An unmanned helicopter photovoltaic solar power plant optimal inspection path planning method

Abstract

The application discloses an optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants, which comprises the following steps: (1) obtaining a grid map package of the photovoltaic power plant; (2) dividing the photovoltaic power plant into a plurality of sub-modules; (3) obtaining the flight height h of the unmanned helicopter; (4) calling the grid map package of the photovoltaic power plant, and respectively obtaining the shortest inspection path through all the photovoltaic panel geographic arrays in each sub-module by using an improved ant colony algorithm; (5) determining the inspection sequence of the photovoltaic panel geographic arrays in the sub-modules, and performing local inspection path planning on the photovoltaic panel geographic arrays according to the inspection sequence and the local inspection path planning requirement of the "bow" shape, and integrating the local inspection paths in each sub-module to form the final inspection path of the sub-module; and (6) performing overall translation, lifting and integration processing on the final inspection path of the sub-module according to the flight height of the unmanned helicopter and the gimbal parameters to obtain the global optimal inspection path of the unmanned helicopter. The application can comprehensively inspect the photovoltaic power plant and obtain the optimal inspection path of the unmanned helicopter.

Description

Technical Field

[0001] This invention relates to the field of unmanned helicopter path planning technology, specifically to an optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants. Background Technology

[0002] With increasing energy demand and growing environmental awareness worldwide, more and more photovoltaic (PV) power plants are being built and put into operation. Due to the characteristics of PV power generation—large panel area and numerous panels—PV power plants occupy a large area. Currently, the main inspection method for PV power plants is manual on-site sampling. This not only fails to provide a comprehensive understanding of the module's condition but also involves cumbersome testing methods and a low fault identification rate. Furthermore, some PV power plants are located in complex environments, such as areas with complex terrain or waterways, making manual inspection difficult and posing safety risks.

[0003] In this context, the use of unmanned helicopters in conjunction with photovoltaic (PV) inspection systems not only solves the problem of PV power plants being widely distributed and far apart, but also effectively reduces costs. To further reduce energy consumption during inspections, existing technologies employ numerous algorithms for optimizing PV inspection paths. These algorithms significantly improve the efficiency and automation of PV power plant inspections, reduce operating costs, and produce neat and orderly photos for easier subsequent processing. For example, patent application number "202110693216.0" discloses a method for optimizing UAV inspection paths for power transmission lines based on ant colony optimization. This method combines geographic information data of transmission lines, towers, and the local terrain with satellite imagery, integrating the towers to be inspected and the effective link area of ​​the UAV into a road map. It optimizes the selection of UAV take-off and landing points, establishes a UAV inspection path planning model for power transmission lines, and improves the maximum-minimum ant colony algorithm to enhance the solution efficiency of the UAV inspection path planning model. However, the inspection paths planned using this method cannot cover all PV panels, and the inspection efficiency needs further improvement. Summary of the Invention

[0004] Purpose of the invention: To address the above problems, this invention provides an optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants. This method can comprehensively inspect the status of all photovoltaic panels in the photovoltaic power plant and simultaneously obtain the shortest inspection path for the unmanned helicopter, resulting in high inspection efficiency and low inspection cost.

[0005] Technical Solution: To solve the above problems, this invention discloses an optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants, specifically including the following steps:

[0006] (1) Collect topographic data of photovoltaic power plants, location of charging piles, location coordinates of photovoltaic panel geographic arrays, and serial number of photovoltaic panel geographic arrays to construct a map database of photovoltaic power plants; Based on the map database, establish a path planning space using the raster modeling method to generate a raster map package of photovoltaic power plants;

[0007] (2) Divide the photovoltaic power plant into several sub-modules;

[0008] (3) Calculate and obtain the flight altitude h of the unmanned helicopter relative to the photovoltaic panel;

[0009] (4) Retrieve the grid map package of the photovoltaic power plant and use the improved ant colony algorithm to obtain the shortest inspection path that passes through all photovoltaic panel geographic arrays in each submodule without repeating the path.

[0010] (5) For each submodule, determine the inspection sequence of all photovoltaic panel geographic arrays in the submodule based on the corresponding shortest inspection path and take-off point; plan the local inspection path for each individual photovoltaic panel geographic array in sequence according to the inspection sequence, that is, complete the inspection of all photovoltaic panels in the individual photovoltaic panel geographic array according to the "bow" shaped inspection path; integrate all local inspection paths in each submodule to form the final inspection path of the submodule.

[0011] (6) Calculate the geometric distance Δd between the final inspection path of each sub-module and the inspection path of the unmanned helicopter based on the flight altitude and gimbal parameters of the unmanned helicopter. After shifting the final inspection path of each sub-module by the corresponding Δd meters, obtain the optimal inspection path of that sub-module. Integrate the optimal inspection paths of all sub-modules and raise them by h meters to finally obtain the global optimal inspection path of the unmanned helicopter.

