An intelligent inspection method and inspection robot based on photovoltaic power station

CN118713308BActive Publication Date: 2024-11-05QILU INST OF TECH
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202411194832.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-29
Publication Date
2024-11-05
Estimated Expiration
2044-08-29

AI Technical Summary

Technical Problem

Photovoltaic power stations are prone to damage in harsh environments. Traditional manual inspections are low efficiency, high cost and inaccurate data, which affects inspection efficiency and is complex in maintenance and management, which poses safety risks.

Method used

The intelligent inspection method based on photovoltaic power stations is adopted, and environmental parameters and equipment status images are obtained through the data acquisition module and visual inspection module, and analyzed and processed to realize automatic early warning and cleaning, and combined with the intelligent inspection robot for automatic cleaning and navigation.

Benefits of technology

It improves patrol efficiency and accuracy, reduces manpower investment, realizes unmanned and refined management of photovoltaic power stations, and significantly improves the economic and safe operation level of the power station.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118713308B_ABST
    Figure CN118713308B_ABST
Patent Text Reader

Abstract

The present application discloses an intelligent inspection method and inspection robot based on photovoltaic power stations, which relates to the field of electronic computer information technology. The intelligent inspection method includes: autonomous inspection according to the task operation instructions or pre-set tasks of the operator in the monitoring center; obtaining the environmental parameter data of the photovoltaic power station and the operating status image of the photovoltaic equipment through the data acquisition module and the visual detection module respectively; transmitting the environmental parameter data and the operating status image to the monitoring center for analysis and processing; and automatic warning or automatic cleaning according to the abnormal results after the analysis and processing. During the inspection process, automatic navigation and path planning are realized through fuzzy PID-SLAM technology, which effectively improves the inspection efficiency and accuracy. At the same time, the inspection robot can start the automatic cleaning function when it detects that the dust coverage rate exceeds the preset threshold, and timely maintain the performance of the photovoltaic components, so that the photovoltaic power station can achieve unmanned and refined management.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of electronic computer information technology, and specifically to an intelligent inspection method and an inspection robot based on a photovoltaic power station. Background Art

[0002] As global environmental issues become increasingly severe, the demand for clean and renewable energy becomes more urgent. As a representative of green and renewable energy, solar energy can be effectively obtained through photovoltaic power stations. With the maturity of photovoltaic technology, the photovoltaic industry is booming. As the core of photovoltaic power generation systems, photovoltaic power stations are crucial to the stable operation of power systems and energy management.

[0003] Centralized photovoltaic power stations are usually built in deserts and areas rich in solar energy resources. Faced with harsh natural environments such as strong winds and sand, high-intensity light, and large temperature differences between day and night, their equipment is easily damaged, and frequent equipment failures pose a challenge to the operation and economy of the power station. In particular, foreign matter on the surface of photovoltaic modules has a significant impact on their output power and service life. The traditional manual inspection method is inefficient, costly, and the data is not accurate enough. It cannot handle foreign matter in a timely and rapid manner, affecting the inspection efficiency. In addition, the large area and number of equipment in photovoltaic power stations increase the complexity of maintenance and management. High costs and safety risks have also become important factors restricting the safe and stable operation of power stations. Summary of the invention

[0004] In order to solve the above technical problems, this application proposes the following technical solutions:

[0005] In a first aspect, an embodiment of the present application provides an intelligent inspection method based on a photovoltaic power station, comprising:

[0006] Conduct autonomous inspections according to the operator's task instructions or pre-set tasks in the monitoring center;

[0007] The environmental parameter data of the photovoltaic power station and the operating status image of the photovoltaic equipment are obtained through the data acquisition module and the visual detection module respectively;

[0008] Transmitting the environmental parameter data and operating status images to a monitoring center for analysis and processing;

[0009] Automatic warning or automatic cleaning is performed according to the abnormal results after the analysis and processing.

