Photovoltaic panel fault inspection method and device, terminal and storage medium

Abstract

The invention provides a photovoltaic panel fault inspection method and device, a terminal and a storage medium. The method comprises the following steps: acquiring a routing inspection image of a photovoltaic panel of a photovoltaic power station, segmenting the routing inspection image to obtain routing inspection sub-images, inputting the routing inspection sub-images into a YOLOv8n target detection model, outputting fault coordinates corresponding to a fault center, and performing fault detection; wherein the inspection image is segmented to obtain inspection sub-images, so that fault detection of the YOLOv8n target detection model is facilitated; determining a fault matching range according to the fault coordinates and a space division result of the photovoltaic power station; wherein the fault matching range comprises a plurality of grid units in the space division result; and obtaining a mapping relation between the photovoltaic panels in the photovoltaic power station and the space division result, carrying out fault positioning on the fault center and the photovoltaic panels in the fault matching range according to the fault matching range, the mapping relation and the fault coordinates, and recording the number of the photovoltaic panel corresponding to the fault center. According to the invention, the problem of rapidly identifying the photovoltaic panel fault can be solved.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic panel inspection, and in particular to a photovoltaic panel fault inspection method, device, terminal and storage medium. Background Art

[0002] With the development of solar power generation technology, the number of photovoltaic power stations has increased dramatically. However, in the long-term use process, due to weather disasters, foreign objects blocking, component aging and other reasons, photovoltaic panels will produce hot spots, cells, diodes and other faults, affecting the power generation efficiency.

[0003] Traditional fault detection methods usually use infrared image detection technology, which requires matching each detected fault instance with all photovoltaic panels in the photovoltaic power station one by one, which requires huge calculations. When the photovoltaic panel array is large, the matching process of fault instances and photovoltaic panels takes a long time and is inefficient, making it difficult to adapt to the real-time monitoring needs of large-scale photovoltaic panel arrays. Summary of the invention

[0004] The embodiments of the present invention provide a method, device, terminal and storage medium for inspecting photovoltaic panel faults to solve the problem of quickly identifying photovoltaic panel faults.

[0005] In a first aspect, an embodiment of the present invention provides a method for inspecting photovoltaic panel faults, comprising:

[0006] Obtain an inspection image of the photovoltaic panels of the photovoltaic power station, divide the inspection image into inspection sub-images, input the inspection sub-images into the YOLOv8n target detection model, output the fault coordinates corresponding to the fault center, and perform fault detection; wherein, dividing the inspection image into inspection sub-images facilitates the YOLOv8n target detection model to perform fault detection;

[0007] Determining a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station; wherein the fault matching range includes a plurality of grid cells in the spatial division result;

[0008] Obtain a mapping relationship between the photovoltaic panels in the photovoltaic power station and the spatial division results, locate the fault of the photovoltaic panels in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates, and record the photovoltaic panel number corresponding to the fault center.

[0009] In a possible implementation manner, determining a fault matching range according to the fault coordinates and a spatial division result of the photovoltaic power station further includes:

[0010] Divide the space of the photovoltaic power station into multiple grid units;

[0011] Determine the grid unit to which the fault belongs according to the coordinates of the fault;

[0012] The nine-square grid consisting of the grid unit to which the fault coordinates belong and the eight adjacent grid units around it is determined as the fault matching range.

[0013] In a possible implementation, locating the fault of the photovoltaic panel in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates, further includes:

[0014] The fault coordinates corresponding to the fault center are matched one by one with the coordinates of all photovoltaic panels within the fault matching range. If the fault coordinates are within the coordinate range of a photovoltaic panel, the photovoltaic panel number is recorded.

[0015] In a possible implementation manner, before acquiring the inspection image of the photovoltaic panels of the photovoltaic power station, the method further includes:

[0016] Obtain the longitude and latitude coordinate information of the area where the photovoltaic power station is located, and divide the area into a number of grid units according to the longitude and latitude coordinate information;

[0017] Acquire a photovoltaic power station modeling image, and establish photovoltaic panel coordinate file information according to the photovoltaic power station modeling image; wherein the photovoltaic panel coordinate file information includes the photovoltaic panel number, the longitude and latitude coordinate information of the center point of the photovoltaic panel, and the longitude and latitude coordinate information of the vertices of the photovoltaic panel;

[0018] According to the latitude and longitude coordinate information of the center point of the photovoltaic panel, each photovoltaic panel is mapped to the corresponding grid unit.

