Inspection method and device for photovoltaic inspection area, storage medium, electronic device and computer program product

Abstract

The invention discloses an inspection method and device for a photovoltaic inspection area, a storage medium, an electronic device and a computer program product, and relates to the field of photovoltaic power station detection.The inspection method for the photovoltaic inspection area comprises the steps that N inspection points are determined in the photovoltaic inspection area, the photovoltaic inspection area is provided with a plurality of photovoltaic power stations, and N is an integer greater than or equal to 1; images shot by the unmanned aerial vehicle at the N inspection points can cover a plurality of photovoltaic power generation stations, and N is an integer greater than or equal to 2; m target inspection routes are determined according to the positions of preset take-off and landing points and the positions of the N inspection points, the preset take-off and landing points are take-off points and landing points of the M unmanned aerial vehicles, and a passing point set corresponding to the M target inspection routes comprises the N inspection points; the passing points except the preset take-off and landing points in the passing point set corresponding to each target inspection route in the M target inspection routes do not coincide; and controlling the M unmanned aerial vehicles to inspect the photovoltaic inspection area according to the M target inspection routes.

Description

Technical Field

[0001] The present application relates to the field of photovoltaic power station detection, and more specifically, to a photovoltaic inspection area inspection method and device, a storage medium, an electronic device, and a computer program product. Background Art

[0002] In recent years, with the continuous development and maturity of photovoltaic power generation technology, the national photovoltaic installed capacity has achieved a substantial growth. The construction of photovoltaic power stations is highly dispersed, the geographical environment is remote, and the area is large, which makes the on-site inspection work of staff greatly affected by the terrain. Especially in complex terrain conditions such as mountains and water surfaces (including tidal flats and waters with fish-light complementary mode, etc.), manual inspections of photovoltaic sites have additional difficulties. The traffic time of some inspection items accounts for more than 70% of the inspection time, and the inspection efficiency is low. The traditional manual inspection method is no longer suitable for large-scale photovoltaic power station groups. Because drones have the advantages of not being restricted by rugged terrain, being able to take a large range of aerial photos from the air, and having high scanning efficiency, they have been used to a certain extent in the field of photovoltaic inspections and have achieved good inspection results.

[0003] However, in the actual process of inspecting the photovoltaic power station area by drone, a single drone is usually used to scan back and forth in rows to collect image data of all photovoltaic power station strings in the area. This method is currently only applicable to centralized photovoltaic power station areas with relatively regular inspection distribution. The inspection effect is poor in areas where the photovoltaic panels are not distributed regularly due to terrain restrictions. In addition, due to the load and endurance of the drone, it is difficult for a single drone to complete all shooting tasks in a single flight. The round-trip battery replacement process requires a certain amount of flight time, which also affects the inspection efficiency.

[0004] With regard to the problem of low efficiency of single UAV inspection of photovoltaic power station area in related technologies, no effective solution has been proposed so far.

[0005] Therefore, it is necessary to improve the related technology to overcome the above-mentioned defects in the related technology. Summary of the invention

[0006] The embodiments of the present application provide a method and device for inspecting a photovoltaic inspection area, a storage medium, an electronic device, and a computer program product to at least solve the problem of low efficiency of inspecting a photovoltaic power station area by a single drone.

[0007] According to one aspect of an embodiment of the present application, a method for inspecting a photovoltaic inspection area is provided, comprising: determining N inspection points in the photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and images captured by a drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2; determining M target inspection routes according to the positions of preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, and the waypoint sets corresponding to the M target inspection routes include the N inspection points, and the waypoints in the waypoint set corresponding to each of the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2; and controlling the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at the waypoints on the corresponding target inspection route to obtain inspection images.

[0008] In an exemplary embodiment, N inspection points are determined in a photovoltaic inspection area, including: determining the flight altitude of the drone based on parameter information of an image acquisition device in the drone and image requirements of the inspection image; determining the effective area of ​​the inspection image taken by the drone based on the flight altitude of the drone; and determining N inspection points in the photovoltaic inspection area based on the effective area of ​​the inspection image.

[0009] In an exemplary embodiment, M target inspection routes are determined according to the positions of the preset take-off and landing points and the positions of the N inspection points, including: determining an initial set according to the Z waypoints, wherein the initial set contains multiple sequences, each of the multiple sequences is divided into a front area and a rear area, the front area has Z values, the Z values ​​are respectively the numbers of the Z waypoints, the rear area has M-1 values, the M-1 values ​​correspond to the M-1 breakpoint positions of the front area, the M-1 values ​​are not equal to each other, the front area is divided into M segments by the M-1 values, each segment corresponds to an inspection route, and when the N inspection points do not include the preset take-off and landing points, the The Z waypoints are the N inspection points. When the N inspection points include the preset take-off and landing points, the Z waypoints are the N-1 inspection points among the N inspection points excluding the preset take-off and landing points. The initial set is processed using a genetic algorithm to obtain a target set, wherein the genetic algorithm is used to generate a new sequence using one or two sequences in the initial set. A sequence with the highest parameter value is determined from the target set to obtain a target sequence, wherein the parameter value is inversely proportional to a first distance, and the first distance is equal to a route distance of the longest inspection route among the M inspection routes corresponding to the sequence. The M inspection routes corresponding to the target sequence are determined as the M target inspection routes.

[0010] In an exemplary embodiment, the initial set is processed using a genetic algorithm to obtain a target set, including: cyclically performing the following operations on the initial set until a first preset condition is satisfied, and the initial set satisfying the first preset condition is determined as the target set, wherein the first preset condition includes: i equals P, or the initial set is not updated after the following operations are performed, P is a preset number of iterations, and i equals 0 when the following operations are performed for the first time: randomly selecting a sequence from the initial set to obtain a first initial sequence; using the genetic algorithm to generate a first target sequence based on the first initial sequence; adding the first target sequence to the initial set; and / or randomly selecting two sequences from the initial set to obtain two second initial sequences; using the genetic algorithm to generate a second target sequence based on the two second initial sequences, and adding the second target sequence to the initial set; when the number of sequences in the initial set is greater than a first preset number, deleting the sequence set in the initial set that exceeds the first preset number from the initial set, wherein the parameter value of the sequence in the sequence set is less than the parameter value of the sequence in the initial set after the sequence set is deleted; updating the value of i to i+1.

