Offshore Photovoltaic Cleaning Method, Device, Equipment and Medium

Abstract

The present application discloses a method, device, equipment and medium for cleaning off - grid photovoltaic power generation, which is applied to the field of new energy technologies. The method includes: obtaining a first photovoltaic image of a photovoltaic platform; analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform, where the pollution result is used to characterize whether the photovoltaic platform is polluted; when the pollution result indicates that the photovoltaic platform is polluted, determining the type of pollutant and the polluted area corresponding to the pollutant type; sending the pollutant type and the polluted area to a cleaning system so that the cleaning system adjusts its pose according to the pollutant type and the polluted area to clean the polluted area; obtaining a second photovoltaic image of the photovoltaic platform; when the second photovoltaic image does not meet the cleaning standard, sending an adjustment instruction to the cleaning system so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. This solution reduces the cleaning cost and improves the cleaning efficiency of the off - grid photovoltaic platform.

Description

Technical Field

[0001] The present invention generally relates to the field of new energy technologies, and particularly to a method, device, equipment and medium for cleaning offshore photovoltaic panels. Background Art

[0002] With the continuous development of new energy technologies, solar energy, as a clean, pollution-free, renewable and sustainable energy source, can convert sunlight into electrical energy through photovoltaic panels and has been increasingly applied in various different scenarios, such as on land, in the sea area, and on the roofs and surfaces of buildings. However, for offshore photovoltaic panels, since they are long-term exposed to the marine environment, they are easily contaminated by pollutants such as salt spray, sea waves, and suspended solids, thereby affecting their power generation efficiency and service life. In order to reduce the pollution of corrosive substances on the surface of photovoltaic modules and extend the service life of photovoltaic panels, it is particularly important to study how to clean offshore photovoltaics.

[0003] Currently, in related technologies, the artificial cleaning method is adopted to clean offshore photovoltaics. However, this method requires a large amount of human resources for large-scale offshore photovoltaic power stations, resulting in high labor costs and poor cleaning efficiency. Summary of the Invention

[0004] In view of the above-mentioned defects or deficiencies in the prior art, it is desirable to provide a method, device, equipment and medium for cleaning offshore photovoltaics.

[0005] In a first aspect, an embodiment of the present application provides a method for cleaning offshore photovoltaics, the method comprising:

[0006] Obtaining a first photovoltaic image of a photovoltaic platform;

[0007] Analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted;

[0008] When the pollution result indicates that the photovoltaic platform is polluted, determining the type of pollutant and the polluted area corresponding to the type of pollutant;

[0009] Sending the type of pollutant and the polluted area to a cleaning system so that the cleaning system adjusts its pose according to the type of pollutant and the polluted area and cleans the polluted area;

[0010] Obtaining a second photovoltaic image of the photovoltaic platform;

[0011] When the second photovoltaic image does not meet the cleaning standard, sending an adjustment instruction to the cleaning system so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0012] In one embodiment, the first photovoltaic image is analyzed and processed to determine the pollution result of the photovoltaic platform, including:

[0013] Obtain the environmental information of the area where the photovoltaic platform is located; the environmental information includes wind speed information and weather information;

[0014] Preprocess the first photovoltaic image according to the environmental information to obtain a preprocessed photovoltaic image;

[0015] Extract features from the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform.

[0016] In one embodiment, preprocessing the first photovoltaic image according to the environmental information to obtain a preprocessed photovoltaic image includes:

[0017] Perform denoising processing on the first photovoltaic image to obtain a denoised image;

[0018] Obtain the color values of each pixel point in the RGB channels of the denoised image;

[0019] Perform weighted summation processing on the color values and preset weight coefficients, calculate the gray mean value of each pixel point, and generate a gray image according to the gray mean value;

[0020] Perform chromaticity correction and brightness correction processing on the gray image according to the environmental information to obtain a preprocessed photovoltaic image.

[0021] In one embodiment, extracting features from the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform includes:

[0022] Perform filtering processing on the preprocessed photovoltaic image to obtain a filtered image, and construct a high-acrylic pyramid and a DoG scale space according to the filtered image; the high-acrylic pyramid includes multiple layers, each layer is provided with an image, the sizes of the images in each layer decrease sequentially from bottom to top, the image in the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images in the other layers except the bottom layer of the multiple layers are filtered images;

[0023] Locate the extreme points in the filtered image to obtain initial extreme points;

[0024] Compare the relationship between the local features and the overall features of the initial extreme points at different scales in the DoG scale space, remove the interfering extreme points in the initial extreme points, and obtain a segmented image;

[0025] Perform detection processing on the segmented image to determine the pollution result of the photovoltaic platform.

[0026] In one embodiment, performing detection processing on the segmented image to determine the pollution result of the photovoltaic platform includes:

[0027] Obtain the area to be detected from the segmented image, and obtain the shape matching library of the area to be detected;

[0028] Perform image registration processing on each shape match in the shape matching library;

[0029] Traverse each pixel point in the area to be detected according to the registration result, and determine the detection result of the area where the pixel point is located; the detection result is used to characterize whether there is pollution in the area where the pixel point is located;

[0030] According to the detection result, accumulate the areas of the areas where all the polluted pixel points are located to obtain the total area of the polluted areas;

[0031] When the total area of the polluted areas is greater than the preset area threshold, it is determined that the area to be detected is polluted; the preset area threshold is the total window area of the preset coefficient.

[0032] In one embodiment, traversing each pixel point in the area to be detected to determine the detection result of the area where the pixel point is located includes:

[0033] For each pixel point in the area to be detected, obtain the color of the pixel point, the gray difference between the pixel point and the adjacent pixel points, and determine whether there are connected black pixel points;

[0034] When the color of the pixel point is black, the gray difference between the pixel point and the adjacent pixel points is greater than the preset gray difference and there are no connected black pixel points, it is determined that there is pollution in the area where the pixel point is located.

[0035] In one embodiment, sending the pollutant type and the polluted area to the cleaning system includes:

[0036] Perform horizontal and vertical comparison on the pollution results and calculate the variance of the pollution results;

[0037] Judge whether the variance is the minimum value among all the current variances;

[0038] When the variance is the minimum value, generate the pose of the cleaning system according to the pollution result with the minimum variance; the pose includes angle information and position information;

[0039] Package the pose, pollutant type and polluted area into a control instruction and send it to the cleaning system.

[0040] In a second aspect, the present application provides an offshore photovoltaic cleaning device, and the device includes:

[0041] A first acquisition module, configured to acquire a first photovoltaic image of a photovoltaic platform;

[0042] An analysis and processing module for analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted.

[0043] An area determination module for determining the pollutant type and the pollution area corresponding to the pollutant type when the pollution result indicates that the photovoltaic platform is polluted.

[0044] A first sending module for sending the pollutant type and the pollution area to the cleaning system, so that the cleaning system adjusts its pose according to the pollutant type and the pollution area and cleans the pollution area.

[0045] A second acquisition module for acquiring the second photovoltaic image of the photovoltaic platform.