[0012] Furthermore, step (4) involves using an improved ant colony algorithm to obtain the shortest inspection path that traverses all photovoltaic panel geographic arrays in each submodule without repetition. This specifically includes the following steps:

[0013] (4.1) Calculate the distance D between each photovoltaic panel geographical array, and scale down the distance D proportionally to speed up the ant colony algorithm; the specific formula is:

[0014]

[0015] In the formula, D ij The distance between the photovoltaic (PV) geographic array numbered i and the PV geographic array numbered j is represented by m; m represents the scaling factor; C i C represents the coordinates of the center point of the photovoltaic panel geographic array numbered i. j d represents the coordinates of the center point of the photovoltaic panel geographic array numbered j. iThe estimated distance to the geographical array of photovoltaic panels with inspection number i is represented by the following formula:

[0016]

[0017] In the formula, l j b represents the length of the j-th row in the photovoltaic panel geographic array; b represents the total number of rows of photovoltaic panels in the photovoltaic panel geographic array; a j Indicates a length of l j The proportion of photovoltaic panel rows to the total number of rows, B i W1 represents the width of the photovoltaic panel geographic array; W2 represents the width of the photovoltaic panel; alfa represents the installation angle of the photovoltaic panel; n represents the number of rows of photovoltaic panels within the shooting range.

[0018] (4.2) Set the number of ants M, the pheromone heuristic factor α, the expected heuristic factor β, the pheromone volatility coefficient ρ, the pheromone concentration Q, and the maximum number of iterations NC. max Initialize the allowed and tabu lists of the photovoltaic panel geographic arrays to be accessed, and the pheromone concentration τ.

[0019] (4.3) Each ant randomly selects a starting photovoltaic (PV) panel geographic array, calculates the probability of an unselected PV panel geographic array being selected, uses a roulette wheel algorithm to select the next PV panel geographic array, and adds the current PV panel geographic array to its tabu list (tabu). The allowed PV panel geographic array to be visited is updated until all ants have visited all PV panel geographic arrays. The specific formula for the probability of an unselected PV panel geographic array being selected is:

[0020]

[0021] In the formula, τ represents the probability that the photovoltaic panel geographic array not selected by the k-th ant will be selected in the next step; ij Indicates the pheromone concentration from photovoltaic panel geographic array i to target photovoltaic panel geographic array j; η ij Let D represent the visibility of the target photovoltaic array j to the photovoltaic array i. ij The reciprocal of τ; is Indicates the pheromone concentration from photovoltaic panel geographic array i to the target photovoltaic panel geographic array s; η is D represents the visibility of the target photovoltaic (PV) array s to the PV array i. is The reciprocal of;

[0022] (4.4) After all ants have completed the traversal task, calculate the total length L of the path traveled by each ant during this iteration. kThe longest path is compared with the shortest path, and the shortest path obtained in the current iteration is compared with the shortest path in the current completed iteration to retain the optimal shortest path.

[0023] (4.5) Update pheromones, the specific formula is:

[0024] τ ij =(1-ρ)τ ij +Δτ ij

[0025]

[0026]

[0027] In the formula, Q represents the total amount of pheromone in this iteration cycle; N represents the number of ants from photovoltaic panel geographic array i to photovoltaic panel geographic array j; L α L represents the length of the optimal shortest path. β σ represents the length of the longest path in this iteration; σ represents the shortest path L. α The number of ants;

[0028] (4.6) Determine whether the current iteration count has reached the maximum. If the maximum iteration count has been reached, output the current optimal shortest path. If not, return to step (4.2).

[0029] (4.7) Determine whether the shortest inspection path of all sub-modules has been calculated. If all are completed, end the calculation and output the shortest inspection path of each sub-module. If not completed, retrieve the grid map of the next sub-module and go to step (4.1).

[0030] Furthermore, step (5) specifically involves:

[0031] (5.1) Determine the relative distance between the unmanned helicopter and each photovoltaic array in a single submodule based on the takeoff point, select the photovoltaic array with the shortest relative distance as the starting photovoltaic array for inspection, and determine the inspection order of the photovoltaic arrays in the submodule based on the shortest inspection path corresponding to the submodule; construct a path table to store waypoints;

[0032] (5.2) Retrieve the grid map of the first photovoltaic panel geographic array in the inspection sequence and put the take-off point into the path table;

[0033] (5.3) Set the last point in the path table as the entry point for the inspection of the photovoltaic panel geographic array. The last point in the path table is the departure point or take-off point of the previous photovoltaic panel geographic array. For the current photovoltaic panel geographic array to be inspected, select the corner point of the photovoltaic panel geographic array closest to the entry point and set it as the planning starting point.

[0034] (5.4) The unmanned helicopter moves along the set inspection direction from the planned starting point; and sets the planned starting point as the current node and adds it to the path table;

[0035] (5.5) Starting from the current node, move p1 meters to the new node according to the "bow" shaped inspection path requirements. The new node is the inflection point of the "bow" shaped trajectory. Replace the current node with the new node and put it into the path table. The size of p1 is the length of the longest row in the photovoltaic panel geographic array within the current shooting range.

[0036] (5.6) Determine whether all photovoltaic panels in the photovoltaic array to be inspected have been inspected. If the inspection is completed, set the current node as the outbound point of the photovoltaic array to be inspected and also as the inbound point of the next photovoltaic array to be inspected, and proceed to step (5.7). If the inspection is not completed, move to a new node and replace the current node with the current node based on the current node and the position information of the longest row in the next shooting range, put it in the path table, change the inspection direction, and proceed to step (5.5).