[0011] In an exemplary embodiment, the genetic algorithm is used to generate a first target sequence based on the first initial sequence, including: shifting the Z values ​​in the front region of the first initial sequence by one position to the left or one position to the right as a whole to obtain the first target sequence, wherein, when the Z values ​​are shifted by one position to the left as a whole, the first value among the Z values ​​becomes the last value among the Z values; when the Z values ​​are shifted by one position to the right as a whole, the last value among the Z values ​​becomes the first value among the Z values; and / or randomly selecting two segments of equal length in the front region of the first initial sequence and exchanging the values ​​on the two segments to obtain the first target sequence; and / or, when a second preset condition is satisfied, randomly adding or subtracting a preset threshold value from each of the M-1 values ​​in the rear region of the first initial sequence to obtain the first target sequence, wherein the second preset condition includes: the M-1 values ​​in the rear region of the first initial sequence are all greater than or equal to 1 and less than Z, and the M-1 values ​​are not equal to each other.

[0012] In an exemplary embodiment, the genetic algorithm is used to generate a second target sequence based on the two second initial sequences, including: randomly selecting a segment in the front region of each of the two second initial sequences to obtain two target segments, and exchanging the values ​​on the two target segments to obtain two intermediate sequences, wherein the two target segments have the same length; when there is a conflicting segment in the intermediate sequence, deleting the value on the conflicting segment, and filling the value on the conflicting segment according to the number of the Z waypoints, so that the Z values ​​in the front region of the intermediate sequence are the numbers of the Z waypoints, to obtain the second target sequence, wherein the conflicting segment is a segment in the intermediate sequence having the same value as the corresponding target segment.

[0013] In an exemplary embodiment, the method further includes: calculating the second preset number corresponding to the i-th sequence in the initial set by the following formula, wherein: i is the corresponding second preset number, O max and O min is the preset value, f i is the parameter value corresponding to the i-th sequence, f max is the maximum parameter value of the sequence in the initial set, f min is the minimum parameter value of the sequence in the initial set: In a process of performing one operation on the initial set, the number of sequences generated based on the i-th sequence in the initial set is controlled not to exceed the corresponding second preset number.

[0014] In an exemplary embodiment, the M drones are controlled to inspect the photovoltaic inspection area according to the M target inspection routes, including: writing the j-th target inspection route among the M target inspection routes into the j-th drone among the M drones, where j is 1, ..., M-1, M in sequence; and controlling the M drones to take off at the preset take-off and landing points according to the path distances of the target inspection paths corresponding to the M drones, where the longer the path distance, the earlier the take-off time.

[0015] According to another aspect of an embodiment of the present application, an inspection device for a photovoltaic inspection area is also provided, comprising: a first determination module, used to determine N inspection points in the photovoltaic inspection area, wherein the photovoltaic inspection area has multiple photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the multiple photovoltaic power generation sites, and N is an integer greater than or equal to 2; a second determination module, used to determine M target inspection routes according to the positions of preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the waypoint sets corresponding to the M target inspection routes include the N inspection points, and the waypoints in the waypoint set corresponding to each of the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2; a control module, used to control the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at the waypoints on the corresponding target inspection route to obtain inspection images.

[0016] According to another aspect of the embodiments of the present application, a computer-readable storage medium is provided, the computer-readable storage medium comprising a stored program, wherein the program is configured to execute the above-mentioned inspection method of the photovoltaic inspection area when running.

[0017] According to another aspect of the embodiments of the present application, there is also provided an electronic device, including a memory and a processor, wherein the memory stores a computer program, wherein the processor is configured to execute the above-mentioned inspection method of the photovoltaic inspection area through the computer program.

[0018] According to another aspect of the embodiments of the present application, a computer program product is provided, including a computer program, and the computer program is executed by a processor to implement the above-mentioned inspection method for the photovoltaic inspection area.

[0019] This application determines the inspection points in the photovoltaic inspection area, and determines the inspection routes of multiple drones according to the locations of the inspection points and the preset take-off and landing points, thereby controlling multiple drones to inspect the photovoltaic inspection area according to the determined inspection routes. Since multiple drones are controlled to inspect the photovoltaic inspection area at the same time, the inspection efficiency is improved, thereby solving the problem of low efficiency of a single drone in inspecting the photovoltaic power station area. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0021] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0022] Figure 1 It is a hardware structure block diagram of a mobile terminal of a photovoltaic inspection area inspection method according to an embodiment of the present application;

[0023] Figure 2 is a flow chart of a method for inspecting a photovoltaic inspection area according to an embodiment of the present application;

[0024] Figure 3 is an example diagram of randomly generating an initial sequence set according to an embodiment of the present application;

[0025] Figure 4 This is a schematic diagram of a drone inspection path planning result corresponding to an initial sequence set according to an embodiment of the present application;

[0026] Figure 5 It is a structural block diagram of an inspection device for a photovoltaic inspection area according to an embodiment of the present application. DETAILED DESCRIPTION

[0027] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0029] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal for a photovoltaic inspection area inspection method according to an embodiment of the present application. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor (MP) or a programmable logic device (FPGA)) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0030] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the inspection method of the photovoltaic inspection area in the embodiment of the present application. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories may be connected to the mobile terminal 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.

[0031] The transmission device 106 is used to receive or send data via a network. The specific example of the above network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 can be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0032] In order to solve the above problems, this embodiment provides a method for inspecting a photovoltaic inspection area, including but not limited to being applied to the above mobile terminal. Figure 2 is a flow chart of a method for inspecting a photovoltaic inspection area according to an embodiment of the present application. Figure 2 As shown, the process includes the following steps S202-S206:

[0033] Step S202: determining N inspection points in a photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2;

[0034] It should be noted that N represents the number of inspection points to be determined, and N is an integer greater than or equal to 2.

[0035] Step S204: Determine M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the set of waypoints corresponding to the M target inspection routes includes the N inspection points, the waypoints in the set of waypoints corresponding to each target inspection route in the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2;

[0036] It should be noted that since multiple drones need to be controlled to inspect the photovoltaic inspection area, a common preset take-off and landing point needs to be determined, which is the point where all drones take off and finally return. The selection of this point is usually based on the convenience and safety of drone operation and the relative position to the photovoltaic area to ensure that the drone can take off, land and return smoothly. Optionally, the preset take-off and landing point can be one of the N inspection points.

[0037] It should be noted that each drone will be assigned a target inspection route to ensure that all inspection points can be covered. When planning M target inspection routes, the load capacity, endurance and flight efficiency of each drone must be taken into account to avoid a route that is too long, resulting in too long flight time or excessive energy consumption. When planning the route, in addition to the preset take-off and landing points, the other passing points on each target inspection route are not repeated, which means that when M drones fly at the same time, they will not collide at the inspection points in the middle or repeatedly shoot the same area, but each will be responsible for different inspection points to improve inspection efficiency and comprehensiveness of data collection.

