Photovoltaic panel cleaning path planning method and system based on dirt retention characteristic evaluation

Abstract

The invention discloses a photovoltaic panel cleaning path planning method and system based on dirt retention characteristic evaluation, relates to the technical field of photovoltaic panel cleaning, and solves the problems that the specific position of a corresponding dust area is comprehensively locked on the basis of the gray expression of the overall surface of a corresponding photovoltaic panel, and the specific position of the corresponding dust area is not comprehensively locked on the basis of the specific characteristics of the dust area. According to the method, image confirmation is carried out on the surface of the photovoltaic panel, a standard image associated with the corresponding photovoltaic panel is locked, then the standard image is checked with a preset image, a corresponding to-be-cleaned area is determined, and based on the overall contour features of the to-be-cleaned area, the to-be-cleaned area is determined. According to the method, the characteristic circles associated with the corresponding to-be-cleaned areas are determined, then diameter uniform distribution is performed on the characteristic circles, in the actual treatment process, characteristic diameter determination is performed on the uniformly distributed related diameters, so that the corresponding cleaning route is locked, and by adopting the determination mode of the cleaning route, the surface of the photovoltaic panel can be effectively cleaned, and the cleaning efficiency of the photovoltaic panel is improved. And the cleaning effect is more comprehensive.

Description

Technical Field

[0001] The present invention relates to the technical field of photovoltaic panel cleaning, and specifically to a method and system for planning a cleaning path of a photovoltaic panel based on the evaluation of fouling characteristics. Background Art

[0002] A photovoltaic panel (solar panel) is a device that converts solar energy into electrical energy; its core part is a solar cell, which mainly works based on the photovoltaic effect of semiconductors; when sunlight shines on the photovoltaic panel, photons will excite electrons in the semiconductor material, causing them to transition from the valence band to the conduction band, thereby generating an electric current; this process realizes the direct conversion of solar energy to electrical energy, and different semiconductor materials (such as monocrystalline silicon, polycrystalline silicon, amorphous silicon, etc.) and battery structures (such as PERC cells, HJT cells, etc.) will affect the performance of the photovoltaic panel.

[0003] An application with the publication number CN116611602A discloses a method and system for planning a cleaning path of a photovoltaic panel. The method includes: collecting an image of the photovoltaic panel, calculating the pose of the photovoltaic panel through image processing, and calculating the actual size of the photovoltaic panel based on the pose of the photovoltaic panel; dividing the photovoltaic panel into several unit grids according to the size of the single cleaning range of the cleaning device; calculating the cleaning cost of each grid starting from a preset starting cleaning position based on the dirt type in each grid identified by image recognition, and setting a cleaning priority; comprehensively calculating the cleaning priority based on the cleaning priority, and planning an optimal cleaning path according to the cleaning priority and the cleaning cost. Through this solution, the water consumption and cleaning time of photovoltaic panel cleaning can be reduced, and the cleaning efficiency can be effectively improved based on path planning.

[0004] When the corresponding cleaning planning path of the photovoltaic panel is confirmed, each relevant grid is cleaned separately. During the cleaning process, there is a diffusion range, which will cause waste of the corresponding cleaning liquid. It does not lock the specific position of the corresponding dust area based on the gray-scale performance of the overall surface of the corresponding photovoltaic panel, and does not plan the route and determine the shooting speed based on the specific characteristics of the dust area, so it is impossible to achieve the effects of energy conservation and excellent cleaning. Summary of the Invention

[0005] Aiming at the deficiencies of the prior art, the present invention provides a method and system for planning a cleaning path of a photovoltaic panel based on the evaluation of fouling characteristics, and solves the problem of not planning the route and determining the shooting speed based on the gray-scale performance of the overall surface of the corresponding photovoltaic panel.