[0037] (5.7) Determine whether all photovoltaic panel geographic arrays in the submodule have been inspected; if they have been inspected, proceed to step (5.8); if they have not been inspected, retrieve the next photovoltaic panel geographic array grid map in the submodule according to the inspection order of the photovoltaic panel geographic arrays, and proceed to step (5.3).

[0038] (5.9) Output the path table to obtain the final inspection path of a single submodule.

[0039] Furthermore, the specific method for obtaining the flight altitude h of the unmanned helicopter relative to the photovoltaic panel in step (3) is as follows:

[0040] (3.1) Calculate the ineffective distance L1 blocked by the photovoltaic panel based on the width of the photovoltaic panel geographic array, the installation angle of the photovoltaic panel, the pitch angle of the unmanned helicopter gimbal, and the field of view angle of the camera. The formula is:

[0041] L1=W2×sin(alfa)×tan(90-beta+0.5×ang)

[0042] In the formula, W2 represents the width of the photovoltaic panel; alfa represents the installation angle of the photovoltaic panel; beta represents the pitch angle of the unmanned helicopter gimbal; and ang represents the field of view angle.

[0043] (3.2) Calculate the distance L2 projected from the photovoltaic panel geographic array to the ground based on the width of the photovoltaic panel geographic array and the installation angle of the photovoltaic panels. The formula is:

[0044] L2 = n × W1 + n × W2 × cos(alfa)

[0045] In the formula, W1 represents the spacing between two consecutive photovoltaic panel arrays; n represents the number of rows of photovoltaic panels within the shooting range;

[0046] (3.3) The flight altitude h of the unmanned helicopter relative to the photovoltaic panel is calculated using the following formula:

[0047]

[0048] Furthermore, step (6) specifically involves:

[0049] (6.1) Calculate the geometric distance Δd between the final inspection path of each submodule and the inspection path of the unmanned helicopter based on the flight altitude and gimbal parameters of the unmanned helicopter. The formula is as follows:

[0050] Δd = h × cos(beta)

[0051] In the formula, Δd represents the geometric distance between the final inspection path of each submodule and the inspection path of the unmanned helicopter; beta represents the pitch angle of the unmanned helicopter gimbal.

[0052] (6.2) The optimal inspection path of each submodule is obtained by shifting the final inspection path of each submodule by the corresponding Δd meters.

[0053] (6.3) Integrate the optimal inspection paths of all sub-modules and raise them by h meters to finally obtain the global optimal inspection path of the unmanned helicopter.

[0054] Furthermore, step (2) specifically involves:

[0055] (2.1) Calculate the inspection path length L of a single photovoltaic panel geographic array i The calculation formula is:

[0056]

[0057] In the formula, l j b represents the length of the j-th row in the photovoltaic panel geographic array; b represents the total number of rows of photovoltaic panels in the photovoltaic panel geographic array; a j Indicates a length of l j The proportion of photovoltaic panel rows to the total number of rows, B i W1 represents the width of the photovoltaic panel geographic array; W2 represents the width of the photovoltaic panel; alfa represents the installation angle of the photovoltaic panel; n represents the number of rows of photovoltaic panels within the shooting range.

[0058] (2) Calculate the maximum range L of the unmanned helicopter. UVA Its expression is:

[0059] L UVA =h UVA ×V UVA

[0060] In the formula, h UVA V represents the endurance of an unmanned helicopter. UVA This indicates the inspection speed of the unmanned helicopter;

[0061] (3) Based on the inspection path length L of each photovoltaic panel geographical array i The location of charging stations and the range of the unmanned helicopter (L) UVA Dividing the entire photovoltaic power plant into several sub-modules minimizes the number of inspections required for unmanned helicopters to simultaneously meet charging needs and inspect all photovoltaic panel arrays in the power plant.

[0062] Furthermore, the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the steps of any of the above-described methods. A computer-readable storage medium storing a computer program thereon, wherein the computer program, when executed by a processor, implements the steps of any of the above-described methods.

[0063] Beneficial Effects: Compared with existing technologies, the optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants described in this invention has the following significant advantages: 1. By proportionally reducing the distance between photovoltaic panels, the ant colony algorithm is improved, thereby accelerating the computation speed of the ant colony algorithm and obtaining the shortest inspection path passing through the geographical array of all photovoltaic panels in the submodule, while improving the speed of path planning; 2. An inspection algorithm is proposed to perform local planning on the geographical array of photovoltaic panels in the submodule, planning a "bow"-shaped path. Finally, the optimized local paths are integrated to obtain the final inspection path. The optimal inspection path for the unmanned helicopter photovoltaic power plant is obtained based on the geometric relationship between the final inspection path and the unmanned helicopter inspection path, realizing the most comprehensive inspection of the status of all photovoltaic panels in the photovoltaic power plant, while improving inspection efficiency. Attached Figure Description

[0064] Figure 1 The diagram shown is a schematic diagram of the photovoltaic power plant described in this invention;

[0065] Figure 2 The diagram shown is a flowchart of the photovoltaic power plant grid modeling process described in this invention.