[0019] In a possible implementation, the step of establishing photovoltaic panel coordinate archive information according to the photovoltaic power station modeling image includes:

[0020] Use PYODM to synthesize the photovoltaic power station modeling image into the photovoltaic power station orthophoto map;

[0021] The orthophoto image of the photovoltaic power station is cut into several orthophoto sub-images, and the orthophoto sub-images are sequentially input into the Real-ESRGAN super-resolution model and the YOLOv8n-seg instance segmentation model to output the photovoltaic panel image; wherein the Real-ESRGAN super-resolution model takes the orthophoto sub-image as input and the orthophoto sub-image with improved quality as output, thereby improving the quality of the orthophoto sub-image and facilitating the YOLOv8n-seg instance segmentation model to segment the photovoltaic panel image; the YOLOv8n-seg instance segmentation model takes the orthophoto sub-image with improved quality as input and the photovoltaic panel image as output, thereby performing the segmentation of the photovoltaic panel image; the orthophoto image of the photovoltaic power station, the orthophoto sub-image and the photovoltaic panel image all contain latitude and longitude coordinate information;

[0022] The photovoltaic panel images are numbered, and photovoltaic panel coordinate file information is established according to the number of the photovoltaic panel image, the longitude and latitude coordinates of its center point, and the longitude and latitude coordinates of its vertices.

[0023] In one possible implementation, the Real-ESRGAN super-resolution model includes a sinc filter, and the sinc filter kernel is:

[0024]

[0025] Where (i, j) is the coordinate of the kernel, ω c is the cutoff frequency and J1 is the first-order Bessel function.

[0026] In a possible implementation, the size of the grid unit is 1*1 meter, the length of the photovoltaic panel is 1.5-2 meters, and the width of the photovoltaic panel is 0.8-1 meter.

[0027] In a second aspect, an embodiment of the present invention provides a photovoltaic panel fault inspection device, comprising:

[0028] A fault coordinate acquisition module is used to acquire an inspection image of a photovoltaic panel of a photovoltaic power station, divide the inspection image into inspection sub-images, input the inspection sub-images into a YOLOv8n target detection model, output the fault coordinates corresponding to the fault center, and perform fault detection; wherein, dividing the inspection image into inspection sub-images facilitates the YOLOv8n target detection model to perform fault detection;

[0029] A fault matching range determination module, used to determine a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station; wherein the fault matching range includes a plurality of grid cells in the spatial division result;

[0030] The fault location module is used to obtain the mapping relationship between the photovoltaic panels in the photovoltaic power station and the space division results, locate the fault of the photovoltaic panels in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates, and record the photovoltaic panel number corresponding to the fault center.

[0031] In a third aspect, an embodiment of the present invention provides a terminal, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, the steps of the method described in the first aspect or any possible implementation manner of the first aspect are implemented.

[0032] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method described in the first aspect or any possible implementation method of the first aspect are implemented.

[0033] The embodiment of the present invention provides a method, device, terminal and storage medium for inspecting photovoltaic panel faults. The inspection image of the photovoltaic panel of the photovoltaic power station is divided to obtain an inspection sub-image, and the inspection image is divided into smaller inspection sub-images. The smaller inspection sub-images are input into the YOLOv8n target detection model to improve the detection efficiency of the YOLOv8n target detection model and facilitate fault detection. The inspection sub-image is input into the YOLOv8n target detection model for fault target detection, and the fault coordinates corresponding to the fault center are output. The YOLOv8n target detection model quickly detects the fault position and improves the efficiency of photovoltaic panel fault detection. According to the fault coordinates and the spatial division result of the photovoltaic power station, the fault matching range is determined. The fault matching range includes multiple grid units in the spatial division result. It is only necessary to match the fault coordinates corresponding to the fault center with the photovoltaic panels within the fault matching range to locate the fault, reduce the number of photovoltaic panels that need to match the fault coordinates, greatly improve the fault location efficiency, and finally optimize the overall efficiency of photovoltaic panel fault inspection and identification. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.

[0035] Figure 1 is a schematic diagram of deduplication of a photovoltaic panel polygon example provided by an embodiment of the present invention;

[0036] Figure 2 is a flow chart of the implementation of the method for inspecting photovoltaic panel faults provided by an embodiment of the present invention;

[0037] Figure 3 is a structural schematic diagram of a photovoltaic panel failure in a photovoltaic power station provided by an embodiment of the present invention;

[0038] Figure 4 It is a structural schematic diagram of a photovoltaic panel fault inspection device provided by an embodiment of the present invention;

[0039] Figure 5 is a schematic diagram of a terminal provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present invention. However, it should be clear to those skilled in the art that the present invention may be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present invention.

[0041] In order to make the purpose, technical solutions and advantages of the present invention more clear, specific embodiments will be described below in conjunction with the accompanying drawings.

