Photovoltaic power station operation and maintenance method and system capable of realizing intelligent navigation
The vertical edge of the photovoltaic module is identified by the processor, and the direction of the cleaning equipment is calculated and adjusted, which solves the problem of dislocation of the cleaning equipment of the photovoltaic power station and improves the cleaning efficiency and intelligence level.
Patent Information
- Application Number
- CN202410989700.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-23
- Publication Date
- 2024-12-13
AI Technical Summary
When cleaning equipment of photovoltaic power stations is easily deviated due to protrusions and weather factors when walking in a straight line, resulting in reduced cleaning efficiency and poor uniformity, and lacks precise positioning and intelligent control.
The photovoltaic component image is obtained through the processor, the vertical edge is identified as the reference line, the forward direction of the cleaning equipment is calculated and adjusted, and the offset is automatically corrected to ensure that the cleaning equipment is moving along the prescribed path.
It improves the cleaning efficiency and effect of photovoltaic power station cleaning equipment, enhances the precise positioning ability of designated areas, and improves the level of intelligent operation and maintenance.
Smart Images
Figure CN119126774A_ABST
Abstract
Description
Technical Field
[0001] The embodiments of this specification relate to the field of photovoltaic panel cleaning, and more particularly to a photovoltaic power station operation and maintenance method and system capable of realizing intelligent navigation. Background Art
[0002] Photovoltaic power stations are power generation systems that convert solar energy directly into electrical energy through the photovoltaic effect. In order to ensure that the photovoltaic modules in the photovoltaic power station can receive sunlight to the greatest extent, the photovoltaic modules are usually laid out and arranged in large areas in the outdoor environment. When the cleaning equipment of the photovoltaic power station moves and cleans the surface of the photovoltaic modules, it needs to follow a certain path to ensure the cleaning efficiency and cleaning effect of the photovoltaic modules. The conventional path is that the cleaning equipment moves in a straight line along the rows or columns arranged by the photovoltaic modules and cleans them in sequence. Due to the protrusions that may exist on the surface of the photovoltaic modules and the interference of outdoor weather, the straight-line movement of the cleaning equipment is prone to deviation, resulting in the omission of some areas of the photovoltaic modules during cleaning, reducing the cleaning efficiency, and the cleaning effect and cleaning uniformity cannot be guaranteed; at the same time, the deviation of the cleaning equipment is not conducive to the precise positioning of the designated cleaning area, and the intelligence level of the cleaning equipment is low. Summary of the invention
[0003] The embodiments of this specification provide a photovoltaic power station operation and maintenance method and system capable of implementing intelligent navigation, aiming to solve one or more of the above-mentioned problems and other potential problems.
[0004] In order to achieve the above objectives, the following technical solutions are provided: According to a first aspect of the present specification, a photovoltaic power station operation and maintenance method capable of realizing intelligent navigation is provided, the method being implemented based on a processor and comprising: Acquire an image of the photovoltaic module in the direction of the cleaning device; Processing and analyzing the photovoltaic module image to obtain a vertical edge of the photovoltaic module, and using it as a reference line; wherein the vertical edge of the photovoltaic module is an edge of a photovoltaic panel on the photovoltaic module parallel to a preset forward direction of the cleaning device; Calculating the forward offset angle of the cleaning device based on the position of the reference line in a plane rectangular coordinate system; wherein the plane rectangular coordinate system takes the bottom edge of the photovoltaic module image as the x-axis; The advancing direction of the cleaning device is adjusted based on the advancing offset angle until the advancing offset angle is less than an offset threshold.
[0005] The photovoltaic power station operation and maintenance method and system that can realize intelligent navigation provided by the embodiments of this specification, by identifying the reference lines in the photovoltaic modules, and using the reference lines as a reference to calculate the forward deviation angle of the cleaning device, the cleaning device automatically corrects the path deviation caused by various factors during the forward process, thereby improving the cleaning efficiency and ensuring the cleaning effect. When the cleaning device moves forward in a specified direction, the inclination of the edge of the photovoltaic panel captured in front of the cleaning device and consistent with the forward direction of the cleaning device is relatively fixed, that is, the inclination of the reference line is relatively fixed, and the accuracy of the correction using the reference line as a reference is high, and the position of the shooting device on the cleaning device is not high. It is suitable for the identification and navigation of the cleaning device on photovoltaic modules of various sizes, and the operation and maintenance of the photovoltaic power station is more intelligent.
