Unmanned aerial vehicle photovoltaic inspection path planning method and system considering multi-constraint coupling

Abstract

The invention discloses an unmanned aerial vehicle photovoltaic inspection path planning method and system considering multi-constraint coupling, and the method comprises the steps: building a wind speed model of time window segmentation, and defining a target function of a dynamic trajectory optimization problem of an unmanned aerial vehicle; setting an initialization parameter of the acceleration opposition algorithm based on the chaos single candidate; generating a random solution and an opposite solution of the target function according to an accelerated opposite learning mechanism, and selecting a better solution as an initial solution of the target function; entering a main cycle, and updating the initial solution; and checking whether the solution is within the boundary range, correcting the boundary-crossing solution based on a preset correction rule until the maximum iteration number is reached, and outputting the lowest solution of the target function as a routing inspection path. According to the invention, multi-target collaborative optimization of endurance time, routing inspection timeliness and wind resistance can be realized, and a routing inspection path with higher routing inspection quality can be obtained.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicle inspection, and in particular relates to a method and system for unmanned aerial vehicle photovoltaic inspection path planning considering multi-constraint coupling. Background Art

[0002] Rural photovoltaic power stations are usually widely distributed and in complex environments, and photovoltaic components need to be regularly inspected to detect faults. Traditional manual inspections are inefficient and costly, and autonomous inspections by drones have become the mainstream trend. However, existing path planning often focuses on the "shortest path", but drones cannot complete full-area inspections in complex weather conditions, resulting in low inspection quality. Summary of the invention

[0003] The present invention provides a method and system for planning a photovoltaic inspection path for an unmanned aerial vehicle (UAV) taking into account multiple constraint coupling, which is used to solve the technical problem that the UAV cannot complete full-area coverage inspection in complex weather conditions, resulting in low inspection quality.

[0004] In a first aspect, the present invention provides a method for planning a photovoltaic inspection path for an unmanned aerial vehicle considering multi-constraint coupling, comprising: A wind speed model with time window segments is established, and the objective function of the UAV dynamic trajectory optimization problem is defined by considering the UAV energy consumption constraint, time penalty constraint, and photovoltaic inspection path planning of the UAV under wind farm conditions; Set the initialization parameters of the accelerated adversarial algorithm based on a chaotic single candidate; Generate a random solution and an adversarial solution of the objective function according to an accelerated adversarial learning mechanism, and select a better solution as an initial solution of the objective function; Enter the main loop and perform the initial solution update; Check whether the solution is within the boundary range, and correct the out-of-bounds solution based on the preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

[0005] In a second aspect, the present invention provides a UAV photovoltaic inspection path planning system considering multi-constraint coupling, comprising: The definition module is configured to establish a wind speed model with time window segments, and the objective function of the UAV dynamic trajectory optimization problem is defined by considering the UAV energy consumption constraint, time penalty constraint, and photovoltaic inspection path planning of the UAV under wind farm conditions; A setting module configured to set initialization parameters of an accelerated opposition algorithm based on a chaotic single candidate; A selection module configured to generate a random solution and an adversarial solution of the objective function according to an accelerated adversarial learning mechanism, and select a better solution as an initial solution of the objective function; A loop module, configured to enter a main loop and perform the initial solution update; The output module is configured to check whether the solution is within the boundary range, and to correct the out-of-bounds solution based on a preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

[0006] According to a third aspect, an electronic device is provided, comprising: at least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the steps of the unmanned aerial vehicle photovoltaic inspection path planning method considering multi-constraint coupling according to any embodiment of the present invention.

[0007] In a fourth aspect, the present invention further provides a computer-readable storage medium having a computer program stored thereon. When the program instructions are executed by a processor, the processor executes the steps of the unmanned aerial vehicle photovoltaic inspection path planning method considering multi-constraint coupling according to any embodiment of the present invention.