[0042] This application cuts the inspection image of the photovoltaic panels of the photovoltaic power station into smaller inspection sub-images to facilitate fault detection by the YOLOv8n target detection model. The inspection sub-image is input into the YOLOv8n target detection model for fault target detection, and the fault coordinates corresponding to the fault center are output. According to the fault coordinates and the results of the spatial division of the photovoltaic power station, the fault matching range is determined, and the fault coordinates corresponding to the fault center are matched with the photovoltaic panels within the fault matching range to locate the fault, thereby reducing the number of photovoltaic panels that need to match the fault coordinates, greatly improving the efficiency of fault location, and ultimately optimizing the overall efficiency of photovoltaic panel fault inspection and identification.

[0043] Before implementing the photovoltaic panel fault inspection and identification of the present application, the following preparations need to be made first.

[0044] In a possible implementation manner, before acquiring the inspection image of the photovoltaic panels of the photovoltaic power station, the method further includes:

[0045] Obtain the longitude and latitude coordinate information of the area where the photovoltaic power station is located, and divide the area into a number of grid units according to the longitude and latitude coordinate information;

[0046] Acquire a photovoltaic power station modeling image, and establish photovoltaic panel coordinate file information according to the photovoltaic power station modeling image; wherein the photovoltaic panel coordinate file information includes the photovoltaic panel number, the longitude and latitude coordinate information of the center point of the photovoltaic panel, and the longitude and latitude coordinate information of the apex of the photovoltaic panel;

[0047] According to the latitude and longitude coordinate information of the center point of the photovoltaic panel, each photovoltaic panel is mapped to the corresponding grid unit.

[0048] According to the location range of the photovoltaic power station, the area where the photovoltaic power station is located is divided into a rectangular area, and the longitude and latitude coordinate information of the four vertices of the rectangular area is obtained. Through the inventor's investigation, it is found that the 5 decimal places of the longitude and latitude coordinates can be accurate to about 1m. After verification, it is found that on a real map, adding 1 to the 5 decimal places of the longitude means moving 1m eastward, and subtracting 1 means moving 1m westward. Adding 1 to the 5 decimal places of the latitude means moving 1m northward, and subtracting 1 means moving 1m southward.

[0049] The grid units are divided according to the values ​​of the last five digits of the longitude and latitude, and the rectangular area containing the photovoltaic power station is divided into several grid unit areas. The divided grid units are square areas of 1m*1m.

[0050] In this embodiment, the rectangular area of ​​the photovoltaic power station is divided into a number of grid units, so as to map the photovoltaic panels to the corresponding grid units, establish the photovoltaic panel coordinate file information, and attribute all photovoltaic panels to the corresponding grid units.

[0051] In a possible implementation, establishing photovoltaic panel coordinate file information according to the photovoltaic power station modeling image includes:

[0052] Use PYODM to synthesize the photovoltaic power station modeling image into the photovoltaic power station orthophoto map;

[0053] The orthophoto image of the photovoltaic power station is cut into several orthophoto sub-images, and the orthophoto sub-images are sequentially input into the Real-ESRGAN super-resolution model and the YOLOv8n-seg instance segmentation model to output the photovoltaic panel image; wherein the Real-ESRGAN super-resolution model takes the orthophoto sub-image as input and the orthophoto sub-image with improved quality as output, thereby improving the quality of the orthophoto sub-image and facilitating the YOLOv8n-seg instance segmentation model to segment the photovoltaic panel image; the YOLOv8n-seg instance segmentation model takes the orthophoto sub-image with improved quality as input and the photovoltaic panel image as output, thereby performing the segmentation of the photovoltaic panel image; the orthophoto image of the photovoltaic power station, the orthophoto sub-image and the photovoltaic panel image all contain latitude and longitude coordinate information;

[0054] The photovoltaic panel images are numbered, and photovoltaic panel coordinate file information is established according to the number of the photovoltaic panel image, the longitude and latitude coordinates of its center point, and the longitude and latitude coordinates of its vertices.

[0055] The images of the photovoltaic power station were collected by drones, and the overlap rate of the photovoltaic power station images was 83%, ensuring the integrity of the collected photovoltaic power station images.

[0056] The size of the photovoltaic power station orthophoto map synthesized by PYODM is 10000*10000. Since the image is relatively large, it is necessary to first divide the photovoltaic power station orthophoto map into orthophoto sub-images. The size of the divided orthophoto sub-images is 640*640. Since the synthesized photovoltaic power station orthophoto map is very large, the quality of the synthesized orthophoto sub-images is not clear and the image quality needs to be improved. The orthophoto sub-images need to be input into the Real-ESRGAN super-resolution model to improve the quality of the orthophoto sub-images, so as to improve the accuracy of the YOLOv8n-seg instance segmentation model in segmenting photovoltaic panel images.

