Offshore photovoltaic cleaning method, device, equipment and medium

Abstract

The invention discloses an offshore photovoltaic cleaning method, device and equipment and a medium, and is applied to the technical field of new energy, and the method comprises the steps: obtaining a first photovoltaic image of a photovoltaic platform; analyzing and processing the first photovoltaic image, and determining a pollution result of the photovoltaic platform; the pollution result is used for representing whether the photovoltaic platform is polluted; when the pollution result represents that the photovoltaic platform is polluted, determining a pollutant type and a polluted area corresponding to the pollutant type; the pollutant type and the polluted area are sent to a cleaning system, so that the cleaning system adjusts the pose according to the pollutant type and the polluted area, and the polluted area is cleaned; acquiring a second photovoltaic image of the photovoltaic platform; and 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 reaches the cleaning standard. According to the scheme, the cleaning cost is reduced, and the cleaning efficiency of the offshore photovoltaic platform is improved.

Description

Technical Field

[0001] The present invention generally relates to the field of new energy technology, and in particular to an offshore photovoltaic cleaning method, device, equipment and medium. 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. It has been increasingly used in various scenarios, such as land, sea, roofs and building surfaces. However, for offshore photovoltaic panels, due to long-term exposure to the marine environment, they are easily contaminated by pollutants such as salt spray, waves, and suspended matter, which affects their power generation efficiency and life. In order to reduce the contamination 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] At present, the relevant technology adopts manual cleaning method to clean offshore photovoltaics. However, this method requires a lot 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 an offshore photovoltaic cleaning method, device, equipment and medium.

[0005] In a first aspect, an embodiment of the present application provides an offshore photovoltaic cleaning method, the method comprising: acquiring a first photovoltaic image of the photovoltaic platform; Analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to indicate whether the photovoltaic platform is polluted; When the pollution result indicates that the photovoltaic platform is polluted, the pollutant type and the polluted area corresponding to the pollutant type are determined; Send the pollutant type and polluted area to the cleaning system, so that the cleaning system adjusts its position according to the pollutant type and polluted area and cleans the polluted area; acquiring a second photovoltaic image of the photovoltaic platform; When the second photovoltaic image does not meet the cleaning standard, an adjustment instruction is sent to the cleaning system, so that the cleaning system adjusts the posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0006] In one embodiment, analyzing and processing the first photovoltaic image to determine the contamination result of the photovoltaic platform includes: Obtain 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; Feature extraction is performed on the preprocessed photovoltaic image to determine the contamination results of the photovoltaic platform.

[0007] In one embodiment, preprocessing the first photovoltaic image according to the environmental information to obtain the preprocessed photovoltaic image includes: Performing denoising on the first photovoltaic image to obtain a denoised image; Get the color value of each pixel in the RGB channel of the denoised image; The color value and the preset weight coefficient are weighted and summed to calculate the grayscale mean of each pixel, and a grayscale image is generated based on the grayscale mean; The grayscale image is subjected to chromaticity correction and brightness correction according to the environmental information to obtain a preprocessed photovoltaic image.

[0008] In one embodiment, feature extraction is performed on the preprocessed photovoltaic image to determine the contamination result of the photovoltaic platform, including: The preprocessed photovoltaic image is filtered to obtain a filtered image, and a high acrylic pyramid and a DoG scale space are constructed according to the filtered image; the high acrylic pyramid includes multiple layers, each layer is provided with an image, and the size of each layer of the image decreases from bottom to top, the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images of the remaining layers except the bottom layer of the multiple layers are filtered images; Locate 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, and obtain the segmented image; The segmented images are detected and processed to determine the pollution results of the photovoltaic platform.

[0009] In one embodiment, the segmented image is detected and processed to determine the pollution result of the photovoltaic platform, including: Obtaining the area to be detected from the segmented image, and obtaining a shape matching library of the area to be detected; Performing image registration processing on each shape match in the shape matching library; According to the registration result, each pixel point in the area to be detected is traversed 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; According to the detection results, the areas of all the polluted pixels are accumulated to obtain the total area of ​​the polluted area; When the total area of ​​the polluted regions is greater than a 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.

[0010] In one embodiment, traversing each pixel point in the area to be detected and determining the detection result of the area where the pixel point is located includes: For each pixel in the area to be detected, obtain the color of the pixel, the grayscale difference between the pixel and its adjacent pixels, and determine whether the pixel is connected to a black pixel; When the color of a pixel is black, the grayscale difference between the pixel and adjacent pixels is greater than a preset grayscale difference, and there are no connected black pixels, it is determined that the area where the pixel is located is polluted.

[0011] In one embodiment, the type of contaminant and the contaminant area are sent to a cleaning system, including: Compare the pollution results horizontally and vertically, and calculate the variance of the pollution results; Determine whether the variance is the minimum value among all current variances; When the variance is the minimum value, the posture of the cleaning system is generated according to the pollution result with the minimum variance; the posture includes angle information and position information; The position, contaminant type and contamination area are packaged into control instructions and sent to the cleaning system.

[0012] In a second aspect, the present application provides an offshore photovoltaic cleaning device, the device comprising: A first acquisition module, used to acquire a first photovoltaic image of the photovoltaic platform; An analysis and processing module is used to analyze and process the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to indicate whether the photovoltaic platform is polluted; An area determination module, used to determine the type of pollutant and the polluted area corresponding to the type of pollutant when the pollution result indicates that the photovoltaic platform is polluted; A first sending module is used to send the pollutant type and the polluted area to the cleaning system, so that the cleaning system adjusts the posture according to the pollutant type and the polluted area to clean the polluted area; A second acquisition module, used to acquire a second photovoltaic image of the photovoltaic platform; The second sending module is used 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 posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0013] In a third aspect, an embodiment of the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the offshore photovoltaic cleaning method of the first aspect described above is implemented.

[0014] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is used to implement the offshore photovoltaic cleaning method as described in the first aspect above.