[0006] In some embodiments, processing and analyzing the photovoltaic assembly image to obtain the vertical edge of the photovoltaic assembly and using it as a reference line includes: Processing the photovoltaic module image by an edge detection algorithm to obtain an edge image; Extracting lines in the edge image using a line extraction function; wherein the lines in the edge image include vertical edges of photovoltaic modules; The lines outside the preset slope range are eliminated, and the reference line is obtained by fitting the remaining lines using the least square method.
[0007] In some embodiments, after obtaining the edge image, the method further includes: The lateral edge of the photovoltaic component in the edge image is hidden by a mask image shielding matrix; wherein the lateral edge is an edge of a photovoltaic panel on the photovoltaic component perpendicular to a preset forward direction of the cleaning device.
[0008] In some embodiments, the forward direction of the cleaning device is adjusted based on the forward offset angle until the forward offset angle is less than an offset threshold, and also includes: starting timing when the forward offset angle exceeds the angle threshold, if the forward offset angle still exceeds the angle threshold when the timing time exceeds the timing threshold, the cleaning device stops running and uploads abnormal information to the server.
[0009] In some embodiments, the method further includes: when the length of the reference line, the coordinates of the midpoint of the reference line, and the forward offset angle are not within a standard range, determining that the reference line recognition is abnormal, and refitting the reference line.
[0010] In some embodiments, it also includes: when the photovoltaic component image is abnormal, the cleaning device stops moving forward, and searches whether there is an identification in the image in front of the cleaning device at this time. If there is, the cleaning device is controlled to act based on the identification, otherwise the abnormal information is uploaded to the server.
[0011] In some embodiments, it also includes: when it is determined that the power of the cleaning device is less than a power threshold, acquiring an image around the cleaning device, identifying a position of a charging device in the image around the cleaning device, calculating a distance and a path between the cleaning device and the charging device based on the position of the charging device, and controlling the cleaning device to move toward the charging device.
[0012] In some embodiments, the method further includes: continuously acquiring the photovoltaic component image at a certain time interval, and when the temperature of the computing module of the processor is too high, the computing module performs frequency reduction protection and lengthens the time interval.
[0013] According to the second aspect of the present specification, a photovoltaic power station operation and maintenance system capable of realizing intelligent navigation is provided, comprising photovoltaic modules and cleaning equipment, wherein the cleaning equipment is provided with a shooting device, a processor and a walking device, the shooting device is used to shoot images of the photovoltaic modules around the cleaning equipment, and the processor is used to control the movement of the cleaning equipment on the photovoltaic modules based on the above-mentioned photovoltaic power station operation and maintenance method capable of realizing intelligent navigation.
[0014] In some embodiments, a plurality of marks are provided on the end edge of the photovoltaic assembly, and the marks contain turning information that can be recognized by the cleaning device. BRIEF DESCRIPTION OF THE DRAWINGS
[0015] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the embodiments of the present specification will become readily understood. In the accompanying drawings, several embodiments of the present specification are shown by way of example and not limitation.
[0016] Figure 1 A flow chart of a photovoltaic power station operation and maintenance method capable of realizing intelligent navigation according to an embodiment of this specification is shown; Figure 2 A schematic diagram of a reference line according to an embodiment of the present specification is shown.
[0017] 1-Left reference line, 2-Right reference line.
[0018] In the various drawings, the same or corresponding reference numerals represent the same or corresponding parts. DETAILED DESCRIPTION
[0019] The preferred embodiments of the present specification will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present specification are shown in the accompanying drawings, it should be understood that the present specification can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present specification more thorough and complete, and to fully convey the scope of the present specification to those skilled in the art.