[0008] The method and system for UAV photovoltaic inspection path planning considering multi-constraint coupling of the present application have the following beneficial effects: The energy consumption constraints of drones, time window penalty functions and dynamic wind field effects are coupled and modeled. By building an objective function integrating fixed cost, dynamic energy cost and penalty cost, multi-objective collaborative optimization of flight time, inspection timeliness and wind resistance is achieved; An accelerated adversarial algorithm based on chaotic single candidate is designed, and the solution convergence efficiency is improved through a three-stage optimization strategy, thereby achieving a faster production drone photovoltaic inspection path. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0010] Figure 1 A flowchart of a method for unmanned aerial vehicle photovoltaic inspection path planning considering multi-constraint coupling provided in one embodiment of the present invention; Figure 2 A structural block diagram of a UAV photovoltaic inspection path planning system considering multi-constraint coupling provided by an embodiment of the present invention; Figure 3 It is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0011] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are 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.

[0012] See also Figure 1 , which shows a flow chart of a UAV photovoltaic inspection path planning method considering multi-constraint coupling in the present application.

[0013] like Figure 1 As shown in the figure, the UAV photovoltaic inspection path planning method considering multi-constraint coupling specifically includes the following steps: Step S101, establish a wind speed model with time window segments, consider the energy consumption constraint of the UAV, the time penalty constraint and the photovoltaic inspection path planning of the UAV under the wind field condition, and define the objective function of the UAV dynamic trajectory optimization problem.

[0014] In this step, the wind speed model is built as follows: Divide the wind direction into 15 directional intervals, equally spaced, with each interval being 24°; Construct a probability distribution for wind direction: , in, It's the wind direction. is the mean wind speed, It is a concentration parameter used to describe the concentration of wind direction. Its value is generally between 0.5 and 5. is the zero-order Bessel function of the first kind; Construct a probability distribution for wind speed: , , , in, is the wind speed, It is a shape parameter used to describe the shape of wind speed and generally takes values ​​between 1 and 3. It is The scale parameter value of each direction interval is usually between 5 and 15 m / s. Construction of joint distribution: The joint probability density function of wind speed and wind direction can be written as: .

[0015] Obtain wind field data for the next 6 hours from the meteorological data center and divide the time window into 10-minute intervals. Preprocess the data and extract the global wind speed and wind direction mean for each window. During the actual inspection, update the meteorological parameters of the next window every 10 minutes through satellite communication.

[0016] Define the objective function of the UAV dynamic trajectory optimization problem, the expression is: , In the formula, Fixed cost for drones, is the dynamic energy cost, For the penalty cost, is the surface state function of the photovoltaic panel, is a weight coefficient in the range [0,1]; , , , , , In the formula, , , The fixed costs of the first UAV, the second UAV, and the The fixed cost of a drone, is the number of drones, is the unloaded weight of the drone, is the air fluid density, is the blade area of ​​the drone, for Inspection time of the point, Drone from point To point The distance is the actual wind speed of the UAV at time t, is the speed of the drone relative to the air, is the global predicted wind speed value, The angle between the wind direction and due north. The angle between the drone's flight path and true north. is a binary variable, indicating that the drone Whether the task is completed, and are the upper and lower bounds of the time window, respectively. and is the penalty coefficient for exceeding the time window, Arrival The actual time of the point, is the light intensity in the current period, is the reference light intensity, is the current PV panel contamination rate, Cleaning status reference value, is the environmental sensitivity coefficient, is the number of time windows, is the acceleration due to gravity.

[0017] Step S102, setting initialization parameters of the accelerated opposition algorithm based on a chaotic single candidate.

[0018] In this step, the initialization parameters include: the adaptive maximum number of iterations based on the convergence rate of the objective function , the initial maximum number of iterations , the overall dimension of the problem , control factors , Change window time , the number of successful attempts threshold , Iteration number threshold in the exploration phase Dimensions of the problem , maximum jump probability , minimum jump probability and regulatory factors .

[0019] Step S103: generating a random solution and an opponent solution of the objective function according to the accelerated adversarial learning mechanism, and selecting a better solution as the initial solution of the objective function.