[0057] In order to ensure that a complete photovoltaic panel can be segmented out by an instance when cutting the image, the overlap rate of the cutting image is set to 25%. Due to the existence of the overlap rate, the same photovoltaic panel may appear in different cutting images and be detected multiple times, thereby obtaining multiple segmentation instances. The present invention proposes an arbitrary polygon deduplication algorithm based on non-maximum suppression (NMS) to remove these overlapping redundant instances and retain only the optimal one.

[0058] The basic principle is to first find all detected photovoltaic panel instance polygon areas in the image, sort them according to their confidence, and then start from the polygon with the highest confidence and traverse the polygons that intersect with it. If it is found that the overlapping area of ​​the current polygon and the previous reference polygon is greater than a threshold, the current polygon will be discarded. This will minimize the number of prediction boxes retained in the end, while ensuring the accuracy of detection. The specific implementation method includes the following steps:

[0059] 1. Sort by the confidence of the prediction box, and take the prediction box with the highest confidence as the benchmark;

[0060] 2. Filter out all polygonal boxes that intersect with the reference box from the remaining polygonal prediction boxes;

[0061] 3. Traverse the filtered polygon boxes and delete them if their overlapping area is greater than a certain threshold.

[0062] In this way, NMS can filter out all redundant frames whose overlapping area with the reference frame is greater than the threshold, thereby optimizing the detection results.

[0063] like Figure 1 As shown, the left picture is a photovoltaic panel instance detected by the YOLOv8n-seg instance segmentation model, and the right picture is two photovoltaic panel instances obtained after the deduplication algorithm.

[0064] In summary, the principle of non-maximum suppression is to use a target box with a higher confidence level as a benchmark to filter out target boxes with a higher degree of overlap, thereby removing redundant target boxes and improving accuracy.

[0065] The YOLOv8n-seg instance segmentation model can effectively handle the instance segmentation task of photovoltaic power station modeling images, providing support for the automated monitoring and maintenance of photovoltaic power stations.

[0066] The orthophotos of photovoltaic power stations synthesized by PYODM, the orthophoto sub-images with improved quality output by the Real-ESRGAN super-resolution model, and the photovoltaic panel images segmented by the YOLOv8n-seg instance segmentation model all contain longitude and latitude coordinate information, which facilitates accurate geographic positioning and subsequent analysis.

[0067] In this embodiment, the Real-ESRGAN super-resolution model is used to improve the quality of the orthophoto sub-image, making the orthophoto sub-image clear, which is conducive to the accuracy of the photovoltaic panel coordinate file information. The YOLOv8n-seg instance segmentation model segments the photovoltaic panel image, which is convenient for numbering the photovoltaic panels so that the photovoltaic panel numbers correspond to the longitude and latitude coordinates of their center points and the longitude and latitude coordinates of the vertices, thereby establishing the photovoltaic panel coordinate file information.

[0068] The above-mentioned photovoltaic panels are mapped to the corresponding grid units according to the longitude and latitude coordinate information of their center points, so that the longitude and latitude coordinate information of the center point of the photovoltaic panels is within the longitude and latitude coordinate range of the corresponding grid units. Each grid unit stores the photovoltaic panel number, the longitude and latitude coordinate information of the center point corresponding to the photovoltaic panel, and the longitude and latitude coordinate information of the vertex.

[0069] In this embodiment, archive information is created for all photovoltaic panels to facilitate matching of the fault coordinates with the photovoltaic panels within the fault matching range after the fault matching range is subsequently determined.

[0070] In other possible implementations, a plane rectangular coordinate system is established, and the area where the photovoltaic power station is located is divided using the plane rectangular coordinates, and the area where the photovoltaic power station is located is divided into a number of grid units.

[0071] See also Figure 2 , which shows a flowchart of the implementation of the method for inspecting photovoltaic panel faults provided by an embodiment of the present invention, and is described in detail as follows:

[0072] Step 101: Obtain an inspection image of the photovoltaic panels of the photovoltaic power station, divide the inspection image into inspection sub-images, input the inspection sub-images into the YOLOv8n target detection model, output the fault coordinates corresponding to the fault center, and perform fault detection; wherein, dividing the inspection image into inspection sub-images facilitates the YOLOv8n target detection model to perform fault detection.

[0073] By dividing the acquired inspection image into multiple smaller inspection sub-images, the image size is reduced, which is convenient for the subsequent YOLOv8n target detection model to perform fault detection. The inspection sub-image is input into the YOLOv8n target detection model for fault detection. When the fault is detected, the fault coordinates corresponding to the fault center are output to facilitate subsequent fault matching.

