is the detection power of drone i, d ij ( t ) represents the distance from the detection grid target j at time t, d min To meet the minimum distance for detection quality, is the interference coefficient of environmental factors on UAV i’s detection of power grid target j at time t. Environmental factors include weather attenuation, terrain shielding, and obstacle density. is the preset interference threshold, x ij Indicates whether the power grid target j is detected by UAV i. When the power grid target j is detected by UAV i, x ij =1, otherwise x ij =0; The range cost generated by the drone performing the detection task satisfies the following relationship: ; In the formula, represents the range cost of UAV i detecting power grid target j, β is the cost weight factor, t j ( t ) represents the terrain complexity coefficient of the location of the power grid target j, c 1, c 2 is the normalization coefficient, W j ( t ) is the real-time wind speed at the location of the power grid target j, T j is the terrain undulation parameter where the power grid target j is located, d max represents the maximum Euclidean distance of all UAVs relative to the grid target j.

[0011] Furthermore, the dynamic benefits of drones include: Obtain the current heading angle of the UAV and the optimal approach angle taking into account environmental factors, and give the deviation coefficient of the UAV; The detection accuracy of the UAV and the inspection value of the power grid target are obtained, and combined with the deviation coefficient of the UAV, the dynamic benefits of the UAV detection of power grid targets are determined.

[0012] Furthermore, the dynamic benefits of the drone satisfy the following relationship:

[0013] In the formula, R ij ( t ) represents the dynamic benefit of UAV i detecting power grid target j at time t, or 0 is the equity weight factor, l i Indicates the detection accuracy of the sensor carried by drone i, F j is the inspection value of power grid target j, i i ( t ) is the current heading angle of UAV i, is the optimal approach angle under environmental constraints.

[0014] Furthermore, the multi-UAV collaborative mission planning objective function satisfies the following relationship:

[0015] In the formula, F ( t ) is the objective function for multi-UAV collaborative mission planning, Ω( t ) represents the dynamic weight vector, which is updated in real time through the Kalman filter. , ω1, ω2, ω3 are all dynamic weights, ω1+ω2+ω3=1, [] T is the transpose operation, Δ E ( t ) is the change of environmental parameters, K,Q is the adjustment matrix in the Kalman algorithm.

[0016] Furthermore, the UAV takes into account the constraints of environmental factors and satisfies the following relationship:

[0017] In the formula, Indicates the current maximum flight speed of drone i. is the default maximum flight speed of the drone. k is the wind speed influencing factor, W max is the maximum ambient wind speed, d safe ( t ) is the maximum safe flight distance, μ is the safety margin adjustment factor, ρ(t) is the real-time obstacle density, and T is the power grid target set of the detection task; Furthermore, the objective function dynamically adjusts the mechanism to satisfy the following relationship:

[0018] Where δ(t) is the rate of change of environmental parameters, δ thresholdis the threshold of the environmental parameter change rate, E(t) is the environmental parameter at time t, Δt is the time change, ω'1 and ω'2 are the dynamic weights of ω1 and ω2 update adjustment, ΔΦ is the communication interference change, ΔV is the power grid value fluctuation amplitude, V max is the maximum grid value, sigmoid() is the activation function, sigmoid(ΔΦ)=1 / (1+e^-ΔΦ).

[0019] Furthermore, based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, nonlinear multi-objective problem iterative optimization is performed to determine the optimal task allocation for multi-UAV collaborative power grid detection, including: S21. Establish a mapping between the positions of individual wild dogs in the dingo algorithm and the target allocation scheme of the power grid corresponding to multiple drones, determine the evolutionary strategies and coefficient decision factors of different hunting methods of the dingo algorithm, and initialize the positions of wild dogs in the dingo algorithm; S22, selecting the corresponding coefficient decision factor to update the evolution strategy based on the multi-UAV collaborative decision-making mechanism at the current wild dog position, generating a new wild dog position according to the updated evolution strategy, and calculating the corresponding power grid detection evaluation index; S23. Repeat iterative step S22 until the preset conditions are met, output the wild dog position with the best power grid detection evaluation index, and obtain the optimal task allocation for multi-UAV collaborative power grid detection.

[0020] Furthermore, the corresponding coefficient decision factor updating evolution strategy is selected based on the multi-UAV collaborative decision-making mechanism at the current wild dog location, including: Based on the multi-UAV collaborative task planning model and environmental factors, the state space of the reinforcement learning algorithm is determined, and the action space of the reinforcement learning algorithm is determined based on the coefficient decision factor; Determine the reward function of reinforcement learning based on the multi-UAV collaborative mission planning model; Based on the current position of the wild dog, combined with the state space, action space and reward function, determine the probability of selecting an action in the action space under the current state space, and based on the maximum probability, determine the coefficient decision factor corresponding to the selected action; The evolution strategy is updated according to the coefficient decision factor corresponding to the selected action.

[0021] Furthermore, the hunting methods of the dingo algorithm include group hunting, individual hunting, and scavenging; The evolutionary strategy of collective hunting satisfies the following relationship:

[0022] The evolutionary strategy of single predation satisfies the following relationship:

[0023] The evolutionary strategy of scavenging satisfies the following relationship:

[0024] In the formula, are the new positions of wild dogs when participating in collective hunting, individual predation and scavenging, respectively. α(t), β(t) and γ(t) are the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation and the scavenging trigger threshold of scavenging, respectively. k∈{1,2,...,P}, P is the number of Australian wild dogs, X k represents the initial position of the selected wild dog, is the time-varying coupling coefficient, , a3 is the time-varying coupling weight, T max is the maximum duration of power grid detection, X best is the best individual among all wild dogs, that is, it has the current optimal allocation plan, F1 represents the adaptive hunting scaling factor, X m , X n are two random wild dogs participating in collective hunting, λ(t) is the dynamic learning rate, , are two random individuals among all dingo individuals, is a random individual among all dingo individuals, a1 and a2 are the weight of individual predators and the threshold for updating scavenging positions, respectively, and rand is a random number between (0,1).