[0020] In this step, the maximum number of adaptive iterations based on the convergence rate of the objective function is determined. , the expression is: , In the formula, is the initial maximum number of iterations, is the control factor, For the The best objective function value at the iteration, For the The best objective function value at the iteration, is the change window time; A mixed probability distribution is used to generate random solutions, expressed as: , In the formula, For Dimension The updated solution of For Dimension The lower bound of For Dimension The lower bound of is a uniformly distributed random number, is the standard normal distribution number, is the random probability, is a random probability, satisfying ; For each random solution , with the jump probability Generate an opposite solution to the current solution , the expression is: , This expression is the main idea of ​​the adversarial learning mechanism in the algorithm. The probability of generating the corresponding item opposite to the current random solution is used, and the one with the smallest objective function value is selected as the initial solution. This initialization step improves the applicability of the subsequent optimization process.

[0021] , , In the formula, is the maximum jump probability, is the minimum jump probability, is the regulating factor, For the The speedup factor for the iteration, For the dimension of the problem, is the total dimension of the problem; Comparing random solutions The objective function value and the opposite solution The objective function value of is retained as the initial solution.

[0022] Step S104, entering the main loop to perform the initial solution update.

[0023] In this step, when the number of iterations is less than the set value, the exploration phase begins, where the solution update rule in the exploration phase is: Calculating weights and adaptive weights , the expression is: , , In the formula, is an exponential function, Indicates Iterations, is the adaptive maximum number of iterations based on the convergence rate of the objective function; With adaptive weight Use the original SCO algorithm to update the solution, the expression is: , In the formula, is the updated solution, is a random number; 1- The probability uses a new strategy, which introduces sine / cosine perturbations to prevent the updated solution from converging quickly. The expression is:

[0024] In the formula, is the imaginary number symbol, is a random number, is a step function; When the number of iterations is greater than the set value, it enters the development phase. The solution update rule in the development phase is: If continuous If the update fails, the preset mutation strategy is used to update the solution. The mutation strategy is: Cauchy variation Trigger mutation, the expression is: , , , In the formula, is a random number between 0 and 1. is the average fitness change of the previous iteration, is the mean change of the optimal fitness of the previous iteration, is the mutation trigger probability factor, is the probability of triggering Cauchy mutation, For Dimension The updated solution of .

[0025] Gaussian variation Trigger mutation, the expression is: , , In the formula, is the probability of triggering Gaussian mutation, Indicates sampling from a Gaussian distribution; Levy mutation Probability trigger mutation, the expression is: , , In the formula, The probability of triggering Levi mutation, for , is the gamma function, for ; If the mutation is not triggered, the candidate solution is updated normally, and the expression is: , In the formula, Is a random number.

[0026] Step S105 , checking whether the solution is within the boundary range, and correcting the out-of-bounds solution based on a preset correction rule until the maximum number of iterations is reached, and outputting the solution with the lowest objective function as the inspection path.

[0027] In this step, the expression of the correction rule is: , , , In the formula, For Dimension The updated solution of For Dimension The lower bound of is the dynamic fractal contraction coefficient, For Dimension The upper bound of It is a very small number, usually 2.22e-16. is the fractal noise function, is the Riemann function, For infinity, is the initial value of the dynamic fractal contraction coefficient, is the attenuation factor, Indicates Iterations, is the adaptive maximum number of iterations based on the convergence rate of the objective function.

[0028] It should be noted that the lowest solution refers to the objective function value recorded after each iteration. After reaching the maximum number of iterations, the objective function value after each iteration is compared, and the drone inspection path sequence with the minimum objective function value is the optimal solution.

[0029] The lowest solution refers to the drone inspection path plan with the smallest objective function value during the algorithm iteration process. The objective function is a quantitative indicator of the multi-constrained optimization problem: , In summary, the lowest solution is the path plan corresponding to the minimum value of the objective function obtained by the algorithm through global search in the feasible solution space.

[0030] In summary, the method of the present application comprehensively considers the UAV energy consumption constraints, time penalty constraints, and UAV photovoltaic inspection path planning under wind field conditions, introduces a wind speed model with time window segments, and uses an accelerated opposition algorithm based on a chaotic single candidate to solve the optimal inspection path of the UAV. It solves the problem that existing path planning often takes the "shortest path" as the core, ignores the UAV energy consumption constraints, time window penalty functions, and the impact of wind speed and wind direction on the UAV's endurance and inspection quality, and solves the problem that traditional algorithms need to frequently adjust parameters when solving, resulting in slow convergence and easy to fall into local optimality. It can achieve multi-objective collaborative optimization of endurance time, inspection timeliness and wind resistance, and obtain an inspection path with higher inspection quality.

