A method and device for identifying foreign matter on a photovoltaic panel suitable for use in a microgrid system

Abstract

A kind of photovoltaic panel foreign matter identification method and cleaning device suitable for micro-grid system, including the coordinate positioning of photovoltaic panel;Industrial camera collects the photovoltaic panel picture of positioning area;Photograph is identified using YOLOv3 algorithm and is identified again positioning;The focal length of industrial camera is adjusted using new positioning information to enlarge photovoltaic panel to the central part of industrial camera for picture capture;After the complete photovoltaic panel information is identified, the outline of processed photovoltaic panel picture is extracted, the area of photovoltaic panel is extracted using region growing method, and whether photovoltaic panel has foreign matter is judged;The positioning and identification problem of photovoltaic panel are realized by using the image recognition analysis technology of deep learning and convolutional neural network in the application, and the world coordinate system conversion method is used as the basis to enlarge and capture the corresponding position of photovoltaic panel by industrial camera, which has the advantage of accurate foreign matter identification, secondly, the cleaning device provided by the application has the advantages of simple structure, energy saving and high cleaning efficiency.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of photovoltaic panel foreign matter identification, and particularly to a photovoltaic panel foreign matter identification method suitable for a micro-grid system. BACKGROUND

[0002] Photovoltaic arrays are used in outdoor environments, covered with different forms of dust particles in the atmospheric environment, especially in the Midwest, which is disturbed by dust weather every year, and the surface of the photovoltaic module is easy to deposit dust.

[0003] On the one hand, the dust falling on the surface of the photovoltaic module will block sunlight, reduce the transmittance of the glass surface of the photovoltaic module, and reduce the solar radiation on the surface of the battery; on the other hand, the heat transfer form on the surface of the product will change, the dust will absorb solar radiation and convert it into its own heat energy, blocking the external heat dissipation of the cover glass of the photovoltaic module. In short, dust is one of the main problems affecting the performance ratio of photovoltaic power stations, and the specific performance is as follows:

[0004] Influence of dust calcification deposition on light transmission:

[0005] Dust contains a large amount of calcium and magnesium oxides. When the dust on the surface of the photovoltaic module rains, a small amount of calcium and magnesium ions will dissolve in the rain and re-attach to the glass surface of the photovoltaic module. If not cleaned in time, with the passage of time, a thick and hard layer of calcium and magnesium scale will be formed on the surface of the photovoltaic module. Once the scale is formed, it is not easy to remove, which will seriously affect the power generation effect of the photovoltaic module, and even cause some photovoltaic panels to generate hot spots and other problems.

[0006] Influence of dust on solar radiation:

[0007] The dust on the surface of the glass will not only seriously affect the transmittance of solar radiation, but also cause a large amount of diffusion of solar radiation on the surface of the photovoltaic module. According to experimental measurements, the dust accumulated on the surface of the glass will cause a loss of solar radiation of 5%-30%, and the loss is mainly caused by the absorption, reflection and scattering of solar radiation by the dust attached to the surface of the glass.

[0008] Influence of dust on the temperature of the photovoltaic module:

[0009] The dust attached to the surface of the toughened glass can make part of the solar radiation absorbed by the dust to be converted into heat energy, thereby increasing the working temperature of the photovoltaic module. Meanwhile, the dust on the surface of the module also has a covering and heat preservation effect on the surface of the photovoltaic module, affecting the heat dissipation of the photovoltaic module, further increasing the thermal temperature effect of the photovoltaic module and affecting the photovoltaic power generation effect. The photovoltaic module is the power generation component of the photovoltaic power generation system and is the core unit of the entire photovoltaic system. The photoelectric conversion efficiency of the photovoltaic module is inversely proportional to the temperature, and the photovoltaic efficiency of the module will decrease with the increase of the temperature. Since the dust attached to the surface of the module absorbs part of the heat energy, the internal temperature of the battery panel is easily increased by heat, thereby reducing the photovoltaic efficiency. In the photovoltaic power generation system, the dust on the surface of the module not only affects the solar radiation on the surface of the photovoltaic cell, but also increases the temperature rise of the photovoltaic module. If the dust on the surface of the glass is not cleaned in time, the power generation performance of the photovoltaic module will be greatly affected. Under a certain dust thickness, the larger the sunlight incidence angle, the longer the sunlight path in the dust layer, and the easier the reflection, scattering and absorption. Therefore, special attention should be paid to the influence of the dust on the surface of the photovoltaic module on the photovoltaic power generation during the operation and maintenance of the photovoltaic module, and the surface of the photovoltaic module should be cleaned in time to enable the power station to operate efficiently. SUMMARY