[0015] The offshore photovoltaic cleaning method, device, equipment and medium provided in the embodiments of the present application include: acquiring a first photovoltaic image of a photovoltaic platform, analyzing and processing the first photovoltaic image, determining the pollution result of the photovoltaic platform, and 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 its posture according to the pollutant type and the pollution area, cleans the pollution area, and acquires 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 posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. Compared with the existing technology, this solution does not require manual inspection. Through analysis and processing of the first photovoltaic image, the pollution results of the photovoltaic platform can be accurately determined. After determining the type of pollutant and the pollution area, the cleaning system is automatically controlled to adjust its posture according to the type of pollutant and the pollution area, so as to achieve efficient and rapid cleaning of the offshore photovoltaic platform. After the initial cleaning treatment, the second photovoltaic image of the photovoltaic platform is obtained again, so as to evaluate the cleaning effect in real time. According to the evaluation results, the cleaning system is controlled to adjust its posture again to clean the offshore photovoltaic platform until the cleaning standard is met, which reduces the cleaning cost, reduces the risks caused by human operation, ensures the cleaning quality, improves the cleaning efficiency of the offshore photovoltaic platform, and further improves the availability and power generation efficiency of the photovoltaic panels. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 A system architecture diagram of an application system of the offshore photovoltaic cleaning method provided in an embodiment of the present application; Figure 2 A schematic diagram of the process of the offshore photovoltaic cleaning method provided in the embodiment of the present application; Figure 3 A schematic diagram of the process of the offshore photovoltaic cleaning method provided in the embodiment of the present application; Figure 4 A schematic flow chart of a method for extracting features from a preprocessed photovoltaic image and determining a contamination result of a photovoltaic platform provided in an embodiment of the present application; Figure 5 A schematic diagram of the process of the offshore photovoltaic cleaning method provided in the embodiment of the present application; Figure 6 A schematic diagram of a flow chart of a method for traversing all pixels in an image and determining whether there is contamination provided in an embodiment of the present application; Figure 7 A schematic diagram of a process for obtaining an optimal solution through iterative updating provided in an embodiment of the present application; Figure 8 A schematic diagram of the structure of an offshore photovoltaic cleaning device provided in an embodiment of the present application; Fig. 9 A schematic diagram of the structure of a computer device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0018] The present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the relevant invention, rather than to limit the invention. It is also necessary to explain that, for ease of description, only the parts related to the invention are shown in the accompanying drawings.

[0019] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0020] As mentioned in the background technology, one method in the related technology is to use natural cleaning method to achieve cleaning, by utilizing the wind speed and waves of the ocean to clean the offshore photovoltaic equipment, however, this method is easily affected by the weather and the cleaning effect is poor; another method is to use artificial cleaning method, but this method requires a lot of human resources for large-scale offshore photovoltaic power stations, resulting in high labor costs and poor cleaning efficiency.

[0021] Based on the above-mentioned defects, the present application provides an offshore photovoltaic cleaning method, device, equipment and medium. Compared with the prior art, this solution does not require 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 pollutant and the pollution area, the cleaning system is automatically controlled to adjust the posture according to the type of pollutant and the pollution area, so as to achieve efficient and rapid cleaning of the offshore photovoltaic platform. After the initial cleaning treatment, the second photovoltaic image of the photovoltaic platform is obtained again to evaluate the cleaning effect in real time, so as to control the cleaning system to adjust the posture again according to the evaluation results to clean the offshore photovoltaic platform until the cleaning standard is met, thereby 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.

[0022] Figure 1 This is an implementation environment architecture diagram of an offshore photovoltaic cleaning method provided by an embodiment of the present application. Figure 1 As shown, the implementation environment architecture includes: a computer device 10 and a cleaning system 20, and the computer device may include a terminal 100 and a server 200. The computer device 10 may establish a communication connection with the cleaning system 20.

[0023] The terminal 100 may be a terminal device in various AI application scenarios. For example, the terminal 100 may be a smart home device such as a smart TV or a smart TV set-top box, or the terminal 100 may be a mobile portable terminal such as a smart phone, a tablet computer, and an e-book reader, or the terminal 100 may be a smart wearable device such as smart glasses or a smart watch, which is not specifically limited in this embodiment.

[0024] The server 200 may be a single server, or a server cluster consisting of a plurality of servers, or the server 200 may include one or more virtualization platforms, or the server 200 may be a cloud computing service center.

[0025] The server 200 may be a server device that provides background services for the AI ​​application installed in the terminal 100 .

[0026] The terminal 100 and the server 200 establish a communication connection via a wired or wireless network. Optionally, the wireless network or wired network uses standard communication technology and / or protocols. The network is usually the Internet, but can also be any network, including but not limited to 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 any combination of a virtual private network.

[0027] Optionally, the terminal 100 is used to obtain 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 contamination 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 the cleaning standard is met. If it does not meet the cleaning standard, it continues to adjust the posture to clean the photovoltaic platform until the cleaning standard is met.

[0028] The cleaning system 20 may include an automatic cleaning robot, a drone cleaning device or other mechanical devices capable of cleaning. The cleaning system 20 is used to receive and respond to control instructions sent by the computer device 10, adjust the posture, and clean the photovoltaic platform.

[0029] For ease of understanding and explanation, the following Figures 2 to 9 The offshore photovoltaic cleaning method, device, equipment and medium provided in the embodiments of the present application are described in detail.

[0030] Figure 2 The flowchart of the offshore photovoltaic cleaning method according to an embodiment of the present application is shown. The method can be executed by a computer device, which can be the above-mentioned Figure 1 The server 200 or the terminal 100 in the system shown, or the computer device can also be a combination of the terminal 100 and the server 200. Figure 2 As shown, the method includes: S101 , acquiring a first photovoltaic image of a photovoltaic platform.