[0006] To achieve the above object, the present invention is realized through the following technical solutions: A method for planning a cleaning path of a photovoltaic panel based on the evaluation of fouling characteristics, including the following steps:

[0007] Step 1: Obtain the complete surface image of the photovoltaic panel to be cleaned based on the machine vision device, and then determine the standard image of this photovoltaic panel to be cleaned from the complete surface image based on the edge feature detection method. The specific method is as follows:

[0008] S11. From the obtained complete surface image, identify the horizontal gradient and vertical gradient of the internal image points, and label the horizontal gradient associated with the corresponding image point as T i and label the associated vertical gradient as Z i where i represents different image points;

[0009] S12. Use to confirm the gradient eigenvalue G associated with the corresponding image point i and compare G i with the preset threshold Y1, where Y1 is a preset value. When G i > Y1, label the corresponding image point as an edge contour point. When G i ≤ Y1, no relevant marking of the edge contour point is performed;

[0010] S13. Based on the sequentially labeled edge contour points, connect several edge contour points to confirm the overall edge contour of the corresponding photovoltaic panel, and then intercept and label the associated image inside the overall edge contour as the standard image;

[0011] Step 2: Based on the confirmed standard image of the photovoltaic panel to be cleaned, convert this standard image into a grayscale image, and confirm the current light intensity. Determine the preset image based on the light intensity, and compare the grayscale image with the preset image to lock the area to be cleaned. The specific method is as follows:

[0012] S21. Based on the relevant light intensity sensor, confirm the light intensity associated with the current moment, and label the confirmed light intensity as Q k where k represents different moments, and then confirm the preset image associated with this light intensity Q k and the preset image is set with the associated center point;

[0013] S22. Based on the overall edge contour of the standard image, lock the center point of this standard image, place the overall edge contour in a two-dimensional coordinate system, and based on the two-dimensional coordinates associated with the corresponding contour points, confirm the mean coordinates of several contour points associated with the overall edge contour, and perform the calibration of the center point inside the standard image based on the position of the mean coordinates;

[0014] S23. Coincide the center point of the standard image with the center point of the preset image, and identify whether the standard image and the preset image completely overlap. If they completely overlap, no processing is performed. If they do not completely overlap, rotate and scale the standard image until the standard image and the preset image completely overlap, and determine the overlapping points. Mark the overlapping points as the same feature points;

[0015] S24. Identify whether the gray values associated with the same feature points are consistent in the grayscale image converted from the standard image or the preset image. If they are not consistent, mark the same feature points within the standard image as abnormal points, and mark the specific area covered by multiple abnormal points as the area to be cleaned, and perform a marking process on the area to be cleaned within the standard image; if the gray values associated with the same feature points are consistent, no processing is performed;

[0016] Step 3. Based on the areas to be cleaned sequentially locked within the standard image, perform a regional contour analysis on each area to be cleaned. Based on the center point of the area to be cleaned and the farthest point of the regional contour, draw a set of characteristic circles, and lock the cleaning route within the characteristic circles. The specific method is as follows:

[0017] S31. Based on the marked area to be cleaned, confirm the regional contour of the corresponding area to be cleaned. Place the regional contour in a two-dimensional coordinate system. Based on the two-dimensional coordinates of the internal contour points of the regional contour, confirm the average coordinates, and then mark the center point within the area to be cleaned based on the confirmed average coordinates;

[0018] S32. Based on the center point marked within the area to be cleaned, find the associated point on the regional contour that is the farthest from the center point in terms of the straight-line distance. Mark this associated point as the farthest point, connect this center point and the farthest point to determine a set of connecting line segments. Based on the center point and the connecting line segments, use the center point as the center and the connecting line segments as the radius to generate a characteristic circle for the area to be cleaned, and construct the internal diameter within the characteristic circle. Each diameter is arranged in an equally divided state during the arrangement process. During the equal division process, confirm the outer line length L associated with adjacent diameters. The outer line length is the part of the circumferential length intercepted between the corresponding diameter endpoints on the same side. Divide the characteristic circle equally according to the number of associated diameters. Stop when the equal division reaches L ≤ Y2, where Y2 is a preset value;

[0019] S33. Based on the several groups of equally divided diameters confirmed within the characteristic circle, mark the part of the diameter within the area to be cleaned corresponding to the equally divided diameter as the characteristic diameter, and based on the position of the other endpoint of the characteristic diameter associated with the center of the circle, use the direction from the other endpoint to the center of the circle as the traveling direction to confirm the cleaning route associated with this characteristic diameter;

[0020] Step 4: Based on the confirmed cleaning route in the area to be cleaned, and based on the pixel value difference between the associated points inside the cleaning route in the standard image associated grayscale image and the preset image, determine the instantaneous shooting speed of the associated points, combine the successively confirmed instantaneous shooting speeds with the cleaning route to generate an execution data packet, and control the execution of the execution data packet through the associated control terminal to complete the cleaning process of the corresponding photovoltaic panel surface; the specific sub-steps are as follows:

[0021] S41: Based on the associated cleaning route in the area to be cleaned, mark the points associated inside the cleaning route as cleaning points, mark the grayscale value of this cleaning point in the standard image associated grayscale as HD q , mark the grayscale value of this cleaning point in the preset image as Hz q , where q represents different cleaning points, and use: SS q =(Hz q -HD q )×C1 + Bz to confirm the instantaneous shooting speed SS of its corresponding cleaning point q , where C1 is a preset fixed coefficient factor and Bz is a preset standard shooting speed;

[0022] S42: Successively confirm the cleaning points associated with different cleaning routes in each area to be cleaned, and confirm the instantaneous shooting speeds associated with the cleaning points. Integrate the corresponding cleaning routes and the instantaneous shooting speeds of the different cleaning points associated therewith to generate an execution data packet. Subsequently, the corresponding control terminal controls the associated spray gun based on this execution data packet to enable the corresponding spray gun to effectively clean the surface of the photovoltaic panel.

[0023] Preferably, the photovoltaic panel cleaning path planning system based on fouling characteristic evaluation includes:

[0024] An image processing terminal that processes the surface complete image of the photovoltaic panel to be cleaned obtained by the machine vision device, and determines the standard image of this photovoltaic panel to be cleaned from the surface complete image based on the edge feature detection method;

[0025] An area to be cleaned confirmation terminal that, based on the confirmed standard image of the photovoltaic panel to be cleaned, converts this standard image into a grayscale image, confirms the current light intensity, determines a preset image based on the light intensity, compares the grayscale image with the preset image, and locks the area to be cleaned;

[0026] A route planning terminal that, based on the areas to be cleaned successively locked in the standard image, analyzes the regional contour of each area to be cleaned, and based on the center point of the area to be cleaned and the farthest point of the regional contour, draws up a set of characteristic circles and locks the cleaning route in the characteristic circles;

[0027] The data packet generation end determines the instantaneous shooting speed of the associated points based on the cleaning route confirmed in the area to be cleaned and the pixel value difference between the associated points inside the cleaning route in the standard image and the preset image, and combines the sequentially confirmed instantaneous shooting speeds with the cleaning route to generate an execution data packet.

[0028] The present invention provides a method and system for planning a cleaning path of a photovoltaic panel based on fouling characteristics evaluation. Compared with the prior art, it has the following beneficial effects:

[0029] In the present invention, by performing image confirmation on the surface of the photovoltaic panel, the standard image associated with the corresponding photovoltaic panel is locked, and then the standard image is compared with the preset image to determine the corresponding area to be cleaned. Subsequently, based on the overall contour characteristics of the area to be cleaned, the characteristic circles associated with the area to be cleaned are confirmed, and then the diameters of the characteristic circles are evenly distributed. In the actual processing process, the relevant diameters of the even distribution are confirmed as the characteristic diameters, so as to lock the corresponding cleaning route. By using this method of determining the cleaning route, the surface of the photovoltaic panel can be effectively cleaned, and the cleaning effect is more comprehensive, achieving a better processing effect;

[0030] Subsequently, according to the gray value difference of different points in the cleaning route, the shooting speed associated with the specified point is determined, and the shooting speeds of the corresponding cleaning route and the corresponding points are integrated to confirm the corresponding execution data packet and perform associated execution, which can not only save the relevant cleaning liquid, but also fully guarantee the cleaning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 It is a schematic flow chart of the method of the present invention;

[0032] Figure 2 It is a schematic diagram for determining the equal division diameter of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0034] The First Embodiment

[0035] Please refer to Figure 1 , this application provides a method for planning a cleaning path of a photovoltaic panel based on fouling characteristics evaluation, including the following steps;

[0036] Step 1: Obtain a complete surface image of the photovoltaic panel to be cleaned based on a machine vision device (the front end of the corresponding cleaning device is equipped with a relevant machine vision device, and a high-definition probe is installed inside to obtain the overall image of the photovoltaic panel surface). Then, determine the standard image of this photovoltaic panel to be cleaned from the complete surface image based on edge feature detection (there may be other irrelevant areas in the complete surface image, so the corresponding areas need to be removed). The specific method for determining the standard image is as follows:

[0037] S11: From the obtained complete surface image, identify the horizontal gradient and vertical gradient of the internal image points, and label the horizontal gradient associated with the corresponding image point as T i and label the associated vertical gradient as Z i where i represents different image points, and their gradient values can be directly obtained from the acquired image and can be directly associated and collected;

[0038] S12: Use to confirm the gradient eigenvalue G associated with the corresponding image point i and compare G i with the preset threshold Y1. Here, Y1 is a preset value, and its specific value is determined by the operator according to experience. When G i > Y1, label the corresponding image point as an edge contour point; otherwise, do not make relevant markings for the edge contour point;

[0039] S13: Based on the sequentially labeled edge contour points, connect several edge contour points to confirm the overall edge contour of the corresponding photovoltaic panel, and then intercept and label the associated image inside the overall edge contour as the standard image;

[0040] This processing method is relatively common in the prior art, so it will not be elaborated here. Based on the relevant edge contour of the corresponding object, directly determine the standard image of the corresponding object to complete the interception processing of the corresponding standard image;

[0041] Step 2: Based on the confirmed standard image of the photovoltaic panel to be cleaned, convert this standard image into a grayscale image, and confirm the current light intensity. Determine the preset image based on the light intensity, and compare the grayscale image with the preset image to lock the area to be cleaned. The specific method for locking the area to be cleaned is as follows:

[0042] S21: Based on a relevant light intensity sensor, confirm the light intensity associated with the current moment, and label the confirmed light intensity as Q k where k represents different moments, and then confirm this light intensity Q kThe associated preset image, which is pre-determined by relevant operators. Different light intensities correspond to different preset images. The preset image is the specific grayscale value representation associated with the corresponding photovoltaic panel in different environments, which is a preset standard grayscale value representation image, and the associated center point is set within the preset image;

[0043] S22. Based on the overall edge contour of the standard image, lock the center point of this standard image, place the overall edge contour in a two-dimensional coordinate system, and based on the two-dimensional coordinates associated with the corresponding contour points, confirm the mean coordinates of several contour points associated with the overall edge contour. Calibrate the center point within the standard image based on the position of the mean coordinates;

[0044] S23. Overlap the center point of the standard image with the center point of the preset image, and identify whether the standard image and the preset image completely overlap. If they completely overlap, no processing is performed. If they do not completely overlap, rotate and scale the standard image until the standard image and the preset image completely overlap, and determine the overlapping points. Mark the overlapping points as the same feature points;

[0045] S24. Identify whether the grayscale values associated with the same feature points are consistent in the grayscale image or preset image converted from the standard image. If they are consistent, no processing is performed. If they are not consistent, mark the same feature points within the standard image as abnormal points, and mark the specific area covered by multiple abnormal points as the area to be cleaned, and perform a marking process on the area to be cleaned within the standard image (that is, mark the grayscale abnormal area on the standard image to facilitate subsequent related processing of this grayscale abnormal area, and subsequently plan the specific route from the associated grayscale abnormal area);

[0046] Step 3. Based on the areas to be cleaned successively locked within the standard image, perform a regional contour analysis on each area to be cleaned. Based on the center point and the farthest point of the regional contour of the area to be cleaned, draw a set of characteristic circles, and lock the cleaning route within the characteristic circles. The specific method for determining the cleaning route is as follows:

[0047] S31. Based on the marked area to be cleaned, confirm the regional contour of the corresponding area to be cleaned, place the regional contour in a two-dimensional coordinate system, based on the two-dimensional coordinates of the internal contour points of the regional contour, confirm the mean coordinates, and then calibrate the center point within the area to be cleaned based on the confirmed mean coordinates;

[0048] S32. Based on the calibrated center point in the area to be cleaned, find the associated point on the area contour that is the farthest from the center point in terms of the straight-line distance. Mark this associated point as the farthest point, connect this center point and the farthest point to determine a set of connecting line segments. Based on the center point and the connecting line segments, take the center point as the center of the circle and the connecting line segment as the radius to generate a characteristic circle for the area to be cleaned. And construct the inner diameter within the characteristic circle, and each diameter is arranged in an equally divided state during the arrangement (that is, the areas associated with each adjacent diameter are equal). And during the equal division arrangement process, confirm the outer perimeter line length L between adjacent diameters (the part of the perimeter intercepted between the corresponding diameter endpoints on the same side, and this part of the perimeter is the corresponding outer perimeter line length. Since each diameter is evenly arranged, the outer perimeter line length L associated with each diameter is also the same). Divide the characteristic circle equally according to the number of associated diameters increased. Stop when the equal division reaches L ≤ Y2, where Y2 is a preset value, and its specific value is determined by the operator according to experience;