[0031] In a specific embodiment, to further verify the effectiveness of the proposed method, the present invention uses the traveling salesman problem eil51 test set to compare the average path length, average solution time and average cost of three different algorithms running independently for 30 times, and uses MATLAB software to verify the proposed mathematical model.

[0032] Table 1 Comparison of the results of three algorithms , The algorithm of the present invention performs best in terms of average path length, average solution time and average cost, which are 447.97, 16.75 seconds and 4231.3 respectively, which are better than the Improved Sparrow Search Algorithm (ISSA) (467.31, 18.14 seconds, 4481.5) and the Grey Wolf Optimization (GWO) (458.56, 16.96 seconds, 4314.9). This shows that the algorithm of the present invention has higher efficiency and effect in path optimization problems, can find shorter and lower-cost paths in a shorter time, and is suitable for application scenarios with high requirements on path length, solution time and cost.

[0033] At the same time, in order to solve the multi-constraint coupling problem in photovoltaic inspection tasks, the present invention constructs a dynamic path optimization model by integrating multi-dimensional constraints such as wind speed, time window, and energy consumption. Experimental results show that the method of the present invention performs significantly better than traditional methods in complex constraint scenarios. The specific data are as follows: (1) Wind speed constraint satisfaction rate In the simulation scenario (average wind speed 8m / s, gusts up to 12m / s), the path planned by the algorithm of the present invention: Only 3.2% of the flight segments exceeded the maximum wind resistance level of the drone (the designed wind resistance is 10m / s), while the excess rates of the traditional genetic algorithm (GA) and particle swarm optimization (PSO) were 15.7% and 18.4% respectively.

[0034] Wind energy utilization rate increased by 21%: By optimizing the flight attitude through the real-time wind speed segmentation model, the cruising speed of the drone in the downwind section increased by 15%, and the energy consumption in the headwind section decreased by 18%.

[0035] (2) Proportion of time window penalty cost In time-sensitive inspection tasks (time window width ±10 minutes): The number of nodes violating the time window is reduced by 45%: through the constraint of the penalty function (the expression of the penalty cost), only 2.3% of the nodes exceed the upper and lower limits of the time window, while the violation rates of the SSA and GWO algorithms are 7.8% and 9.1%, respectively.

[0036] (3) Balanced energy consumption among multiple machines The standard deviation of energy consumption of a single UAV is reduced to 8.7%: Compared with 15.2% of the traditional method, the present invention optimizes path allocation through a dynamic energy cost model (an expression of dynamic energy cost), thus avoiding some UAVs from running out of power prematurely.

[0037] In order to verify the responsiveness of the method of the present invention under dynamic meteorological conditions, the following two types of extreme scenario tests were designed: (1) Wind speed constraint satisfaction rate The simulated wind speed suddenly increased from 8m / s to 15m / s (for 3 minutes) and then returned to the initial value.

[0038] The algorithm in this paper only needs 2 iterations (about 0.8 seconds) to complete the path adjustment, and the increase in path length is controlled within 4.2% of the original path. In terms of energy consumption stability, the use of a dynamic wind speed model reduces the energy consumption fluctuation of the drone in gusts by 51%, avoiding energy waste caused by frequent speed changes.

[0039] (2) Scenarios with dynamic changes in light intensity and contamination rate The contamination rate of photovoltaic panels quickly increased from the initial 20% to 50% (simulating a sandstorm), while the local light intensity dropped sharply by 30%.

[0040] Based on the surface state function of the photovoltaic panel, the algorithm automatically shifts the path to areas with high contamination rate and sufficient light, and the fault detection completeness rate is increased to 98.7%. After the contamination rate suddenly changes, the path update delay is less than 1.2 seconds, ensuring that the inspection task is not affected by sudden changes in the environment.

[0041] In order to compare the solution of the method of the present application with other algorithms under multi-constraint conditions, 100 simulation iterations were performed in MATLAB based on a 1km² photovoltaic power station. The comparison table of multi-constraint satisfaction is as follows, where "improvement range" is the optimization ratio of the method of the present invention relative to other algorithms.