[0046] A second sending module for sending an adjustment instruction to the cleaning system when the second photovoltaic image does not meet the cleaning standard, so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0047] In a third aspect, an embodiment of the present application provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the offshore photovoltaic cleaning method as described in the first aspect above.

[0048] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored, and the computer program is used to implement the offshore photovoltaic cleaning method as described in the first aspect above.

[0049] The marine photovoltaic cleaning method, device, equipment and medium provided in the embodiments of the present application, the method includes: obtaining a first photovoltaic image of a photovoltaic platform, analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform, when the pollution result indicates that the photovoltaic platform is polluted, determining the pollutant type and the pollution area corresponding to the pollutant type, and then sending the pollutant type and the pollution area to a cleaning system, so that the cleaning system adjusts its pose according to the pollutant type and the pollution area, cleans the pollution area, and obtains a second photovoltaic image of the photovoltaic platform. When the second photovoltaic image does not meet the cleaning standard, sending an adjustment instruction to the cleaning system, so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. Compared with the prior art, in this solution, there is no need for manual inspection. Through the analysis and processing of the first photovoltaic image, the pollution result of the photovoltaic platform can be accurately determined. After determining the pollutant type and the pollution area, the cleaning system is automatically controlled to adjust its pose according to the pollutant type and the pollution area, realizing the efficient and rapid cleaning of the marine photovoltaic platform. And after the initial cleaning process, a second photovoltaic image of the photovoltaic platform is obtained again, so as to evaluate the cleaning effect in real time. According to the evaluation result, the cleaning system is controlled to adjust its pose again to clean the marine photovoltaic platform until the cleaning standard is reached, reducing the cleaning cost, reducing the risks caused by human operations, ensuring the cleaning quality, improving the cleaning efficiency of the marine photovoltaic platform, and further improving the availability and power generation efficiency of the photovoltaic panels. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required to be used in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0051] Figure 1 It is a system architecture diagram of an application system for the marine photovoltaic cleaning method provided in the embodiments of the present application;

[0052] Figure 2 It is a flowchart of the marine photovoltaic cleaning method provided in the embodiments of the present application;

[0053] Figure 3 It is a flowchart of the marine photovoltaic cleaning method provided in the embodiments of the present application;

[0054] Figure 4 It is a flowchart of the method for extracting features from the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform provided in the embodiments of the present application;

[0055] Figure 5Schematic flow chart of the offshore photovoltaic cleaning method provided by the embodiment of the present application;

[0056] Figure 6 Schematic flow chart of the method for traversing all pixel points in the image and determining whether there is pollution provided by the embodiment of the present application;

[0057] Figure 7 Schematic flow chart of the method for obtaining the optimal solution through update and iteration provided by the embodiment of the present application;

[0058] Figure 8 Schematic structural diagram of the offshore photovoltaic cleaning device provided by the embodiment of the present application;

[0059] Figure 9 Schematic structural diagram of a computer device shown in the embodiment of the present application. Detailed implementation manners

[0060] The present application will be further described in detail below with reference to the drawings and embodiments. It can be understood that the specific embodiments described herein are only used to explain the related invention and not to limit the invention. In addition, it should be noted that, for the convenience of description, only the parts related to the invention are shown in the drawings.

[0061] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other. The present application will be described in detail below with reference to the drawings and embodiments.

[0062] As mentioned in the background art, in one way of the related art, natural cleaning is used to achieve cleaning. By utilizing the wind speed and waves of the ocean to clean the offshore photovoltaic equipment, however, this method is extremely vulnerable to weather and the cleaning effect is poor; in another way, manual cleaning is used, but this method requires a large amount of human resources for large-scale offshore photovoltaic power stations, resulting in high labor costs and poor cleaning efficiency.

[0063] Based on the above defects, the present application provides a method, device, equipment and medium for cleaning marine photovoltaic systems. Compared with the prior art, in this solution, there is no need for manual inspection. By analyzing and processing the first photovoltaic image, the pollution result of the photovoltaic platform can be accurately determined. After determining the type of pollutants and the polluted area, the cleaning system is automatically controlled to adjust its pose according to the type of pollutants and the polluted area, so as to achieve efficient and rapid cleaning of the marine photovoltaic platform. After the initial cleaning process, the second photovoltaic image of the photovoltaic platform is obtained again to evaluate the cleaning effect in real time. Then, according to the evaluation result, the cleaning system is controlled to adjust its pose again to clean the marine photovoltaic platform until the cleaning standard is reached. This reduces the cleaning cost, reduces the risk caused by human operation, ensures the cleaning quality, improves the cleaning efficiency of the marine photovoltaic platform, and further improves the availability and power generation efficiency of the photovoltaic panels.

[0064] Figure 1 It is an implementation environment architecture diagram of a method for cleaning marine photovoltaic systems provided by an embodiment of the present application. As Figure 1 shown, the implementation environment architecture includes: a computer device 10 and a cleaning system 20. The computer device may include a terminal 100 and a server 200. The computer device 10 can establish a communication connection with the cleaning system 20.

[0065] The terminal 100 can be a terminal device in various AI application scenarios. For example, the terminal 100 can be a smart home device such as a smart TV or a smart TV set-top box, or the terminal 100 can be a mobile portable terminal such as a smart phone, a tablet computer, and an e-book reader. Alternatively, the terminal 100 can be a smart wearable device such as smart glasses or a smart watch. This embodiment does not make specific limitations in this regard.

[0066] The server 200 can be a single server, or a server cluster composed of several servers. Alternatively, the server 200 can include one or more virtualization platforms, or the server 200 can be a cloud computing service center.

[0067] Among them, the server 200 can be a server device that provides background services for the AI applications installed in the above terminal 100.

[0068] A communication connection is established between the terminal 100 and the server 200 through a wired or wireless network. Optionally, the above-mentioned wireless network or wired network uses standard communication technologies and / or protocols. The network is usually the Internet, but can also be any network, including but not limited to any combination of a local area network (LAN), a metropolitan area network (MAN), a wide area network (WAN), a mobile, wired or wireless network, a private network or a virtual private network.

[0069] Optionally, the terminal 100 is used to acquire a photovoltaic image of the photovoltaic platform and send the photovoltaic image to the server 200, so that the server 200 analyzes and processes the photovoltaic image, determines the pollution result of the photovoltaic platform, and generates a control instruction to send to the cleaning system 20, so that the cleaning system 20 cleans the photovoltaic platform according to the control instruction, and after the cleaning process, the server 200 determines in real time whether it meets the cleaning standard. When it does not meet the cleaning standard, it continues to adjust the pose to clean the photovoltaic platform until the cleaning standard is reached.

[0070] The above-mentioned cleaning system 20 may include an automatic cleaning robot, a drone cleaning device or other mechanical devices capable of performing cleaning. The cleaning system 20 is used to receive and respond to the control instruction sent by the computer device 10, and adjust the pose to clean the photovoltaic platform.

[0071] For the convenience of understanding and illustration, the following Figures 2 to 9 elaborates in detail the offshore photovoltaic cleaning method, device, equipment and medium provided by the embodiments of the present application.