[0031] It should be noted that the photovoltaic platform is located at sea, and the photovoltaic platform may include a floating photovoltaic platform, a fixed photovoltaic platform, etc. The floating photovoltaic platform includes: a floating body, a photovoltaic panel, a bracket and other components; the fixed photovoltaic platform includes: a steel pipe pile fixed platform, a pipe frame fixed photovoltaic platform, etc., and its components include photovoltaic panels, brackets, etc. The first photovoltaic image of the photovoltaic platform may include a photovoltaic panel image, and the photovoltaic panel image may be arranged in a square array or in a queue.

[0032] Optionally, the first photovoltaic image of the photovoltaic platform may be acquired by importing through an external device, or acquired through real-time detection by a camera, or acquired through a database or blockchain. This embodiment does not impose any limitation on the method for acquiring the first photovoltaic image of the photovoltaic platform.

[0033] Exemplarily, a high-definition surveillance camera is installed on the array area of ​​the offshore photovoltaic platform, and the high-definition surveillance camera is used to collect real-time images of the photovoltaic platform. Specifically, the computer device may send an image acquisition instruction to the surveillance camera, and after receiving the image acquisition instruction, the surveillance camera shoots and processes the offshore photovoltaic platform, thereby collecting a first photovoltaic image and sending it to the computer device, so that the computer device obtains the first photovoltaic image of the photovoltaic platform. Among them, the above-mentioned high-definition surveillance camera may have functions such as waterproof, dustproof, corrosion-resistant, and salt spray-resistant, which can make the collected first photovoltaic image clearer and facilitate subsequent precise image processing.

[0034] S102: Analyze and process the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to indicate whether the photovoltaic platform is polluted.

[0035] S103: When the pollution result indicates that the photovoltaic platform is polluted, the pollutant type and the pollution area corresponding to the pollutant type are determined.

[0036] Specifically, after acquiring the first photovoltaic image of the photovoltaic platform, an image recognition algorithm and an image analysis algorithm may be used to perform image analysis processing on the first photovoltaic image to determine whether the photovoltaic platform is polluted. The photovoltaic platform may be polluted by pollutants such as salt spray, waves, and suspended matter. The image recognition algorithm may be an image recognition algorithm based on deep learning, which is trained and learned from a large amount of sample data to enable it to have the ability to identify common types of pollutants on offshore photovoltaic platforms.

[0037] It is understandable that the pollution results of the above-mentioned photovoltaic platform may include that the photovoltaic panels in the photovoltaic platform are polluted or that the photovoltaic panels are not polluted. When pollution occurs, the pollutant type and the polluted area corresponding to the pollutant type are determined. Since the environment of the area where the photovoltaic platform is located will affect the generated image, the light intensity varies greatly at different times and weather. Strong light will make the photovoltaic image too bright and some details will be lost; weak light will make the image dim, the signal-to-noise ratio will be reduced, and the details will be difficult to distinguish, increasing the difficulty of photovoltaic component status recognition and fault detection in the image. As the position of the sun moves and the angle of illumination changes, there will be differences in light and dark on the surface of the photovoltaic panel, resulting in uneven grayscale of the image, interference in image segmentation and feature extraction, and affecting the accurate assessment of the overall state of the photovoltaic panel. Therefore, in the process of analyzing and processing the first photovoltaic image and determining the pollution results of the photovoltaic platform, it is necessary to consider the environmental information of the area where the photovoltaic platform is located.

[0038] The process of analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform includes: obtaining 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; and extracting features from the preprocessed photovoltaic image to determine the pollution result of the photovoltaic platform.

[0039] Among them, since the surface of offshore photovoltaic panels is often swept by the forces of waves, sea breezes, etc., and the complete structural characteristics of the panels themselves often cause attachments on their surface, the image needs to be processed before analyzing the first photovoltaic image. Among them, the above-mentioned 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 pre-processed according to the environmental information, and the pre-processed image can be obtained through denoising, grayscale correction, chromaticity correction, brightness correction and other processing, and then the pre-processed photovoltaic image is subjected to feature extraction, such as edge detection, texture analysis, shape matching and other processing to determine the pollution result of the photovoltaic platform.

[0040] After determining the first photovoltaic image, the first photovoltaic image is denoised to obtain a denoised image, and the color value of each pixel in the denoised image in the RGB channel is obtained; the color value and the preset weight coefficient are weighted and summed to calculate the grayscale mean of each pixel, and a grayscale image is generated according to the grayscale mean; the grayscale image is chromatically corrected and brightness corrected according to the environmental information to obtain a preprocessed photovoltaic image.

[0041] In this embodiment, after the first photovoltaic image is denoised to obtain the denoised image, the color value of each pixel in the denoised image in the RGB (red, green and blue) channel is determined. The color of each of the three channels is usually between 0 and 255, and the weight coefficient of each color channel is obtained. The weight coefficient is customized according to actual needs. For example, it can be determined according to the sensitivity of the human eye to different colors. Then, the color value and the preset weight coefficient are weighted and summed to calculate the grayscale mean of each pixel, and a grayscale image is generated according to the grayscale mean. For example, the weight coefficient of the R channel is determined to be 0.299, the weight coefficient of the B channel is 0.114, and the weight coefficient of the G channel is 0.587. The weighted sum is used to obtain the grayscale mean, which is expressed by the following formula: Gray=0.299×R+0.587×G+0.114×B; Among them, R is the color value of the pixel in the R channel, G is the color value of the pixel in the G channel, B is the color value of the pixel in the B channel, and Gray is the grayscale mean.

[0042] In this embodiment, by obtaining the color value of each pixel in the RGB channel and using weighted summation to calculate the grayscale mean of each pixel, and then generating a grayscale image, the image data processing time can be greatly reduced and the image processing efficiency can be improved.