[0042] Table 2 Comparison of satisfaction of multiple constraints .

[0043] See also Figure 2 , which shows a structural block diagram of a UAV photovoltaic inspection path planning system considering multi-constraint coupling in the present application.

[0044] like Figure 2 As shown, the UAV photovoltaic inspection path planning system 200 includes a definition module 210, a setting module 220, a selection module 230, a loop module 240 and an output module 250.

[0045] Among them, the definition module 210 is configured to establish a wind speed model with time window segments, and define the objective function of the UAV dynamic trajectory optimization problem by considering the UAV energy consumption constraint, time penalty constraint and photovoltaic inspection path planning of the UAV under wind field conditions; the setting module 220 is configured to set the initialization parameters of the accelerated opponent algorithm based on a chaotic single candidate; the selection module 230 is configured to generate a random solution and an opponent solution of the objective function according to the accelerated opponent learning mechanism, and select a better solution as the initial solution of the objective function; the loop module 240 is configured to enter the main loop and update the initial solution; the output module 250 is configured to check whether the solution is within the boundary range, and correct the out-of-bounds solution based on a preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

[0046] It should be understood that Figure 2 Modules and references documented in Figure 1 Therefore, the operations and features described above for the method and the corresponding technical effects are also applicable to Figure 2 The modules in it will not be described in detail here.

[0047] In other embodiments, the embodiments of the present invention further provide a computer-readable storage medium having a computer program stored thereon, wherein when the program instructions are executed by a processor, the processor is caused to execute the UAV photovoltaic inspection path planning method considering multi-constraint coupling in any of the above method embodiments; As an implementation mode, the computer-readable storage medium of the present invention stores computer-executable instructions, and the computer-executable instructions are configured as follows: A wind speed model with time window segments is established, and the objective function of the UAV dynamic trajectory optimization problem is defined by considering the UAV energy consumption constraint, time penalty constraint, and photovoltaic inspection path planning of the UAV under wind farm conditions; Set the initialization parameters of the accelerated adversarial algorithm based on a chaotic single candidate; Generate a random solution and an adversarial solution of the objective function according to an accelerated adversarial learning mechanism, and select a better solution as an initial solution of the objective function; Enter the main loop and perform the initial solution update; Check whether the solution is within the boundary range, and correct the out-of-bounds solution based on the preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

[0048] The computer-readable storage medium may include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application required by at least one function; the data storage area may store data created according to the use of the unmanned aerial vehicle photovoltaic inspection path planning system considering multiple constraint coupling, etc. In addition, the computer-readable storage medium may include a high-speed random access memory, and may also include a memory, such as at least one disk storage device, a flash memory device, or other non-volatile solid-state storage device. In some embodiments, the computer-readable storage medium may optionally include a memory remotely arranged relative to the processor, and these remote memories may be connected to the unmanned aerial vehicle photovoltaic inspection path planning system considering multiple constraint coupling via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0049] Figure 3 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, such as Figure 3 As shown, the device includes: a processor 310 and a memory 320. The electronic device may also include: an input device 330 and an output device 340. The processor 310, the memory 320, the input device 330 and the output device 340 may be connected via a bus or other means. Figure 3 The example of the bus connection is taken. The memory 320 is the above-mentioned computer-readable storage medium. The processor 310 executes various functional applications and data processing of the server by running the non-volatile software programs, instructions and modules stored in the memory 320, that is, the method for planning the path of photovoltaic inspection of unmanned aerial vehicles considering multiple constraints coupling in the above-mentioned method embodiment is realized. The input device 330 can receive input digital or character information, and generate key signal input related to user settings and function control of the path planning system of photovoltaic inspection of unmanned aerial vehicles considering multiple constraints coupling. The output device 340 may include display devices such as display screens.

[0050] The electronic device can execute the method provided by the embodiment of the present invention, and has the functional modules and beneficial effects corresponding to the execution method. For technical details not described in detail in this embodiment, please refer to the method provided by the embodiment of the present invention.