[0072] Figure 2 The following shows a schematic flow chart of the offshore photovoltaic cleaning method provided by the embodiments of the present application. This method can be executed by a computer device, and the computer device can be the server 200 or the terminal 100 in the above Figure 1 shown system, or the computer device can also be a combination of the terminal 100 and the server 200. As Figure 2 shown, the method includes:

[0073] S101. Acquire a first photovoltaic image of the photovoltaic platform.

[0074] It should be noted that the above photovoltaic platform is located at sea. The photovoltaic platform can include a floating photovoltaic platform, a fixed photovoltaic platform, etc. The floating photovoltaic platform includes components such as a floating body, photovoltaic panels, and brackets; the fixed photovoltaic platform includes a steel pipe pile fixed platform, a jacket fixed photovoltaic platform, etc., and its components include photovoltaic panels, brackets, etc. The first photovoltaic image of the above photovoltaic platform can include a photovoltaic panel image, and the photovoltaic panel image can be arranged in a square matrix form or in a queue form.

[0075] Optionally, the first photovoltaic image of the above photovoltaic platform can be obtained by importing through an external device, or can be obtained by real-time detection by a camera, or can also be obtained from a database or blockchain. This embodiment does not impose any limitations on the acquisition method of the first photovoltaic image of the photovoltaic platform.

[0076] Exemplarily, a high-definition surveillance camera is installed in the square matrix area of the offshore photovoltaic platform, and real-time image acquisition of the photovoltaic platform is performed through this high-definition surveillance camera. Specifically, a computer device can send an image acquisition instruction to the surveillance camera. After receiving the image acquisition instruction, the surveillance camera performs shooting processing on the offshore photovoltaic platform, thereby collecting the first photovoltaic image and sending it to the computer device, enabling the computer device to obtain the first photovoltaic image of the photovoltaic platform. Among them, the above high-definition surveillance camera can have functions such as waterproof, dustproof, anti-corrosion, and anti-salt spray, which can make the collected first photovoltaic image clearer and facilitate subsequent accurate image processing.

[0077] S102. Analyze and process the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted.

[0078] S103. When the pollution result characterizes that the photovoltaic platform is polluted, determine the pollutant type and the pollution area corresponding to the pollutant type.

[0079] Specifically, after obtaining the first photovoltaic image of the photovoltaic platform, an image recognition algorithm and an image analysis algorithm can be used to perform image analysis and processing on the first photovoltaic image to determine whether the photovoltaic platform is polluted. Among them, the situation where the photovoltaic platform is polluted can be pollution by pollutants such as salt spray, sea waves, and suspended matter. The image recognition algorithm can be an image recognition algorithm based on deep learning. This algorithm is trained and learned through a large number of sample data to enable it to have the ability to recognize common pollutant types of offshore photovoltaic platforms.

[0080] It can be understood that the pollution results of the above photovoltaic platform may include pollution of the photovoltaic panels in the photovoltaic platform or no pollution of the photovoltaic panels. When pollution occurs, the type of pollutant and the pollution area corresponding to the pollutant type are determined. Since the environment of the area where the photovoltaic platform is located will affect the generated images, there are significant differences in light intensity at different times and weather conditions. Strong light will make the photovoltaic images too bright and some details will be lost; weak light will make the images dim, the signal-to-noise ratio will decrease, and details will be difficult to distinguish, increasing the difficulty of identifying the state of photovoltaic components and detecting faults in the images. As the sun moves, the light angle changes, and light and dark differences appear on the surface of the photovoltaic panels, resulting in uneven image grayscale, interfering with image segmentation and feature extraction, and affecting the accurate assessment of the overall state of the photovoltaic panels. Therefore, environmental information of the area where the photovoltaic platform is located needs to be considered in the process of analyzing and processing the first photovoltaic image to determine the pollution results of the photovoltaic platform.

[0081] In the process of analyzing and processing the first photovoltaic image to determine the pollution results of the photovoltaic platform, it includes: obtaining the environmental information of the area where the photovoltaic platform is located; the environmental information includes wind speed information and weather information; preprocessing the first photovoltaic image according to the environmental information to obtain the preprocessed photovoltaic image; extracting features from the preprocessed photovoltaic image to determine the pollution results of the photovoltaic platform.

[0082] Among them, since the surface of the offshore photovoltaic panels is often blown by the forces of sea waves, sea winds, etc., and due to its own complete structural characteristics, there are often attachments on its surface. Therefore, the image needs to be processed before analyzing the first photovoltaic image. Among them, the above attachments may include bird droppings, foam, driftwood, etc. After obtaining the environmental information of the area where the photovoltaic platform is located, the first photovoltaic image can be preprocessed according to the environmental information. Through processing such as denoising, gray correction, chromaticity correction, and brightness correction, the preprocessed image is obtained, and then features are extracted from the preprocessed photovoltaic image, such as edge detection, texture analysis, shape matching, etc., to determine the pollution results of the photovoltaic platform.

[0083] After determining the first photovoltaic image, perform denoising processing on the first photovoltaic image to obtain the denoised image, and obtain the color values of each pixel point in the RGB channels of the denoised image; perform weighted summation processing on the color values and the preset weight coefficients, calculate the gray mean value of each pixel point, and generate a gray image according to the gray mean value; perform chromaticity correction and brightness correction processing on the gray image according to the environmental information to obtain the preprocessed photovoltaic image.

[0084] In this embodiment, after denoising the first photovoltaic image to obtain the denoised image, the color values of each pixel point in the RGB (red, green, blue) channels of the denoised image are determined. The color of each of the three channels is usually between 0 and 255, and the weight coefficients of each color channel are obtained. The weight coefficients are custom-set according to actual requirements. For example, they can be determined according to the sensitivity of the human eye to different colors. Then, the color values and the preset weight coefficients are subjected to weighted summation processing to calculate the gray mean value of each pixel point, and a grayscale image is generated based on the gray mean value. For example, if the weight coefficient of the R channel is 0.299, the weight coefficient of the B channel is 0.114, and the weight coefficient of the G channel is 0.587, the gray mean value is obtained by weighted summation, which is expressed by the following formula:

[0085] Gray = 0.299×R + 0.587×G + 0.114×B;

[0086] Wherein, R is the color value of the pixel point in the R channel, G is the color value of the pixel point in the G channel, B is the color value of the pixel point in the B channel, and Gray is the gray mean value.

[0087] In this embodiment, by obtaining the color values of each pixel point in the RGB channels and calculating the gray mean value of each pixel point by weighted summation, and then generating a grayscale image, the image data processing time can be greatly reduced, and the image processing efficiency can be improved.

[0088] Further, when the pollution result indicates that the photovoltaic platform is polluted, a neural network model can be used to identify the first photovoltaic image to determine the pollutant type and the pollution area corresponding to the pollutant type. The pollutant type can include bird droppings, foam, driftwood, etc., and the pollution area can be a partial area in the first photovoltaic image. Among them, the neural network model can be, for example, a convolutional neural network model. The neural network model is trained according to the sample image data and the image annotation results, and is used to identify the pollutant type and the corresponding pollution area in the image, and has the ability to accurately identify the common pollutant types on the offshore photovoltaic platform.