[0010] The purpose of the present application is to overcome the shortcomings of the prior art and provide a photovoltaic panel foreign matter identification method suitable for a micro-grid system. The method can accurately identify dust and foreign matter on the photovoltaic panel, and the cleaning device can clean the photovoltaic foreign matter on the photovoltaic panel in time.

[0011] The purpose of the present application is achieved by the following technical solutions:

[0012] A photovoltaic panel foreign matter identification method suitable for a micro-grid system, the method comprising the following steps:

[0013] S1, an industrial camera calibrates the position information of the photovoltaic panel arrangement, and saves the industrial camera control coordinates corresponding to the photovoltaic panel arrangement position in the form of a structure to an edge computer;

[0014] S2, the edge computer controls the industrial camera to rotate to the coordinates saved in step S1 to capture pictures, and sends the captured picture data to a data analysis server; wherein the coordinates include a horizontal angle p, a vertical angle t, and a magnification z;

[0015] S3, the data analysis server identifies the photovoltaic panel in the picture by using a YOLOv3 algorithm; then identifies and locates the photovoltaic panel in the picture by using a deep learning tracking algorithm, and then adjusts the angle of the industrial camera by using the obtained positioning information to identify and locate the photovoltaic panel again;

[0016] S4, adjusting the focal length of the industrial camera according to the positioning information obtained again in step S3, the edge computer controlling the industrial camera to magnify the photovoltaic panel to the central part of the industrial camera to capture a picture again, then sending the captured picture to the data analysis server for filtering processing and then sending to the edge computer again;

[0017] S5, the edge computer uses the Canny algorithm to extract the edge of the picture, and then uses the Hough line detection and Harris corner point detection to identify the photovoltaic panel, judges whether there is complete photovoltaic panel information in the picture, if there is complete photovoltaic panel information, the edge computer forwards the picture data to the data analysis server, if not, returns to step S2;

[0018] S6, the data analysis server performs gray processing, denoising filtering processing, binary processing and morphological closing operation processing on the received picture;

[0019] S7, the edge computer repeats step S5, extracts the contour of the processed photovoltaic panel picture, and uses the region growing method to extract the area of the photovoltaic panel, if there is a photovoltaic shadow area surrounded by a square or a rectangle, then the area has foreign matter; if not, return to step S2;

[0020] S8, the data analysis server controls the cleaning device of the area to clean the photovoltaic panel of the area.

[0021] In the above invention content, further, the edge computer and the data analysis server are communicated through a gigabit switch.

[0022] In the above invention content, further, in the Densenet-53 structure of the YOLOv3 algorithm, the OSA module in VovNetV2 is replaced, and an eSE module is added.

[0023] In the above invention content, further, in step S4, the filtering processing adopts an average filtering method.

[0024] In the above invention content, further, in step S6, the Cvtcolor function is used to convert the image rgb to gray, and then convert the picture to a gray image.

[0025] In the above invention content, further, in step S6, the GaussianBlur function is used to denoise the picture.

[0026] In the above invention content, further, in step S6, the Gaussianfilter function is used to perform Gaussian filtering processing on the picture.

[0027] In the above invention, further, in step S6, the Threshold function is used to binarize the picture.

[0028] In the above invention, further, in step S6, the Morphologyex function is used to perform a morphological closing operation.

[0029] A cleaning device for foreign matter of a photovoltaic panel suitable for a micro-grid system;

[0030] The cleaning device comprises a transmission belt arranged on both sides of the photovoltaic panel, a cleaning contact, and a motor in driving connection with the transmission belt. The transmission belts on the two sides are provided with a fixing frame, the cleaning contact is fixedly connected with the fixing frame through a support, and the cleaning surface of the cleaning contact is in abutment with the surface of the photovoltaic panel.