[0049] S33. Based on the several groups of equally divided diameters confirmed within the characteristic circle, mark the part of the diameter of the corresponding equally divided diameter that is inside the area to be cleaned as the characteristic diameter. And based on the position of the characteristic diameter associated with the center of the circle and the other endpoint, take the direction from the other endpoint to the center of the circle as the traveling direction to confirm the cleaning route associated with this characteristic diameter (the walking path of this cleaning route is the covering path of the characteristic diameter, and its relevant cleaning direction is determined based on the two endpoints of the characteristic diameter. One of the endpoints coincides with the center of the circle, then mark this endpoint as the end point, and the other endpoint is the starting point. Take the direction from the starting point to the end point as the cleaning direction of the cleaning route to complete the association and formulation of the corresponding cleaning route);

[0050] Combined with Figure 2 , based on the confirmed area to be cleaned and the corresponding area contour, confirm the associated center point inside it. Then, based on the farthest point, determine a set of connecting line segments. Based on the center point and the connecting line segments, confirm the corresponding characteristic circle. Based on the characteristic circle and the internal relevant diameters, perform equal division processing on the equally divided diameters. And during the equal division process, gradually increase the number of corresponding equally divided diameters. Stop when the outer perimeter line length L between adjacent equally divided diameters meets the corresponding evaluation conditions.

[0051] Second Embodiment

[0052] In the specific implementation process of this embodiment, compared with the above embodiment, this embodiment is for the association and formulation of the instantaneous shooting speed at the corresponding cleaning points within the corresponding cleaning route;

[0053] It further includes the following steps:

[0054] Step 4. Based on the confirmed cleaning route in the area to be cleaned, and based on the pixel value difference between the associated points inside the cleaning route in the standard image associated grayscale image and the preset image, determine the instantaneous shooting speed of the associated points, and combine the sequentially confirmed instantaneous shooting speeds with the cleaning route to generate an execution data packet, and control the execution of the execution data packet through the associated control terminal to complete the cleaning process of the corresponding photovoltaic panel surface. The specific sub-steps for generating the execution data packet are as follows:

[0055] S41. Based on the associated cleaning route in the area to be cleaned, calibrate the points associated in the cleaning route as cleaning points, and calibrate the grayscale value of this cleaning point in the standard image associated grayscale as HD q , and calibrate the grayscale value of this cleaning point in the preset image as Hz q , where q represents different cleaning points, and use: SS q =(Hz q -HD q )×C1 + Bz to confirm the instantaneous shooting speed SS of its corresponding cleaning point q , where C1 is a preset fixed coefficient factor, and Bz is a preset standard shooting speed (the more dust, the lower the grayscale value of the corresponding area. Because dust will interfere with the light reflection situation in the corresponding area, it will cause the grayscale value of the corresponding point to be lower. The lower the grayscale value, the higher the instantaneous shooting speed is required for cleaning. The instantaneous shooting speed is the flow rate of the corresponding cleaning liquid sprayed by the corresponding spray gun);

[0056] S42. Sequentially confirm the cleaning points associated with different cleaning routes in each area to be cleaned, and confirm the instantaneous shooting speeds associated with the cleaning points. Integrate the corresponding cleaning routes and the instantaneous shooting speeds of the different cleaning points associated therewith to generate an execution data packet. Subsequently, the corresponding control terminal controls the associated spray gun based on this execution data packet to enable the corresponding spray gun to effectively clean the surface of the photovoltaic panel.