[0025] Furthermore, the coefficient decision factors include the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation, and the scavenging trigger threshold of scavenging, satisfying the following relationship:

[0026] In the formula, Δ f is the fitness variance of the wild dog population, σ is the preset variance, β min and β max are the minimum exploration degree and the maximum exploration degree, respectively, and γ base is the basic scavenging trigger value, and ε is the greed coefficient.

[0027] Further, based on the current wild dog position, the probability of selecting an action in the action space under the current state space is determined by combining the state space, the action space and the reward function, and based on the maximum probability, the coefficient decision factor corresponding to the selected action is determined, including; Based on the current position of the wild dog, the reinforcement learning algorithm is used to evaluate the expected long-term benefits of each action in the action space under all states in the current state space before the update, and the expected long-term benefits corresponding to the next moment are determined; Select the maximum value from the expected long-term benefits corresponding to the next moment, and update the expected long-term benefits of each action in the action space under all states in the current state space in combination with the reward function; Determine the probability of selecting each action in the action space under all states in the current state space based on the expected long-term benefits of selecting each action in the action space under all states in the updated current state space; Screen and determine the action selected when the probability is the highest, and based on the relationship between the coefficient decision factor and the action space, determine the coefficient decision factor of the hunting method corresponding to the action selected when the probability is the highest.

[0028] Furthermore, the probability of selecting each action in the action space under all states in the current state space satisfies the following relationship:

[0029] In the formula, π(a|s) is the current state space S t Select action space A in state s t The probability of action a in , Q'(s,a) is the updated current state space S t Select action space A in state s t The expected long-term benefit of action a in the above example is Q(s,a), and Q(s,a) is the current state space S before the update. t Select action space A in state s t The expected long-term benefit of action a, T is the total number of grid targets, l r is the learning rate, r(t) is the reward function, η is the discount factor, s', a' are the state at the next moment reached by executing action a in the current state s and the action selected by the state at the next moment, Δ F ( t ) is the difference between the objective function of multi-UAV collaborative mission planning at time t and time t-1, Ω opt is a small perturbation value of the dynamic weight vector, a small value close to 0, collision_rate is the collision rate, a4 and a5 are the benefit weights of minimizing the target weight and minimizing the dynamic weight vector, respectively. F ( t ) is the objective function for multi-UAV collaborative mission planning, W j ( t ) is the real-time wind speed at the location of the power grid target j, is the interference coefficient of environmental factors on UAV i’s detection of power grid target j at time t, ρ(t) is the real-time obstacle density, α(t), β(t), and γ(t) are the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation, and the scavenging trigger threshold of scavenging, respectively. Δα, Δβ, and Δγ are the discrete adjustment values ​​of the hunting intensity coefficient, the exploration rate parameter, and the scavenging trigger threshold, respectively.

[0030] In a second aspect, the present invention further provides a multi-UAV collaborative power grid detection device, which adopts the multi-UAV collaborative power grid detection method, and the device includes: A model building module, which is used to establish a multi-UAV collaborative mission planning model based on the dynamic cost and dynamic benefit required for UAV detection considering environmental factors; The task allocation module is used to iteratively optimize the nonlinear multi-objective problem based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, and determine the optimal task allocation for multi-UAV collaborative power grid detection; The power grid detection module is used to coordinate multiple UAVs to perform power grid detection based on optimal task allocation.

[0031] Furthermore, the model building module includes: Based on the detection cost and flight cost generated by the drone performing the detection task, determine the dynamic cost corresponding to the drone detection; Determine the dynamic benefits of the drone based on the drone’s detection accuracy and the detection value of the power grid target; According to the dynamic cost and dynamic benefit of UAVs, the objective function of multi-UAV collaborative mission planning is given; The constraints of UAVs taking environmental factors into account and the dynamic adjustment mechanism of the objective function are determined, and a multi-UAV collaborative mission planning model is established in combination with the objective function of multi-UAV collaborative mission planning.

[0032] Furthermore, the task allocation module includes: S21. Establish a mapping between the positions of individual wild dogs in the dingo algorithm and the target allocation scheme of the power grid corresponding to multiple drones, determine the evolutionary strategies and coefficient decision factors of different hunting methods of the dingo algorithm, and initialize the positions of wild dogs in the dingo algorithm; S22, selecting the corresponding coefficient decision factor to update the evolution strategy based on the multi-UAV collaborative decision-making mechanism at the current wild dog position, generating a new wild dog position according to the updated evolution strategy, and calculating the corresponding power grid detection evaluation index; S23. Repeat iterative step S22 until the preset conditions are met, output the wild dog position with the best power grid detection evaluation index, and obtain the optimal task allocation for multi-UAV collaborative power grid detection.

[0033] Furthermore, the task allocation module includes: Based on the multi-UAV collaborative task planning model and environmental factors, the state space of the reinforcement learning algorithm is determined, and the action space of the reinforcement learning algorithm is determined based on the coefficient decision factor; Determine the reward function of reinforcement learning based on the multi-UAV collaborative mission planning model; Based on the current position of the wild dog, combined with the state space, action space and reward function, determine the probability of selecting an action in the action space under the current state space, and based on the maximum probability, determine the coefficient decision factor corresponding to the selected action; The evolution strategy is updated according to the coefficient decision factor corresponding to the selected action.