[0010] In a possible implementation, the step of acquiring environmental parameter data of a photovoltaic power station and an operating status image of a photovoltaic device by means of a data acquisition module and a visual detection module respectively includes:

[0011] The data acquisition module is arranged outside the photovoltaic power station to collect the temperature, humidity and solar radiation intensity of the photovoltaic power station;

[0012] High-definition cameras are used to collect dust images of photovoltaic panels, exposed wires and instrument panels in photovoltaic power plants.

[0013] In a possible implementation, transmitting the environmental parameter data to a monitoring center for analysis and processing includes:

[0014] After filtering, denoising and calibrating the environmental parameter data, the temperature distribution is analyzed to identify abnormal temperature areas;

[0015] By comparing the temperature difference and change rate parameters of different abnormal temperature areas, it is possible to determine whether there is local overheating or excessive temperature difference.

[0016] In a possible implementation, transmitting the operating status image to a monitoring center for analysis and processing includes:

[0017] Performing grayscale conversion on the running status image to increase the dynamic range of the image;

[0018] The noise in the image is removed by adjusting the contrast, and the edge is enhanced to detect and measure the required target in the image;

[0019] The adjusted image is segmented using an image segmentation model based on an edge information extraction module of octave convolution;

[0020] The dust region of the solar photovoltaic panel in the segmented image is marked and the area of ​​the dust region is calculated.

[0021] In a possible implementation manner, marking the dust region of the solar photovoltaic panel in the segmented image and calculating the area of ​​the dust region includes:

[0022] The threshold cutting method is used to screen the pixel area belonging to the dust part;

[0023] After converting the dust image of the photovoltaic panel into a binary image, a closing operation is performed to eliminate noise, fill holes in the image, and make the edge of the target part smooth;

[0024] After performing the image region closing operation, background information irrelevant to the target part is eliminated to obtain the required binary image of the target part;

[0025] The area occupied by dust on each solar photovoltaic panel is automatically calculated using the formula embedded in the algorithm.

[0026] In a possible implementation, automatic warning or automatic cleaning is performed according to the abnormal result after the analysis and processing, including:

[0027] When it is determined that the area occupied by the dust exceeds 1 / 4 of the overall area of ​​the photovoltaic panel, controlling the intelligent inspection robot to clean the photovoltaic panel;

[0028] When an abnormal temperature area is detected or exposed wires or abnormal dial data are identified, the intelligent inspection robot is controlled to trigger an alarm mechanism.

[0029] In a possible implementation, when it is determined that the area occupied by the dust exceeds 1 / 4 of the total area of ​​the photovoltaic panel, controlling the intelligent inspection robot to clean the photovoltaic panel includes:

[0030] After determining the target photovoltaic panel to be cleaned, obtaining path environment data to reach the target photovoltaic panel;

[0031] Determine the current position of the inspection robot and the position of the target photovoltaic panel, and use the A* path search algorithm combined with the path environment data to obtain the optimal path;

[0032] When the inspection robot is running along the optimal path, a visually assisted navigation method is used to perform deviation correction control on the robot;

[0033] When the target photovoltaic panel is reached, a cleaning area is determined and cleaning is performed.

[0034] In a possible implementation, when the inspection robot is running along the optimal path, a visually assisted navigation method is used to perform deviation correction control on the robot, including:

[0035] The road image acquired by the high-definition camera is calibrated and then subjected to distortion correction;

[0036] The detection range is narrowed by cropping the ROI area, and the lane lines with high saturation are separated by converting to the HSV color space;

[0037] The road image noise is removed by image filtering, and the lane line coordinates are obtained by using the Canny operator and Hough line detection to draw the lane area;

[0038] Then, the bird's-eye view of the road image is obtained through inverse perspective transformation, and the coordinates of the corner points in the lane area are obtained to calculate the navigation lateral deviation and directional deviation;

[0039] Finally, the fuzzy PID controller is imported to perform correction simulation for different initial deviation conditions.