[0066] Figure 3 The diagram shown is a schematic diagram of the sub-modules of the photovoltaic power plant described in this invention;

[0067] Figure 4The image shown is a digital map model of the first submodule in the submodules described in this invention;

[0068] Figure 5 The diagram shown is a flowchart of the ant colony algorithm described in this invention.

[0069] Figure 6 The diagram shown is a schematic diagram of the shortest inspection path for the first submodule of the present invention.

[0070] Figure 7 The diagram shows the flowchart of the local planning algorithm for each submodule described in this invention.

[0071] Figure 8 The diagram shown is a local inspection path of a single photovoltaic geographic array in the first sub-module of the present invention.

[0072] Figure 9 The diagram shows the optimal inspection path of the unmanned helicopter for a single photovoltaic geographic array in the first submodule according to the present invention.

[0073] Figure 10 The diagram shows the optimal inspection path of the unmanned helicopter in the first sub-module according to the present invention. Detailed Implementation

[0074] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0075] This invention provides an optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants, the specific steps of which include:

[0076] Step 1, such as Figure 1 As shown, based on actual infrastructure data, including: topographic data of photovoltaic power plants, locations of charging piles, coordinates of photovoltaic panel geographic array locations, and photovoltaic panel geographic array serial numbers, a map database for photovoltaic power plants is established. A planning space is constructed from this map database, and the planning space is discretized using a raster grid to obtain a raster map package for the photovoltaic power plants. For example... Figure 2 As shown, the specific steps include:

[0077] (1) Scan the overall site plan of the photovoltaic power plant and establish a map database, which includes the boundaries of the photovoltaic power plant, terrain data, the size of the photovoltaic panels, the location of the photovoltaic panels, etc.

[0078] (2) Using the Northeast-Eastern Celestial Rectangular Coordinate System, a rectangular planning space is established based on the boundary of the entire photovoltaic power plant. The boundary of this planning space is larger than the boundary of the entire photovoltaic power plant. The origin of the coordinate system is the southwest reference point of the terrain data, and the height is relative height, with the height of the reference point being the height of the photovoltaic power plant.

[0079] (3) Convert the topographic data of the photovoltaic power plant and the location of the photovoltaic panels into coordinates in the northeast-central rectangular coordinate system;

[0080] (4) Determine the size of the grid cell based on the size of the unmanned helicopter, the terrain data of the photovoltaic power plant, and other information;

[0081] (5) Discretize the three dimensions of the planning space according to the corresponding grid length to form a space composed of i×j×k grids, which is mathematically represented by a three-dimensional matrix array;

[0082] (6) Determine the location based on whether there is a photovoltaic panel in each grid. Figure 3 The elements of the dimensional matrix array are either 1 or 0;

[0083] (7) Form a complete raster map package, which includes the planned space size, raster size, number of rasters, terrain data, photovoltaic panel location information, photovoltaic panel geographic array information, and the map composed of all rasters. Figure 3 3D matrix.

[0084] Step Two: Based on the established map database of the photovoltaic power plant and considering the unmanned helicopter's endurance, flight speed, and other performance characteristics, the original photovoltaic power plant is divided into several modules to ensure that the unmanned helicopter achieves optimal inspection efficiency under performance constraints. Specifically, this includes:

[0085] (1) Calculate the length of each row of a single photovoltaic panel geographic array, and then, in conjunction with the camera's shooting range, calculate the estimated "bow"-shaped photovoltaic inspection path, and calculate the inspection path length L of a single photovoltaic panel geographic array. i The calculation formula is:

[0086]

[0087] In the formula, l j b represents the length of the j-th row in the photovoltaic panel geographic array; b represents the total number of rows of photovoltaic panels in the photovoltaic panel geographic array; a j Indicates a length of l j The proportion of photovoltaic panel rows to the total number of rows, B i W1 represents the width of the photovoltaic panel geographic array; W2 represents the width of the photovoltaic panel; alfa represents the installation angle of the photovoltaic panel; n represents the number of rows of photovoltaic panels within the shooting range.

[0088] (2) Calculate the maximum range L of the unmanned helicopter based on its performance parameters such as endurance and flight speed. UVA Its expression is:

[0089] L UVA =h UVA ×V UVA

[0090] In the formula, hUVA V represents the endurance of an unmanned helicopter. UVA This indicates the inspection speed of the unmanned helicopter;

[0091] (3) Based on the inspection path length L of each photovoltaic panel geographical array i The location of charging stations and the range of the unmanned helicopter (L) UVA , using L i / L UVA The number of unmanned helicopter inspections is estimated, and based on the number of inspections, the number of inspections required to inspect all photovoltaic panels in the photovoltaic power plant's geographical array while meeting the charging needs of the unmanned helicopter is minimized.

[0092] like Figure 3 As shown, in this embodiment, the photovoltaic power plant is divided into four modules, designated as sub-modules 1 to 4, numbered 1-4. The locations of the charging piles are marked with the letter A; modules closer to the charging piles contain more photovoltaic panels. Figure 4 The image shown is a raster map of the first submodule, containing several geographic arrays of photovoltaic panels.