[0034] The present invention provides a multi-UAV collaborative power grid detection method and device, which at least have the following beneficial effects: (1) The present invention combines the diverse predation mechanisms of the dingo algorithm to establish an allocation scheme evolution strategy that includes collective hunting, individual predation, scavenging and other evolutionary strategies to replace the single evolutionary mode of the traditional swarm intelligence algorithm, giving the population individuals more autonomy and avoiding the algorithm from falling into local optimality.

[0035] (2) The present invention models the power grid detection characteristics, designs a dynamic adjustment mechanism for the objective function and power grid detection evaluation indicators, thereby ensuring that the population individuals are optimized toward the correct allocation plan and improving the algorithm optimization efficiency.

[0036] (3) The present invention can quickly and effectively solve the problem of multi-UAV collaborative power grid detection task allocation based on the Dingo algorithm, providing a new solution to the problem of multi-UAV collaborative power grid detection task allocation in complex environments. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 A flowchart of a multi-UAV collaborative power grid detection method provided by the present invention; Figure 2 A schematic diagram of the principle of a multi-UAV collaborative power grid detection method provided by an embodiment of the present invention; Figure 3 A flowchart for obtaining optimal task allocation provided by an embodiment of the present invention; Figure 4 A flowchart of an update evolution strategy provided for an embodiment of the present invention; Figure 5 A schematic diagram of a multi-UAV collaborative power grid detection device provided by the present invention. DETAILED DESCRIPTION

[0038] In order to better understand the above technical solution, the above technical solution will be described in detail below in conjunction with the accompanying drawings and specific implementation methods. 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.

[0039] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0040] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0041] The present invention, based on the construction of a multi-UAV collaborative task planning model, combines the dingo algorithm to optimize and solve the optimal task allocation plan for multiple UAVs, and finally realizes the collaborative power grid detection planning of multiple UAVs taking into account environmental factors. Although the existing dingo algorithm is applied to the path planning of power inspection UAVs, it is only a simple application, using the dingo algorithm for single-objective path optimization, and does not have the detailed cost and objective function dynamic adjustment mechanism and power grid detection evaluation index of the dingo optimization algorithm for power grid detection task allocation in the present invention, and finally realizes the multi-objective collaborative power grid detection planning by taking into account the objective function of environmental factors, and the corresponding evolutionary strategy and coefficient decision factor of the dingo algorithm, which is better suitable for the application scenario of multi-UAV power grid detection.

[0042] like Figure 1 As shown, the present invention provides a multi-UAV collaborative power grid detection method, comprising: Based on the dynamic cost and dynamic benefit required for drone detection considering environmental factors, a multi-drone collaborative task planning model is established; Based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, the nonlinear multi-objective problem is iteratively optimized to determine the optimal task allocation for multi-UAV collaborative power grid detection. Coordinate multiple UAVs for power grid inspection based on optimal task allocation.

[0043] In actual application scenarios, such as Figure 2 As shown, it is a principle structure block diagram of the present invention, and the detailed steps of the present invention are as follows: Step 1: Establish a multi-UAV collaborative mission planning model; Step 1.2: Determine the dynamic cost: Obtain the distance between the UAV and the power grid target and the minimum distance that meets the detection quality, and give the interference distance; Based on the interference of environmental factors on drone detection and the detection power of drones, combined with the interference distance, the detection cost of drones is determined; The real-time wind speed and terrain undulation parameters at the power grid target are obtained and normalized to give the terrain complexity coefficient at the power grid target; Obtain the maximum Euclidean distance of all UAVs relative to the power grid target, and combine the terrain complexity coefficient at the power grid target and the distance between the UAV and the power grid target to give the UAV's range cost; The detection cost and flight cost of the drone are superimposed to give the dynamic cost corresponding to the drone detection.

[0044] That is, the cost required for the drone detection of the present invention mainly includes the detection cost and the flight cost generated by the drone performing the detection task: 1) The detection cost generated by the drone performing the detection task: It is determined according to the distance of the drone detection target and environmental factors, specifically:

[0045] In the formula, represents the detection cost of UAV i detecting power grid target j at time t, α is the detection weight factor, is the detection power of drone i, d ij ( t ) represents the distance from the detection grid target j at time t, d min In order to meet the minimum distance for detection quality, that is, the minimum distance required to ensure that the detection quality meets the predetermined requirements, is the interference coefficient of environmental factors on UAV i’s detection of power grid target j at time t. Environmental factors include weather attenuation, terrain shielding, and obstacle density. is the preset interference threshold, x ij Indicates whether the power grid target j is detected by UAV i. When the power grid target j is detected by UAV i, x ij =1, otherwise x ij =0; 2) The range cost of the drone performing the detection task: The smaller the range cost of the drone, the greater the probability that the power grid target is assigned to the drone, and the range cost is:

[0046] In the formula, represents the range cost of UAV i to detect power grid target j, β is the cost weight factor, t j ( t ) represents the terrain complexity coefficient of the location of the power grid target j, c 1, c 2 is the normalization coefficient, W j ( t ) is the real-time wind speed at the location of the power grid target j, T j is the terrain undulation parameter where the power grid target j is located, d max represents the maximum Euclidean distance of all UAVs relative to the grid target j; Step 1.3, determine dynamic benefits: The dynamic benefits of a drone refer to the detection benefits of the drone taking environmental factors into account when performing detection tasks through sensors, including: Obtain the current heading angle of the drone and the optimal approach angle taking into account environmental factors, and give the deviation coefficient of the drone; The detection accuracy of the drone and the inspection value of the power grid target are obtained, and combined with the deviation coefficient of the drone, the dynamic benefits of the drone detecting the power grid target are determined to meet the following relationship:

[0047] In the formula, R ij ( t ) represents the dynamic benefit of UAV i detecting power grid target j at time t, or 0 is the income weight factor, l i Indicates the detection accuracy of the sensor carried by drone i, F j is the inspection value of power grid target j, i i ( t ) is the current heading angle of UAV i, is the optimal approach angle under environmental constraints; Step 1.4, determine the objective function of multi-UAV collaborative task planning: use the cost of minimizing detection under environmental factors to determine the final allocation plan. The smaller the objective function, the better the allocation plan. The objective function is:

[0048] In the formula, F ( t ) is the objective function for multi-UAV collaborative mission planning, Ω( t ) represents the dynamic weight vector, which is updated in real time through the Kalman filter. , ω1, ω2, ω3 are all dynamic weights, corresponding to the dynamic weight of dynamic cost, the dynamic weight of dynamic benefit and the dynamic weight of dynamic weight, ω1+ω2+ω3=1, [] T is the transpose operation, Δ E ( t ) is the change of environmental parameters, K,Q is the adjustment matrix in the Kalman algorithm.

[0049] Step 1.5: Determine the constraints that take environmental factors into account: The maximum flight speed and maximum safe flight distance of all drones are the same, and each power grid target must be detected, so the constraints are:

[0050] In the formula, Indicates the current maximum flight speed of drone i. is the default maximum flight speed of the drone. k is the wind speed influencing factor, W max is the maximum ambient wind speed, d safe ( t ) is the maximum safe flight distance, μ is the safety margin adjustment factor, ρ(t) is the real-time obstacle density, and T is the power grid target set of the detection task; The multi-UAV collaborative power grid detection task planning model of the present invention is:

[0051] Step 1.6: Determine the dynamic adjustment mechanism of the objective function taking environmental factors into account: adjust the weights of dynamic cost and dynamic benefit respectively according to the communication conditions and environmental parameters during drone detection, specifically: By comparing the environmental parameter change rate with the corresponding environmental parameter change rate threshold, when the environmental parameter change rate is large, that is, the environmental parameter change rate is greater than the environmental parameter change rate threshold, the dynamic weight adjustment of the dynamic cost and dynamic benefit in the objective function is performed to satisfy the following relationship:

[0052] Where δ(t) is the rate of change of environmental parameters, δ thresholdis the threshold of the environmental parameter change rate, E(t) is the environmental parameter at time t, Δt is the time change, ω'1 and ω'2 are the dynamic weights of ω1 and ω2 update adjustment, ΔΦ is the communication interference change, ΔV is the power grid value fluctuation amplitude, V max is the maximum grid value.

[0053] Step 2: Establish a diverse evolutionary strategy based on dingoes: Different wild dogs choose different hunting methods based on probability, and solve nonlinear multi-constraint and multi-optimization objectives based on different hunting methods to determine the optimal task allocation for multi-UAV collaborative power grid detection; According to different hunting strategies, corresponding evolutionary strategies are designed. The hunting methods of the dingo algorithm include collective hunting, individual hunting and scavenging. The evolutionary strategy of collective hunting satisfies the following relationship:

[0054] The evolutionary strategy of single predation satisfies the following relationship:

[0055] The evolutionary strategy of scavenging satisfies the following relationship:

[0056] In the formula, are the new positions of wild dogs when participating in collective hunting, individual predation and scavenging, respectively. α(t), β(t) and γ(t) are the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation and the scavenging trigger threshold of scavenging, respectively. k∈{1,2,...,P}, P is the number of Australian wild dogs, X k represents the initial position of the selected wild dog, is the time-varying coupling coefficient, , a3 is the time-varying coupling weight, T max is the maximum duration of power grid detection, X best is the best individual among all wild dogs, that is, it has the current optimal allocation plan, F1 represents the adaptive hunting scaling factor, X m , X n are two random wild dogs participating in collective hunting, λ(t) is the dynamic learning rate, , are two random individuals among all dingo individuals, is a random individual among all dingo individuals, a1 and a2 are the weight of individual predators and the threshold for updating scavenging positions, respectively, and rand is a random number between (0,1).

[0057] in Indicates that when rand is greater than a2, calculate X best -X k , otherwise calculate .

[0058] Step 3: Nonlinear multi-objective optimization solution based on the dingo algorithm, such as Figure 3 As shown, including: Establish a mapping between the positions of individual dingoes in the dingo algorithm and the target allocation scheme of the power grid corresponding to multiple drones, determine the evolutionary strategies and coefficient decision factors of different hunting methods of the dingo algorithm, and initialize the positions of dingoes in the dingo algorithm; At the current wild dog position, the corresponding coefficient decision factor is selected based on the multi-UAV collaborative decision-making mechanism to update the evolutionary strategy, and a new wild dog position is generated according to the updated evolutionary strategy, and the corresponding power grid detection evaluation index is calculated; The iterative update and calculation process is repeated until the preset conditions are met, and the wild dog positions with the best power grid detection evaluation indicators are output, and the optimal task allocation for multi-UAV collaborative power grid detection is obtained.

[0059] Specifically: Step 3.1, Wild Dog Individual Coding: The present invention adopts a mapping real number vector encoding method to encode individuals in the dingo algorithm, and establishes a mapping between dingo individuals and a multi-UAV collaborative task planning model. The number of UAVs is M, and the number of tasks performed, i.e., the number of power grid targets detected, is T. Step 3.2, population initialization: The wild dog population positions are initialized as follows:

[0060] in, represents the initial position of the wild dog, and express The upper and lower boundary constraints of Represents a random number between 0 and 1; Step 3.3, hunting phase: During the hunting phase, the wild dog population uses the predation evolutionary strategy established in step 2 to conduct neighborhood searches (encirclement and individual predation) or global random searches (scavenging) at the current wild dog location to develop new predation locations. In order to achieve accurate predation position updates that adapt to dynamic changes in environmental factors, the evolutionary strategy also needs to be updated.