[0031] In the above invention, further, the cleaning surface of the cleaning contact comprises a non-adhesive area composed of non-woven fabric and an adhesive area composed of solidified fibers, and the adhesive area is arranged in a dot or line shape in the non-adhesive area.

[0032] In the above invention, further, the power p of the motor is:

[0033] p=f 摩 ×v

[0034] Wherein, f 摩 m is the sliding frictional force between the cleaning contact and the photovoltaic panel, and the calculation formula is:

[0035] f 摩 =m×g / tanθ

[0036] m is the mass of the cleaning device; g is the local gravitational acceleration; and θ is the included angle between the photovoltaic panel and the horizontal plane.

[0037] v is the average speed of the transmission belt, and the value range is v=0.03-0.08m / s.

[0038] The beneficial effects of the present application are: the present application realizes the positioning and identification of the photovoltaic panel by adopting the image recognition and analysis technology of deep learning and convolutional neural network, and then uses the world coordinate system conversion method to make the industrial camera magnify and capture the corresponding position of the photovoltaic panel, which has the advantages of accurate foreign matter identification. Secondly, the cleaning device provided by the present application has the advantages of simple structure, energy saving, and high cleaning efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0039] Figure 1 The flowchart of the present application;

[0040] Figure 2 The flowchart of the picture processing of the data analysis server of the present application;

[0041] Figure 3 Structure diagram of the cleaning device of the present application. DETAILED DESCRIPTION

[0042] The present application can be implemented or applied in other different embodiments, and the details in the specification can be modified or changed based on different views and applications without departing from the spirit of the present application. It should be noted that the following examples and features in the examples can be combined with each other without conflict.

[0043] Embodiment:

[0044] A foreign matter identification method for photovoltaic panels suitable for micro-grid systems, as shown in the accompanying drawings, the method comprises the following steps: Figure 1 The method comprises the following steps:

[0045] S1, the industrial camera calibrates the position information of the photovoltaic panel arrangement, saves the industrial camera control coordinates corresponding to the photovoltaic panel arrangement position in the form of a structure to the edge computer, wherein the coordinates include horizontal angle p, vertical angle t, magnification z

[0046] S2, the edge computer controls the industrial camera to rotate to the ptz coordinates saved in step S1 to capture pictures, and sends the captured picture data to the data analysis server;

[0047] S3, the data analysis server uses the YOLOv3 algorithm to identify the photovoltaic panel in the picture; then uses the tracking algorithm of deep learning to identify and position the photovoltaic panel in the picture, and then uses the obtained positioning information to adjust the angle of the industrial camera, and identifies and positions the photovoltaic panel again;

[0048] In this step, in order to suppress gradient disappearance and improve the indicators AP and mAP, in the present application, the Darknet-53 in the YOLOv3 algorithm is improved, specifically: in the Densenet-53 structure of the YOLOv3 algorithm, the

[0049] The OSA module in VovNetV2 is replaced, and the eSE module is added, and the improved structure is shown in the following table:

[0050]

[0051]

[0052] S4, adjusting the focal length of the industrial camera according to the positioning information obtained again in step S3, the edge computer controlling the industrial camera to magnify the photovoltaic panel to the central part of the industrial camera to capture a picture again, then sending the captured picture to the data analysis server for filtering, and then sending the filtered picture to the edge computer again;

[0053] In this step, the data analysis server uses an average filtering method in the filtering process, and the filtering function is as follows:

[0054]

[0055]

[0056] wherein the value of M is 3

[0057] S5, the edge computer uses the Canny algorithm to perform edge extraction on the picture, and uses the Hough line detection and Harris corner point detection to identify the photovoltaic panel, and judges whether the picture contains complete photovoltaic panel information. If the picture contains complete photovoltaic panel information, the edge computer forwards the picture data to the data analysis server. If the picture does not contain complete photovoltaic panel information, the edge computer returns to step S2.

[0058] S6, the data analysis server processes the received picture, please refer to the attached Figure 2 picture processing process, which specifically includes: grayscale processing, denoising filtering processing, binary processing and morphological closing operation processing.

[0059] Preferably, when processing the picture, the Cvtcolor function is used to convert the image rgb to gray, and then convert the picture to a gray image; the GaussianBlur function is used to denoise the picture; the Gaussianfilter function is used to perform Gaussian filtering on the picture; the Threshold function is used to perform binary processing on the picture; and the Morphologyex function is used to perform morphological closing operation processing.