[0020] The term "including" and its variations used in this document represent open inclusion, that is, "including but not limited to". Unless otherwise stated, the term "or" means "and / or". The term "based on" means "based at least in part on". The terms "an example embodiment" and "an embodiment" mean "at least one example embodiment". The term "another embodiment" means "at least one other embodiment". Terms such as "upper", "lower", "front", and "rear" indicating placement or positional relationships are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the principles of this specification, and do not indicate or imply that the referred elements must have a specific orientation, be constructed or operate in a specific orientation, and therefore should not be understood as limitations on this specification.
[0021] A photovoltaic power station operation and maintenance method and system capable of realizing intelligent navigation according to an embodiment of the present specification will be described in detail below with reference to the accompanying drawings. Figure 1 A flow chart of a photovoltaic power station operation and maintenance method capable of realizing intelligent navigation in an embodiment of this specification is shown. The photovoltaic power station operation and maintenance method capable of realizing intelligent navigation in an embodiment of this specification is implemented based on a processor and includes the following steps: S1, obtaining an image of the photovoltaic module in the direction of the cleaning device; S3, processing and analyzing the photovoltaic module image to obtain the vertical edge of the photovoltaic module, and using it as a reference line; S5, calculating the forward deviation angle of the cleaning device based on the position of the reference line in the plane rectangular coordinate system; S7. Adjusting the forward direction of the cleaning device based on the forward offset angle until the forward offset angle is less than an offset threshold.
[0022] The cleaning device captures images or videos of photovoltaic components in the forward direction of the cleaning device through a camera or other shooting device and caches the images. In step S1, the processor obtains the images of the photovoltaic components at a certain time interval, or captures the images of the photovoltaic components in the video at a certain time interval to reduce computing power consumption; the time interval is determined by the program loop duration inside the processor. Under normal circumstances, the real-time image cached by the shooting device is read once every 0.1 seconds. When the temperature of the computing module of the processor is too high, the computing module performs frequency reduction protection and lengthens the time interval. The lengthened time interval generally does not exceed 0.3 seconds.
[0023] When the photovoltaic component image obtained in step S1 is abnormal, the cleaning device stops moving forward and searches to see whether there is an identification in the image in front of the cleaning device at this time. If so, the cleaning device is controlled to move based on the identification, otherwise the abnormal information is uploaded to the server. The photovoltaic component image is abnormal, which means that the photovoltaic component in the image captured by the shooting device is incomplete or the photovoltaic component cannot be captured. At this time, the photovoltaic component image is identified as an abnormal image, that is, the cleaning device may move to the end of the photovoltaic component at this time. Photovoltaic components are generally laid out by photovoltaic panels arranged in rows and columns. Therefore, when the cleaning device advances to the end of a row or column of photovoltaic components, it needs to turn or make a U-turn to clean the surface of photovoltaic components in other areas; in order to increase the intelligence level of the cleaning device, realize automatic walking and reduce manual operation, the end of the photovoltaic component is provided with the identification, which is a QR code or other special symbol containing position and steering information. When the identification is a QR code, the processor can parse the information contained in the QR code and find the corresponding pre-stored instructions based on the parsed information. When there is a QR code in the image in front of the cleaning device, the processor can control the cleaning device to turn left, turn right, turn left, turn right, etc. based on the pre-stored instructions; the QR code can also contain other instructions, so that the cleaning device can perform more actions based on the instructions. Similarly, when the identifier is a symbol or image in other forms, the processor can also find the corresponding pre-stored instructions based on the identifier. When the area of the photovoltaic module is large, multiple QR codes can be set, and the multiple QR codes are arranged along the end edge of the photovoltaic module. When there are multiple QR codes in the image in front of the cleaning device, the QR code closest to the cleaning device is selected for parsing and executing the instructions in the QR code. When there is no QR code in the image in front of the cleaning device, it means that the cleaning device has not moved to the end of the photovoltaic module at this time, or the movement and image shooting of the cleaning device on the photovoltaic module are abnormal, and it is necessary to upload abnormal information to the server. At this time, the cleaning device stops running. After receiving the abnormal information, the staff can remotely observe the status of the cleaning device through the shooting device and make corresponding adjustments.