[0038] Optionally, the set of waypoints corresponding to each of the M target inspection routes includes preset take-off and landing points.

[0039] Step S206: Control the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at a passing point on the corresponding target inspection route to obtain an inspection image.

[0040] In the above steps, by determining the inspection points in the photovoltaic inspection area, the inspection routes of multiple drones are determined according to the locations of the inspection points and the preset take-off and landing points, thereby controlling multiple drones to inspect the photovoltaic inspection area according to the determined inspection routes. Since multiple drones are controlled to inspect the photovoltaic inspection area at the same time, the inspection efficiency is improved, thereby solving the problem of low efficiency of a single drone in inspecting the photovoltaic power station area.

[0041] In an exemplary embodiment, determining N inspection points in a photovoltaic inspection area may be achieved by following the steps S11-S13:

[0042] Step S11: determining the flight altitude of the UAV according to the parameter information of the image acquisition device in the UAV and the image requirements of the inspection image;

[0043] Optionally, the parameter information of the image acquisition device in the UAV includes the resolution of the camera carried by the UAV pod, and the image requirements of the inspection image include requirements for the clarity of the inspection image.

[0044] It should be noted that the resolution of the camera carried by the drone is one of the key factors that determine the clarity of the captured image. The higher the resolution, the richer the details of the image. When planning the flight altitude, it is necessary to ensure that the camera can capture sufficiently clear images at that altitude to meet the inspection needs. Usually, the resolution of the camera will give an image clarity indicator that can be achieved at different flight altitudes, such as the Ground Sample Distance (GSD), which is the actual ground distance represented by each pixel in the image taken from the air to the ground.

[0045] It should be noted that because clearer images are needed, the drone is required to fly at a lower altitude to obtain a smaller GSD value. At the same time, the coverage of the inspection image will also affect the choice of flight altitude. If a single image is required to cover as many photovoltaic panels as possible at an inspection point, the drone needs to fly higher. Therefore, the final flight altitude of the drone needs to ensure that the drone can capture as many photovoltaic panels as possible and the images of the photovoltaic panels captured are clear.

[0046] Step S12: determining the effective area of ​​the inspection image captured by the drone according to the flight altitude of the drone;

[0047] Optionally, trigonometric functions are used to calculate the effective area of ​​the inspection image captured by the drone at the determined flight altitude.

[0048] Step S13: determining N inspection points in the photovoltaic inspection area according to the effective area of ​​the inspection image.

[0049] Optionally, the effective area of ​​the inspection image is analyzed, and the center point position of the inspection image to be taken is planned in sequence from north to south and from west to east. The planned center point position of the inspection image is the inspection point. When planning the first inspection point, ensure that when the drone is directly above the first inspection point for inspection and shooting, the northernmost photovoltaic panel within the effective area of ​​the inspection image taken does not exceed the northern edge of the photovoltaic inspection area, thereby determining the vertical coordinate of the first inspection point, and then ensure that the westernmost photovoltaic panel on the same row of the first inspection point within the effective area of ​​the inspection image taken does not exceed the western edge of the photovoltaic inspection area, thereby determining the horizontal coordinate of the first inspection point, and then determining the specific location of the first inspection point. Next, based on the effective area of ​​the inspection image taken when the drone is at the first inspection point, the inspection points are planned eastward in sequence until the photovoltaic panels in the row in the photovoltaic inspection area are fully covered by the sum of the effective areas of all inspection images in the row. After determining the inspection points of each row, go to the next row to continue selecting inspection points, and repeat the steps of determining the inspection points of each row until the southernmost photovoltaic panel in the photovoltaic inspection area is also covered by the sum of the effective areas of all inspection images.

[0050] Optionally, the N inspection points are recorded as a two-dimensional coordinate array, each row of the two-dimensional array contains a horizontal coordinate and a vertical coordinate of a passing point, and the row number of each row is the number of the inspection point corresponding to the row.

[0051] Optionally, the Euler distance between each inspection point needs to be calculated to generate a distance moment with Z as the number of rows and columns, so as to determine the path length corresponding to each inspection path according to the distance matrix.

[0052] In an exemplary embodiment, determining M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points can be achieved by the following steps S21-S24:

[0053] Step S21: determining an initial set according to the Z waypoints, wherein the initial set contains a plurality of sequences, each of the plurality of sequences is divided into a front area and a rear area, the front area has Z values, the Z values ​​are numbers of the Z waypoints, respectively, the rear area has M-1 values, the M-1 values ​​correspond to the M-1 breakpoint positions of the front area, the M-1 values ​​are not equal to each other, the front area is divided into M segments by the M-1 values, each segment corresponds to a patrol route, when the N patrol points do not include the preset take-off and landing points, the Z waypoints are the N patrol points, when the N patrol points include the preset take-off and landing points, the Z waypoints are the N-1 patrol points among the N patrol points excluding the preset take-off and landing points;

[0054] Optionally, assuming that the N inspection points are numbered 1-10, where the inspection point numbered 1 is the preset take-off and landing point, and the inspection points numbered 2-10 are the above-mentioned Z waypoints, then a sequence in the initial set can be determined based on the Z waypoints as shown in 3. Figure 3 As shown, Figure 3 The front region of the sequence is (5, 6, 9, 4, 2, 8, 10, 3, 7), Figure 3 The rear region of the sequence is (4, 6), Figure 3 The three segments corresponding to the front region of the sequence are (5, 6, 9, 4), (2, 8), and (10, 3, 7). The inspection route corresponding to the segment (5, 6, 9, 4) is (1->5->6->9->4->1), which is Figure 4 Path 1 in (first from the preset take-off and landing point to the inspection point numbered 5, then to the inspection point numbered 6, then to the inspection point numbered 9, then to the inspection point numbered 4, and finally back to the preset take-off and landing point).

[0055] Assuming that the preset take-off and landing point is x, and a segment of the front area of ​​the sequence is (y, z, ..., k), the inspection route corresponding to the segment is (x->y->z->...->k->x).

[0056] It should be noted that, assuming that a value in the rear area of ​​the sequence is i, a breakpoint is set between the i-th value and the i+1 value in the front area of ​​the sequence.