[0060]

[0061] S7, the edge computer repeats step S5, extracts the contour of the processed photovoltaic panel picture, and uses the region growing method to extract the area of the photovoltaic panel. If there is a photovoltaic shadow area surrounded by a square or a rectangle, the area contains foreign matter. If there is no photovoltaic shadow area, the edge computer returns to step S2.

[0062] S8, the data analysis server controls the cleaning device of the area to clean the photovoltaic panel of the area.

[0063] ​​It needs to be explained that the above embodiment is that the communication between the edge computer and the data analysis server is realized through the gigabit switch. In the specific implementation, the communication between the edge computer and the data analysis server can also be realized through other communication modes, including but not limited to the communication realized through the gigabit switch.

[0064] Embodiment 2:

[0065] A cleaning device for foreign matter of a photovoltaic panel suitable for a micro-grid system; the cleaning device comprises a conveyor belt 2 arranged on both sides of the photovoltaic panel 1, a cleaning contact 3, and a motor 6 in driving connection with the conveyor belt, a fixed frame 4 is arranged between the two conveyor belts 2, the cleaning contact 3 is fixedly connected with the fixed frame 4 through a support 5, and the cleaning surface of the cleaning contact 3 abuts against the surface of the photovoltaic panel 1. After receiving the foreign matter identification signal of the data analysis server, the data analysis server controls the rotating time and direction of the motor 6, so as to achieve the purpose of cleaning the surface of the photovoltaic panel 1 by the cleaning device. After cleaning is completed, the cleaning contact 3 returns to its initial position.

[0066] In the above embodiment, more specifically, the cleaning surface of the cleaning contact comprises a non-adhesive area composed of non-woven fabric and an adhesive area composed of solidified fibers, the adhesive area is arranged in a dot or line shape in the non-adhesive area, the non-woven fabric in the non-adhesive area can wipe off fine dirt such as oil film, and the solidified fibers in the adhesive area can wipe off stubborn dirt that is firmly stuck, so that the cleaning contact can exhibit good wiping performance on various dirt. Since the solidified fibers in the adhesive area are harder than the non-adhesive part, stubborn dirt can be easily wiped off, and the cleaning performance can be improved. At the same time, the flexibility of the non-woven fabric in the non-adhesive part can softly wipe against the concave-convex of the wiping object. The material has good washing durability and can be repeatedly used, which is conducive to reducing costs.

[0067] Embodiment 3:

[0068] In order to reduce the power consumption of the cleaning device, the selection of the motor 6 needs to consider the actual power required to drive the cleaning contact, so we need to calculate the power of the motor required to drive the conveyor belt and the cleaning device to move.

[0069] According to the power calculation formula: P = F 摩 × V, the calculation of the sliding friction needs to calculate the inclination angle of the photovoltaic panel and the horizontal plane. The inclination angle is an important factor affecting the power generation capacity of the photovoltaic panel, and the determination of the inclination angle is affected by many factors such as the latitude, altitude, climate conditions and installation position of the installation site, which needs to be considered comprehensively. Prediction model of solar hourly radiation intensity:

[0070]

[0071] where W(q) is the weather factor, representing the influence of weather on the intensity of radiation; q is the weather type variable; A(Y, G) is the projected area of the photovoltaic panel per unit area in the direction of the perpendicular incidence of sunlight; Y is the position of the photovoltaic panel, which can be represented by the angle a between the plane of the photovoltaic panel and the horizontal plane g and the angle β between the plane of the photovoltaic panel and the east-west direction g . H0 represents the intensity of radiation per unit area at the Earth's surface L at time T, and the position L at the surface can be represented by the latitude φ and the altitude h. The time T can be represented by t1 and t2, t1 being the hour variable and t2 being the time of day variable, N = 355 or 366.