[0074] For example, planning a drone inspection route for a photovoltaic power station, and collecting inspection images of photovoltaic panels in the photovoltaic power station through drone inspection. Since the inspection images collected by drones are very clear, there is no need to improve the quality of the inspection images. The inspection images can be directly segmented and input into the YOLOv8n target detection model.

[0075] The YOLOv8n target detection model adds a Coordinate Attention mechanism to optimize the accuracy of infrared fault small target detection. The coordinate attention mechanism pools along the width and height directions to obtain two one-dimensional vectors M w =avgpool w (X), M h =avgpool h (X), where X represents the image input data, avgpool w Indicates pooling along the width direction, M w That is, the one-dimensional vector obtained by pooling along the width direction; similarly, M h Represents the one-dimensional vector obtained by pooling along the height direction. Two one-dimensional vectors generate the attention map A h =σ(W h M h +b h ) and A w =σ(W w M w +b w ), where W h represents the weight along the height direction, b h represents the bias along the height direction, σ represents the sigmoid activation function, A h represents the attention map generated along the height direction; similarly, W w , b w and A w The coordinate attention mechanism uses the generated attention map to adjust the channel weights in the feature map to highlight important features and suppress irrelevant features, thereby improving the detection performance of the YOLOv8n target detection model.

[0076] Step 102: determining a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station; wherein the fault matching range includes a plurality of grid cells in the spatial division result.

[0077] The detected faults are associated with the specific location of the PV plant, so that subsequent fault processing, such as repairing or replacing the faulty equipment, can be carried out more effectively.

[0078] The spatial partitioning divides the entire area of ​​the photovoltaic power station into multiple grid cells, each grid cell represents a part of the power station. The fault matching range is included in the result of the spatial partitioning and is a part of the result of the spatial partitioning.

[0079] The fault matching range is determined according to the fault coordinates and the spatial division results of the photovoltaic power station. The fault coordinates can be quickly located to the fault matching range. The fault matching range includes multiple grid units in the spatial division results. It is only necessary to match the fault coordinates corresponding to the fault center with the photovoltaic panels within the fault matching range to locate the fault. It is not necessary to match the fault coordinates with all photovoltaic panels in the photovoltaic power station, which reduces the number of photovoltaic panels that need to be matched with the fault coordinates, greatly improves the fault location efficiency, and ultimately optimizes the overall efficiency of photovoltaic fault inspection and identification.

[0080] Step 103: Obtain the mapping relationship between the photovoltaic panels in the photovoltaic power station and the spatial division results, locate the fault of the photovoltaic panels in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates, and record the photovoltaic panel number corresponding to the fault center.

[0081] The mapping relationship includes the coordinate position information of each photovoltaic panel and the corresponding spatial grid unit.

[0082] In a possible implementation manner, locating the fault of the photovoltaic panel in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates also includes:

[0083] The fault coordinates corresponding to the fault center are matched one by one with the coordinates of all photovoltaic panels within the fault matching range. If the fault coordinates are within the coordinate range of a photovoltaic panel, the photovoltaic panel number is recorded.

[0084] By matching the fault coordinates with the PV panels within the fault matching range, it is possible to quickly and accurately find and record the faulty PV panels, thereby improving the operation and maintenance efficiency of the PV power plant. This systematic approach helps reduce the time for fault detection and repair, ensuring the stable operation of the PV power plant.

[0085] The embodiment of the present invention obtains inspection sub-images by dividing the inspection image of the photovoltaic panel of the photovoltaic power station, divides the inspection image into inspection sub-images of smaller size, and inputs the inspection sub-images of smaller size into the YOLOv8n target detection model, thereby improving the detection efficiency of the YOLOv8n target detection model and facilitating fault detection. The inspection sub-image is input into the YOLOv8n target detection model for fault target detection, and the fault coordinates corresponding to the fault center are output. The YOLOv8n target detection model quickly detects the fault location and improves the efficiency of photovoltaic panel fault detection. According to the fault coordinates and the spatial division result of the photovoltaic power station, the fault matching range is determined. The fault matching range includes multiple grid units in the spatial division result. It is only necessary to match the fault coordinates corresponding to the fault center with the photovoltaic panels within the fault matching range to locate the fault, thereby reducing the number of photovoltaic panels that need to match the fault coordinates, greatly improving the fault location efficiency, and finally optimizing the overall efficiency of photovoltaic panel fault inspection and identification.

[0086] In a possible implementation manner, determining the fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station further includes:

[0087] Divide the space of the photovoltaic power station into multiple grid units;

[0088] Determine the grid unit to which the fault belongs according to the coordinates of the fault;

[0089] The nine-square grid consisting of the grid unit to which the fault coordinates belong and the eight adjacent grid units around it is determined as the fault matching range.