[0061] Specifically, Figure 4 As shown, the corresponding coefficient decision factor update evolution strategy is selected based on the multi-UAV collaborative decision-making mechanism at the current wild dog position, which may include: Based on the multi-UAV collaborative task planning model and environmental factors, the state space of the reinforcement learning algorithm is determined, and the action space of the reinforcement learning algorithm is determined based on the coefficient decision factor; Determine the reward function of reinforcement learning based on the multi-UAV collaborative mission planning model; Based on the current position of the wild dog, combined with the state space, action space and reward function, determine the probability of selecting an action in the action space under the current state space, and based on the maximum probability, determine the coefficient decision factor corresponding to the selected action; The evolution strategy is updated according to the coefficient decision factor corresponding to the selected action.

[0062] Among them, based on the current wild dog position, combined with the state space, action space and reward function, the probability of selecting an action in the action space under the current state space is determined, and based on the maximum probability, the coefficient decision factor corresponding to the selected action is determined, including; Based on the current position of the wild dog, the reinforcement learning algorithm is used to evaluate the expected long-term benefits of each action in the action space under all states in the current state space before the update, and the expected long-term benefits corresponding to the next moment are determined; Select the maximum value from the expected long-term benefits corresponding to the next moment, and update the expected long-term benefits of each action in the action space under all states in the current state space in combination with the reward function; Determine the probability of selecting each action in the action space under all states in the current state space based on the expected long-term benefits of selecting each action in the action space under all states in the updated current state space; Screen and determine the action selected when the probability is the highest, and based on the relationship between the coefficient decision factor and the action space, determine the coefficient decision factor of the hunting method corresponding to the action selected when the probability is the highest.

[0063] The probability of selecting each action in the action space under all states in the current state space satisfies the following relationship:

[0064] In the formula, π(a|s) is the current state space S t Select action space A in state s t The probability of action a in , Q'(s,a) is the updated current state space S t Select action space A in state s t The expected long-term benefit of action a in the above example, the corresponding max indicates that the selected action is the best estimated action, and Q(s,a) is the current state space S before the update. t Select action space A in state s t The expected long-term benefit of action a, T is the total number of grid targets, lr is the learning rate, r(t) is the reward function, η is the discount factor, s', a' are the state at the next moment reached by executing action a in the current state s and the action selected by the state at the next moment, Δ F ( t ) is the difference between the objective function of multi-UAV collaborative mission planning at time t and time t-1, Ω opt is a small perturbation value of the dynamic weight vector, a small value close to 0, for example, [0.01, 0.01, 0.01], which can avoid problems such as gradient vanishing or gradient exploding when Ω(t) is reduced to 0. collision_rate is the collision rate, a4 and a5 are the weights of minimizing the target weight and minimizing the benefit of the dynamic weight vector, respectively, a4=0.6, a5=0.3, F ( t ) is the objective function for multi-UAV collaborative mission planning, W j ( t ) is the real-time wind speed at the location of the power grid target j, is the interference coefficient of environmental factors on UAV i’s detection of power grid target j at time t, ρ(t) is the real-time obstacle density, α(t), β(t), and γ(t) are the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation, and the scavenging trigger threshold of scavenging, respectively. Δα, Δβ, and Δγ are the discrete adjustment values ​​of the hunting intensity coefficient, the exploration rate parameter, and the scavenging trigger threshold, respectively.

[0065] Among them, for the current state space S before updating t Select action space A in state s t The expected long-term benefit Q(s,a) of action a in the game can be determined using the DQN network algorithm, which includes: 1. Initialize the DQN network: Initialize the DQN network, whose input is state s and action a, and output is Q(s,a); The parameters of the DQN network are randomly initialized; 2. Observe the current state and action: At each time step t, observe the current state s and select an action a; 3. Forward propagation: Take the current state s and action a as input, perform forward propagation through the DQN network, and obtain the output Q(s,a) value, which is the expected long-term benefit before updating.

[0066] The present invention also includes establishing a dual-channel encoding mechanism: Main encoding channel: Represents the task allocation relationship between UAV and target; Auxiliary encoding channel: represents the set of coefficient decision factors, Φ α , Φ β , Φ γIt is a network that controls the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation, and the scavenging trigger threshold of scavenging. That is, strategy u actually controls the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation, and the scavenging trigger threshold of scavenging through these three networks.

[0067] By establishing a dual-channel encoding mechanism, X d and Y d Realization from S t To A t The mapping, that is, At=f(X d ,Y d ,S t ), the mapping function MLP neural network.