[0089] S104. Send the pollutant type and the pollution area to the cleaning system so that the cleaning system adjusts its pose according to the pollutant type and the pollution area and cleans the pollution area.

[0090] After determining the pollutant type and the pollution area, the pollution type and the pollution area can be encapsulated into a control instruction and sent to the cleaning system. After receiving the control instruction, the cleaning system responds to the control instruction, adjusts its pose, and cleans the pollution area. For example, the pose of a robot is controlled or a drone is adjusted to clean the pollution area on the photovoltaic platform.

[0091] S105. Obtain the second photovoltaic image of the photovoltaic platform.

[0092] S106. When the second photovoltaic image does not meet the cleaning standard, send an adjustment instruction to the cleaning system so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0093] In this embodiment, the second photovoltaic image is an image of the photovoltaic platform collected after the cleaning system performs cleaning on the photovoltaic platform. The above cleaning standard is custom - set according to actual needs. For example, it can be that the proportion of the polluted area in the total area is greater than a preset proportion, and the preset proportion can be adjusted adaptively.

[0094] Please refer to Figure 3 As shown, the method for cleaning a marine photovoltaic platform provided in the embodiment of the present application is applied to a computer device and a cleaning system, and includes the following steps:

[0095] S1001. The computer device obtains the first photovoltaic image of the photovoltaic platform.

[0096] S1002. The computer device uses an intelligent algorithm to process and analyze the first photovoltaic image to determine whether pollution has occurred.

[0097] S1003. When no pollution has occurred, the computer device waits for the execution of the next timing task.

[0098] S1004. When pollution has occurred, the computer device determines the type of pollutant and the polluted area.

[0099] S1005. The computer device sends the type of pollutant and the polluted area to the cleaning system.

[0100] S1006. The cleaning system adjusts its pose according to the type of pollutant and the polluted area to perform cleaning on the photovoltaic platform.

[0101] S1007. The computer device obtains the second photovoltaic image of the photovoltaic platform.

[0102] S1008. The computer device determines whether the second photovoltaic image meets the cleaning standard.

[0103] S1009. When the cleaning standard is met, the process ends.

[0104] S1010. When the cleaning standard is not met, the computer device sends an adjustment instruction to the cleaning system.

[0105] S1011. The cleaning system receives and responds to the adjustment instruction, and readjusts the cleaning method of the cleaning system to perform cleaning on the photovoltaic platform until the cleaning standard is met.

[0106] Specifically, a monitoring camera is installed on the offshore photovoltaic platform. The camera can be a high-definition camera. The first photovoltaic image of the photovoltaic platform is collected by the camera and sent to the computer device. Then, the computer device obtains the first photovoltaic image of the photovoltaic platform. An intelligent algorithm is used to process and analyze the first photovoltaic image to determine whether pollution has occurred. When no pollution has occurred, wait for the execution of the next scheduled task. When pollution has occurred, determine the type of pollutant and the polluted area, and determine the location information of the polluted area through the positioning system. Then, send the type of pollutant and the polluted area to the cleaning system, so that the cleaning system adjusts the angle and position of the cleaning system according to the received polluted area and the type of pollutant, and performs cleaning treatment on the polluted area on the surface of the photovoltaic panel. Then, the photovoltaic platform is photographed again by the camera, so that the computer device obtains the second photovoltaic image, and the second photovoltaic image is evaluated to determine whether it meets the cleaning standard. When the cleaning standard is met, the process ends. When the cleaning standard is not met, readjust the cleaning method of the cleaning system to perform re-cleaning treatment on the photovoltaic platform until the cleaning standard is met. The cleaning method includes adjusting the position or attitude of the cleaning system.

[0107] Among them, after receiving the control instruction, the above-mentioned cleaning system can automatically adjust the position and direction of the cleaning equipment according to the position information of the polluted area, so as to realize the cleaning treatment of the polluted area in the offshore photovoltaic platform.

[0108] The offshore photovoltaic cleaning method provided in the embodiments of the present application includes: obtaining a first photovoltaic image of a photovoltaic platform, analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform. When the pollution result indicates that the photovoltaic platform is polluted, determining the pollutant type and the pollution area corresponding to the pollutant type, and then sending the pollutant type and the pollution area to a cleaning system, so that the cleaning system adjusts its pose according to the pollutant type and the pollution area, cleans the pollution area, and obtains a second photovoltaic image of the photovoltaic platform. When the second photovoltaic image does not meet the cleaning standard, sending an adjustment instruction to the cleaning system, so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. Compared with the prior art, in this solution, there is no need for manual inspection. By analyzing and processing the first photovoltaic image, the pollution result of the photovoltaic platform can be accurately determined. After determining the pollutant type and the pollution area, the cleaning system is controlled to automatically adjust its pose according to the pollutant type and the pollution area, realizing efficient and rapid cleaning of the offshore photovoltaic platform. After the initial cleaning process, the second photovoltaic image of the photovoltaic platform is obtained again to evaluate the cleaning effect in real time. According to the evaluation result, the cleaning system is controlled to adjust its pose again to clean the offshore photovoltaic platform until the cleaning standard is reached, reducing the cleaning cost, reducing the risks caused by human operation, ensuring the cleaning quality, improving the cleaning efficiency of the offshore photovoltaic platform, and further improving the availability and power generation efficiency of the photovoltaic panels.

[0109] In one of the embodiments, the present application also provides a specific implementation manner for extracting features from the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform. Please refer to Figure 4 as shown, the method includes:

[0110] S201. Perform filtering processing on the preprocessed photovoltaic image to obtain a filtered image, and construct a high-acrylic pyramid and a DoG scale space according to the filtered image; the high-acrylic pyramid includes multiple layers, each layer is provided with an image, the sizes of the images in each layer decrease sequentially from bottom to top, the image in the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images in the remaining layers except the bottom layer of the multiple layers are filtered images.

[0111] S202. Locate the extreme points in the filtered image to obtain initial extreme points.

[0112] S203. Compare the relationship between the local features and the overall features of the initial extreme points at different scales in the DoG scale space, and remove the interfering extreme points in the initial extreme points to obtain a segmented image.

[0113] S204. Perform detection processing on the segmented image to determine the pollution result of the photovoltaic platform.

[0114] It should be noted that the above-mentioned high-acrylic pyramid is a multi-scale representation method of images. It is constructed by performing multiple downsamplings (reducing the image size) on the preprocessed photovoltaic image and combining Gaussian filtering operations. Figuratively speaking, it is like building a pyramid. Starting from the preprocessed photovoltaic image at the bottom layer, each layer is processed based on the previous layer, thereby constructing the high-acrylic pyramid.