[0093] Step 3: Calculate the geometric position of the unmanned helicopter relative to the center of the image of the photovoltaic panel being photographed;

[0094] (1) Calculate the ineffective distance L1 blocked by the photovoltaic panel based on the width of the photovoltaic panel geographic array, the installation angle of the photovoltaic panel, the pitch angle of the unmanned helicopter gimbal, and the field of view angle of the camera. The formula is:

[0095] L1=W2×sin(alfa)×tan(90-beta+0.5×ang)

[0096] In the formula, invalid distance L1 represents the distance without photovoltaic panels within the shooting range; W2 represents the width of the photovoltaic panel; alfa represents the installation angle of the photovoltaic panel; beta represents the pitch angle of the unmanned helicopter gimbal; and ang represents the field of view angle.

[0097] (2) Calculate the distance L2 projected from the photovoltaic panel geographic array to the ground based on the width of the photovoltaic panel geographic array and the installation angle of the photovoltaic panels. The formula is:

[0098] L2 = n × W1 + n × W2 × cos(alfa)

[0099] In the formula, W1 represents the spacing between two consecutive photovoltaic panel arrays; n represents the number of rows of photovoltaic panels within the shooting range;

[0100] (3.3) The flight altitude h of the unmanned helicopter relative to the photovoltaic panel is calculated using the following formula:

[0101]

[0102] In this embodiment, W2 is 3.327, alfa is 28°, beta is 80°, and ang is 34°. W1 is 6.2, and n is 3.

[0103] Step 4: Retrieve the grid map package of the photovoltaic power plant and use an improved ant colony algorithm to obtain the shortest inspection path that passes through all photovoltaic panel geographic arrays in each submodule without repetition; for example... Figure 5 As shown, the specific process is as follows:

[0104] (1) Calculate the center point CP of all photovoltaic array geographic arrays within the submodule. i Then CP i Store in array C;

[0105] (2) Calculate the estimated distance d for inspecting a single photovoltaic panel array. i The formula is:

[0106]

[0107] In the formula, l j b represents the length of the j-th row in the photovoltaic panel geographic array; b represents the total number of rows of photovoltaic panels in the photovoltaic panel geographic array; a j Indicates a length of l j The proportion of photovoltaic panel rows to the total number of rows, B i W1 represents the width of the photovoltaic panel geographic array; W2 represents the width of the photovoltaic panel; alfa represents the installation angle of the photovoltaic panel; n represents the number of rows of photovoltaic panels within the shooting range.

[0108] (3) Calculate the distance D between each photovoltaic panel geographical array, and scale down the distance D proportionally to speed up the ant colony algorithm; the specific formula is:

[0109]

[0110] In the formula, D ij The distance between the photovoltaic (PV) geographic array numbered i and the PV geographic array numbered j is represented by m; m represents the scaling factor; C i C represents the coordinates of the center point of the photovoltaic panel geographic array numbered i. j d represents the coordinates of the center point of the photovoltaic panel geographic array numbered j. i This represents the estimated distance traveled after inspecting the photovoltaic panel geographic array numbered i.

[0111] (4) Set the number of ants M, pheromone heuristic factor α, expected heuristic factor β, pheromone volatility coefficient ρ, pheromone concentration Q, and maximum number of iterations NC. max Initialize the allowed and tabu lists of the photovoltaic panel geographic arrays to be accessed, and the pheromone concentration τ.

[0112] (5) M ants randomly select the starting photovoltaic (PV) panel geographical array. Based on path heuristics and pheromone concentration, they calculate the probability of an unselected PV panel geographical array being selected. They then use a roulette wheel algorithm to select the next PV panel geographical array and add the selected array to their respective tabu lists (tabu). The list of PV panel geographical arrays to be visited (allowed) is updated until all ants have traversed all PV panel geographical arrays. The specific formula for the probability of an unselected PV panel geographical array being selected is:

[0113]

[0114] In the formula, τ represents the probability that the photovoltaic panel geographic array not selected by the k-th ant will be selected in the next step; ij Indicates the pheromone concentration from photovoltaic panel geographic array i to target photovoltaic panel geographic array j; η ij Let D represent the visibility of the target photovoltaic array j to the photovoltaic array i. ij The reciprocal of τ; is Indicates the pheromone concentration from photovoltaic panel geographic array i to the target photovoltaic panel geographic array s; η is D represents the visibility of the target photovoltaic (PV) array s to the PV array i. is The reciprocal of;

[0115] (6) After all ants have completed their traversal, calculate the total length L of the path traveled by each ant during this iteration. k The longest path is compared with the shortest path, and the shortest path obtained in the current iteration is compared with the shortest path in the current completed iteration to retain the optimal shortest path.

[0116] (7) Calculate the newly added pheromone Δτ in the path search during this iteration. ij Then, some of the previous pheromones are evaporated, and the newly added pheromones are added. The iteration count NC is then incremented by 1. The formula for the newly added pheromones is:

[0117]

[0118] The formula for updating pheromones is:

[0119] τ ij =(1-ρ)τij +Δτ ij

[0120]

[0121] In the formula, Q represents the total amount of pheromone in this iteration cycle; N represents the number of ants from photovoltaic panel geographic array i to photovoltaic panel geographic array j; L α L represents the length of the optimal shortest path. β σ represents the length of the longest path in this iteration; σ represents the shortest path L. α The number of ants;

[0122] (8) Determine whether the current iteration count has reached the maximum. If the maximum iteration count has been reached, output the current optimal shortest path. If not, return to step (4).