[0040] In a second aspect, an embodiment of the present application provides an intelligent inspection robot based on a photovoltaic power station, comprising:

[0041] An inspection robot body, and an intelligent inspection device, a communication module and a monitoring center electrically connected to the inspection robot body, wherein the intelligent inspection device includes a data acquisition module and a visual detection module, and the environmental parameter data obtained by the data acquisition module and the operating status image of the photovoltaic equipment obtained by the visual detection module are transmitted to the monitoring center through the communication module. The monitoring center executes the method described in any possible implementation method to realize intelligent inspection of the photovoltaic power station and performs abnormal processing based on the inspection results.

[0042] Compared with the prior art, the beneficial effects of this application are:

[0043] This application uses the robot's intelligent inspection system, combined with high-definition cameras, multiple sensors and advanced visual detection algorithms, to achieve accurate detection and positioning of dust accumulation on the surface of photovoltaic modules. The inspection data is transmitted to the monitoring center through the communication module, and automatic navigation and path planning are achieved through fuzzy PID-SLAM technology, which effectively improves the inspection efficiency and accuracy. At the same time, the robot can start the automatic cleaning function when it detects that the dust coverage rate exceeds the preset threshold, and timely maintain the performance of photovoltaic modules.

[0044] The application of intelligent inspection robots has greatly reduced manpower input, enabling unmanned and refined management of photovoltaic power stations, significantly improving the economic and safe operation level of the power stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 A schematic diagram of a process flow of an intelligent inspection method based on a photovoltaic power station provided in an embodiment of the present application;

[0046] Figure 2 The inspection workflow diagram provided for the embodiment of the present application;

[0047] Figure 3 A detection flow chart provided for an embodiment of the present application;

[0048] Figure 4 A structural diagram of an image segmentation model provided in an embodiment of the present application;

[0049] Figure 5 A schematic diagram of the structure of an edge information extraction module is provided for an embodiment of the present application;

[0050] Figure 6 Provide a flow chart of automatic navigation of an intelligent inspection robot for the embodiment of the present application;

[0051] Figure 7 A schematic diagram of the system structure of an intelligent inspection robot based on a photovoltaic power station provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The present solution is described below in conjunction with the accompanying drawings and specific implementation methods.

[0053] Figure 1 A schematic diagram of a process flow of an intelligent inspection method based on a photovoltaic power station provided in an embodiment of the present application is shown in FIG. Figure 1 In this embodiment, an intelligent inspection method based on a photovoltaic power station includes:

[0054] S101, performing autonomous inspection according to the task operation instruction of the operator in the monitoring center or the pre-set task.

[0055] In this embodiment, the photovoltaic power station can be unmanned and can carry out inspection work independently according to pre-set tasks.

[0056] S102, obtaining environmental parameter data of the photovoltaic power station and operating status images of the photovoltaic equipment through a data acquisition module and a visual detection module respectively.

[0057] In this embodiment, the data acquisition module is set outside the photovoltaic power station to collect environmental parameters of the photovoltaic power station, such as temperature, humidity, solar radiation intensity and other data, to perceive environmental anomalies and severe weather conditions in real time.

[0058] See also Figure 2 and Figure 3 In order to effectively calculate the dust percentage, a high-resolution HD camera with the ability to provide clear images and accurately capture in different light environments is used. The camera is used to collect images of the operating status of the photovoltaic equipment. The camera parameters are adjusted to an appropriate state in advance, and a fixed height is set to ensure that the captured images are clear, so as to collect the original images of the photovoltaic panels. For example, the dust image of the photovoltaic panel, the image of the exposed wires, and the image of the instrument panel, etc. When an abnormality is identified, an alarm signal is issued to remind the staff to grasp the actual solar energy resource status.

[0059] S103, transmitting the environmental parameter data and the operating status image to the monitoring center for analysis and processing.

[0060] The monitoring center will not only save the inspection information, but also analyze and process the data to extract valuable information. Comprehensive analysis of historical inspection information can help evaluate the operating status of the photovoltaic power station over a period of time, and also provide a reliable basis for professionals to fully understand the operating status of the equipment. The monitoring center has the function of receiving abnormal alarms from the inspection robot, which can promptly remind the staff to solve the fault problem.