[0072] The solar radiation on the surface of the surface object is composed of direct radiation, scattered radiation and reflected radiation. The direct radiation is the main part of the radiation amount. It has the greatest influence on the photovoltaic power generation relative to the other two radiations. The scattered radiation is included in the weather factor. Ignoring the reflected radiation, we only care about the direct radiation. The total amount of daily astronomical radiation refers to the total radiation perpendicularly incident on the upper layer of the atmosphere in a day, and the calculation formula is as follows:

[0073]

[0074] The relative distance ρ between the sun and the earth is:

[0075]

[0076] In the formula: T is the period (24×24×60s); Gsc is the solar constant (1.367×10 -3 MJ / m 2 s); ω s is the sunset angle on the horizontal plane; δ is the solar declination angle; J is the number of days in a year, i.e. the day sequence, from January 1 to December 31.

[0077] The total radiation H h per unit area per hour on the horizontal plane is calculated as follows:

[0078]

[0079] In the formula:

[0080]

[0081]

[0082] In the formula, ω s is the solar azimuth angle. In this way, the intensity of solar radiation per unit area in any 1h of a day can be known.

[0083] Classification of meteorological conditions and determination of influencing factors

[0084] In actual situation, the amount of radiation reaching the surface of photovoltaic panels has a strong correlation with weather factors. Cloudy, sandstorms, rain and snow weather will cause a large gap between the actual value and the theoretical value of radiation intensity, even lower than the minimum radiation intensity required by the power generation system in the ideal large radiation intensity period, so that the photovoltaic power generation system is in a shutdown state. According to the influence degree of weather on radiation intensity, the daily maximum and minimum temperature difference, humidity, and precipitation are used to cluster analyze the weather type. Considering the differences of weather conditions in different seasons and different regions, the data is preprocessed first, and then the weather variables are normalized. Finally, the distance between the weather of a day and each type of weather is calculated according to the interval clustering principle to obtain the type of the weather of the day.

[0085] The daily power generation of photovoltaic panels is counted, and the power generation under each type of weather in a year can be obtained. Taking the power generation of weather type 1 as the maximum value and 0 as the minimum value, the power generation loss percentage under each type of weather is counted, and the weather factor value corresponding to each type of weather is calculated.

[0086] Using meteorological statistical data, the distribution of each type of weather in a year can be obtained. The weather type is simulated by using the Monte Carlo method, and the specific calculation process is as follows:

[0087] 1) Data initialization, give the change range of installation angle, determine β g whether it needs to change; determine the number of Monte Carlo simulation N1; give the latitude, altitude and weather variables of the calculation area, and preprocess the data;

[0088] 2) Cluster the weather factors, give the distribution of each type of weather in each season, and determine the probability of occurrence of each type of weather.

[0089] 3) Calculate the hourly solar radiation intensity.

[0090] 4) Determine the calculation angles α g and β g , let i = 1, F 1j = 0;

[0091] 5) Generate a random number, and use the roulette strategy to determine the weather type of the day;

[0092] 6) Calculate the total amount of solar radiation received by the photovoltaic panel in a year;

[0093] 7) Calculate the average annual radiation received under the installation angles α g and β g ;

[0094] 8) Determine whether all calculation angles have been traversed.

[0095] 9) output the optimal angle. Output the installation angle corresponding to the maximum annual radiation received.

[0096] By the above manner, taking Shanghai as an example, the weather factors, altitude, latitude and other factors of Shanghai are comprehensively considered, and the optimal inclination angle of the photovoltaic panel and the horizontal plane is calculated to be 25° for the maximum power generation.

[0097] In order to simplify the problem, it is agreed that the conveyor belt is always in uniform motion, and the following can be obtained:

[0098] f 摩 = m x g / tan theta

[0099] In the formula, f 摩 is the sliding friction; m is the mass of the cleaning device (m = 3 kg); g is the local gravitational acceleration g = 9.8; theta is the angle between the photovoltaic panel and the horizontal plane (theta = 25°). The data is brought into the calculation, and the average value is obtained by multiple calculations f = 63.04 N.

[0100] Through the power calculation formula:

[0101] p = f 摩 x v

[0102] In the formula, p is the power, f 摩 is the sliding friction; v is the average speed of the conveyor belt; the data is brought into the calculation, and the power range is 1.89-5.04 W. In practice, we should design a 20% redundancy of the motor power, and the power range is calculated to be 2.363-6.3 W. Considering the aging power loss of the motor, a motor with a power of 3 W can be selected in this area.

[0103] Further, it is found through experiments that the speed of the conveyor belt cannot be too fast or too slow. Too fast will make leaves, paper, dust, plastic and other light objects float, and too slow cannot achieve the purpose of timely cleaning. After many experiments, it is found that the speed should be controlled within the range of 0.03-0.08 m / s.