[0043] Furthermore, when the pollution result indicates that the photovoltaic platform is polluted, a neural network model may be used to identify the first photovoltaic image to determine the type of pollutant and the polluted area corresponding to the pollutant type. The pollutant type may include bird droppings, foam, driftwood, etc., and the polluted area may be a partial area in the first photovoltaic image. The neural network model may be, for example, a convolutional neural network model, which is trained based on sample image data and image annotation results, and is used to identify the type of pollutant in the image and the corresponding polluted area, and has the ability to accurately identify common pollutant types on offshore photovoltaic platforms.

[0044] S104: Send the pollutant type and the polluted area to the cleaning system, so that the cleaning system adjusts its position according to the pollutant type and the polluted area to clean the polluted area.

[0045] After determining the pollutant type and the contaminated area, the pollutant type and the contaminated area can be packaged into control instructions and sent to the cleaning system. After receiving the control instructions, the cleaning system responds to the control instructions, adjusts its posture, and cleans the contaminated area. For example, the contaminated area on the photovoltaic platform can be cleaned by controlling the posture of the robot or adjusting the drone.

[0046] S105 , acquiring a second photovoltaic image of the photovoltaic platform.

[0047] S106: When the second photovoltaic image does not meet the cleaning standard, an adjustment instruction is sent to the cleaning system, so that the cleaning system adjusts the posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0048] In this embodiment, the second photovoltaic image is an image of the photovoltaic platform collected after the cleaning system cleans the photovoltaic platform. The above cleaning standard is customized according to actual needs, for example, the proportion of the contaminated area to the total area can be greater than a preset proportion, and the preset proportion can be adaptively adjusted.

[0049] See also Figure 3 As shown, the offshore photovoltaic platform cleaning method provided in the embodiment of the present application is applied to computer equipment and a cleaning system, comprising the following steps: S1001. A computer device acquires a first photovoltaic image of a photovoltaic platform.

[0050] S1002. The computer device uses an intelligent algorithm to process and analyze the first photovoltaic image to determine whether pollution occurs.

[0051] S1003: When no pollution occurs, the computer device waits for the next scheduled task to be executed.

[0052] S1004. When pollution occurs, the computer device determines the type of pollutant and the polluted area.

[0053] S1005. The computer device sends the pollutant type and polluted area to the cleaning system.

[0054] S1006. The cleaning system adjusts its position according to the type of pollutants and the polluted area to clean the photovoltaic platform.

[0055] S1007. The computer device acquires a second photovoltaic image of the photovoltaic platform.

[0056] S1008. The computer device determines whether the second photovoltaic image meets the cleaning standard.

[0057] S1009: When the cleaning standard is reached, the process ends.

[0058] S1010: When the cleaning standard is not met, the computer device sends an adjustment instruction to the cleaning system.

[0059] S1011. The cleaning system receives and responds to the adjustment instruction, and readjusts the cleaning method of the cleaning system to clean the photovoltaic platform until the cleaning standard is met.

[0060] Specifically, a monitoring camera is set on the offshore photovoltaic platform, which can be a high-definition camera. The camera collects the first photovoltaic image of the photovoltaic platform and sends the first photovoltaic image to the computer device, and then the computer device obtains the first photovoltaic image of the photovoltaic platform, uses an intelligent algorithm to process and analyze the first photovoltaic image, and judges whether pollution occurs. When pollution does not occur, wait for the next scheduled task to be executed; when pollution occurs, determine the type of pollutant and the polluted area, and determine the location information of the polluted area through the positioning system, and 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 pollutant type, and cleans the polluted area on the surface of the photovoltaic panel. Then, the camera takes a photo of the photovoltaic platform again, so that the computer device obtains the second photovoltaic image, and evaluates the second photovoltaic image to judge whether it meets the cleaning standard; when the cleaning standard is met, the process execution ends; when the cleaning standard is not met, readjust the cleaning method of the cleaning system to re-clean the photovoltaic platform until the cleaning standard is met. The cleaning method includes adjusting the position or posture of the cleaning system.

[0061] Among them, after receiving the control command, the above-mentioned cleaning system can automatically adjust the position and direction of the cleaning equipment according to the location information of the contaminated area, so as to realize the cleaning of the contaminated area in the offshore photovoltaic platform.

[0062] An offshore photovoltaic cleaning method provided in an embodiment of the present application includes: acquiring a first photovoltaic image of a photovoltaic platform, analyzing and processing the first photovoltaic image, determining a contamination result of the photovoltaic platform, and when the contamination result indicates that the photovoltaic platform is contaminated, determining the pollutant type and the contaminated area corresponding to the pollutant type, and then sending the pollutant type and the contaminated area to a cleaning system so that the cleaning system adjusts its posture according to the pollutant type and the contaminated area, cleans the contaminated area, and acquires 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 posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. Compared with the existing technology, this solution does not require manual inspections. Through analysis and processing of the first photovoltaic image, the pollution results of the photovoltaic platform can be accurately determined. After determining the type of pollutant and the pollution area, the cleaning system is controlled to automatically adjust its posture according to the type of pollutant and the pollution area, so as to achieve 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 results, the cleaning system is controlled to adjust its posture again to clean the offshore photovoltaic platform until the cleaning standard is met, which reduces the cleaning cost, reduces the risks caused by human operation, ensures the cleaning quality, improves the cleaning efficiency of the offshore photovoltaic platform, and further improves the availability and power generation efficiency of the photovoltaic panels.

[0063] In one of the embodiments, the present application also provides a specific implementation method for extracting features from the pre-processed photovoltaic image to determine the contamination result of the photovoltaic platform, see Figure 4 As shown, the method includes: S201, filtering the preprocessed photovoltaic image to obtain a filtered image, and constructing a high acrylic pyramid and DoG scale space according to the filtered image; the high acrylic pyramid includes multiple layers, each layer is provided with an image, and the size of each layer of the image decreases from bottom to top, the image of the bottom layer in the multiple layers is the preprocessed photovoltaic image, and the images of the remaining layers except the bottom layer in the multiple layers are filtered images.