[0057] The photovoltaic panel cleaning path planning system based on the dirt accumulation characteristic evaluation includes:

[0058] An image processing terminal that processes the surface complete image of the photovoltaic panel to be cleaned obtained by the machine vision device, and determines the standard image of this photovoltaic panel to be cleaned from the surface complete image based on the edge feature detection method;

[0059] An area to be cleaned confirmation terminal that, based on the confirmed standard image of the photovoltaic panel to be cleaned, converts this standard image into a grayscale image, confirms the current light intensity, determines a preset image based on the light intensity, compares the grayscale image with the preset image, and locks the area to be cleaned;

[0060] The route planning terminal analyzes the regional contour of each area to be cleaned based on the areas to be cleaned sequentially locked in the standard image. Based on the center point of the area to be cleaned and the farthest point of the regional contour, a set of characteristic circles are drawn up, and the cleaning route is locked in the characteristic circles;

[0061] The data packet generation terminal determines the instantaneous firing rate of the associated points based on the pixel value differences between the standard image and the preset image of the associated points inside the cleaning route based on the confirmed cleaning route in the area to be cleaned, and combines the sequentially confirmed instantaneous firing rates with the cleaning route to generate an execution data packet.

[0062] The third embodiment

[0063] In the specific implementation process of this embodiment, it includes all the implementation processes of the above two sets of embodiments.

[0064] Some data in the above formula are numerically calculated by removing their dimensions, and the content not described in detail in this specification belongs to the prior art well known to those skilled in the art.

[0065] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation, characterized in that: The following steps are involved: Step 1: Obtain a complete surface image of the photovoltaic panel to be cleaned based on a machine vision device, and then determine a standard image of the photovoltaic panel to be cleaned from the complete surface image based on edge feature detection; Step 2: based on the confirmed standard image of the photovoltaic panel to be cleaned, convert the standard image into a grayscale image, confirm the light intensity at the current moment, determine the preset image based on the light intensity, compare the grayscale image with the preset image, and lock the area to be cleaned; Step 3: Based on the areas to be cleaned locked in sequence in the standard image, perform regional contour analysis on each area to be cleaned, formulate a set of characteristic circles based on the center point of the area to be cleaned and the farthest point of the regional contour, and lock the cleaning route in the characteristic circles; Step 4: Based on the cleaning route confirmed in the area to be cleaned, and based on the difference in pixel values ​​between the associated points in the standard image, the associated grayscale image, and the preset image, determine the instantaneous firing rate of the associated points, combine the confirmed instantaneous firing rates with the cleaning route in sequence, generate an execution data packet, and control the execution of the execution data packet through the associated control terminal to complete the cleaning process of the corresponding photovoltaic panel surface.

2. The photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation according to claim 1 is characterized in that: In step 1, the specific method of determining the standard image is: S11, from the acquired complete surface image, identify the horizontal gradient and vertical gradient of the internal image point, and calibrate the horizontal gradient associated with the corresponding image point as T i , and the associated vertical gradient is calibrated as Z i , where i represents different image points; S12, use Confirm the gradient eigenvalue G associated with the corresponding image point i , and G i It is checked against the preset threshold value Y1, where Y1 is the preset value. i >Y1, the corresponding image point is marked as an edge contour point; S13, based on the edge contour points calibrated in sequence, several edge contour points are connected to confirm the overall edge contour of the corresponding photovoltaic panel, and then the associated image inside the overall edge contour is intercepted and calibrated as a standard image.

3. The photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation according to claim 2 is characterized in that: In step S12, when G i When ≤Y1, no relevant marking of edge contour points is performed.

4. The photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation according to claim 1 is characterized in that: In step 2, the specific method of locking the area to be cleaned is: S21. Based on the relevant light intensity sensor, confirm the light intensity associated with the current moment, and calibrate the confirmed light intensity as Q k , where k represents different moments, and then confirm the light intensity Q k The associated preset image, wherein the associated center point is set in the preset image; S22, based on the overall edge contour of the standard image, locking the center point of the standard image, placing the overall edge contour in a two-dimensional coordinate system, and based on the two-dimensional coordinates associated with the corresponding contour points, confirming the mean coordinates of a plurality of contour points associated with the overall edge contour, and calibrating the center point in the standard image based on the location of the mean coordinates; S23, aligning the center point of the standard image with the center point of the preset image, and identifying whether the standard image and the preset image completely overlap, if they completely overlap, no processing is performed, if they do not completely overlap, rotating and scaling the standard image until the standard image and the preset image completely overlap, and determining the overlapping points, and marking the overlapping points as the same feature points; S24. Identify whether the grayscale values ​​associated with the same feature points are consistent from the grayscale image converted from the standard image or the preset image. If not, mark the same feature points located in the standard image as abnormal points, and mark the specific area covered by multiple abnormal points as an area to be cleaned, and mark the area to be cleaned in the standard image.