[0115] Please refer to Figure 5 As shown, high-definition surveillance cameras and cleaning systems are set on the phalanx area of the offshore photovoltaic platform. The first photovoltaic image of the photovoltaic platform is collected by the cameras and sent to the computer device. Then, the computer device obtains the first photovoltaic image of the photovoltaic platform and uses intelligent algorithms to process and analyze the first photovoltaic image to determine whether pollution has occurred. For example, an image recognition algorithm based on deep learning is used. When the first photovoltaic image is not polluted, no cleaning treatment is required, and the monitoring of the photovoltaic platform continues; when the first photovoltaic image is polluted, the type of pollutant and the polluted area are determined, and then the type of pollutant and the polluted area are sent to the cleaning system, so that the cleaning system cleans the polluted area of the photovoltaic platform according to the received polluted area and the type of pollutant. Then, the photovoltaic platform is photographed again by the cameras, so that the computer device obtains the second photovoltaic image, and the cleaning effect of the second photovoltaic image is evaluated to determine whether it meets the cleaning standard; when the cleaning standard is met, it indicates that the cleaning of the offshore photovoltaic platform is completed and the process execution ends; when the cleaning standard is not met, the cleaning system re-adjusts the intelligent algorithm parameters and re-cleans the photovoltaic platform according to the adjusted parameters until the cleaning standard is reached. The intelligent algorithm parameters may include adjusting the position or posture of the cleaning system.

[0116] In this embodiment, after obtaining the preprocessed photovoltaic image, global image feature extraction is performed on it. The Speeded Up Robust Features (SURF) algorithm can be used for feature extraction processing to determine the pollution result of the photovoltaic platform. The SURF algorithm is based on the Scale-Invariant Feature Transform (SIFT) algorithm and uses integral images and the Hessian matrix to achieve feature extraction and description. Its core idea is to detect representative feature points in different scale spaces and calculate the descriptors of these feature points to achieve image feature extraction and matching. The SURF algorithm is divided into two parts: low-level and high-level analysis. For the low-level part, it includes two steps: the first step is to perform Gaussian filtering on the preprocessed photovoltaic image to obtain the filtered image. This Gaussian filtering is an image smoothing method used to reduce image noise and details; the second step is to construct a Gaussian pyramid based on the filtered image and construct a Difference of Gaussian (DoG) scale space based on the Gaussian pyramid to locate the extreme points in the filtered image to obtain the initial extreme points. For the high-level analysis part, it is to further refine and distinguish the initial extreme points. By comparing the relationship between the local features and the overall features at different scales in the DoG scale space, the misjudged points after binarization are excluded, thereby obtaining the segmented image. The SURF algorithm also includes: a fast decision algorithm and a dynamic value function module. The fast decision algorithm is used to quickly exclude irrelevant extreme points, and the dynamic value function module is a heuristic criterion that excludes those extreme points that do not bring useful information based on the local features of the extreme points and their corresponding scales. The image after the above processing has removed the influence of interfering objects and retained the main features.

[0117] Specifically, after obtaining the preprocessed photovoltaic image, Gaussian filtering needs to be performed on the image, which can be expressed by the following formula:

[0118] ;

[0119] where σ is the standard deviation used to control the width of the Gaussian function, and x and y are the horizontal and vertical coordinates of the feature points in the image, respectively.

[0120] After performing Gaussian filtering, the feature points in the image can be detected by using the Hessian matrix. For a feature point (x, y) in the image, its Hessian matrix is defined as:

[0121] ;

[0122] where Lxx, Lxy, Lyx, and Lyy are the second-order partial derivatives of the image at the feature point (x, y), respectively, and σ is the scale parameter.

[0123] It is understandable that in the actual calculation process, the SURF algorithm can use the integral image to quickly calculate the determinant of the Hessian matrix. The approximate formula for the determinant is as follows:

[0124] ;

[0125] where Lxx, Lxy, Lyx, and Lyy are the second-order partial derivatives of the image at the feature point (x, y), and 0.9 here is an empirical coefficient used to balance the error caused by using the box filter approximation.

[0126] In the process of constructing the high Acrylic pyramid, the preprocessed photovoltaic image can be used as the bottom layer (layer 0) of the high Acrylic pyramid, and then Gaussian filtering is performed on the image of this layer. The function of Gaussian filtering is to smooth the image and reduce the noise in the image. After that, the image of the upper layer (layer 1) is obtained through the downsampling operation (usually removing the pixels of even rows and even columns). This process will be repeated continuously, and the size of the image of each layer gradually becomes smaller, forming a pyramid-shaped structure. For example, assume that the original image is an image of 512×512 pixels, which is the layer 0 of the high Acrylic pyramid. After the first Gaussian filtering and downsampling, an image of 256×256 pixels is obtained as the layer 1 of the high Acrylic pyramid, and through the same operation, an image of 128×128 pixels of the layer 2 can be obtained, etc. These image layers of different sizes can be used to analyze the image at different scales.

[0127] In this embodiment, the scale space of the high Acrylic pyramid of the SURF algorithm can be selected as 7 layers, and the initial extreme points are calculated at different scales for each layer. The initial scale of the high Acrylic pyramid can be d = 4.5833... Repeat this process until the bottom layer is decomposed to obtain 7 groups of initial extreme points in the scale space.

[0128] After constructing the high Acrylic pyramid, the DoG (Difference of Gaussian) scale space can be constructed based on the high Acrylic pyramid, which is obtained by subtracting the images of adjacent layers in the high Acrylic pyramid. In this way, a series of images representing the change differences of the image at different scales can be obtained, and these difference images constitute the DoG scale space.

[0129] In this embodiment, constructing the DoG scale space based on the high Acrylic pyramid can detect the feature points in the image, especially those positions that appear as extreme points in the scale space. At different scales, the features (such as edges, corners, etc.) in the image will have different manifestations. The DoG scale space can highlight the changes of these features at different scales, making the feature points easier to be detected.

[0130] For example, when detecting extreme points in an image, the response of extreme points in the DoG scale space is usually more obvious than that in the original image or a simple Gaussian pyramid image. Because DoG can capture the rapid change of pixel values of corner points between adjacent scales, and the position of this rapid change is likely to be the key feature points in the image. By searching for extreme points (maximum and minimum values) in the DoG scale space, possible feature points can be initially located, providing important clues for subsequent feature extraction and analysis.

[0131] After constructing the DoG scale space and locating the extreme points in the filtered image to obtain the initial extreme points, compare the relationship between the local features and the overall features of the initial extreme points at different scales in the DoG scale space. Remove the interfering extreme points in the initial extreme points quickly through a fast determination algorithm, and use a dynamic value function to exclude the interfering extreme points according to the local features of the initial extreme points and their corresponding scales, so as to retain the main features and obtain the segmented image.

[0132] Furthermore, in the process of detecting and processing the segmented image to determine the pollution result of the photovoltaic platform, it can be to obtain the area to be detected from the segmented image and obtain the shape matching library of the area to be detected; perform image registration processing on each shape matching in the shape matching library; traverse each pixel point in the area to be detected to determine the detection result of the area where the pixel point is located; the detection result is used to represent whether there is pollution in the area where the pixel point is located; according to the detection result, accumulate the areas of all the areas where the polluted pixel points are located to obtain the total area of the polluted areas; when the total area of the polluted areas is greater than the preset area threshold, it is determined that the area to be detected is polluted; the preset area threshold is the total window area of the preset coefficient.