[0057] Step S22: Processing the initial set using a genetic algorithm to obtain a target set, wherein the genetic algorithm is used to generate a new sequence using one or two sequences in the initial set;

[0058] Step S23: determining a sequence with the highest parameter value from the target set to obtain a target sequence, wherein the parameter value is inversely proportional to the first distance, and the first distance is equal to the route distance of the longest inspection route among the M inspection routes corresponding to the sequence;

[0059] It should be noted that the parameter value is the fitness corresponding to each sequence. The longer the first distance is, the longer the inspection time consumed by the inspection route corresponding to the sequence is (when implementing the embodiment of the present application, the inspection route with the longest route distance among the determined M inspection routes can be preferentially selected to launch the drone for inspection, and then the inspection routes with shorter route distances can be selected in turn to launch the drone for inspection, so as to ensure that when the first drone flies back to the preset take-off and landing point, the other drones have already flown back to the preset take-off and landing point to complete the inspection work, that is, the inspection time consumed by the M target inspection routes corresponding to the target sequence is the time consumed by the first drone to complete the longest inspection route among the target inspection routes), which means that the fitness of the sequence is lower.

[0060] Optionally, assuming that there are three sequences in the target set, and the inspection time for completing the longest inspection path among the M target inspection routes corresponding to each sequence is 1 hour, 2 hours, and 3 hours respectively, then the sequence with an inspection time of 1 hour is selected as the target sequence.

[0061] Step S24: determining the M inspection routes corresponding to the target sequence as the M target inspection routes.

[0062] Alternatively, assuming Figure 3 The sequence shown is the target sequence, and there are three segments. A patrol route can be determined based on these three segments (such as Figure 4 As shown, the three inspection routes are path 1, path 2 and path 3).

[0063] In an exemplary embodiment, the initial set is processed using a genetic algorithm to obtain a target set, which can be achieved by the following steps: the following steps S1-S3 are cyclically performed on the initial set until a first preset condition is satisfied, and the initial set satisfying the first preset condition is determined as the target set, wherein the first preset condition includes: i is equal to P, or the initial set is not updated after the following operations are performed, P is a preset number of iterations, and i is equal to 0 when the following operations are performed for the first time:

[0064] S1: randomly selecting a sequence from the initial set to obtain a first initial sequence; using the genetic algorithm to generate a first target sequence based on the first initial sequence; adding the first target sequence to the initial set; and / or randomly selecting two sequences from the initial set to obtain two second initial sequences; using the genetic algorithm to generate a second target sequence based on the two second initial sequences, and adding the second target sequence to the initial set;

[0065] S2: when the number of sequences in the initial set is greater than a first preset number, deleting a sequence set in the initial set that exceeds the first preset number from the initial set, wherein a parameter value of the sequence in the sequence set is less than a parameter value of the sequence in the initial set after the sequence set is deleted;

[0066] S3: Update the value of i to i+1.

[0067] Optionally, assuming that there are 150 sequences in the initial set and the first preset number is 50, the sequences in the initial set need to be sorted in descending order according to the parameter value, so as to delete the last 50 sequences from the initial set.

[0068] In an exemplary embodiment, using the genetic algorithm to generate a first target sequence based on the first initial sequence can be achieved by the following steps S31 and / or S32 and / or S33:

[0069] Step S31: Shifting the Z values ​​in the front area of ​​the first initial sequence by one position to the left or one position to the right as a whole, to obtain the first target sequence, wherein, when the Z values ​​are shifted by one position to the left as a whole, the first value among the Z values ​​becomes the last value among the Z values; when the Z values ​​are shifted by one position to the right as a whole, the last value among the Z values ​​becomes the first value among the Z values;

[0070] It should be noted that the execution object of step S31 is a sequence. After step S31, the first and last waypoints of each inspection route in the M inspection routes corresponding to the sequence are exchanged with the other inspection routes in the M routes.

[0071] by Figure 3 Take the sequence shown as an example, Figure 3 The first area of ​​the sequence is (5, 6, 9, 4, 2, 8, 10, 3, 7). Then, after the Z values ​​in the first area of ​​the first initial sequence are shifted to the left by one position, based on Figure 3 The front region of the first target sequence generated by the sequence is (6, 9, 4, 2, 8, 10, 3, 7, 5).

[0072] Step S32: randomly selecting two segments of equal length in the front region of the first initial sequence, and exchanging the values ​​of the two segments to obtain the first target sequence;

[0073] It should be noted that the execution object of step S32 is a sequence, and the two selected segments of equal length should not contain the value indicated by the break point in the rear area of ​​the sequence to ensure that there will be no exchange of inspection points across two or more inspection routes in step S32.

[0074] by Figure 3 Take the sequence shown as an example, Figure 3 The first region of the sequence is (5, 6, 9, 4, 2, 8, 10, 3, 7), then in executing step S32 (replacing Figure 3 After exchanging the fragment (5, 6) with the fragment (9, 4) in Figure 3 The front region of the first target sequence generated by the sequence is (9, 4, 5, 6, 2, 8, 10, 3, 7).

[0075] Step S33: When a second preset condition is met, each of the M-1 values ​​in the rear area of ​​the first initial sequence is randomly added or subtracted by a preset threshold to obtain the first target sequence, wherein the second preset condition includes: the M-1 values ​​in the rear area of ​​the first initial sequence are all greater than or equal to 1 and less than Z, and the M-1 values ​​are not equal to each other.

[0076] It should be noted that the execution object of step S33 is a sequence.

[0077] by Figure 3 Taking the sequence shown as an example, Figure 3 The rear area of ​​the sequence is (4, 6), then in executing step S32 (replace Figure 3 Each of the two values ​​in the rear area of ​​the sequence is randomly added by 2), and then based on Figure 3 The rear region of the first target sequence generated by the sequence is (6, 8).

[0078] Optionally, the preset threshold may be 1, and each of the M-1 values ​​in the rear area of ​​the first initial sequence is randomly added or subtracted by 1, so that the length of each of the M inspection routes determined by the first initial sequence varies.

[0079] In an exemplary embodiment, using the genetic algorithm to generate the second target sequence based on the two second initial sequences can be achieved by the following steps S41-S42:

[0080] Step S41: randomly selecting a segment in the front region of each of the two second initial sequences to obtain two target segments, and exchanging the values ​​of the two target segments to obtain two intermediate sequences, wherein the two target segments have the same length;

[0081] Step S42: When there is a conflicting segment in the intermediate sequence, the value on the conflicting segment is deleted, and the value on the conflicting segment is filled according to the numbers of the Z waypoints, so that the Z values ​​in the front area of ​​the intermediate sequence are the numbers of the Z waypoints, and the second target sequence is obtained, wherein the conflicting segment is a segment in the intermediate sequence having a value equal to that of the corresponding target segment.

[0082] It should be noted that the execution objects of steps S41 - S42 are two sequences.