[0090] In a possible implementation, the size of the grid unit is 1*1 meter, the length of the photovoltaic panel is 1.5-2 meters, and the width of the photovoltaic panel is 0.8-1 meter.

[0091] Since the size of the grid unit divided by the photovoltaic power station is 1*1 meter, the length of the photovoltaic panel is 1.5-2 meters, and the width of the photovoltaic panel is 0.8-1 meter, there is a situation where a photovoltaic panel occupies two or three horizontally adjacent grid units at the same time. When establishing the coordinate file information, it is assigned to the corresponding grid unit according to the longitude and latitude coordinates of the center point of the photovoltaic panel. When the fault coordinates corresponding to the fault center are located at the edge of a photovoltaic panel, the photovoltaic panel at this time is an adjacent grid unit of the grid unit to which the fault coordinates belong. The grid unit to which the fault coordinates belong does not contain the coordinate information of the photovoltaic panel to which the fault coordinates belong. If the photovoltaic panel is matched only according to the fault coordinates and the grid unit to which it belongs, the photovoltaic panel cannot be matched. Therefore, the fault matching range needs to be set to the nine-square grid composed of the grid unit to which the fault coordinates belong and the eight adjacent grid units around it, such as Figure 3 shown.

[0092] In this embodiment, the fault coordinates are matched to the grid unit to which it belongs and the eight grid units around it, and the fault coordinates do not need to be matched with all photovoltaic panels in the photovoltaic power station, which significantly reduces unnecessary calculations and greatly improves the matching efficiency. The area of ​​the nine-square grid in the real geographic space is approximately a 3*3 meter square area. Define the parameter d to represent the direction, d=-1 represents left or downward, d=1 represents right or upward, and the parameters dx and dy represent the direction changes of the x-axis and y-axis respectively, dx={-1,0,1}, dy={-1,0,1}, so the coordinates of the nine-square grid grid unit can be expressed as (x+dx, y+dy).

[0093] In one possible implementation, the Real-ESRGAN super-resolution model includes a sinc filter, and the sinc filter kernel is:

[0094]

[0095] Where (i, j) is the coordinate of the kernel, ω c is the cutoff frequency and J1 is the first-order Bessel function.

[0096] The Real-ESRGAN super-resolution model uses a sinc filter to enable the model to better learn and understand overshoot artifacts and effectively suppress them in practical applications. In this way, the model can improve the quality of image super-resolution while reducing the generation of unnatural artifacts.

[0097] In other possible ways, one or more stages of PYODM, Real-ESRGAN super-resolution model, YOLOv8n-seg instance segmentation model and YOLOv8n target detection model can adopt other ways. This application mainly uses PYODM, Real-ESRGAN super-resolution model, YOLOv8n-seg instance segmentation model and YOLOv8n target detection model as examples for explanation.

[0098] The present invention utilizes the Real-ESRGAN super-resolution model to improve the resolution of orthophoto sub-images of photovoltaic power stations, thereby improving the accuracy of photovoltaic panel segmentation and ensuring the quality of photovoltaic panel archive information.

[0099] The present invention is based on the YOLOv8n target detection model and adds a Coordinate attention mechanism to improve the accuracy of photovoltaic panel fault detection.

[0100] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0101] The following is an embodiment of the device of the present invention. For details not described in detail therein, reference may be made to the corresponding method embodiment described above.

[0102] Figure 4 The schematic diagram of the structure of the photovoltaic panel fault inspection device provided by the embodiment of the present invention is shown. For the convenience of explanation, only the part related to the embodiment of the present invention is shown, which is described in detail as follows:

[0103] like Figure 4 As shown, the photovoltaic panel fault inspection device 4 includes:

[0104] The fault coordinate acquisition module 41 is used to acquire the inspection image of the photovoltaic panel of the photovoltaic power station, divide the inspection image into inspection sub-images, input the inspection sub-images into the YOLOv8n target detection model, output the fault coordinates corresponding to the fault center, and perform fault detection; wherein, the inspection image is divided into inspection sub-images to facilitate the YOLOv8n target detection model to perform fault detection;

[0105] A fault matching range determination module 42 is used to determine a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station; wherein the fault matching range includes a plurality of grid cells in the spatial division result;

[0106] The fault location module 43 is used to obtain the mapping relationship between the photovoltaic panels in the photovoltaic power station and the space division results, locate the fault of the photovoltaic panels in the fault center and the fault matching range according to the fault matching range, mapping relationship and fault coordinates, and record the photovoltaic panel number corresponding to the fault center.