[0051] As an implementation mode, the electronic device is applied to a UAV photovoltaic inspection path planning system considering multi-constraint coupling, and is used for a client, comprising: at least one processor; and a memory connected to the at least one processor in communication; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can: A wind speed model with time window segments is established, and the objective function of the UAV dynamic trajectory optimization problem is defined by considering the UAV energy consumption constraint, time penalty constraint, and photovoltaic inspection path planning of the UAV under wind farm conditions; Set the initialization parameters of the accelerated adversarial algorithm based on a chaotic single candidate; Generate a random solution and an adversarial solution of the objective function according to an accelerated adversarial learning mechanism, and select a better solution as an initial solution of the objective function; Enter the main loop and perform the initial solution update; Check whether the solution is within the boundary range, and correct the out-of-bounds solution based on the preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

[0052] Through the description of the above implementation modes, those skilled in the art can clearly understand that each implementation mode can be implemented by means of software plus a necessary general hardware platform, or of course by hardware. Based on such an understanding, the above technical solution can essentially or in other words be embodied in the form of a software product that contributes to the prior art. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods of each embodiment or some parts of the embodiment.

[0053] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for photovoltaic inspection path planning by unmanned aerial vehicle considering multi-constraint coupling, characterized in that: include: A wind speed model with time window segments is established, and the objective function of the UAV dynamic trajectory optimization problem is defined by considering the UAV energy consumption constraint, time penalty constraint, and photovoltaic inspection path planning of the UAV under wind farm conditions; Set the initialization parameters of the accelerated adversarial algorithm based on a chaotic single candidate; Generate a random solution and an adversarial solution of the objective function according to an accelerated adversarial learning mechanism, and select a better solution as an initial solution of the objective function; Enter the main loop and perform the initial solution update; Check whether the solution is within the boundary range, and correct the out-of-bounds solution based on the preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

2. According to claim 1, a method for planning a photovoltaic inspection path of an unmanned aerial vehicle considering multi-constraint coupling is characterized in that: The wind speed model with time window segments is established, and the objective function of the dynamic trajectory optimization problem of the drone is defined by considering the drone energy consumption constraint, time penalty constraint and the photovoltaic inspection path planning of the drone under the wind field condition. The objective function includes: Obtain wind field data for a preset time period in the future, divide the time window into preset time windows, and extract the global wind speed and wind direction mean of each window; Define the objective function of the UAV dynamic trajectory optimization problem, the expression is: ， In the formula, Fixed cost for drones, is the dynamic energy cost, For the penalty cost, is the surface state function of the photovoltaic panel, is a weight coefficient in the range [0,1]; ， ， ， ， ， In the formula, , , The fixed costs of the first UAV, the second UAV, and the The fixed cost of a drone, is the number of drones, is the unloaded weight of the drone, is the air fluid density, is the blade area of ​​the drone, for Inspection time of the point, Drone from point To point The distance is the actual wind speed of the UAV at time t, is the speed of the drone relative to the air, is the global predicted wind speed value, The angle between the wind direction and due north. The angle between the drone's flight path and true north. is a binary variable, indicating that the drone Whether the task is completed, and are the upper and lower bounds of the time window, respectively. and is the penalty coefficient for exceeding the time window, Arrival The actual time of the point, is the light intensity in the current period, is the reference light intensity, is the current PV panel contamination rate, Cleaning status reference value, is the environmental sensitivity coefficient, is the number of time windows, is the acceleration due to gravity.

3. The method for photovoltaic inspection path planning of unmanned aerial vehicle considering multi-constraint coupling according to claim 1 is characterized in that: The initialization parameters include: the adaptive maximum number of iterations based on the convergence rate of the objective function , the initial maximum number of iterations , the overall dimension of the problem , control factors , Change window time , the number of successful attempts threshold , Iteration number threshold in the exploration phase Dimensions of the problem , maximum jump probability , minimum jump probability and regulatory factors .