[0064] S202: Locate the extreme points in the filtered image to obtain initial extreme points.

[0065] S203, comparing the relationship between the local features and the overall features of the initial extreme value points at different scales in the DoG scale space, removing the interfering extreme value points in the initial extreme value points, and obtaining a segmented image.

[0066] S204: Detect and process the segmented image to determine the pollution result of the photovoltaic platform.

[0067] It should be noted that the above-mentioned high acrylic pyramid is a multi-scale representation method of an image. It is constructed by multiple downsampling (reducing the image size) of the preprocessed photovoltaic image and combining it with a Gaussian filter operation. Figuratively speaking, it is like building a pyramid, starting from the bottom preprocessed photovoltaic image, each layer is processed on the basis of the previous layer, thereby constructing a high acrylic pyramid.

[0068] See also Figure 5 As shown, a high-definition monitoring camera and a cleaning system are set on the square array area of ​​the offshore photovoltaic platform, and the first photovoltaic image of the photovoltaic platform is collected by the camera and sent to the computer device, and then the computer device obtains the first photovoltaic image of the photovoltaic platform, and uses an intelligent algorithm to process and analyze the first photovoltaic image to determine whether pollution occurs, for example, using an image recognition algorithm based on deep learning. When the first photovoltaic image is not polluted, there is no need for cleaning, and the photovoltaic platform continues to be monitored; 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 type of pollutant. Then the photovoltaic platform is photographed again by the camera, 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 readjusts the intelligent algorithm parameters, and re-cleans the photovoltaic platform according to the adjusted parameters until the cleaning standard is met. The intelligent algorithm parameters may include adjusting the position or posture of the cleaning system.

[0069] In this embodiment, after obtaining the pre-processed photovoltaic image, global image feature extraction is performed on it, and the accelerated robust feature extraction algorithm (Speeded Up Robust Features, SURF) can be used for feature extraction processing to determine the pollution results of the photovoltaic platform. The SURF algorithm is based on the scale-invariant feature change (SIFT) algorithm, and uses the integral image and Hessian matrix to realize 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 realize image feature extraction and matching. The SURF algorithm is divided into two parts: low-level and high-level analysis. The low-level part includes two steps: the first step is to perform Gaussian filtering on the pre-processed photovoltaic image to obtain a filtered image. The Gaussian filter is an image smoothing method used to reduce image noise and details; the second step is to construct a high acrylic pyramid based on the filtered image, and construct a DoG (Difference of Gaussian) scale space based on the high acrylic pyramid, locate the extreme points in the filtered image, and obtain the initial extreme points. The high-level analysis part is to further refine and distinguish the initial extreme points. By comparing the relationship between local features and overall features at different scales in the DoG scale space, the points that are misjudged after binarization are eliminated, thereby obtaining the segmented image. The SURF algorithm also includes: a fast judgment algorithm and a dynamic value function module. The fast judgment algorithm is used to quickly eliminate irrelevant extreme points. The dynamic value function module is a heuristic criterion that excludes extreme points that do not bring useful information based on the local features of the extreme points and their corresponding scales. After the above processing, the image has removed the influence of interference and retained the main features.

[0070] Specifically, after obtaining the preprocessed photovoltaic image, Gaussian filtering needs to be performed on the image, which can be expressed by the following formula: ; Among them, σ is the standard deviation, which is 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.

[0071] After 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: ; Among them, Lxx, Lxy, Lyx and Lyy are the second-order partial derivatives of the image at the feature point (x, y), and σ is the scale parameter.

[0072] It is understandable that in the actual calculation process, the SURF algorithm can use the integral graph to quickly calculate the determinant of the Hessian matrix. The approximate formula of the determinant is: ; Among them, Lxx, Lxy, Lyx and Lyy are the second-order partial derivatives of the image at the feature point (x, y), respectively. The 0.9 here is an empirical coefficient used to balance the error caused by the use of box filter approximation.

[0073] 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 the image of this layer is Gaussian filtered. The function of Gaussian filtering is to smooth the image and reduce the noise in the image. After that, the image of the previous layer (layer 1) is obtained by downsampling operation (usually removing pixels of even rows and even columns). This process will be repeated continuously, and the size of each layer of the image will gradually decrease, forming a pyramid-shaped structure. For example, suppose the original image is a 512×512 pixel image, which is the 0th layer of the high acrylic pyramid. After the first Gaussian filtering and downsampling, a 256×256 pixel image is obtained as the 1st layer of the high acrylic pyramid. After the same operation, a 128×128 pixel second layer image can be obtained. These image layers of different sizes can be used to analyze images at different scales.

[0074] 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 value points are calculated at different scales in each layer. The initial high acrylic pyramid scale can be d=4.5833... Repeat this process until the bottom layer is decomposed to obtain 7 groups of initial extreme value points on the scale space.

[0075] 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 two adjacent layers in the high acrylic pyramid. In this way, a series of images representing the difference in the changes of images at different scales can be obtained, and these difference images constitute the DoG scale space.

[0076] In this embodiment, the DoG scale space is constructed based on the high acrylic pyramid, which can detect feature points in the image, especially those positions that appear as extreme points in the scale space. At different scales, features in the image (such as edges, corners, etc.) will have different performances. The DoG scale space can highlight the changes of these features at different scales, making feature points easier to detect.

[0077] For example, when detecting extreme points in an image, the response of the extreme points in the DoG scale space is usually more obvious than in the original image or a simple high acrylic pyramid image. This is because DoG can capture the rapid changes in the pixel values ​​of corner points between adjacent scales, and the location of such rapid changes is likely to be the key feature point in the image. By finding the extreme points (maximum and minimum values) in the DoG scale space, possible feature points can be preliminarily located, providing important clues for subsequent feature extraction and analysis.