[0024] Specifically, step S3 includes: S31, processing the photovoltaic module image by an edge detection algorithm to obtain an edge image; S33, extracting lines in the edge image using a line extraction function; S35, eliminating the lines outside the preset slope range, and fitting the remaining lines using the least square method to obtain the reference line.
[0025] In step S31, the photovoltaic module image is first preprocessed, and the preprocessing includes: converting the photovoltaic module image into an image matrix in BGR format through the opencv library; reducing the size of the image matrix to reduce the subsequent calculation amount; denoising the image after the size reduction through a Gaussian filter matrix; then converting the color image matrix in BGR format into a grayscale image to complete the preprocessing; finally, extracting the edge of the grayscale image through two sets of edge detection filter matrices to obtain the edge image.
[0026] The photovoltaic module is paved with photovoltaic panels arranged in rows and columns, and there are horizontal edges or vertical edges between several photovoltaic panels. The offset of the cleaning device is judged with reference to the vertical edge parallel to the preset forward direction of the cleaning device. The lines in the edge image include the vertical and horizontal edges of the photovoltaic module. In order to avoid the interference of the horizontal edge on the extraction of the vertical edge and increase the recognition accuracy of the vertical edge, after step S31 and before step S33, it is necessary to hide the area where the horizontal edge of the photovoltaic module in the edge image is located through a mask image shielding matrix, and filter out the noise and short lines in the remaining area through erosion and expansion algorithms.
[0027] In step S35, the lines outside the preset slope range are eliminated, that is, the vertical lines that are far or close to the cleaning device are eliminated, so that the slope of the fitted baseline line is appropriate and easy to calculate. When the length of the baseline line, the midpoint coordinates of the baseline line and the forward offset angle are not within the standard range, the baseline recognition is judged to be abnormal, and the baseline is re-fitted. For example: the shooting device shoots a video with a clarity of 720P, the resolution of the photovoltaic component image obtained by the processor is 1280*720, and the pixel resolution after preprocessing is 480*320. When the length of the baseline obtained by fitting is less than 50 pixels, the midpoint coordinates of the baseline line are higher than the midpoint of the photovoltaic component image, and the forward offset angle exceeds 20°, the above situations will judge that the baseline recognition is abnormal. When the baseline is re-fitted multiple times and is judged to be abnormal, abnormal information is sent to the service end.
[0028] In step S5, the plane rectangular coordinate system takes the bottom edge of the photovoltaic component image as the x-axis, and the y-axis is perpendicular to the x-axis and points to the forward direction of the cleaning device. Based on the slope of the reference line in the plane rectangular coordinate system, the angle between the reference line and the x-axis can be obtained. For example, when the slope is 1, the angle between the reference line and the x-axis is 45°; the forward offset angle is obtained by subtracting the angle from the standard angle. The standard angle can be set manually or automatically generated during the initialization stage before the cleaning device moves. For example, the robot is powered on by remote control, the main controller completes the startup in about 20 seconds, and the delay relay powers on the motor driver at the 30th second. After the automatic operation switch is remotely turned on, the robot will perform image recognition initialization in about 8 seconds, specifically acquiring the photovoltaic component image, identifying the reference line in the photovoltaic component image, and setting the angle between the reference line identified this time and the x-axis as the standard angle. Then the cleaning device starts to walk and clean automatically. The cleaning device automatically generates the standard angle each time it is started, so that the value of the standard angle is more accurate and applicable, can adapt to different cleaning environments, and reduce the impact of a slight tilt of the camera on the cleaning device. Figure 2 As shown, the two reference lines on both sides of the cleaning device are fitted, and the standard angles corresponding to the two reference lines are 45° and 135° respectively. When the angle between the left reference line 1 and the x-axis is greater than 45°, and the angle between the right reference line 2 and the x-axis is greater than 135°, it means that the cleaning device is deflected to the right at this time, and a higher pwm signal should be provided to the drive motor on the right side of the cleaning device, so that the speed of the walking device on the right side of the cleaning device is greater than the speed of the walking device on the left side, and the cleaning robot turns left to correct the direction; wherein, when the forward offset angle is 1°, the speed of the drive motor increases by 1%. The speeds of the drive motors on the left and right sides of the cleaning device are respectively determined by the forward offset angles of the left and right reference lines, avoiding the influence of the difference between the forward offset angles of the two reference lines; when the difference between the forward offset angles of the two reference lines exceeds the normal range, the reference lines also need to be refitted.