[0133] After obtaining the segmented image, an arbitrary area can be selected from the segmented image as the area to be detected. Considering that the common pollutants on the photovoltaic panel are bird droppings, for example, stains arranged in a "one" shape or a "||" shape on the photovoltaic panel are less common. Therefore, a matching method based on shape matching and energy histogram is adopted to determine whether the photovoltaic panel is contaminated by stains. First, obtain the shape matching library of the area to be detected, and perform image registration processing on each shape match in the shape matching library to obtain the registration result, which is used to characterize the matching result between the area to be detected and the corresponding shape in the shape matching library, and can include successful matching or failed matching. Then, retrieve whether there is pollution in the image one by one according to the registration result. When the registration result is that the area to be detected matches successfully with the corresponding shape in the shape matching library, traverse each pixel point in the area to be detected to determine the detection result of the area where the pixel point is located, including: for each pixel point in the area to be detected, obtain the color of the pixel point, the gray difference between the pixel point and its neighboring pixel points, and determine whether there are connected black pixel points; when the color of the pixel point is black, the gray difference between the pixel point and its neighboring pixel points is greater than the preset gray difference, and there are no connected black pixel points, it is determined that there is pollution in the area where the pixel point is located.

[0134] Please refer to Figure 6 and Figure 7 As shown, first load the segmented image, initialize the total sum of the polluted area quantity Polluted AreaSum = 0, traverse each pixel point in the image. For each current pixel point, first determine whether the current pixel point is black. When it is black, calculate the gray difference between the current pixel point and its neighboring pixel points, and determine whether the gray difference is greater than the preset gray difference A. When it is greater than the preset gray difference A, check whether there are connected black pixel points for the current pixel point. When there are no connected black pixel points, it indicates that there is pollution in the area where the current pixel point is located. Mark the area where the current pixel point is located as a polluted area, and set Polluted AreaSum = 1. Then move to the next pixel point to perform the above pollution judgment process until all pixel points in the image have been checked; if the current pixel point is not black, or the current gray difference is not greater than the preset gray difference A, or there are connected black pixel points for the current pixel point, it indicates that there is no pollution in the area where the current pixel point is located. Skip the current pixel point and move to the next pixel point to perform the pollution judgment process until all pixel points in the area to be detected have been checked.

[0135] After all pixel points in the area to be detected have been inspected, the areas of all regions where polluted pixel points exist in the area to be detected are accumulated to obtain the total polluted area sum, P_Polluted AreaSum. Then, it is determined whether the total polluted area sum is greater than a preset area threshold. When the total polluted area sum, P_Polluted AreaSum, is greater than the preset area threshold, for example, taking the preset coefficient as 80%, that is, when P_Polluted AreaSum > 80% * Window Area (the total window area of the area to be detected), it is determined that the area to be detected is polluted and needs to be cleaned; otherwise, it is determined that the area to be detected is not polluted and does not need to be cleaned. Since there are multiple areas to be detected in the segmented image, the above process is repeated for each area to be detected to determine whether each area to be detected is polluted, thereby obtaining the pollution result of the photovoltaic platform. For example, this pollution result can be represented by a number. When the pollution result corresponding to the area to be detected is 1, it indicates that the area to be detected is polluted; when the pollution result corresponding to the area to be detected is 0, it indicates that the area to be detected is not polluted.

[0136] After obtaining the pollution result, the pollution results are compared horizontally and vertically, and the variance of the pollution results is calculated; it is determined whether the variance is the minimum among all variances; when the variance is the minimum, the pose information of the cleaning system is generated according to the pollution result with the minimum variance; the pose information includes angle information and position information;

[0137] The rotation angle, pollutant type, and polluted area are encapsulated into a control instruction and sent to the cleaning system. Among them, in the process of horizontally and vertically comparing the pollution results, it can be a horizontal and vertical comparison of the images corresponding to the areas to be detected with pollution. When the pollution result indicates that the area to be detected is polluted, an image detection model can be used for image detection processing to obtain the pollutant type. The image detection model can be, for example, a convolutional neural network model.

[0138] Exemplarily, after obtaining the pollution result of the i-th time, where i = 1, 2,..., nm, there is a corresponding result set. Then, a horizontal comparison is performed, and the mean and horizontal variance of all pollution results in the result set are calculated. The horizontal variance reflects the degree of dispersion of the data within the pollution result of the i-th time. The vertical variance is calculated, which reflects the degree of dispersion of the data at the same position in different calculations. Then, considering the horizontal and vertical variances comprehensively, a comprehensive variance index is obtained. This index can include the weight coefficients corresponding to the horizontal and vertical variances to balance the importance of the horizontal and vertical variances. When the comprehensive variance index is the minimum value, the pose of the cleaning system is generated according to the pollution result with the minimum comprehensive variance index. This pose includes angle information a and position information, which is the optimal solution. Then, the pose, pollutant type, and pollution area are encapsulated into a control instruction and sent to the cleaning system. The cleaning system receives and responds to the control instruction, adjusts the pose, and realizes the cleaning process of the photovoltaic platform.

[0139] Optionally, after cleaning the pollution area, the computer device collects the image information of the photovoltaic platform again to obtain the second photovoltaic image, and then evaluates the cleaning effect of the photovoltaic platform to ensure the cleaning quality. According to the evaluation result, the cleaning system can automatically adjust the intelligent algorithm parameters for updating and optimizing the model. Through continuous learning, the algorithm can more accurately judge the pollution degree, optimize the cleaning decision, and improve the cleaning efficiency.

[0140] In this embodiment, by real-time monitoring of the photovoltaic platform, it is automatically determined whether the cleaning standard is reached according to the image information, and automatic cleaning is performed when the cleaning standard is not reached, reducing the equipment maintenance cost and failure rate. Compared with the traditional manual inspection and cleaning method, it can realize the automatic and efficient cleaning of the photovoltaic platform, greatly improving the availability and power generation efficiency of the photovoltaic platform, reducing the maintenance cost and safety risk of the offshore photovoltaic platform, and further improving the stability and reliability of the entire system.

[0141] On the other hand, Figure 8 This is a schematic structural diagram of an offshore photovoltaic cleaning device provided by an embodiment of the present application. This device can be a device within a terminal or a server, such as Figure 8 As shown, the device includes:

[0142] A first acquisition module 710, configured to acquire a first photovoltaic image of the photovoltaic platform;

[0143] An analysis and processing module 720, configured to analyze and process the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted;

[0144] An area determination module 730, configured to determine the type of pollutant and the pollution area corresponding to the pollutant type when the pollution result indicates that the photovoltaic platform is polluted;

[0145] A first sending module 740, configured to send the pollutant type and the pollution area to the cleaning system, so that the cleaning system adjusts its pose according to the pollutant type and the pollution area to clean the pollution area;

[0146] A second acquisition module 750, configured to acquire a second photovoltaic image of the photovoltaic platform;

[0147] A second sending module 760, configured to send an adjustment instruction to the cleaning system when the second photovoltaic image does not meet the cleaning standard, so that the cleaning system adjusts its pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0148] Optionally, the analysis and processing module 720 is specifically configured to:

[0149] Acquire the environmental information of the area where the photovoltaic platform is located; the environmental information includes wind speed information and weather information;

[0150] Preprocess the first photovoltaic image according to the environmental information to obtain a preprocessed photovoltaic image;

[0151] Extract features from the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform.