[0078] After constructing the DoG scale space and locating the extreme points in the filtered image to obtain the initial extreme points, the relationship between the local features and the overall features of the initial extreme points at different scales in the DoG scale space is compared. The interfering extreme points in the initial extreme points are quickly removed through a fast judgment algorithm, and the dynamic value function is used to exclude the interfering extreme points according to the local features of the initial extreme points and their corresponding scales, thereby retaining the main features and obtaining the segmented image.

[0079] Furthermore, in the process of detecting and processing the segmented image to determine the pollution result of the photovoltaic platform, the area to be detected can be obtained from the segmented image, and a shape matching library of the area to be detected can be obtained; image registration processing can be performed on each shape matching in the shape matching library; each pixel point in the area to be detected can be traversed 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; according to the detection result, the areas of all the areas where the polluted pixels are located are accumulated to obtain the total area of ​​the polluted areas; when the total area of ​​the polluted areas is greater than a preset area threshold, it is determined that the area to be detected is polluted; the preset area threshold is the total window area with a preset coefficient.

[0080] After obtaining the segmented image, you can select an area from the segmented image as the area to be detected. Considering that bird droppings are the most common contaminants on photovoltaic panels, for example, stains arranged in a "one" shape or "||" shape on photovoltaic panels are rare, so 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 matching in the shape matching library to obtain the registration result. The registration result is used to characterize the matching result between the area to be detected and the corresponding shape in the shape matching library, which may include a successful match or a failed match. Then, according to the registration result, search one by one to see if there is contamination in the image. When the registration result is that the area to be detected successfully matches the corresponding shape in the shape matching library, traverse each pixel in the area to be detected to determine the detection result of the area where the pixel is located, including: for each pixel in the area to be detected, obtain the color of the pixel, the grayscale difference between the pixel and its neighboring pixels, and determine whether the pixel is connected by black pixels; when the color of the pixel is black, the grayscale difference between the pixel and its neighboring pixels is greater than the preset grayscale difference and there are no black pixels connected, it is determined that the area where the pixel is located is contaminated.

[0081] See also Figure 6 and Figure 7 As shown, first load the segmented image, initialize the total number of polluted area areas Polluted AreaSum=0, traverse each pixel in the image, for each current pixel, first determine whether the current pixel is black, when it is black, calculate the grayscale difference between the current pixel and the adjacent pixel, and determine whether the grayscale difference is greater than the preset grayscale difference A, when it is greater than the preset grayscale difference A, check whether the current pixel is connected to a black pixel, when there is no black pixel connected, it indicates that the area where the current pixel is located is polluted, mark the area where the current pixel is located as a polluted area, and set Polluted AreaSum=1, then move to the next pixel to perform the above pollution judgment process until all pixels in the image are checked; if the current pixel is not black, or the current grayscale difference is not greater than the preset grayscale difference A, or the current pixel is connected to a black pixel, it indicates that the area where the current pixel is located is not polluted, skip the current pixel, and move to the next pixel to perform the pollution judgment process until all pixels in the area to be detected are checked.

[0082] After all pixels in the area to be detected are checked, the areas of all the pixels in the area to be detected that are polluted are accumulated to obtain the total area of ​​the polluted area P_Polluted AreaSum. Then, it is determined whether the total area of ​​the polluted area is greater than the preset area threshold. When the total area of ​​the polluted area P_Polluted AreaSum is greater than the preset area threshold, for example, taking the preset coefficient as 80%, that is, 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, the 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.

[0083] After obtaining the pollution results, 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 value among all variances; when the variance is the minimum value, the posture information of the cleaning system is generated according to the pollution result with the minimum variance; the posture information includes angle information and position information; The rotation angle, pollutant type and polluted area are packaged into a control instruction and sent to the cleaning system. In the process of comparing the pollution results horizontally and vertically, the images corresponding to the area to be detected where pollution exists can be compared horizontally and vertically. When the pollution result indicates that there is pollution in the area to be detected, an image detection model can be used to perform image detection processing to obtain the pollutant type. The image detection model can be, for example, a convolutional neural network model.

[0084] Exemplarily, after obtaining the i-th pollution result, i=1, 2, ..., nm all have corresponding result sets, and then a horizontal comparison is performed to calculate the mean and horizontal variance of all pollution results in the result set. The horizontal variance reflects the degree of discreteness of the internal data of the i-th pollution result; and the vertical variance is calculated. The vertical variance reflects the degree of discreteness of the data at the same position in different calculations, and then the horizontal and vertical variances are comprehensively considered to obtain a comprehensive variance index. The index may include weight coefficients corresponding to the horizontal variance and the vertical variance, which are used to balance the importance of the horizontal variance and the vertical variance. When the comprehensive variance index is the minimum value, the posture of the cleaning system is generated according to the pollution result with the minimum comprehensive variance index. The posture includes angle information a and position information, which is the optimal solution, and then the posture, pollutant type and pollution area are packaged into a control instruction, and the control instruction is sent to the cleaning system. The cleaning system receives and responds to the control instruction, adjusts the posture, and realizes the cleaning process of the photovoltaic platform.

[0085] Optionally, after cleaning the contaminated area, the computer equipment collects image information of the photovoltaic platform again to obtain a second photovoltaic image, and then evaluates the cleaning effect of the photovoltaic platform to ensure the cleaning quality. According to the evaluation results, the cleaning system can automatically adjust the intelligent algorithm parameters to update and optimize the model. Through continuous learning, the algorithm can more accurately judge the degree of contamination, optimize cleaning decisions, and improve cleaning efficiency.

[0086] In this embodiment, the photovoltaic platform is monitored in real time, and it is automatically determined whether the cleaning standard is met based on the image information. If the cleaning standard is not met, it is automatically cleaned, thereby reducing the equipment maintenance cost and failure rate. Compared with the traditional manual inspection and cleaning method, it can achieve 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.