[0123] (9) Determine whether the shortest inspection path of all sub-modules has been calculated. If all are completed, end the calculation and output the shortest inspection path of each sub-module. If not completed, retrieve the grid map of the next sub-module and proceed to step (1).

[0124] In this embodiment, the shortest inspection path that passes through all photovoltaic panel geographic arrays in the first module without repetition is obtained, as follows: Figure 6 As shown.

[0125] Step 5: For each submodule, by determining the relative position of the unmanned helicopter and the photovoltaic panel geographic array, calculate the entry point for the unmanned helicopter to enter the photovoltaic panel geographic array. Then, based on the image acquisition requirements, plan a "bow"-shaped local inspection path. Integrate the local inspection paths from each submodule to form the final inspection path for that submodule. The final inspection path for that submodule is the path to the center point of the acquired image. For example... Figure 7 As shown, specifically:

[0126] (1) Determine the relative distance between the unmanned helicopter and each photovoltaic array in a single submodule based on the takeoff point, select the photovoltaic array with the shortest relative distance as the starting photovoltaic array for inspection, and determine the inspection order of the photovoltaic arrays in the submodule based on the shortest inspection path corresponding to the submodule; construct a path table to store waypoints;

[0127] (2) Retrieve the grid map of the first photovoltaic panel geographic array in the inspection sequence and put the take-off point into the path table;

[0128] (3) Set the last point in the path table, i.e. the departure point or take-off point of the previous photovoltaic array, as the entry point for the inspection of the photovoltaic array; for the current photovoltaic array to be inspected, select the corner point of the photovoltaic array closest to the entry point and set it as the planning starting point;

[0129] (4) The unmanned helicopter moves along the set inspection direction from the planned starting point; and sets the planned starting point as the current node and puts it into the path table; where the inspection direction is from east to west or from west to east, depending on the relative position of the planned starting point and the photovoltaic array.

[0130] (5) Starting from the current node, move p1 meters to the new node according to the "bow" shaped inspection path requirements. The new node is the inflection point of the "bow" shaped trajectory. Replace the current node with the new node and put it into the path table. The size of p1 is the length of the longest row in the photovoltaic panel geographic array within the current shooting range.

[0131] (6) Determine whether the current photovoltaic array to be inspected has been inspected. If the inspection is completed, set the new current node as the outbound point of the current photovoltaic array to be inspected, and also as the inbound point of the next photovoltaic array to be inspected, and proceed to step (7); if the inspection is not completed, move to a new node and replace it with the current node according to the current node and the position information of the longest row in the next shooting range, put it into the path table, change the inspection direction, and proceed to step (5).

[0132] (7) Determine whether all photovoltaic panel geographic arrays in the submodule have been inspected; if they have been inspected, proceed to step (8); if they have not been inspected, retrieve the next photovoltaic panel geographic array grid map in the submodule according to the inspection order of the photovoltaic panel geographic arrays, and proceed to step (3).

[0133] (8) Output the path table and integrate the local inspection paths of the photovoltaic geographic array to form the final inspection path of a single sub-module.

[0134] In this embodiment, the first submodule performs local path planning for a single photovoltaic geographic array according to the "bow" shaped inspection requirements, as follows: Figure 8 As shown.

[0135] Step Six: Obtain the globally optimal inspection path for the unmanned helicopter. This includes:

[0136] (1) Calculate the geometric distance Δd between the final inspection path of each submodule and the inspection path of the unmanned helicopter based on the flight altitude and gimbal parameters of the unmanned helicopter. The formula is as follows:

[0137] Δd = h × cos(beta)

[0138] In the formula, Δd represents the geometric distance between the final inspection path of each submodule and the inspection path of the unmanned helicopter; beta represents the pitch angle of the unmanned helicopter gimbal.

[0139] (2) The optimal inspection path of each sub-module is obtained by shifting the final inspection path of each sub-module by the corresponding Δd meters.

[0140] (3) Integrate the optimal inspection paths of all sub-modules and raise them by h meters to finally obtain the global optimal inspection path of the unmanned helicopter.

[0141] In this embodiment, as Figure 9 As shown, in the first submodule, after local path planning is performed for a single photovoltaic geographic array according to the "bow" shaped inspection requirements, the local path is shifted as a whole by Δd meters to obtain the optimal inspection path for a single photovoltaic geographic array in this submodule. Figure 10 As shown, the optimal inspection path for the unmanned helicopter within the first submodule is obtained. After completing the inspection of a submodule, the unmanned helicopter returns to the charging station for charging. After charging, it flies to other submodules to continue inspection. In summary, the method described in this invention can inspect all photovoltaic panels in a photovoltaic power plant, and the inspection path is the optimal one.