[0107] In a possible implementation manner, the fault matching range determination module 42 is used to:

[0108] Determine the fault matching range based on the fault coordinates and the spatial division results of the photovoltaic power station, and also include:

[0109] Divide the space of the photovoltaic power station into multiple grid units;

[0110] Determine the grid unit to which the fault belongs according to the coordinates of the fault;

[0111] The nine-square grid consisting of the grid unit to which the fault coordinates belong and the eight adjacent grid units around it is determined as the fault matching range.

[0112] In a possible implementation manner, the fault location module 43 is used to:

[0113] According to the fault matching range, mapping relationship and fault coordinates, the fault center and the photovoltaic panel in the fault matching range are located, which also includes:

[0114] The fault coordinates corresponding to the fault center are matched one by one with the coordinates of all photovoltaic panels within the fault matching range. If the fault coordinates are within the coordinate range of a photovoltaic panel, the photovoltaic panel number is recorded.

[0115] The embodiment of the present invention obtains inspection sub-images by dividing the inspection image of the photovoltaic panel of the photovoltaic power station, divides the inspection image into inspection sub-images of smaller size, and inputs the inspection sub-images of smaller size into the YOLOv8n target detection model, thereby improving the detection efficiency of the YOLOv8n target detection model and facilitating fault detection. The inspection sub-image is input into the YOLOv8n target detection model for fault target detection, and the fault coordinates corresponding to the fault center are output. The YOLOv8n target detection model quickly detects the fault location and improves the efficiency of photovoltaic panel fault detection. According to the fault coordinates and the spatial division result of the photovoltaic power station, the fault matching range is determined. The fault matching range includes multiple grid units in the spatial division result. It is only necessary to match the fault coordinates corresponding to the fault center with the photovoltaic panels within the fault matching range to locate the fault, thereby reducing the number of photovoltaic panels that need to match the fault coordinates, greatly improving the fault location efficiency, and finally optimizing the overall efficiency of photovoltaic panel fault inspection and identification.

[0116] Figure 5 is a schematic diagram of a terminal provided by an embodiment of the present invention. Figure 5 As shown, the terminal 5 of this embodiment includes: a processor 50, a memory 51, and a computer program 52 stored in the memory 51 and executable on the processor 50. When the processor 50 executes the computer program 52, the steps in the above-mentioned various photovoltaic panel fault inspection method embodiments are implemented, such as Figure 2 Alternatively, when the processor 50 executes the computer program 52, the functions of the modules in the above-mentioned device embodiments are realized, for example Figure 4 The functions of each module are shown.

[0117] Exemplarily, the computer program 52 may be divided into one or more modules, which are stored in the memory 51 and executed by the processor 50 to implement the present invention. The one or more modules may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program 52 in the terminal 5. For example, the computer program 52 may be divided into Figure 4 The modules shown.

[0118] The terminal 5 may be a computing device such as a desktop computer, a notebook, a PDA, or a cloud server. The terminal 5 may include, but is not limited to, a processor 50 and a memory 51. Those skilled in the art will appreciate that Figure 5 It is only an example of terminal 5 and does not constitute a limitation on terminal 5. It may include more or fewer components than shown in the figure, or a combination of certain components, or different components. For example, the terminal may also include input and output devices, network access devices, buses, etc.

[0119] The processor 50 may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.

[0120] The memory 51 may be an internal storage unit of the terminal 5, such as a hard disk or memory of the terminal 5. The memory 51 may also be an external storage device of the terminal 5, such as a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the terminal 5. Further, the memory 51 may also include both an internal storage unit of the terminal 5 and an external storage device. The memory 51 is used to store the computer program and other programs and data required by the terminal. The memory 51 may also be used to temporarily store data that has been output or is to be output.

[0121] The technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example for illustration. In practical applications, the above-mentioned function allocation can be completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above. The functional units and modules in the embodiment can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of this application. The specific working process of the units and modules in the above-mentioned system can refer to the corresponding process in the aforementioned method embodiment, which will not be repeated here.

[0122] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0123] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0124] In the embodiments provided by the present invention, it should be understood that the disclosed devices / terminals and methods can be implemented in other ways. For example, the device / terminal embodiments described above are only schematic. For example, the division of the modules or units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0125] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0126] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0127] If the integrated module is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by the processor, the steps of the above-mentioned various inspection photovoltaic panel fault method embodiments can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0128] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.

Claims

1. A method for inspecting photovoltaic panel failures, characterized in that: include: Obtain an inspection image of the photovoltaic panels of the photovoltaic power station, divide the inspection image into inspection sub-images, input the inspection sub-images into the YOLOv8n target detection model, output the fault coordinates corresponding to the fault center, and perform fault detection; wherein, dividing the inspection image into inspection sub-images facilitates the YOLOv8n target detection model to perform fault detection; Determining a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station; wherein the fault matching range includes a plurality of grid cells in the spatial division result; Obtain a mapping relationship between the photovoltaic panels in the photovoltaic power station and the spatial division results, locate the fault of the photovoltaic panels in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates, and record the photovoltaic panel number corresponding to the fault center.