[0061] In this embodiment, the acquired environmental parameter data is pre-processed by filtering, denoising, and calibration, and then the temperature distribution is analyzed to identify abnormal temperature areas. By comparing the temperature difference, change rate and other parameters of different parts, it can be determined whether there is local overheating, excessive temperature difference and other conditions. These analysis results will provide important reference for the decision-making of inspection personnel.

[0062] The collected running status image is then transformed into grayscale. As a form of image enhancement, grayscale transformation can increase the dynamic range of the image, expand the contrast of the pigment, make the image clearer, highlight the main features of the dust, and the storage space required for the gray image is small, thereby increasing the image calculation speed. By adjusting the contrast, the original image that is blurred or even indistinguishable after grayscale is processed into a clear image rich in useful information, effectively removing the noise in the image, and enhancing the edges or other required areas in the image, so that it is easier to detect and measure the required targets in the image. After preprocessing operations, the target and the background are segmented. Among them, image segmentation is applicable to images where the target and background occupy different grayscale ranges. It can not only reduce the amount of data, but also simplify the analysis and processing process. In this embodiment, an image segmentation model based on an edge information extraction module of octave convolution is used to segment the adjusted image, see. Figure 4 and Figure 5 , image feature information is divided into high-frequency information and low-frequency information. Octave convolution is a lightweight convolution, which can pay more attention to the feature expression of high-frequency edge information without adding a large number of parameters, reduce the proportion of low-frequency information, and suppress the feature expression of image redundant information. The edge information extraction module constructed in this embodiment is composed of 1×1 convolution and octave convolution. From the channel dimension, the channel is first reduced to 1 / 4, and then the information is extracted by octave convolution and then upgraded to the original value.

[0063] The dust area of ​​the solar photovoltaic panel in the segmented image is marked and the area of ​​the dust area is calculated. In this embodiment, the pixel area belonging to the dust part is screened out by means of threshold cutting (based on Gray, RGB, and HSV values ​​respectively), and the dust image of the photovoltaic panel is converted into a binary image (represented by 0-1 pixels) and then a closing operation is performed, that is, a combination of expansion and corrosion operations to eliminate noise, fill some small holes in the image, and make the edge of the target part smooth. After the image area closing operation is performed, the background information irrelevant to the target part is eliminated to obtain the final required binary image of the target part, and the area occupied by the dust of each solar photovoltaic panel is automatically calculated using the formula embedded in the algorithm.

[0064] S104, automatic warning or automatic cleaning is performed according to the abnormal results after analysis and processing.

[0065] In this embodiment, the monitoring center stores and performs corresponding data analysis and processing for the data collected by the inspection robot. According to the results of data processing and analysis, once abnormal temperature areas, exposed wires, abnormal dial data, etc. are detected, the intelligent inspection system will immediately trigger the alarm mechanism. The purpose of the alarm function is to promptly warn relevant personnel to deal with abnormal conditions to prevent more serious losses. The alarm method covers a variety of forms such as sound, light, and electricity, which can be selected according to actual needs. At the same time, the alarm information will be recorded in the system, including the location and time of the abnormal temperature, which is convenient for subsequent tracing and investigation.

[0066] When it is determined based on the results of data processing and analysis that the area occupied by the dust exceeds 1 / 4 of the overall area of ​​the photovoltaic panel, the intelligent inspection robot is controlled to clean the photovoltaic panel.

[0067] Among them, based on SLAM technology, a safe and efficient path can be explored according to the current position and target position of the robot. In order to improve the problems of inaccurate positioning and poor path control stability of SLAM, this embodiment adopts visual assisted navigation to perform deviation correction control. Based on the deviation correction control method of fuzzy PID, the incremental PID control method has better dynamic performance, and the PID output at the next moment is obtained by accumulation, and the cumulative error is small. Based on the improved PID-SLAM algorithm, the robot can identify obstacles in time during the movement of the predetermined path, adjust the travel path, and then achieve obstacle avoidance and obstacle crossing, ensuring the flexibility of path movement.