5. The photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation according to claim 4 is characterized in that: In step S24, if the grayscale values ​​associated with the feature points are consistent, no processing is performed.

6. The photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation according to claim 1 is characterized in that: In step 3, the specific method of determining the cleaning route is: S31, based on the calibrated area to be cleaned, confirm the area contour corresponding to the area to be cleaned, place the area contour in a two-dimensional coordinate system, confirm the mean coordinates based on the two-dimensional coordinates of the contour points inside the area contour, and then calibrate the center point in the area to be cleaned based on the confirmed mean coordinates; S32, based on the center point marked in the area to be cleaned, find the associated point on the area contour that is farthest from the center point in a straight line, mark the associated point as the farthest point, connect the center point and the farthest point, determine a set of connecting line segments, and based on the center point and the connecting line segments, take the center point as the center of the circle and the connecting line segments as the radius to generate a characteristic circle about the area to be cleaned, and construct the internal diameter within the characteristic circle, and each diameter is arranged in an equally divided state, and in the process of equal division and arrangement, confirm the outer circle length L associated with adjacent diameters, and divide the characteristic circle equally according to the number of associated diameters, and stop when the equal division reaches L≤Y2, where Y2 is a preset value; S33. Based on several groups of equally divided diameters confirmed within the characteristic circle, the partial diameters of the corresponding equally divided diameters located inside the area to be cleaned are calibrated as characteristic diameters, and based on the positions of the center of the circle and the other end point associated with the characteristic diameter, the direction from the other end point to the center of the circle is used as the travel direction to confirm the cleaning route associated with this characteristic diameter.

7. The photovoltaic panel cleaning path planning method based on pollution characteristics evaluation according to claim 6 is characterized in that: In step S32, the outer circle line length is the circumference of the portion cut off between the corresponding diameter endpoints on the same side.

8. The photovoltaic panel cleaning path planning method based on pollution accumulation characteristic evaluation according to claim 1 is characterized in that: In step 4, the specific sub-steps of generating the execution data packet are: S41, based on the cleaning route associated with the area to be cleaned, mark the points associated with the cleaning route as cleaning points, and mark the grayscale value of the cleaning point in the grayscale associated with the standard image as HD q , the gray value of this cleaning point in the preset image is calibrated as Hz q , where q represents different cleaning points, using: SS q =(Hz q -HD q )×C1+Bz confirms the instantaneous firing rate SS of the corresponding cleaning point q , where C1 is the preset fixed coefficient factor, and Bz is the preset standard firing rate; S42. Confirm the cleaning points associated with different cleaning routes in each area to be cleaned in turn, and confirm the instantaneous shooting speed associated with the cleaning points, integrate the corresponding cleaning routes and the instantaneous shooting speeds of the different associated cleaning points, and generate an execution data packet. Subsequently, the corresponding control end controls the associated spray gun based on this execution data packet, so that the corresponding spray gun effectively cleans the surface of the photovoltaic panel.

9. A photovoltaic panel cleaning path planning system based on pollution characteristics assessment, the planning system is used to execute the photovoltaic panel cleaning path planning method based on pollution characteristics assessment according to any one of claims 1 to 8, characterized in that: include: The image processing end processes the complete surface image of the photovoltaic panel to be cleaned obtained by the machine vision device, and determines the standard image of the photovoltaic panel to be cleaned from the complete surface image based on edge feature detection; The to-be-cleaned area confirmation end converts the standard image into a grayscale image based on the confirmed standard image of the photovoltaic panel to be cleaned, confirms the light intensity at the current moment, determines a preset image based on the light intensity, compares the grayscale image with the preset image, and locks the to-be-cleaned area; On the route planning side, based on the areas to be cleaned that are locked in sequence in the standard image, the regional contour analysis is performed on each area to be cleaned, and a set of feature circles is drawn up based on the center point of the area to be cleaned and the farthest point of the regional contour, and the cleaning route is locked in the feature circles; The data packet generation end determines the instantaneous firing rate of the associated points based on the cleaning route confirmed in the area to be cleaned and the difference in pixel values ​​between the standard image and the preset image at the associated points within the cleaning route, and combines the instantaneous firing rates confirmed in sequence with the cleaning route to generate an execution data packet.

Images