4. The method for photovoltaic inspection path planning of unmanned aerial vehicle considering multi-constraint coupling according to claim 1 is characterized in that: Generating a random solution and an opposing solution of the objective function according to the accelerated adversarial learning mechanism, and selecting a better solution as the initial solution of the objective function includes: Determine the maximum number of adaptive iterations based on the convergence rate of the objective function , the expression is: ， In the formula, is the initial maximum number of iterations, is the control factor, For the The best objective function value at the iteration, For the The best objective function value at the iteration, is the change window time; A mixed probability distribution is used to generate random solutions, expressed as: ， In the formula, For Dimension The updated solution of For Dimension The lower bound of For Dimension The lower bound of is a uniformly distributed random number, is the standard normal distribution number, is the random probability, is a random probability, satisfying ; For each random solution , with the jump probability Generate an opposite solution to the current solution , the expression is: ， ， ， In the formula, is the maximum jump probability, is the minimum jump probability, is the regulating factor, For the The speedup factor for the iteration, For the dimension of the problem, is the total dimension of the problem; Comparing random solutions The objective function value and the opposite solution The objective function value of is retained as the initial solution.

5. The method for photovoltaic inspection path planning by unmanned aerial vehicle considering multi-constraint coupling according to claim 1 is characterized in that: The entering of the main loop and performing the initial solution update comprises: When the number of iterations is less than the set value, it enters the exploration phase, where the solution update rule in the exploration phase is: Calculating weights and adaptive weights , the expression is: ， ， In the formula, is an exponential function, Indicates Iterations, is the adaptive maximum number of iterations based on the convergence rate of the objective function; With adaptive weight Use the original SCO algorithm to update the solution, the expression is: ， In the formula, is the updated solution, is a random number; 1- The probability uses a new strategy, which introduces sine / cosine perturbations to prevent the updated solution from converging quickly. The expression is: ， In the formula, is the imaginary number symbol, is a random number, is a step function; When the number of iterations is greater than the set value, it enters the development phase. The solution update rule in the development phase is: If continuous If the update fails, the preset mutation strategy is used to update the solution. The mutation strategy is: The Cauchy variation Trigger mutation, the expression is: ， ， ， In the formula, is a random number between 0 and 1. is the average fitness change of the previous iteration, is the mean change of the optimal fitness of the previous iteration, is the mutation trigger probability factor, is the probability of triggering Cauchy mutation, For Dimension The updated solution of ; Gaussian variation Trigger mutation, the expression is: ， ， In the formula, is the probability of triggering Gaussian mutation, Indicates sampling from a Gaussian distribution; Levy mutation Probability trigger mutation, the expression is: ， ， In the formula, The probability of triggering Levi mutation, for , is the gamma function, for ; If the mutation is not triggered, the candidate solution is updated normally, and the expression is: ， In the formula, Is a random number.

6. The method for planning a photovoltaic inspection path for an unmanned aerial vehicle considering multiple constraints coupling according to claim 1 is characterized in that: The expression of the correction rule is: ， ， ， In the formula, For Dimension The updated solution of For Dimension The lower bound of is the dynamic fractal contraction coefficient, For Dimension The upper bound of It is a very small number, usually 2.22e-16. is the fractal noise function, is the Riemann function, For infinity, is the initial value of the dynamic fractal contraction coefficient, is the attenuation factor, Indicates Iterations, is the adaptive maximum number of iterations based on the convergence rate of the objective function.

7. A UAV photovoltaic inspection path planning system considering multi-constraint coupling, characterized in that: include: The definition module is configured to establish a wind speed model with time window segments, and the objective function of the UAV dynamic trajectory optimization problem is defined by considering the UAV energy consumption constraint, time penalty constraint, and photovoltaic inspection path planning of the UAV under wind farm conditions; A setting module configured to set initialization parameters of an accelerated opposition algorithm based on a chaotic single candidate; A selection module configured to generate a random solution and an adversarial solution of the objective function according to an accelerated adversarial learning mechanism, and select a better solution as an initial solution of the objective function; A loop module, configured to enter a main loop and perform the initial solution update; The output module is configured to check whether the solution is within the boundary range, and to correct the out-of-bounds solution based on a preset correction rule until the maximum number of iterations is reached, and output the solution with the lowest objective function as the inspection path.

8. An electronic device, characterized in that: include: At least one processor, and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to perform the method described in any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.

Images