[0068] After updating the evolution strategy and generating new wild dog positions, the power grid detection evaluation index calculated by the present invention is as follows:

[0069] Where P u (t) is the power grid detection and evaluation index, Q u (t) is the long-term cumulative reward value of strategy u, R u (t) is the current immediate environment reward, strategy u is to select an action in the action space under a certain state in the state space, η=0.8 is the discount factor, Q v (t), R v (t) represents the cumulative reward value and immediate environmental return under the recorded historical strategy v, where the historical strategy v is all strategies u adopted in the historical position update; In addition, the short-term target value and long-term target value of predation quality can be used to establish power grid detection evaluation indicators, which specifically include the following steps: Determine the fitness value calculated by the allocation scheme corresponding to the individual of the previous iteration and the fitness value calculated by the allocation scheme corresponding to the individual of the current iteration; select the minimum fitness value from the fitness value calculated by the allocation scheme corresponding to the individual of the previous iteration and the fitness value calculated by the allocation scheme corresponding to the individual of the current iteration, and perform proportional calculation on the selected minimum fitness value and the fitness value calculated by the allocation scheme corresponding to the individual of the previous iteration; based on the difference between the base threshold of the allocation scheme and the proportional calculation result, determine the short-term target value under the allocation scheme corresponding to the individual of the current iteration; wherein the base threshold of the allocation scheme is 1; The adjustment value is determined based on the number of iterations, and the balance item value is determined in combination with the number of times that the individual obtains a fitness value less than the benchmark threshold by using the individual's corresponding allocation scheme in the current iteration and before; the number of times that the individual obtains a fitness value less than the benchmark threshold by using the individual's corresponding allocation scheme in the current iteration and before is proportional to the total number of times that the individual's corresponding allocation scheme is used in the current iteration and before, and the balance item value is superimposed to give the long-term target value under the individual's corresponding allocation scheme in the current iteration; Determine the weight parameters of the short-term target value and the long-term target value, and combine the short-term target value and the long-term target value to give the current iteration of power grid detection evaluation indicators; The power grid detection evaluation index of the previous iteration is compared with the power grid detection evaluation index of the current iteration, and when the power grid detection evaluation index of the previous iteration is better than the power grid detection evaluation index of the current iteration, the parameters of the evolutionary strategy in the current iteration are updated.

[0070] The specific power grid detection evaluation indicators under this method are:

[0071] In formula (11), Indicates The iteration uses u The comprehensive evaluation value under the allocation scheme corresponding to each strategy is: , are the probability weight parameters for strategy selection, , Respectively represent The iteration uses u The short-term target value and long-term target value under the allocation plan corresponding to each strategy. It represents the fitness value calculated by the allocation scheme corresponding to the individual in the previous iteration. It represents the fitness value calculated by the allocation scheme corresponding to the current iteration individual. Indicates that the individual t The first iteration and the previous u The number of times the allocation scheme corresponding to a strategy obtains a fitness value less than 1, Indicated in t The first iteration and the previous u The total number of allocation schemes corresponding to the strategy, is the balance coefficient, and log(t) is the natural logarithm with t as the independent variable.

[0072] The preset condition for stopping the update iteration of the present invention is the maximum number of iterations, or the power grid detection evaluation index after several consecutive iterations has changed little or deteriorated. After completing the update iteration, by comparing all the power grid detection evaluation indicators, the wild dog position with the largest power grid detection evaluation index is selected, and the optimal task allocation corresponding to the multiple drones is determined according to the wild dogs, and the multiple drones are coordinated to realize power grid detection.

[0073] The present invention has at least the following beneficial effects: (1) The present invention combines the diverse predation mechanisms of the dingo algorithm to establish an allocation scheme evolution strategy that includes collective hunting, individual predation, scavenging and other evolutionary strategies to replace the single evolutionary mode of the traditional swarm intelligence algorithm, giving the population individuals more autonomy and avoiding the algorithm from falling into local optimality.

[0074] (2) The present invention models the power grid detection characteristics, designs a dynamic adjustment mechanism for the objective function and power grid detection evaluation indicators, thereby ensuring that the population individuals are optimized toward the correct allocation plan and improving the algorithm optimization efficiency.

[0075] (3) The present invention can quickly and effectively solve the problem of multi-UAV collaborative power grid detection task allocation based on the Dingo Sink algorithm, providing a new solution to the problem of multi-UAV collaborative power grid detection task allocation in complex environments.

[0076] like Figure 5 As shown, the present invention also provides a multi-UAV collaborative power grid detection device, which adopts the multi-UAV collaborative power grid detection method, and the device includes: A model building module, which is used to establish a multi-UAV collaborative mission planning model based on the dynamic cost and dynamic benefit required for UAV detection considering environmental factors; The task allocation module is used to iteratively optimize the nonlinear multi-objective problem based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, and determine the optimal task allocation for multi-UAV collaborative power grid detection; The power grid detection module is used to coordinate multiple UAVs to perform power grid detection based on optimal task allocation.

[0077] Furthermore, the model building module includes: Based on the detection cost and flight cost generated by the drone performing the detection task, determine the dynamic cost corresponding to the drone detection; Determine the dynamic benefits of the drone based on the drone’s detection accuracy and the detection value of the power grid target; According to the dynamic cost and dynamic benefit of UAVs, the objective function of multi-UAV collaborative mission planning is given; The constraints of UAVs taking environmental factors into account and the dynamic adjustment mechanism of the objective function are determined, and a multi-UAV collaborative mission planning model is established in combination with the objective function of multi-UAV collaborative mission planning.

[0078] Furthermore, the task allocation module includes: S21. Establish a mapping between the positions of individual wild dogs in the dingo algorithm and the target allocation scheme of the power grid corresponding to multiple drones, determine the evolutionary strategies and coefficient decision factors of different hunting methods of the dingo algorithm, and initialize the positions of wild dogs in the dingo algorithm; S22, selecting the corresponding coefficient decision factor to update the evolution strategy based on the multi-UAV collaborative decision-making mechanism at the current wild dog position, generating a new wild dog position according to the updated evolution strategy, and calculating the corresponding power grid detection evaluation index; S23. Repeat iterative step S22 until the preset conditions are met, output the wild dog position with the best power grid detection evaluation index, and obtain the optimal task allocation for multi-UAV collaborative power grid detection.