[0104] The above-described embodiments only express the specific implementation of the present application, and the description is more specific and detailed, but it cannot be understood as a limitation on the scope of the patent of the present application. It should be noted that for ordinary skilled persons in the art, without departing from the concept of the present application, a number of modifications and improvements can be made, which are within the scope of protection of the present application.

Claims

1. A method for identifying foreign objects on photovoltaic panels suitable for microgrid systems, characterized in that, Includes the following steps: S1. The industrial camera calibrates the position information of the photovoltaic panel arrangement and saves the industrial camera control coordinates corresponding to the photovoltaic panel arrangement position to the edge computer in the form of a structure. S2. The edge computer controls the industrial camera to rotate to the coordinates saved in step S1 to capture images, and sends the captured image data to the data analysis server; wherein, the coordinates include horizontal angle p, vertical angle t, and magnification z; S3, the data analysis server uses the YOLOv3 algorithm to identify the photovoltaic panels in the image; then it uses a deep learning tracking algorithm to identify and locate the photovoltaic panels in the image, and then uses the obtained positioning information to adjust the angle of the industrial camera to identify and locate the photovoltaic panels again; S4. Adjust the focal length of the industrial camera according to the positioning information obtained again in step S3. The edge computer controls the industrial camera to magnify the photovoltaic panel to the center of the industrial camera to capture the image again. Then, the captured image is transferred to the data analysis server for filtering and then sent back to the edge computer. S5. The edge computer uses the Canny algorithm to extract edges from the image, and then uses Hough line detection and Harris corner detection to identify the photovoltaic panel. It determines whether the image contains complete photovoltaic panel information. If complete photovoltaic panel information exists, the edge computer forwards the image data to the data analysis server. If not, it returns to step S2. S6. The data analysis server performs grayscale conversion, noise reduction filtering, binarization, and morphological closing operations on the received images. S7. The edge computer repeats step S5 to extract the outline of the processed photovoltaic panel image and uses the region growing method to extract the area of ​​the photovoltaic panel. If there is a photovoltaic shadow area that forms a square or rectangle, then there is a foreign object in the area; if not, return to step S2. S8. The data analysis server controls the cleaning device in this area to clean the photovoltaic panels in this area.

2. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S3, the Densenet-53 structure of the YOLOv3 algorithm is replaced with the OSA module from VovNetV2, and the eSE module is added.

3. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S4, the filtering process is performed using an average filtering method.

4. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S6, the Cvtcolor function is used to convert the image from RGB to grayscale, thus converting the image into a grayscale image.

5. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S6, the GaussianBlur function is used to denoise the image.

6. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S6, the image is processed by Gaussian filtering using the Gaussianfilter function.

7. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S6, the image is binarized using the Threshold function.

8. The method for identifying foreign objects on photovoltaic panels in a microgrid system according to claim 1, characterized in that, In step S6, morphological closing operations are performed using the Morphologyex function.

9. A cleaning device based on the photovoltaic panel foreign object identification method for microgrid systems according to claim 1, characterized in that, The cleaning device includes conveyor belts on both sides of the photovoltaic panel, cleaning contacts, and a motor that is driven by the conveyor belts. A fixed frame is provided between the two conveyor belts. The cleaning contacts are fixedly connected to the fixed frame through a bracket, and the cleaning surface of the cleaning contacts abuts against the surface of the photovoltaic panel.

10. A cleaning device according to claim 9, characterized in that, The cleaning surface of the cleaning contact includes a non-adhesive area made of non-woven fabric and an adhesive area made of cured fibers. The adhesive area is arranged in a dotted or linear manner within the non-adhesive area.

11. A cleaning device according to claim 9, characterized in that, The power p of the motor is: p=f 摩 ×v; Among them, f 摩 This is the sliding friction force between the cleaning contact and the photovoltaic panel, and its calculation formula is: in 摩 =m×g / tanθ; m is the mass of the removal device; g is the local gravitational acceleration. It is the angle between the photovoltaic panel and the horizontal plane; v is the average speed of the conveyor belt, and its value ranges from v = 0.03 to 0.08 m / s.

Images