[0083] Optionally, there are two second initial sequences, wherein the front region of the first second initial sequence is (5, 6, 9, 4, 2, 8, 10, 3, 7), and the front region of the second second initial sequence is (3, 2, 4, 8, 9, 6, 5, 10, 7), then after executing step S41 (selecting the fragment (5, 6) in the front region of the first second initial sequence, and selecting the fragment (3, 2) in the front region of the second second initial sequence, and exchanging the two selected fragments), the front region of the intermediate sequence generated by the first second initial sequence is ( 3，2， 9, 4, 2, 8, 10, 3, 7, the front region of the intermediate sequence generated by the second initial sequence is ( 5，6 , 4, 8, 9, 6, 5, 10, 7). Since there are conflicting segments between the two intermediate sequences, step S42 is executed. After executing step S42, the first second target sequence has a front region of (3, 2, 9, 4, 5, 8, 10, 6, 7) and the second second target sequence has a front region of (5, 6, 4, 8, 9, 3, 2, 10, 7). The first second target sequence is obtained by executing the above steps S41-S42 according to the first second initial sequence, and the second second target sequence is obtained by executing the above steps S41-S42 according to the second second initial sequence.

[0084] It should be noted that, in the case where there are conflicting segments in the intermediate sequence, if the values ​​on the conflicting segments are not deleted, it may cause the values ​​in some segments of the final target sequence to be repeated with the values ​​of another segment, which in turn causes the final drone inspection path planning to exist, such as a waypoint being assigned to two different drone inspection paths. The values ​​on the conflicting segments are thus deleted to ensure that each waypoint is inspected by only one drone, avoiding waste of resources and inefficiency. At the same time, in order for the number of each inspection point to appear completely in the drone inspection path (each inspection point is inspected), after deleting the values ​​of the conflicting segments, the values ​​on the conflicting segments need to be filled in according to the numbers of the Z waypoints, so that each drone's inspection task can be guaranteed to include all the points that it should inspect, ensuring the comprehensiveness of the inspection work.

[0085] In an exemplary embodiment, the method further comprises the following steps: calculating the second preset number corresponding to the i-th sequence in the initial set by the following formula, wherein: i is the corresponding second preset number, O max and O min is the preset value, f i is the parameter value corresponding to the i-th sequence, f max is the maximum parameter value of the sequence in the initial set, f min is the minimum parameter value of the sequence in the initial set:

[0086]

[0087] In a process of performing one operation on the initial set, the number of sequences generated based on the i-th sequence in the initial set is controlled not to exceed the corresponding second preset number.

[0088] Optionally, when the second preset number corresponding to the first sequence in the initial set is 10, in the process of performing one operation on the initial set, the number of sequences generated based on the first sequence in the initial set does not exceed 10.

[0089] In an exemplary embodiment, controlling the M drones to inspect the photovoltaic inspection area according to the M target inspection routes can be achieved by following the steps S51-S52:

[0090] Step S51: writing the j-th target inspection route among the M target inspection routes into the j-th drone among the M drones, where j is 1, ..., M-1, M in sequence;

[0091] Step S52: controlling the M drones to take off at the preset take-off and landing points according to the path distances of the target inspection paths corresponding to the M drones, wherein the longer the path distance, the earlier the take-off time.

[0092] Optionally, the planned M target inspection routes are distributed to the drone remote control systems with corresponding numbers through the network, and the drone corresponding to the target inspection route with the longest path distance is launched first.

[0093] In an exemplary embodiment, the method further includes the following steps: summarizing and storing the image data collected by the drone in ascending order of serial numbers, establishing a correspondence between the inspection points and the photovoltaic panel numbers, and sorting and renaming the images according to the sequence. Verifying at the edge of each image whether it can be spliced ​​with the theoretical adjacent images, and determining whether the actual inspection points of the image are consistent with the planned points. Finally, the entire process of collecting and collating the drone collaborative inspection data is completed, and a complete set of image data sets obtained from this collaborative inspection is generated to provide a data source for the image recognition algorithm module.

[0094] Obviously, the above-described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. In order to better understand the above method, the above process is described below in conjunction with the embodiments, but it is not intended to limit the technical solutions of the embodiments of the present invention. Figure 3 The overall flow chart of a photovoltaic inspection area inspection method of an embodiment of the application is shown, specifically:

[0095] 1. Input the 2D model of the photovoltaic inspection area and determine the coordinates of each center point (i.e. the above inspection points) to be photographed;

[0096] Firstly, the PV inspection area map established by UAV forward photography is used as the data source for analysis. The flight altitude is determined by combining parameters such as the resolution of the camera carried by the UAV pod and the clarity requirements of the inspection image, and then the effective coverage area of ​​a single rectangular inspection image is obtained.

[0097] Secondly, the photovoltaic inspection area model is established and analyzed, and the center point positions of the inspection images to be taken are planned in sequence from north to south and from west to east. Starting from the first point, ensure that the northernmost photovoltaic string does not exceed the northern edge of the shooting area, so as to determine the vertical coordinate of the first point; then ensure that the westernmost photovoltaic string does not exceed the western edge of the shooting area on this row, and determine the specific position of this point. Based on this area, the shooting area is planned eastward in sequence until all the photovoltaic strings in this row are covered. After each row is placed, go to the next row to continue selecting the center point. This process continues to cycle until the southernmost string of the photovoltaic inspection area to be analyzed is covered. After completing this operation, the photovoltaic inspection area is divided into multiple shooting blocks, which are recorded as a two-dimensional array of center point coordinates. Each row contains a horizontal coordinate and a vertical coordinate in sequence, and the row number is the center point number. In the subsequent path planning process of the drone flight inspection, the pre-placed shooting center point is the passing point of the flight path. Calculate the Euler distance between each point and generate a distance matrix with the same number of rows and columns as the number of center points.

[0098] 2. Encoding the primary chromosomes (i.e. the above sequences) and population reproduction and replacement;

[0099] Get the number m of drones that can perform collaborative inspections, and the number n of waypoints. Based on this, randomly shuffle the waypoint numbers within the range and encode them into a sequence with a total length of n+m-1 as a primary chromosome, with each element as a gene fragment of the chromosome. The first n genes represent the waypoints with corresponding serial numbers, and the last m-1 genes mark the genes at which the first half of the chromosome is broken. The first half of the chromosome is divided into m fragments, and each fragment corresponds to the waypoints included in the flight path planned for the inspection of one of the drones, as well as the order of the points passed through. Taking the case of n=9 and m=3 as an example, the take-off and landing point of the drone is numbered 1, and the primary chromosome is randomly generated as follows: Figure 3 As shown;

[0100] This chromosome corresponds to a possible solution for drone inspection path planning, such as Figure 4 shown.