[0029] In step S7, the timing starts when the forward offset angle exceeds the angle threshold. If the forward offset angle still exceeds the angle threshold when the timing time exceeds the timing threshold, it means that the cleaning device fails to automatically correct the forward route within the timing threshold, and the image recognition or return action of the cleaning device is abnormal. At this time, the cleaning device needs to stop running and upload abnormal information to the server, and the staff remotely monitors the status of the cleaning device. The angle threshold is generally about 15°, and the timing threshold is several seconds. If the forward offset angle is adjusted back to the angle threshold within the timing threshold, the timing process stops and the timing time is reset.
[0030] In the above process, when it is determined that the power level of the cleaning device is less than the power threshold, the image around the cleaning device is acquired, the position of the charging device in the image around the cleaning device is identified, the distance and path between the cleaning device and the charging device are calculated based on the position of the charging device, and the cleaning device is controlled to move toward the charging device. The cleaning device is constantly moving during the cleaning process and may be located at any position on the photovoltaic module. Therefore, when the power level is too low and needs to be charged, it is necessary to acquire the image around the cleaning device to find the position of the charging device. The charging device is provided with an identification QR code or a special identification mark. The position of the charging device in the image around the cleaning device is determined by identifying the QR code or the special identification mark. The vertical offset distance and the lateral offset distance between the cleaning device and the charging device are calculated based on the position of the charging device in the image around the cleaning device. The cleaning device is controlled to move toward the charging device and dock with the charging device to complete the automatic charging of the cleaning device.
[0031] The embodiments of this specification also provide a photovoltaic power station operation and maintenance system capable of realizing intelligent navigation, including photovoltaic modules and cleaning equipment. The cleaning equipment is provided with a shooting device, a processor and a walking device. The shooting device is used to shoot images of photovoltaic modules around the cleaning equipment. The processor is used to control the movement of the cleaning equipment on the photovoltaic modules based on the above-mentioned photovoltaic power station operation and maintenance method capable of realizing intelligent navigation. The shooting device includes a plurality of cameras, which can at least obtain images in four directions, namely, front, back, left, and right, of the cleaning equipment; at least two cameras are provided in the forward direction of the cleaning equipment, and the two cameras respectively shoot images of photovoltaic modules in front of and behind the cleaning equipment; the processor respectively identifies respective reference lines based on the images of photovoltaic modules in front of and behind the cleaning equipment, and calculates the forward offset angle, and the forward offset angles in front of and behind the cleaning equipment can be mutually verified; when the staff remotely monitors the cleaning equipment, they also observe the situation around the cleaning equipment through the shooting device. The processor can also identify stains and obstacles in the image based on the photovoltaic module image. When an obstacle is identified, the cleaning device is controlled to turn to avoid the obstacle. During the obstacle avoidance process, the fitting of the reference line and the calculation of the deflection angle are stopped. When stains are identified, the cleaning device is controlled to move back and forth repeatedly on the stains to clean. Several marks are provided on the end edge of the photovoltaic module. The marks contain steering information that can be recognized by the cleaning device. The distance between the several marks should be such that the number of marks displayed in the photovoltaic module image does not exceed 3, so as to avoid identifying non-corresponding marks, which causes the processor to control the cleaning device to perform an incorrect steering action.