Claims

1. A method for optimal inspection path planning in an unmanned helicopter photovoltaic solar power plant, characterized in that, Includes the following steps: (1) Collect topographic data of photovoltaic power plants, location of charging piles, location coordinates of photovoltaic panel geographic arrays, and serial number of photovoltaic panel geographic arrays to construct a map database of photovoltaic power plants; Based on the map database, establish a path planning space using the raster modeling method to generate a raster map package of photovoltaic power plants; (2) Divide the photovoltaic power plant into several sub-modules; (3) Calculate and obtain the flight altitude of the unmanned helicopter relative to the photovoltaic panel. ; (4) Retrieve the grid map package of the photovoltaic power plant and use the improved ant colony algorithm to obtain the shortest inspection path that passes through all photovoltaic panel geographic arrays in each sub-module without repeating the path. (5) For each submodule, determine the inspection sequence of all photovoltaic panel geographic arrays in that submodule based on the corresponding shortest inspection path and takeoff point; According to the inspection sequence, local inspection paths are planned for each individual photovoltaic panel geographic array, that is, the inspection of all photovoltaic panels in the individual photovoltaic panel geographic array is completed according to the "bow" shaped inspection path; all local inspection paths in each sub-module are integrated to form the final inspection path of the sub-module. (6) Calculate the geometric distance between the final inspection path of each sub-module and the inspection path of the unmanned helicopter based on the flight altitude of the unmanned helicopter and the gimbal parameters of the unmanned helicopter. The final inspection path of each submodule will be shifted as a whole to the corresponding... After obtaining the optimal inspection path for this submodule, integrate the optimal inspection paths of all submodules and raise them. The optimal global inspection path for the unmanned helicopter was ultimately obtained by measuring meters. Step (4) involves using an improved ant colony algorithm to obtain the shortest inspection path that traverses all photovoltaic panel geographic arrays in each submodule without repetition. This specifically includes the following steps: (4.1) Calculate and obtain the distance between each photovoltaic panel geographic array. , distance Scaling down proportionally to speed up the ant colony algorithm; the specific formula is: In the formula, Indicates the number is The geographical array of photovoltaic panels is numbered as The distance of the geographical array of photovoltaic panels; Indicates the scaling factor; Indicates the number is The coordinates of the center point of the photovoltaic panel geographic array. Indicates the number is The coordinates of the center point of the photovoltaic panel geographic array. Indicates the inspection number is The estimated path of the photovoltaic panel geographical array is calculated using the following formula: In the formula, Indicates the first in the geographic array of photovoltaic panels The length of the line; This represents the total number of rows of photovoltaic panels in the photovoltaic panel geographic array; Indicates length is The proportion of photovoltaic panel rows to the total number of rows. Indicates the width of the photovoltaic panel geographic array; This indicates the spacing between two consecutive photovoltaic panel geographic arrays. Indicates the width of the photovoltaic panel; Indicates the angle at which the photovoltaic panel is installed; Indicates the number of rows of photovoltaic panels within the shooting range; (4.2) Set the number of ants Pheromones as heuristic factors Expected heuristic factor pheromone volatility coefficient pheromone concentration Maximum number of iterations Initialize the geographic array of photovoltaic panels to be accessed. Contraindications list pheromone concentration ; (4.3) Each ant randomly selects the starting photovoltaic panel geographic array, calculates the probability of the unselected photovoltaic panel geographic array being selected, uses the roulette wheel algorithm to select the next photovoltaic panel geographic array, and puts the current photovoltaic panel geographic array into its respective tabu list. In the middle, update the geographic array of photovoltaic panels to be visited. The process continues until all ants have traversed all the photovoltaic panel geographic arrays; the specific formula for the probability of an unselected photovoltaic panel geographic array being selected is as follows: In the formula, Indicates the first The probability that a photovoltaic panel geographic array that was not selected by an ant will be selected next; Indicates the geographic array of photovoltaic panels To the target photovoltaic panel geographic array pheromone concentration; Indicates the geographic array of the target photovoltaic panel Geographic array of photovoltaic panels Visibility, specifically The reciprocal of; Indicates the geographic array of photovoltaic panels To the target photovoltaic panel geographic array pheromone concentration; Indicates the geographic array of the target photovoltaic panel Geographic array of photovoltaic panels Visibility, specifically The reciprocal of; (4.4) After all ants have completed the traversal task, calculate the total length of the path traveled by each ant during this iteration. The longest path is compared with the shortest path, and the shortest path obtained in the current iteration is compared with the shortest path in the current completed iteration to retain the optimal shortest path. (4.5) Update pheromones, the specific formula is as follows: In the formula, This indicates the total amount of pheromones in this iteration cycle; Indicates the geographic array of photovoltaic panels To photovoltaic geographic array The number of ants; This represents the length of the optimal shortest path; This indicates the length of the longest path in the current iteration loop; Indicates taking the shortest path The number of ants; (4.6) Determine whether the current iteration count has reached the maximum. If the maximum iteration count has been reached, output the current optimal shortest path. If not, return to step (4.2). (4.7) Determine whether the shortest inspection path of all sub-modules has been calculated. If all are completed, end the calculation and output the shortest inspection path of each sub-module. If not completed, retrieve the grid map of the next sub-module and proceed to step (4.1).