[0079] Furthermore, the task allocation module includes: Based on the multi-UAV collaborative task planning model and environmental factors, the state space of the reinforcement learning algorithm is determined, and the action space of the reinforcement learning algorithm is determined based on the coefficient decision factor; Determine the reward function of reinforcement learning based on the multi-UAV collaborative mission planning model; Based on the current position of the wild dog, combined with the state space, action space and reward function, determine the probability of selecting an action in the action space under the current state space, and based on the maximum probability, determine the coefficient decision factor corresponding to the selected action; The evolution strategy is updated according to the coefficient decision factor corresponding to the selected action.

[0080] Furthermore, the task allocation module includes: Determine the fitness value calculated by the allocation scheme corresponding to the individual of the previous iteration and the fitness value calculated by the allocation scheme corresponding to the individual of the current iteration; select the minimum fitness value from the fitness value calculated by the allocation scheme corresponding to the individual of the previous iteration and the fitness value calculated by the allocation scheme corresponding to the individual of the current iteration, and perform proportional calculation on the selected minimum fitness value and the fitness value calculated by the allocation scheme corresponding to the individual of the previous iteration; based on the difference between the base threshold of the allocation scheme and the proportional calculation result, determine the short-term target value under the allocation scheme corresponding to the individual of the current iteration; wherein the base threshold of the allocation scheme is 1; The adjustment value is determined based on the number of iterations, and the balance item value is determined in combination with the number of times that the individual obtains a fitness value less than the benchmark threshold by using the individual's corresponding allocation scheme in the current iteration and before; the number of times that the individual obtains a fitness value less than the benchmark threshold by using the individual's corresponding allocation scheme in the current iteration and before is proportional to the total number of times that the individual's corresponding allocation scheme is used in the current iteration and before, and the balance item value is superimposed to give the long-term target value under the individual's corresponding allocation scheme in the current iteration; Determine the weight parameters of the short-term target value and the long-term target value, and combine the short-term target value and the long-term target value to give the current iteration of power grid detection evaluation indicators.

[0081] Although preferred embodiments of the present invention have been described, additional changes and modifications may be made to these embodiments by those skilled in the art once the basic inventive concepts are known. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention. Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A multi-UAV collaborative power grid detection method, characterized in that: include: Based on the dynamic cost and dynamic benefit required for drone detection considering environmental factors, a multi-drone collaborative task planning model is established; Based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, the nonlinear multi-objective problem is iteratively optimized to determine the optimal task allocation for multi-UAV collaborative power grid detection. Coordinate multiple UAVs for power grid inspection based on optimal task allocation.

2. The multi-UAV collaborative power grid detection method according to claim 1, characterized in that: Based on the dynamic cost and dynamic benefit required for drone detection considering environmental factors, a multi-drone collaborative task planning model is established, including: Based on the detection cost and flight cost generated by the drone performing the detection task, determine the dynamic cost corresponding to the drone detection; Determine the dynamic benefits of the drone based on the drone’s detection accuracy and the detection value of the power grid target; According to the dynamic cost and dynamic benefit of UAVs, the objective function of multi-UAV collaborative mission planning is given; The constraints of UAVs taking environmental factors into account and the dynamic adjustment mechanism of the objective function are determined, and a multi-UAV collaborative mission planning model is established in combination with the objective function of multi-UAV collaborative mission planning.

3. The multi-UAV collaborative power grid detection method according to claim 2, characterized in that: Determine the dynamic cost corresponding to drone detection, including: Obtain the distance between the UAV and the power grid target and the minimum distance that meets the detection quality, and give the interference distance; Based on the interference of environmental factors on drone detection and the detection power of drones, combined with the interference distance, the detection cost of drones is determined; The real-time wind speed and terrain undulation parameters at the power grid target are obtained and normalized to give the terrain complexity coefficient at the power grid target; Obtain the maximum Euclidean distance of all UAVs relative to the power grid target, and combine the terrain complexity coefficient at the power grid target and the distance between the UAV and the power grid target to give the UAV's range cost; The detection cost and flight cost of the drone are superimposed to give the dynamic cost corresponding to the drone detection.

4. The multi-UAV collaborative power grid detection method according to claim 2 or 3, characterized in that: The objective function of multi-UAV collaborative mission planning satisfies the following relationship: ； In the formula, F ( t ) is the objective function for multi-UAV collaborative mission planning, represents the detection cost of UAV i detecting power grid target j at time t, represents the range cost of UAV i detecting power grid target j, R ij ( t ) represents the dynamic benefit of UAV i detecting power grid target j at time t, Ω( t ) represents the dynamic weight vector, which is updated in real time through the Kalman filter. , ω1, ω2, ω3 are all dynamic weights, Δ E ( t ) is the change of environmental parameters, K,Q is the adjustment matrix in the Kalman algorithm, [] T is the transpose operation; The dynamic adjustment mechanism of the objective function satisfies the following relationship: ； Where δ(t) is the rate of change of environmental parameters, δ threshold is the environmental parameter change rate threshold, E(t) is the environmental parameter, ω'1 and ω'2 are the dynamic weights of ω1 and ω2 update adjustment, ΔΦ is the communication interference change, ΔV is the power grid value fluctuation amplitude, V max is the maximum grid value.

5. The multi-UAV collaborative power grid detection method according to claim 1, characterized in that: Based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, the nonlinear multi-objective problem is iteratively optimized to determine the optimal task allocation for multi-UAV collaborative power grid detection, including: S21. Establish a mapping between the location of individual dingoes in the dingo algorithm and the target allocation scheme of the power grid corresponding to multiple drones, and determine the evolutionary strategies and coefficient decision factors of different hunting methods of the dingo algorithm; S22, selecting the corresponding coefficient decision factor to update the evolution strategy based on the multi-UAV collaborative decision-making mechanism at the current wild dog position, generating a new wild dog position according to the updated evolution strategy, and calculating the corresponding power grid detection evaluation index; S23. Repeat iterative step S22 until the preset conditions are met, output the wild dog position with the best power grid detection evaluation index, and obtain the optimal task allocation for multi-UAV collaborative power grid detection.