[0087] on the other hand, Figure 8 The schematic diagram of the structure of an offshore photovoltaic cleaning device provided in an embodiment of the present application is shown in FIG. The device may be a device in a terminal or a server, such as Figure 8 As shown, the device comprises: A first acquisition module 710 is used to acquire a first photovoltaic image of the photovoltaic platform; The analysis and processing module 720 is used to analyze and process the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to indicate whether the photovoltaic platform is polluted; The area determination module 730 is used to determine the type of pollutant and the polluted area corresponding to the type of pollutant when the pollution result indicates that the photovoltaic platform is polluted; A first sending module 740 is used to send the pollutant type and the polluted area to the cleaning system, so that the cleaning system adjusts the posture according to the pollutant type and the polluted area to clean the polluted area; A second acquisition module 750 is used to acquire a second photovoltaic image of the photovoltaic platform; The second sending module 760 is used 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 posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0088] Optionally, the analysis and processing module 720 is specifically used for: Obtain 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; Feature extraction is performed on the preprocessed photovoltaic image to determine the contamination results of the photovoltaic platform.

[0089] Optionally, the analysis and processing module 720 is further configured to: Performing denoising on the first photovoltaic image to obtain a denoised image; Get the color value of each pixel in the RGB channel of the denoised image; The color value and the preset weight coefficient are weighted and summed to calculate the grayscale mean of each pixel, and a grayscale image is generated based on the grayscale mean; The grayscale image is subjected to chromaticity correction and brightness correction according to the environmental information to obtain a preprocessed photovoltaic image.

[0090] Optionally, the analysis and processing module 720 is further configured to: The preprocessed photovoltaic image is filtered to obtain a filtered image, and a high acrylic pyramid and a DoG scale space are constructed according to the filtered image; the high acrylic pyramid includes multiple layers, each layer is provided with an image, and the size of each layer of the image decreases from bottom to top, the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images of the remaining layers except the bottom layer of the multiple layers are filtered images; Locate 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, and obtain the segmented image; The segmented images are detected and processed to determine the pollution results of the photovoltaic platform.

[0091] Optionally, the analysis and processing module 720 is further configured to: Obtaining the area to be detected from the segmented image, and obtaining a shape matching library of the area to be detected; Performing image registration processing on each shape match in the shape matching library; According to the registration result, each pixel point in the area to be detected is traversed 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; According to the detection results, the areas of all the polluted pixels are accumulated to obtain the total area of ​​the polluted area; When the total area of ​​the polluted regions is greater than a 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.

[0092] Optionally, the analysis and processing module 720 is further configured to: For each pixel in the area to be detected, obtain the color of the pixel, the grayscale difference between the pixel and its adjacent pixels, and determine whether the pixel is connected to a black pixel; When the color of a pixel is black, the grayscale difference between the pixel and adjacent pixels is greater than a preset grayscale difference, and there are no connected black pixels, it is determined that the area where the pixel is located is polluted.

[0093] Optionally, the first sending module 740 is specifically configured to: Compare the pollution results horizontally and vertically, and calculate the variance of the pollution results; Determine whether the variance is the minimum value among all current variances; When the variance is the minimum value, the posture of the cleaning system is generated according to the pollution result with the minimum variance; the posture includes angle information and position information; The position, contaminant type and contamination area are packaged into control instructions and sent to the cleaning system.

[0094] It can be understood that the functions of each functional module of the offshore photovoltaic cleaning system of this embodiment can be specifically implemented according to the method in the above method embodiment, and its specific implementation process can refer to the relevant description of the above method embodiment, which will not be repeated here.

[0095] On the other hand, a computer device provided in an embodiment of the present application includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the program, the offshore photovoltaic cleaning method as described above is implemented.

[0096] Reference below Fig. 9 , Fig. 9 A schematic diagram of the structure of a computer system of a terminal device according to an embodiment of the present application.

[0097] like Fig. 9As shown, the computer system 300 includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage part 303 to a 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, the ROM 302, and the RAM 303 are connected to each other through a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0098] 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 a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; 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 a computer program read therefrom is installed into the storage section 308 as needed.

[0099] 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 includes a program code 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-mentioned functions defined in the system of the present application are executed.

[0100] It should be noted that the computer-readable medium shown in the present application may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may 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, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present application, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0101] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each box in the flow chart or block diagram can represent a module, a program segment or a part of the code, and the aforementioned module, program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flow chart, and the combination of the boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0102] The units or modules involved in the embodiments described in the present application may be implemented by software or by hardware. The described units or modules may also be set in a processor, for example, may be described as: a processor, comprising: a first acquisition module, an analysis and processing module, an area determination module, a first sending module, a second acquisition module, and a second sending module. Among them, the names of these units or modules do not constitute a limitation on the units or modules themselves in some cases. For example, the first acquisition module may also be described as "for acquiring a first photovoltaic image of a photovoltaic platform".

[0103] As another aspect, the present application further provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiment; or may exist independently without being assembled into the electronic device. The above computer-readable storage medium stores one or more programs, and when the above programs are used by one or more processors to execute the offshore photovoltaic cleaning method described in the present application: acquiring a first photovoltaic image of the photovoltaic platform; Analyzing and processing the first photovoltaic image to determine a contamination result of the photovoltaic platform; the contamination result is used to indicate whether the photovoltaic platform is contaminated; When the pollution result indicates that the photovoltaic platform is polluted, determining the type of pollutant and the pollution area corresponding to the type of pollutant; Sending the pollutant type and the polluted area to a cleaning system, so that the cleaning system adjusts its position according to the pollutant type and the polluted area, and cleans the polluted area; acquiring a second photovoltaic image of the photovoltaic platform; When the second photovoltaic image does not meet the cleaning standard, an adjustment instruction is sent to the cleaning system, so that the cleaning system adjusts the posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