2. The optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants according to claim 1, characterized in that, Step (5) is as follows: (5.1) Determine the relative distance between the unmanned helicopter and each photovoltaic array within a single submodule based on the takeoff point, select the photovoltaic array with the shortest relative distance as the starting photovoltaic array for inspection, and determine the inspection sequence of the photovoltaic arrays in that submodule based on the shortest inspection path corresponding to that submodule; construct The table is used to store waypoints; (5.2) Retrieve the grid map of the first photovoltaic panel geographic array in the inspection sequence and place the takeoff point into it. In the table; (5.3) will The last point in the table is set as the inbound point for inspecting the geographic array of photovoltaic panels. The last point in the table is the departure point or takeoff point of the previous photovoltaic panel geographic array; for the current photovoltaic panel geographic array to be inspected, select the corner point of the photovoltaic panel geographic array closest to the entry point and set it as the planning starting point; (5.4) The unmanned helicopter moves along the set inspection direction from the planned starting point; and sets the planned starting point as the current node, and places it into the inspection area. In the table; (5.5) Starting from the current node, move according to the "bow" shaped inspection path requirements. Move to a new node, which is the inflection point of the "bow"-shaped trajectory. Replace the current node with the new node and place it... The table; the description The size is the length of the longest row in the geographic array of photovoltaic panels within the current shooting range; (5.6) Determine whether all photovoltaic panels in the current photovoltaic panel geographic array to be inspected have been inspected. If the inspection is completed, set the current node as the outbound point of the current photovoltaic panel geographic array to be inspected, and also as the inbound point of the next photovoltaic panel geographic array to be inspected, and proceed to step (5.7). If the inspection is not completed, based on the current node and the location information of the longest row within the next shooting range, move to a new node and replace the current node, then place it. In the table, change the inspection direction and proceed to step (5.5). (5.7) Determine whether all photovoltaic panel geographic arrays within the submodule have been inspected; If the inspection has been completed, proceed to step (5.8); if the inspection has not been completed, retrieve the next photovoltaic panel geographic array grid map in the sub-module according to the inspection sequence of the photovoltaic panel geographic array, and proceed to step (5.3). (5.9) Output The table retrieves the final inspection path for a single submodule.

3. The optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants according to claim 1, characterized in that, The flight altitude of the unmanned helicopter relative to the photovoltaic panel in step (3) The specific method of obtaining it is as follows: (3.1) Calculate the ineffective distance blocked by the photovoltaic panel based on the width of the photovoltaic panel geographic array, the installation angle of the photovoltaic panel, the pitch angle of the unmanned helicopter gimbal, and the field of view angle of the camera. The formula is: In the formula, Indicates the width of the photovoltaic panel; Indicates the angle at which the photovoltaic panel is installed; This indicates the pitch angle of the unmanned helicopter gimbal; Indicates the field of view angle; (3.2) Calculate the distance from the photovoltaic panel geographic array projected onto the ground based on the width of the photovoltaic panel geographic array and the installation angle of the photovoltaic panels. The formula is: In the formula, This indicates the spacing between two consecutive photovoltaic panel geographic arrays; Indicates the number of rows of photovoltaic panels within the shooting range; (3.3) Calculate the flight altitude of the unmanned helicopter relative to the photovoltaic panel. The formula is: 。 4. The optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants according to claim 1, characterized in that, Step (6) specifically involves: (6.1) Calculate the geometric distance between the final inspection path of each submodule and the inspection path of the unmanned helicopter based on the flight altitude of the unmanned helicopter and the gimbal parameters of the unmanned helicopter. The formula is: In the formula, This represents the geometric distance between the final inspection path of each submodule and the inspection path of the unmanned helicopter; This indicates the pitch angle of the unmanned helicopter gimbal; (6.2) Shift the final inspection path of each sub-module as a whole to the corresponding location. After measuring meters, obtain the optimal inspection path for this submodule; (6.3) Integrate the optimal inspection paths of all sub-modules and raise them. The optimal global inspection path for the unmanned helicopter was finally obtained by measuring meters.

5. The optimal inspection path planning method for unmanned helicopter photovoltaic solar power plants according to claim 1, characterized in that, Step (2) specifically involves: (2.1) Calculating the inspection path length of a single photovoltaic panel geographic array. The calculation formula is: In the formula, Indicates the first in the geographic array of photovoltaic panels The length of the line; This represents the total number of rows of photovoltaic panels in the photovoltaic panel geographic array; Indicates length is The proportion of photovoltaic panel rows to the total number of rows. Indicates the width of the photovoltaic panel geographic array; This indicates the spacing between two consecutive photovoltaic panel geographic arrays. Indicates the width of the photovoltaic panel; Indicates the angle at which the photovoltaic panel is installed; Indicates the number of rows of photovoltaic panels within the shooting range; (2) Calculate the maximum range of the unmanned helicopter. Its expression is: In the formula, This indicates the flight time of the unmanned helicopter. This indicates the inspection speed of the unmanned helicopter; (3) Based on the inspection path length of each photovoltaic panel geographical array Location of charging stations and range of unmanned helicopters Dividing the entire photovoltaic power plant into several sub-modules minimizes the number of inspections required for unmanned helicopters to inspect all photovoltaic panel arrays while simultaneously meeting charging needs.

6. A computer device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 5.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method described in any one of claims 1 to 5.

Images