[0101] Next, a chromosome population of size U0 is initialized randomly, and the maximum size of the population during the iteration is specified as Umax The primary fitness of a chromosome is agreed to be the distance of the longest flight path, and the secondary fitness is the sum of the distances of all the flight paths it contains, which is used as the basis for comparison of genetic evolution. When comparing, the primary fitness is calculated and compared first, and the secondary fitness is compared when they are equal.

[0102] After the primary chromosome population is generated, one or two chromosomes in the population are selected in turn in a roulette wheel manner to perform offspring reproduction operations. During the reproduction process, mutations are added according to probability, mainly including the following types:

[0103] 1. Slide. Select a chromosome and shift the first half of it one position to the left or right. The gene fragment that exceeds the limit of the head moving to the left or the tail moving to the right is looped and supplemented to the other end. Through this mutation, the first and last visited points of each path are exchanged with other paths.

[0104] 2. Flip. Select a chromosome, randomly select two segments of equal length in its first half and swap their positions. Note that the selected segments should not contain the breakpoint position in the second half of the chromosome to ensure that there is no waypoint exchange across two or more paths in the flip operation.

[0105] 3. Crossover. Select two chromosomes and swap the segments of equal length to generate new daughter chromosomes. The crossover segments may conflict with the rest, so the conflicting segments that are not crossed are deleted and replaced to generate the correct daughter chromosomes.

[0106] 4. Drift. Select a chromosome, and on the premise that there are no repeated breakpoints after the change and the breakpoint value is less than , randomly perform +1 and -1 operations on the breakpoint gene in the second half of the chromosome, so that the length of each path changes.

[0107] During the reproduction process, in order to ensure that the genetic characteristics of excellent chromosomes are well preserved while ensuring that genetic evolution has diversity to avoid falling into the local optimum, chromosomes with better fitness should be selected to increase the number of their excellent offspring so that they have genetic advantages; at the same time, chromosomes with poor fitness are allowed to reproduce fewer offspring to protect genetic diversity. i The i-th chromosome controls the number of its offspring in this round of reproduction. i for:

[0108]

[0109] After each round of reproduction, the offspring chromosomes are added to the population, their fitness is calculated, the entire chromosome population is sorted by fitness, chromosomes with poor fitness are eliminated, and the population size is maintained at U max Within.

[0110] 3. Generation of UAV collaborative inspection paths;

[0111] After completing the agreed breeding rounds or reaching a certain round without population updates, the algorithm is considered to have converged. At this time, the chromosome with the highest fitness in the population corresponds to the path planning solution given by this algorithm. Take the values ​​of each gene in the second half of the chromosome as the breakpoint, and cut the first half of the chromosome into m segments, each of which contains the numbers of several waypoints. Starting from the first segment, the specific location of each waypoint is listed in sequence as a stop and shoot point, and the center points are connected by a straight line to generate the operation path of the inspection drone with serial number 1, and finally return to the starting point. And so on, generate the inspection paths of the m inspection drones.

[0112] 4. UAV collaborative inspection operation process;

[0113] The planned path file is distributed to the corresponding numbered drone remote control system through the network, and then the operation and maintenance personnel drive the drone equipment to the starting point. The drones are released in order of the total length of the path. The drones with longer one-way paths take off first, so that they can perform the inspection tasks according to the planned path. After completing their respective inspection tasks, they return to the starting point, and the operation and maintenance personnel recover the drones that return and land. After the last returning drone is recovered, the drone inspection operation is completed.

[0114] 5. Generation of inspection image dataset;

[0115] The image data collected by the drone are summarized, sorted and stored in ascending order of serial numbers, and the corresponding relationship between the shooting points and the string-array numbers is established. The images are sorted and renamed according to the string sequence. At the edge of each image, it is checked whether it can be spliced ​​with the theoretical adjacent images to determine whether the actual shooting point of the image is consistent with the planned point. Finally, the whole process of collecting and sorting the drone collaborative inspection data is completed, and a complete set of image data sets obtained from this collaborative inspection is generated, providing a data source for the image recognition algorithm module.

[0116] It should be noted that the present application also has the following advantages: (1) By carrying several drones to the center of the photovoltaic area and releasing them at the same time, each of them scans once according to the planned path, and then the collected image data is merged and organized in sequence into the inspection image collection results of the entire photovoltaic area. For large-scale centralized photovoltaic power stations with a large photovoltaic area and relatively dispersed sub-arrays, the efficiency of drone inspections can be greatly improved if the drone equipment is sufficiently configured. Compared with the inspection process of the entire photovoltaic area by a single drone taking off multiple times, a large amount of drone return and battery replacement time is saved, and the inspection task can be completed quickly. (2) For the coordinated inspection operation of multiple drones, offline path planning is completed based on this application before the on-site operation is implemented, and each flight path and shooting point are imported in advance, so that multiple drones can synchronously complete their respective inspection tasks without human intervention after taking off from the photovoltaic area, and return to the starting point one after another. After the last drone completes the task and returns, the operation and maintenance personnel complete the drone recovery and evacuate the site.

[0117] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods of each embodiment of the present application.

[0118] In this embodiment, a patrol device for a photovoltaic patrol area is also provided, which is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceivable.

[0119] Figure 5 : is a structural block diagram of a photovoltaic inspection area inspection device according to an embodiment of the present application, the device comprising:

[0120] The first determination module 502 is used to determine N inspection points in a photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2;

[0121] The second determination module 504 is used to determine M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the set of waypoints corresponding to the M target inspection routes includes the N inspection points, the waypoints in the set of waypoints corresponding to each target inspection route in the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2;

[0122] The control module 506 is used to control the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at a passing point on the corresponding target inspection route to obtain an inspection image.

[0123] The above device determines the inspection points in the photovoltaic inspection area, and determines the inspection routes of multiple drones according to the positions of the inspection points and the preset take-off and landing points, thereby controlling multiple drones to inspect the photovoltaic inspection area according to the determined inspection routes. Since multiple drones are controlled to inspect the photovoltaic inspection area at the same time, the inspection efficiency is improved, thereby solving the problem of low efficiency of a single drone in inspecting the photovoltaic power station area.

[0124] In an exemplary embodiment, the first determination module 502 is also used to determine the flight altitude of the UAV based on parameter information of the image acquisition device in the UAV and the image requirements of the inspection image; determine the effective area of ​​the inspection image taken by the UAV based on the flight altitude of the UAV; and determine N inspection points in the photovoltaic inspection area based on the effective area of ​​the inspection image.