[0104] In summary, the offshore photovoltaic cleaning method, device, equipment and medium provided in the embodiments of the present application include: acquiring a first photovoltaic image of a photovoltaic platform, analyzing and processing the first photovoltaic image, determining the pollution result of the photovoltaic platform, and 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 its posture according to the pollutant type and the pollution area, cleans the pollution area, and acquires 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 posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard. Compared with the existing technology, this solution does not require manual inspection. Through analysis and processing of the first photovoltaic image, the pollution results of the photovoltaic platform can be accurately determined. After determining the type of pollutant and the pollution area, the cleaning system is automatically controlled to adjust its posture according to the type of pollutant and the pollution area, so as to achieve efficient and rapid cleaning of the offshore photovoltaic platform. After the initial cleaning treatment, the second photovoltaic image of the photovoltaic platform is obtained again to evaluate the cleaning effect in real time. According to the evaluation results, the cleaning system is controlled to adjust its posture again to clean the offshore photovoltaic platform until the cleaning standard is met, which reduces the cleaning cost, reduces the risks caused by human operation, ensures the cleaning quality, improves the cleaning efficiency of the offshore photovoltaic platform, and further improves the availability and power generation efficiency of the photovoltaic panels.

[0105] The above description is only a preferred embodiment of the present application and an explanation of the technical principles used. 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 a 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 above features are replaced with the technical features with similar functions disclosed in this application (but not limited to) by each other.

Claims

1. A method for cleaning offshore photovoltaic power generation, characterized in that: The offshore photovoltaic cleaning method comprises: acquiring a first photovoltaic image of the photovoltaic platform; Analyzing and processing the first photovoltaic image to determine a contamination result of the photovoltaic platform; the contamination result is used to indicate whether the photovoltaic platform is contaminated; When the pollution result indicates that the photovoltaic platform is polluted, determining the type of pollutant and the pollution area corresponding to the type of pollutant; Sending the pollutant type and the polluted area to a cleaning system, so that the cleaning system adjusts its position according to the pollutant type and the polluted area, and cleans the polluted area; acquiring a second photovoltaic image of the photovoltaic platform; When the second photovoltaic image does not meet the cleaning standard, an adjustment instruction is sent to the cleaning system, so that the cleaning system adjusts the posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

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: Acquiring 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; Feature extraction is performed on the preprocessed photovoltaic image to determine the contamination 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 on the first photovoltaic image to obtain a denoised image; Obtain the color value of each pixel in the RGB channel of the denoised image; Performing weighted sum processing on the color value and a preset weight coefficient, calculating the grayscale mean of each pixel point, and generating a grayscale image according to the grayscale mean; The grayscale image is subjected to chromaticity correction and brightness correction processing according to the environmental information to obtain a preprocessed photovoltaic image.

4. The offshore photovoltaic cleaning method according to claim 2, characterized in that: Extracting features from the preprocessed photovoltaic image to determine the contamination result of the photovoltaic platform includes: The preprocessed photovoltaic image is filtered to obtain a filtered image, and a high acrylic pyramid and a DoG scale space are constructed according to the filtered image; the high acrylic pyramid includes multiple layers, each layer is provided with an image, and the size of each layer of the image decreases from bottom to top, the image of the bottom layer of the multiple layers is the preprocessed photovoltaic image, and the images of the remaining layers of the multiple layers except the bottom layer are the filtered images; Positioning the extreme points in the filtered image to obtain initial extreme points; Comparing the relationship between the local features and the overall features of the initial extreme value points at different scales in the DoG scale space, removing the interfering extreme value points in the initial extreme value points, and obtaining a segmented image; The segmented image is subjected to detection processing 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: Acquire a region to be detected from the segmented image, and acquire a shape matching library of the region to be detected; Performing image registration processing on each shape match in the shape matching library; Traversing each pixel point in the area to be detected according to the registration result, determining 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 results, the areas of all the regions where the polluted pixels are located are accumulated to obtain the total area of ​​the polluted regions; When the total area of ​​the polluted regions is greater than a preset area threshold, it is determined that the area to be detected is polluted; the preset area threshold is the total window area of ​​a preset coefficient.

6. The offshore photovoltaic cleaning method according to claim 5, traversing each pixel point in the area to be detected and determining the detection result of the area where the pixel point is located, comprises: For each pixel in the area to be detected, obtaining the color of the pixel, the grayscale difference between the pixel and adjacent pixels, and determining whether the pixel is connected to a black pixel; When the color of the pixel is black, the grayscale difference between the pixel and adjacent pixels is greater than a preset grayscale difference, and there are no connected black pixels, it is determined that the area where the pixel is located is polluted.

7. The method according to claim 4, characterized in that Sending the contaminant type and the contaminated area to a cleaning system includes: Comparing the pollution results horizontally and vertically, and calculating the variance of the pollution results; Determine whether the variance is the minimum value among all current variances; When the variance is the minimum value, generating a posture of the cleaning system according to the pollution result with the minimum variance; the posture includes angle information and position information; The posture, the pollutant type and the polluted area are packaged into a control instruction and sent to the cleaning system.

8. An offshore photovoltaic cleaning device, characterized in that: The offshore photovoltaic cleaning device comprises: A first acquisition module, used to acquire a first photovoltaic image of the photovoltaic platform; An analysis and processing module, used for analyzing and processing the first photovoltaic image to determine the pollution result of the photovoltaic platform; the pollution result is used to indicate whether the photovoltaic platform is polluted; An area determination module, used for determining the type of pollutant and the polluted area corresponding to the type of pollutant when the pollution result indicates that the photovoltaic platform is polluted; A first sending module, used for sending the pollutant type and the polluted area to the cleaning system, so that the cleaning system adjusts the posture according to the pollutant type and the polluted area, and cleans the polluted area; A second acquisition module, used for acquiring a second photovoltaic image of the photovoltaic platform; The second sending module is used 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 the posture again in response to the adjustment instruction until the second photovoltaic image meets the cleaning standard.

9. A computer device comprising: A memory, a processor, and a computer program stored in 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 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the offshore photovoltaic cleaning method according to any one of claims 1 to 7 are implemented.

Images