2. The method for inspecting photovoltaic panel failures according to claim 1, characterized in that: Determining a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station also includes: Divide the space of the photovoltaic power station into multiple grid units; Determine the grid unit to which the fault belongs according to the coordinates of the fault; The nine-square grid consisting of the grid unit to which the fault coordinates belong and the eight adjacent grid units around it is determined as the fault matching range.

3. The method for inspecting photovoltaic panel failures according to claim 2, characterized in that: According to the fault matching range, the mapping relationship and the fault coordinates, the fault center and the photovoltaic panels in the fault matching range are fault located, further comprising: The fault coordinates corresponding to the fault center are matched one by one with the coordinates of all photovoltaic panels within the fault matching range. If the fault coordinates are within the coordinate range of a photovoltaic panel, the photovoltaic panel number is recorded.

4. The method for inspecting photovoltaic panel failures according to claim 1, characterized in that: Before obtaining the inspection image of the photovoltaic panels of the photovoltaic power station, the method further includes: Obtain the longitude and latitude coordinate information of the area where the photovoltaic power station is located, and divide the area into a number of grid units according to the longitude and latitude coordinate information; Acquire a photovoltaic power station modeling image, and establish photovoltaic panel coordinate file information according to the photovoltaic power station modeling image; wherein the photovoltaic panel coordinate file information includes the photovoltaic panel number, the longitude and latitude coordinate information of the center point of the photovoltaic panel, and the longitude and latitude coordinate information of the vertices of the photovoltaic panel; According to the latitude and longitude coordinate information of the center point of the photovoltaic panel, each photovoltaic panel is mapped to the corresponding grid unit.

5. The method for inspecting photovoltaic panel failures according to claim 4, characterized in that: The photovoltaic panel coordinate file information is established according to the photovoltaic power station modeling image, including: Use PYODM to synthesize the photovoltaic power station modeling image into the photovoltaic power station orthophoto map; The orthophoto image of the photovoltaic power station is cut into several orthophoto sub-images, and the orthophoto sub-images are sequentially input into the Real-ESRGAN super-resolution model and the YOLOv8n-seg instance segmentation model to output the photovoltaic panel image; wherein the Real-ESRGAN super-resolution model takes the orthophoto sub-image as input and the orthophoto sub-image with improved quality as output, thereby improving the quality of the orthophoto sub-image and facilitating the YOLOv8n-seg instance segmentation model to segment the photovoltaic panel image; the YOLOv8n-seg instance segmentation model takes the orthophoto sub-image with improved quality as input and the photovoltaic panel image as output, thereby performing the segmentation of the photovoltaic panel image; the orthophoto image of the photovoltaic power station, the orthophoto sub-image and the photovoltaic panel image all contain latitude and longitude coordinate information; The photovoltaic panel images are numbered, and photovoltaic panel coordinate file information is established according to the number of the photovoltaic panel image, the longitude and latitude coordinates of its center point, and the longitude and latitude coordinates of its vertices.

6. The method for inspecting photovoltaic panel failures according to claim 5, characterized in that: The Real-ESRGAN super-resolution model includes a sinc filter, and the sinc filter kernel is: Where (i, j) is the coordinate of the kernel, ω c is the cutoff frequency and J1 is the first-order Bessel function.

7. The method for inspecting photovoltaic panel failures according to claim 1, characterized in that: The size of the mesh unit is 1*1 meter, the length of the photovoltaic panel is 1.5-2 meters, and the width of the photovoltaic panel is 0.8-1 meter.

8. A photovoltaic panel fault inspection device, characterized in that: include: A fault coordinate acquisition module is used to acquire an inspection image of a photovoltaic panel of a photovoltaic power station, divide the inspection image into inspection sub-images, input the inspection sub-images into a YOLOv8n target detection model, output the fault coordinates corresponding to the fault center, and perform fault detection; wherein, dividing the inspection image into inspection sub-images facilitates the YOLOv8n target detection model to perform fault detection; A fault matching range determination module, used to determine a fault matching range according to the fault coordinates and the spatial division result of the photovoltaic power station; wherein the fault matching range includes a plurality of grid cells in the spatial division result; The fault location module is used to obtain the mapping relationship between the photovoltaic panels in the photovoltaic power station and the space division results, locate the fault of the photovoltaic panels in the fault center and the fault matching range according to the fault matching range, the mapping relationship and the fault coordinates, and record the photovoltaic panel number corresponding to the fault center.

9. A terminal comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the method as claimed in any one of claims 1 to 7 are implemented.

Images