[0125] In an exemplary embodiment, the second determination module 504 is further used to determine an initial set according to the Z waypoints, wherein the initial set includes multiple sequences, each of the multiple sequences is divided into a front area and a rear area, the front area has Z values, the Z values ​​are respectively numbers of the Z waypoints, the rear area has M-1 values, the M-1 values ​​correspond to the M-1 breakpoint positions of the front area, the M-1 values ​​are not equal to each other, the front area is divided into M segments by the M-1 values, each segment corresponds to a patrol route, and when the N patrol points do not include the preset take-off and landing points, the Z waypoints are the N patrol points. Inspection points, in the case where the N inspection points include the preset take-off and landing points, the Z waypoints are N-1 inspection points among the N inspection points excluding the preset take-off and landing points; the initial set is processed using a genetic algorithm to obtain a target set, wherein the genetic algorithm is used to generate a new sequence using one or two sequences in the initial set; a sequence with the highest parameter value is determined from the target set to obtain a target sequence, wherein the parameter value is inversely proportional to a first distance, and the first distance is equal to the route distance of the longest inspection route among the M inspection routes corresponding to the sequence; the M inspection routes corresponding to the target sequence are determined as the M target inspection routes.

[0126] In an exemplary embodiment, the second determination module 504 is further used to cyclically perform the following operations on the initial set until a first preset condition is satisfied, and the initial set satisfying the first preset condition is determined as the target set, wherein the first preset condition includes: i is equal to P, or the initial set after the following operations are performed is not updated, P is a preset number of iterations, and i is equal to 0 when the following operations are performed for the first time: randomly selecting a sequence from the initial set to obtain a first initial sequence; using the genetic algorithm to generate a first target sequence based on the first initial sequence; adding the first target sequence to the initial set; and / or randomly selecting two sequences from the initial set to obtain two second initial sequences; using the genetic algorithm to generate a second target sequence based on the two second initial sequences, and adding the second target sequence to the initial set; when the number of sequences in the initial set is greater than a first preset number, deleting the sequence set in the initial set that exceeds the first preset number from the initial set, wherein the parameter value of the sequence in the sequence set is less than the parameter value of the sequence in the initial set after the sequence set is deleted; updating the value of i to i+1.

[0127] In an exemplary embodiment, the second determination module 504 is further configured to shift the Z values ​​in the front region of the first initial sequence by one position to the left or one position to the right as a whole to obtain the first target sequence, wherein when the Z values ​​are shifted by one position to the left as a whole, the first value among the Z values ​​becomes the last value among the Z values; when the Z values ​​are shifted by one position to the right as a whole, the last value among the Z values ​​becomes the first value among the Z values; and / or randomly select two segments of equal length in the front region of the first initial sequence and exchange the values ​​on the two segments to obtain the first target sequence; and / or when a second preset condition is met, randomly add or subtract a preset threshold value from each value in the M-1 values ​​in the rear region of the first initial sequence to obtain the first target sequence, wherein the second preset condition includes: the M-1 values ​​in the rear region of the first initial sequence are all greater than or equal to 1 and less than Z, and the M-1 values ​​are not equal to each other.

[0128] In an exemplary embodiment, the second determination module 504 is further used to randomly select a segment in the front area of ​​each of the two second initial sequences to obtain two target segments, and exchange the values ​​on the two target segments to obtain two intermediate sequences, wherein the lengths of the two target segments are equal; when there is a conflicting segment in the intermediate sequence, the value on the conflicting segment is deleted, and the value on the conflicting segment is filled according to the number of the Z waypoints, so that the Z values ​​in the front area of ​​the intermediate sequence are the numbers of the Z waypoints, to obtain the second target sequence, wherein the conflicting segment is a segment in the intermediate sequence having the same value as the corresponding target segment.

[0129] In an exemplary embodiment, the second determining module 504 is further configured to calculate the second preset number corresponding to the i-th sequence in the initial set by using the following formula, wherein: i is the corresponding second preset number, O max and O min is the preset value, f i is the parameter value corresponding to the i-th sequence, f max is the maximum parameter value of the sequence in the initial set, f min is the minimum parameter value of the sequence in the initial set: In a process of performing one operation on the initial set, the number of sequences generated based on the i-th sequence in the initial set is controlled not to exceed the corresponding second preset number.

[0130] In an exemplary embodiment, the control module 506 is also used to write the j-th target inspection route among the M target inspection routes into the j-th drone among the M drones, where j is 1,...,M-1,M in sequence; and control the M drones to take off at the preset take-off and landing points according to the path distances of the target inspection paths corresponding to the M drones, where the longer the path distance, the earlier the take-off time.

[0131] An embodiment of the present application further provides a computer-readable storage medium, in which a computer program is stored, wherein the computer program is configured to execute the steps of any of the above method embodiments when running.

[0132] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0133] S1, determining N inspection points in a photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2;

[0134] S2, determining M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the set of waypoints corresponding to the M target inspection routes includes the N inspection points, the waypoints in the set of waypoints corresponding to each target inspection route in the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2;

[0135] S3, controlling the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at a passing point on the corresponding target inspection route to obtain an inspection image.

[0136] In an exemplary embodiment, the computer-readable storage medium may include, but is not limited to, various media that can store computer programs, such as a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk.

[0137] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0138] An embodiment of the present application further provides a computer program product, including a computer program, and when the computer program is executed by a processor, the steps in any one of the above method embodiments are performed.

[0139] An embodiment of the present application further provides an electronic device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor is configured to run the computer program to execute the steps in any one of the above method embodiments.

[0140] Optionally, in this embodiment, the processor may be configured to perform the following steps through a computer program:

[0141] S1, determining N inspection points in a photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2;

[0142] S2, determining M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the set of waypoints corresponding to the M target inspection routes includes the N inspection points, the waypoints in the set of waypoints corresponding to each target inspection route in the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2;

[0143] S3, controlling the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at a passing point on the corresponding target inspection route to obtain an inspection image.

[0144] In an exemplary embodiment, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor, and the input / output device is connected to the processor.

[0145] For specific examples in this embodiment, reference may be made to the examples described in the above embodiments and exemplary implementation modes, and this embodiment will not be described in detail herein.

[0146] Obviously, those skilled in the art should understand that the above modules or steps of the present application can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order from that herein, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present application is not limited to any specific combination of hardware and software.