[0152] Optionally, the analysis and processing module 720 is further configured to:

[0153] Perform denoising processing on the first photovoltaic image to obtain a denoised image;

[0154] Obtain the color values of each pixel point in the RGB channels of the denoised image;

[0155] Perform weighted summation processing on the color values and the preset weight coefficients, calculate the gray mean value of each pixel point, and generate a gray image according to the gray mean value;

[0156] Perform chromaticity correction and brightness correction processing on the gray image according to the environmental information to obtain a preprocessed photovoltaic image.

[0157] Optionally, the analysis and processing module 720 is further configured to:

[0158] Perform filtering processing on the preprocessed photovoltaic image to obtain a filtered image, and construct a high-acrylic pyramid and a DoG scale space according to the filtered image; the high-acrylic pyramid includes multiple layers, each layer is provided with an image, the sizes of the images in each layer decrease sequentially from bottom to top, the image in the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images in the other layers except the bottom layer of the multiple layers are filtered images;

[0159] Locate the extreme points in the filtered image to obtain the initial extreme points;

[0160] Compare the relationship between the local features and the overall features of the initial extreme points at different scales in the DoG scale space, remove the interfering extreme points in the initial extreme points, and obtain the segmented image;

[0161] Perform detection processing on the segmented image to determine the pollution result of the photovoltaic platform.

[0162] Optionally, the analysis and processing module 720 is further configured to:

[0163] Obtain the area to be detected from the segmented image, and obtain the shape matching library of the area to be detected;

[0164] Perform image registration processing on each shape matching in the shape matching library;

[0165] Traverse each pixel point in the area to be detected according to the registration result to determine the detection result of the area where the pixel point is located; the detection result is used to characterize whether there is pollution in the area where the pixel point is located;

[0166] According to the detection result, accumulate the areas of all the areas where the polluted pixel points are located to obtain the total pollution area;

[0167] When the total pollution area is greater than the preset area threshold, it is determined that the area to be detected is polluted; the preset area threshold is the total window area of the preset coefficient.

[0168] Optionally, the analysis and processing module 720 is further configured to:

[0169] For each pixel point in the area to be detected, obtain the color of the pixel point, the gray difference between the pixel point and the adjacent pixel points, and determine whether there are connected black pixel points;

[0170] When the color of the pixel point is black, the gray difference between the pixel point and the adjacent pixel points is greater than the preset gray difference, and there are no connected black pixel points, it is determined that there is pollution in the area where the pixel point is located.

[0171] Optionally, the first sending module 740 is specifically configured to:

[0172] Perform horizontal and vertical comparisons on the pollution result and calculate the variance of the pollution result;

[0173] Determine whether the variance is the minimum value among all the current variances;

[0174] When the variance is the minimum value, generate the pose of the cleaning system according to the pollution result with the minimum variance; the pose includes angle information and position information;

[0175] The pose, pollutant type, and pollution area are encapsulated into control instructions and sent to the cleaning system.

[0176] It can be understood that the functions of the functional modules of the offshore photovoltaic cleaning system in this embodiment can be specifically implemented according to the methods in the above method embodiments. The specific implementation process can refer to the relevant descriptions in the above method embodiments and will not be elaborated here.

[0177] On the other hand, the computer device provided in the embodiment of the present application includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the offshore photovoltaic cleaning method as described above.

[0178] Next, refer to Figure 9 , Figure 9 which is a schematic structural diagram of the computer system of the terminal device in the embodiment of the present application.

[0179] As Figure 9 shown, the computer system 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 302 or the program loaded from the storage section 303 into the random access memory (RAM) 303. In the RAM 303, various programs and data required for the operation of the system 300 are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other via a bus 304. The input / output (I / O) interface 305 is also connected to the bus 304.

[0180] The following components are connected to the I / O interface 305: an input section 306 including a keyboard, a mouse, etc.; an output section 307 including, for example, a cathode ray tube (CRT), a liquid crystal display (LCD), etc. and a speaker; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that the computer program read from it can be installed into the storage section 308 as needed.

[0181] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a machine-readable medium, and the computer program contains program codes for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 303, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, the above functions defined in the system of the present application are executed.

[0182] It should be noted that the computer-readable medium shown in the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the two. A computer-readable storage medium can be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of the computer-readable storage medium can include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present application, the computer-readable storage medium can be any tangible medium that contains or stores a program, and the program can be used by or in combination with an instruction execution system, apparatus, or device. In the present application, a computer-readable signal medium can include a data signal propagated in a baseband or as part of a carrier wave, in which computer-readable program codes are carried. Such a propagated data signal can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium can also be any computer-readable medium other than a computer-readable storage medium, and the computer-readable medium can send, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device. The program codes contained on the computer-readable medium can be transmitted by any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0183] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagram may represent a module, a program segment, or a portion of code that contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order than that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending on the functions involved. It should also be noted that each block in the block diagram and / or flowchart, as well as combinations of blocks in the block diagram and / or flowchart, may be implemented by a dedicated hardware-based system that performs the specified functions or operations, or may be implemented by a combination of dedicated hardware and computer instructions.

[0184] The units or modules involved in the embodiments described in the present application can be implemented in software or in hardware. The described units or modules can also be provided in a processor. For example, it can be described as a processor including a first acquisition module, an analysis and processing module, a region determination module, a first transmission module, a second acquisition module, and a second transmission module. Among them, the names of these units or modules do not, in some cases, constitute a limitation on the units or modules themselves. For example, the first acquisition module can also be described as "used to acquire the first photovoltaic image of the photovoltaic platform".

[0185] As another aspect, the present application also provides a computer-readable storage medium, which can be included in the electronic device described in the above embodiments; or can exist separately without being assembled into the electronic device. The above computer-readable storage medium stores one or more programs, and when the above-mentioned programs are executed by one or more processors to perform the offshore photovoltaic cleaning method described in the present application:

[0186] Acquire the first photovoltaic image of the photovoltaic platform;

[0187] Analyze and process the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted;

[0188] When the pollution result characterizes that the photovoltaic platform is polluted, determine the pollutant type and the pollution area corresponding to the pollutant type;

[0189] Send the pollutant type and the pollution area to the cleaning system so that the cleaning system adjusts its pose according to the pollutant type and the pollution area and cleans the pollution area;

[0190] Obtain a second photovoltaic image of the photovoltaic platform;