6. The multi-UAV collaborative power grid detection method according to claim 5, characterized in that: At the current wild dog location, the corresponding coefficient decision factor update evolution strategy is selected based on the multi-UAV collaborative decision-making mechanism, including: Based on the multi-UAV collaborative task planning model and environmental factors, the state space of the reinforcement learning algorithm is determined, and the action space of the reinforcement learning algorithm is determined based on the coefficient decision factor; Determine the reward function of reinforcement learning based on the multi-UAV collaborative mission planning model; Based on the current position of the wild dog, combined with the state space, action space and reward function, determine the probability of selecting an action in the action space under the current state space, and based on the maximum probability, determine the coefficient decision factor corresponding to the selected action; The evolution strategy is updated according to the coefficient decision factor corresponding to the selected action.

7. The multi-UAV collaborative power grid detection method according to claim 5 or 6, characterized in that: The dingo algorithm's hunting methods include group hunting, individual hunting, and scavenging; The evolutionary strategy of collective hunting satisfies the following relationship: ； The evolutionary strategy of single predation satisfies the following relationship: ； The evolutionary strategy of scavenging satisfies the following relationship: ； In the formula, X k new,1 、X k new,2 、X k new,3 are the new positions of wild dogs when participating in collective hunting, individual predation and scavenging, respectively. α(t), β(t) and γ(t) are the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation and the scavenging trigger threshold of scavenging, i.e., the coefficient decision factors of different hunting methods. k∈{1,2,...,P}, P is the number of Australian wild dogs, X k represents the initial position of the selected wild dog, is the time-varying coupling coefficient, , a3 is the time-varying coupling weight, T max is the maximum duration of power grid detection, X best is the best individual among all wild dogs, that is, it has the current optimal allocation plan, F1 represents the adaptive hunting scaling factor, X m , X n are two random wild dogs participating in collective hunting, λ(t) is the dynamic learning rate, , are two random individuals among all dingo individuals, is a random individual among all dingo individuals, a1 and a2 are the weight of individual predators and the threshold for updating scavenging positions, respectively, and rand is a random number between (0,1).

8. The multi-UAV collaborative power grid detection method according to claim 7, characterized in that: Based on the current position of the wild dog, combined with the state space, action space and reward function, determine the probability of selecting an action in the action space under the current state space, and based on the maximum probability, determine the coefficient decision factor corresponding to the selected action, including; Based on the current position of the wild dog, the reinforcement learning algorithm is used to evaluate the expected long-term benefits of each action in the action space under all states in the current state space before the update, and the expected long-term benefits corresponding to the next moment are determined; Select the maximum value from the expected long-term benefits corresponding to the next moment, and update the expected long-term benefits of each action in the action space under all states in the current state space in combination with the reward function; Determine the probability of selecting each action in the action space under all states in the current state space based on the expected long-term benefits of selecting each action in the action space under all states in the updated current state space; Screen and determine the action selected when the probability is the highest, and based on the relationship between the coefficient decision factor and the action space, determine the coefficient decision factor of the hunting method corresponding to the action selected when the probability is the highest.

9. The multi-UAV collaborative power grid detection method according to claim 8, characterized in that: The probability of selecting each action in the action space under all states in the current state space satisfies the following relationship: ； In the formula, π(a|s) is the current state space S t Select action space A in state s t The probability of action a in , Q'(s,a) is the updated current state space S t Select action space A in state s t The expected long-term benefit of action a in the above example is Q(s,a), and Q(s,a) is the current state space S before the update. t Select action space A in state s t The expected long-term benefit of action a, T is the total number of grid targets, l r is the learning rate, r(t) is the reward function, η is the discount factor, s', a' are the state at the next moment reached by executing action a in the current state s and the action selected by the state at the next moment, Δ F ( t ) is the difference between the objective function of multi-UAV collaborative mission planning at time t and time t-1, Ω opt is the small perturbation value of the dynamic weight vector, collision_rate is the collision rate, a4 and a5 are the benefit weights of minimizing the target weight and minimizing the dynamic weight vector respectively. F ( t ) is the objective function for multi-UAV collaborative mission planning, W j ( t ) is the real-time wind speed at the location of the power grid target j, is the interference coefficient of environmental factors on UAV i’s detection of power grid target j at time t, ρ(t) is the real-time obstacle density, α(t), β(t), and γ(t) are the hunting intensity coefficient of collective hunting, the exploration rate parameter of individual predation, and the scavenging trigger threshold of scavenging, respectively. Δα, Δβ, and Δγ are the discrete adjustment values ​​of the hunting intensity coefficient, the exploration rate parameter, and the scavenging trigger threshold, respectively.

10. A multi-UAV collaborative power grid detection device, characterized in that: The method for multi-UAV collaborative power grid detection according to any one of claims 1 to 9 is adopted, and the device comprises: A model building module, which is used to establish a multi-UAV collaborative mission planning model based on the dynamic cost and dynamic benefit required for UAV detection considering environmental factors; The task allocation module is used to iteratively optimize the nonlinear multi-objective problem based on the multi-UAV collaborative task planning model and multi-UAV collaborative decision-making mechanism, and determine the optimal task allocation for multi-UAV collaborative power grid detection; The power grid detection module is used to coordinate multiple UAVs to perform power grid detection based on optimal task allocation.

Images