[0147] The above is only a preferred implementation of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A method for inspecting a photovoltaic inspection area, characterized in that: include: Determine N inspection points in a photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2; Determine M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the set of waypoints corresponding to the M target inspection routes includes the N inspection points, the waypoints in the set of waypoints corresponding to each target inspection route in the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2; The M drones are controlled to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at a passing point on the corresponding target inspection route to obtain an inspection image.

2. The method according to claim 1, characterized in that: Determine N inspection points in the photovoltaic inspection area, including: Determining the flight altitude of the UAV according to parameter information of an image acquisition device in the UAV and image requirements of the inspection image; Determine the effective area of ​​the inspection image captured by the drone according to the flight altitude of the drone; N inspection points are determined in the photovoltaic inspection area according to the effective area of ​​the inspection image.

3. The method according to claim 1, characterized in that Determining M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points includes: An initial set is determined according to the Z waypoints, wherein the initial set contains a plurality of sequences, each of the plurality of sequences is divided into a front area and a rear area, the front area has Z values, the Z values ​​are numbers of the Z waypoints, respectively, the rear area has M-1 values, the M-1 values ​​correspond to the M-1 breakpoint positions of the front area, the M-1 values ​​are not equal to each other, the front area is divided into M segments by the M-1 values, each segment corresponds to a patrol route, when the N patrol points do not include the preset take-off and landing points, the Z waypoints are the N patrol points, when the N patrol points include the preset take-off and landing points, the Z waypoints are the N-1 patrol points among the N patrol points excluding the preset take-off and landing points; Processing the initial set using a genetic algorithm to obtain a target set, wherein the genetic algorithm is used to generate a new sequence using one or two sequences in the initial set; Determine a sequence with the highest parameter value from the target set to obtain a target sequence, wherein the parameter value is inversely proportional to a first distance, and the first distance is equal to a route distance of the longest inspection route among the M inspection routes corresponding to the sequence; The M inspection routes corresponding to the target sequence are determined as the M target inspection routes.

4. The method according to claim 3, characterized in that The initial set is processed using a genetic algorithm to obtain a target set, including: The initial set is cyclically executed with the following operations until a first preset condition is satisfied, and the initial set satisfying the first preset condition is determined as the target set, wherein the first preset condition includes: i is equal to P, or the initial set after the following operations are executed is not updated, P is a preset number of iterations, and i is equal to 0 when the following operations are executed for the first time: Randomly select a sequence from the initial set to obtain a first initial sequence; use the genetic algorithm to generate a first target sequence based on the first initial sequence; add the first target sequence to the initial set; and / or randomly select two sequences from the initial set to obtain two second initial sequences; use the genetic algorithm to generate a second target sequence based on the two second initial sequences, and add the second target sequence to the initial set; When the number of sequences in the initial set is greater than a first preset number, deleting the sequence sets in the initial set that exceed the first preset number from the initial set, wherein parameter values ​​of the sequences in the sequence set are less than parameter values ​​of the sequences in the initial set after the sequence sets are deleted; Update the value of i to i+1.

5. The method according to claim 4, characterized in that Using the genetic algorithm to generate a first target sequence based on the first initial sequence includes: The first target sequence is obtained by shifting the Z values ​​in the front area of ​​the first initial sequence by one position to the left or one position to the right as a whole, wherein when the Z values ​​are shifted by one position to the left as a whole, the first value among the Z values ​​becomes the last value among the Z values; when the Z values ​​are shifted by one position to the right as a whole, the last value among the Z values ​​becomes the first value among the Z values; and / or Randomly selecting two fragments of equal length in the front region of the first initial sequence, and exchanging the values ​​of the two fragments to obtain the first target sequence; and / or When a second preset condition is met, each of the M-1 values ​​in the rear area of ​​the first initial sequence is randomly added or subtracted from a preset threshold to obtain the first target sequence, wherein the second preset condition includes: the M-1 values ​​in the rear area of ​​the first initial sequence are all greater than or equal to 1 and less than Z, and the M-1 values ​​are not equal to each other.

6. The method according to claim 4, characterized in that Using the genetic algorithm, generating a second target sequence based on the two second initial sequences, comprising: Randomly selecting a segment in the front region of each of the two second initial sequences to obtain two target segments, and exchanging the values ​​of the two target segments to obtain two intermediate sequences, wherein the two target segments have the same length; In the case where there is a conflicting segment in the intermediate sequence, the value on the conflicting segment is deleted, and the value on the conflicting segment is filled according to the numbers of the Z waypoints, so that the Z values ​​in the front area of ​​the intermediate sequence are the numbers of the Z waypoints, and the second target sequence is obtained, wherein the conflicting segment is a segment in the intermediate sequence having a value equal to that of the corresponding target segment.

7. The method according to claim 4, characterized in that The method further comprises: The second preset number corresponding to the i-th sequence in the initial set is calculated by the following formula, where: i is the corresponding second preset number, O max and O min is the preset value, f i is the parameter value corresponding to the i-th sequence, f max is the maximum parameter value of the sequence in the initial set, f min is the minimum parameter value of the sequence in the initial set: In a process of performing one operation on the initial set, the number of sequences generated based on the i-th sequence in the initial set is controlled not to exceed the corresponding second preset number.

8. The method according to claim 1, characterized in that Controlling the M drones to inspect the photovoltaic inspection area according to the M target inspection routes includes: Write the j-th target inspection route among the M target inspection routes into the j-th drone among the M drones, where j is 1, ..., M-1, M in sequence; The M drones are controlled to take off at the preset take-off and landing points according to the path distances of the target inspection paths corresponding to the M drones, wherein the longer the path distance, the earlier the take-off time.

9. A photovoltaic inspection area inspection device, characterized in that: include: A first determination module is used to determine N inspection points in a photovoltaic inspection area, wherein the photovoltaic inspection area has a plurality of photovoltaic power generation sites, and the images captured by the drone at the N inspection points can cover the plurality of photovoltaic power generation sites, and N is an integer greater than or equal to 2; A second determination module is used to determine M target inspection routes according to the positions of the preset take-off and landing points and the positions of the N inspection points, wherein M is the number of drones, the preset take-off and landing points are the take-off points and landing points of the M drones, the set of waypoints corresponding to the M target inspection routes includes the N inspection points, the waypoints in the set of waypoints corresponding to each target inspection route in the M target inspection routes do not overlap except for the preset take-off and landing points, and M is an integer greater than or equal to 2; A control module is used to control the M drones to inspect the photovoltaic inspection area according to the M target inspection routes, wherein each drone collects images at a passing point on the corresponding target inspection route to obtain an inspection image.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 8 when executed.

11. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 8 through the computer program.

12. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

Images