[0032] Although several specific implementation details are included in the above discussion, these should not be interpreted as limiting the scope of this specification. Certain features described in the context of separate embodiments may also be implemented in a single implementation in combination. Conversely, various features described in the context of a single implementation may also be implemented in multiple implementations individually or in any suitable sub-combination.
[0033] Although the subject matter has been described in language specific to structural features and / or methodological logical actions, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. On the contrary, the specific features and actions described above are merely example forms of implementing the claims.
[0034] The embodiments of the present specification have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or technical improvements in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. A photovoltaic power station operation and maintenance method capable of realizing intelligent navigation, the method is implemented based on a processor, and is characterized in that: include: Acquire an image of the photovoltaic module in the direction of the cleaning device; Processing and analyzing the photovoltaic module image to obtain a vertical edge of the photovoltaic module, and using it as a reference line; wherein the vertical edge of the photovoltaic module is an edge of a photovoltaic panel on the photovoltaic module parallel to a preset forward direction of the cleaning device; Calculating the forward offset angle of the cleaning device based on the position of the reference line in a plane rectangular coordinate system; wherein the plane rectangular coordinate system takes the bottom edge of the photovoltaic component image as the x-axis; The advancing direction of the cleaning device is adjusted based on the advancing offset angle until the advancing offset angle is less than an offset threshold.
2. The method according to claim 1, characterized in that Processing and analyzing the photovoltaic module image to obtain the vertical edge of the photovoltaic module and using it as a reference line includes: Processing the photovoltaic module image by an edge detection algorithm to obtain an edge image; Extracting lines in the edge image using a line extraction function; wherein the lines in the edge image include vertical edges of photovoltaic modules; The lines outside the preset slope range are eliminated, and the reference line is obtained by fitting the remaining lines using the least square method.
3. The method according to claim 2, characterized in that After obtaining the edge image, the method further includes: The lateral edge of the photovoltaic component in the edge image is hidden by a mask image shielding matrix; wherein the lateral edge is an edge of a photovoltaic panel on the photovoltaic component perpendicular to a preset forward direction of the cleaning device.
4. The method according to claim 1, characterized in that: The forward direction of the cleaning device is adjusted based on the forward offset angle until the forward offset angle is less than an offset threshold, and also includes: starting timing when the forward offset angle exceeds the angle threshold, if the forward offset angle still exceeds the angle threshold when the timing time exceeds the timing threshold, the cleaning device stops running and uploads abnormal information to the server.
5. The method according to claim 4, characterized in that Also includes: When the length of the reference line, the coordinates of the midpoint of the reference line, and the forward offset angle are not within the standard range, it is determined that the reference line recognition is abnormal, and the reference line is refitted.
6. The method according to claim 4, characterized in that Also includes: When the photovoltaic component image is abnormal, the cleaning device stops moving forward and searches to see whether there is an identification in the image in front of the cleaning device at this time. If so, the cleaning device is controlled to turn based on the identification, otherwise the abnormal information is uploaded to the server.
7. The method according to claim 6, characterized in that Also includes: When it is determined that the power level of the cleaning device is less than a power threshold, an image around the cleaning device is obtained, the position of the charging device in the image around the cleaning device is identified, the distance and path between the cleaning device and the charging device are calculated based on the position of the charging device, and the cleaning device is controlled to move toward the charging device.
8. The method according to claim 1, characterized in that Also includes: The photovoltaic component image is continuously acquired at a certain time interval. When the temperature of the computing module of the processor is too high, the computing module performs frequency reduction protection and prolongs the time interval.
9. A photovoltaic power station operation and maintenance system capable of realizing intelligent navigation, characterized in that: It comprises a photovoltaic component and a cleaning device, wherein the cleaning device is provided with a shooting device, a processor and a walking device, the shooting device is used to shoot images of the photovoltaic components around the cleaning device, and the processor is used to control the movement of the cleaning device on the photovoltaic component based on the method described in claim 1.
10. The system according to claim 9, characterized in that A plurality of marks are arranged on the end edge of the photovoltaic assembly, and the marks contain the turning information which can be identified by the cleaning device.