[0191] When the second photovoltaic image does not meet the cleaning standard, send an adjustment instruction to the cleaning system so that the cleaning system adjusts the pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0192] In summary, the offshore photovoltaic cleaning method, device, equipment and medium provided in the embodiments of the present application, the method includes: obtaining a first photovoltaic image of a photovoltaic platform, analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform, when the pollution result indicates that the photovoltaic platform is polluted, determining the pollutant type and the pollution area corresponding to the pollutant type, and then sending the pollutant type and the pollution area to the cleaning system so that the cleaning system adjusts the pose according to the pollutant type and the pollution area, cleans the pollution area, and obtains a second photovoltaic image of the photovoltaic platform. When the second photovoltaic image does not meet the cleaning standard, send an adjustment instruction to the cleaning system so that the cleaning system adjusts the pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. Compared with the prior art, in this solution, there is no need for manual inspection. Through the analysis and processing of the first photovoltaic image, the pollution result of the photovoltaic platform can be accurately determined. After determining the pollutant type and the pollution area, the cleaning system is automatically controlled to adjust the pose according to the pollutant type and the pollution area, realizing the efficient and rapid cleaning of the offshore photovoltaic platform. And after the initial cleaning process, a second photovoltaic image of the photovoltaic platform is obtained again to evaluate the cleaning effect in real time, and the cleaning system is controlled according to the evaluation result to adjust the pose again to clean the offshore photovoltaic platform until the cleaning standard is reached, reducing the cleaning cost, reducing the risks caused by human operations, ensuring the cleaning quality, improving the cleaning efficiency of the offshore photovoltaic platform, and further improving the availability and power generation efficiency of the photovoltaic panels.

[0193] The above description is only a preferred embodiment of the present application and an explanation of the applied technical principle. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solution formed by the specific combination of the above technical features, but should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the inventive concept. For example, the technical solutions formed by mutually replacing the above features with the (but not limited to) technical features with similar functions disclosed in the present application.

Claims

1. A method for cleaning marine photovoltaic systems, characterized in that, The offshore photovoltaic cleaning method includes: Obtaining a first photovoltaic image of a photovoltaic platform; Analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted; When the pollution result indicates that the photovoltaic platform is polluted, determining the pollutant type and the pollution area corresponding to the pollutant type; Sending the pollutant type and the pollution area to a cleaning system, so that the cleaning system adjusts its pose according to the pollutant type and the pollution area and cleans the pollution area; Obtaining a second photovoltaic image of the photovoltaic platform; When the second photovoltaic image does not meet the cleaning standard, sending an adjustment instruction to the cleaning system, so that the cleaning system adjusts the pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard; Among them, sending the pollutant type and the pollution area to the cleaning system includes: Making horizontal and vertical comparisons of the pollution result and calculating the variance of the pollution result; Judging whether the variance is the minimum value among all current variances; When the variance is the minimum value, generating the pose of the cleaning system according to the pollution result with the minimum variance; the pose includes angle information and position information; Encapsulating the pose, the pollutant type and the pollution area into a control instruction and sending it to the cleaning system.

2. The offshore photovoltaic cleaning method according to claim 1, characterized in that, Analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform includes: Obtaining the environmental information of the area where the photovoltaic platform is located; the environmental information includes wind speed information and weather information; Preprocessing the first photovoltaic image according to the environmental information to obtain a preprocessed photovoltaic image; Performing feature extraction on the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform.

3. The offshore photovoltaic cleaning method according to claim 2, characterized in that Preprocessing the first photovoltaic image according to the environmental information to obtain a preprocessed photovoltaic image includes: Performing denoising processing on the first photovoltaic image to obtain a denoised image; Obtaining the color values of each pixel point in the RGB channels of the denoised image; Performing weighted summation processing on the color values and a preset weight coefficient, calculating the gray mean value of each pixel point, and generating a gray image according to the gray mean value; Performing chromaticity correction and brightness correction processing on the gray image according to the environmental information to obtain a preprocessed photovoltaic image.

4. The offshore photovoltaic cleaning method according to claim 2, wherein, Performing feature extraction on the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform includes: Performing filtering processing on the preprocessed photovoltaic image to obtain a filtered image, and constructing a high Acrylic pyramid and a DoG scale space according to the filtered image; the high Acrylic pyramid includes multiple layers, each layer is provided with an image, the sizes of the images in each layer decrease sequentially from bottom to top, the image in the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images in the other layers except the bottom layer of the multiple layers are the filtered images; Locating the extreme points in the filtered image to obtain initial extreme points; Compare the relationship between the local features and the overall features of the initial extreme points at different scales in the DoG scale space, remove the interfering extreme points in the initial extreme points, and obtain the segmented image. Perform detection processing on the segmented image to determine the pollution result of the photovoltaic platform.

5. The offshore photovoltaic cleaning method according to claim 4, characterized in that, Performing detection processing on the segmented image to determine the pollution result of the photovoltaic platform includes: Obtain the area to be detected from the segmented image, and obtain the shape matching library of the area to be detected. Perform image registration processing on each shape match in the shape matching library. Traverse each pixel point in the area to be detected according to the registration result, and determine the detection result of the area where the pixel point is located; the detection result is used to characterize whether there is pollution in the area where the pixel point is located. According to the detection result, accumulate the areas of all the areas where the polluted pixel points are located to obtain the total area of the polluted areas. When the total area of the polluted areas is greater than the preset area threshold, it is determined that the area to be detected is polluted; the preset area threshold is the total window area of the preset coefficient.

6. The offshore photovoltaic cleaning method according to claim 5, traversing each pixel point in the area to be detected to determine the detection result of the area where the pixel point is located, including: For each pixel point in the area to be detected, obtain the color of the pixel point, the gray difference between the pixel point and the adjacent pixel points, and determine whether there are connected black pixel points. When the color of the pixel point is black, the gray difference between the pixel point and the adjacent pixel points is greater than the preset gray difference, and there are no connected black pixel points, it is determined that there is pollution in the area where the pixel point is located.

7. A marine photovoltaic cleaning device, characterized in that, The offshore photovoltaic cleaning device includes: A first acquisition module, configured to acquire a first photovoltaic image of a photovoltaic platform. An analysis and processing module, configured to perform analysis and processing on the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to characterize whether the photovoltaic platform is polluted. A region determination module, configured to determine the type of pollutant and the polluted region corresponding to the type of pollutant when the pollution result indicates that the photovoltaic platform is polluted. A first sending module, configured to send the type of pollutant and the polluted region to a cleaning system, so that the cleaning system adjusts its pose according to the type of pollutant and the polluted region and cleans the polluted region. A second acquisition module, configured to acquire a second photovoltaic image of the photovoltaic platform. A second sending module, configured to send an adjustment instruction to the cleaning system when the second photovoltaic image does not meet the cleaning standard, so that the cleaning system adjusts the pose again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. The first sending module is specifically configured to: Perform horizontal and vertical comparison on the pollution result, and calculate the variance of the pollution result. Determine whether the variance is the minimum value among all the current variances. When the variance is the minimum value, generate the pose of the cleaning system according to the pollution result with the minimum variance; the pose includes angle information and position information. Encapsulate the pose, the pollutant type, and the contaminated area into a control instruction and send it to the cleaning system.

8. A computer device, comprising: A memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the offshore photovoltaic cleaning method according to any one of claims 1-6.

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

Images