[0068] See also Figure 6 In this embodiment, after first determining the target photovoltaic panel to be cleaned, the path environmental data to the target photovoltaic panel is obtained by the laser radar sensor, and the laser odometer positioning method is used to match the sensor data with the map to obtain the accurate position of the robot. Once the current position of the inspection robot and the position of the target photovoltaic panel are determined, the A* path search algorithm can be used to find a path. The algorithm will consider the obstacles and environmental information in the map to find the shortest or optimal path. For the problem of inaccurate SLAM positioning in complex road environments, the robot is corrected by visual assisted navigation. The road image is obtained by a high-definition camera, and the distortion is corrected after camera calibration. The detection range is reduced by ROI area cropping, and the lane line with high saturation is separated by converting to HSV color space. The road image noise is removed by image filtering, and the lane line coordinates are obtained by Canny operator and Hough line detection, and the lane area is drawn. Then the bird's-eye view of the road image is obtained by inverse perspective transformation, and the corner coordinates of the lane area are obtained to calculate the navigation lateral deviation and directional deviation. Finally, the fuzzy PID controller is introduced to perform correction simulation for different initial deviation conditions. When the target photovoltaic panel is reached, a cleaning area is determined and cleaning is performed.

[0069] Corresponding to the intelligent inspection method based on a photovoltaic power station provided in the above embodiment, the present application also provides an embodiment of an intelligent inspection robot based on a photovoltaic power station.

[0070] See also Figure 7 The intelligent inspection robot based on the photovoltaic power station in the embodiment of the present application includes: an inspection robot body, and an intelligent inspection device, a communication module and a monitoring center electrically connected to the inspection robot body. The inspection robot system can collect solar radiation and meteorological information data and electronic image information in real time, and use the 5G communication module to complete the reception and transmission of data information with high speed and low latency, thereby forming a network information network between the photovoltaic power station inspection robot carrier and the monitoring center, so that the information collected by each sensor is accurately transmitted to the monitoring center, achieving the effect of accurate monitoring and efficient transmission, and also providing data support and guarantee for its subsequent scientific research.

[0071] The monitoring center can timely monitor the status of running equipment, conduct solar energy resource analysis, photoelectric conversion efficiency evaluation, photovoltaic power station comprehensive efficiency evaluation, etc., implement target warnings, and automatically generate inspection reports. In addition, it can automatically save real-time data to facilitate later historical data query, GIS information query and scientific research. When abnormal conditions such as equipment failure, power undervoltage, exposed wires, etc. occur, an alarm signal will be issued to inform the staff of the specific location.

[0072] In addition, the inspection robot is powered by lithium-ion batteries and equipped with a thermal imaging dual-spectrum camera, which has the ability to collect both infrared and visible light information. The robot's inspection methods include fixed-point timed inspection and remote control inspection. Among them, fixed-point timed inspection is to pre-set the time, path, and target for automatic inspection, and remote control inspection is for the operator to control the remote control intelligent inspection robot through electronic equipment to complete the robot's path movement and real-time data collection.

[0073] It should be noted that, in this article, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0074] The above is only a specific implementation of the present application. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. The protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An intelligent inspection method based on a photovoltaic power station, characterized in that: include: Conduct autonomous inspections according to the operator's task instructions or pre-set tasks in the monitoring center; The environmental parameter data of the photovoltaic power station and the operating status image of the photovoltaic equipment are obtained through the data acquisition module and the visual detection module respectively; Transmitting the environmental parameter data and operating status images to a monitoring center for analysis and processing, including: Performing grayscale conversion on the running status image to increase the dynamic range of the image; The noise in the image is removed by adjusting the contrast, and the edge is enhanced to detect and measure the required target in the image; The adjusted image is segmented using an image segmentation model based on an edge information extraction module of octave convolution. The edge information extraction module is composed of 1×1 convolution and octave convolution. The channel dimension is reduced to 1 / 4, and then the information is extracted by octave convolution and then the dimension is increased to the original value. Marking the dust region of the solar photovoltaic panel in the segmented image and calculating the area of ​​the dust region; Marking the dust region of the solar photovoltaic panel in the segmented image and calculating the area of ​​the dust region includes: The threshold cutting method is used to screen the pixel area belonging to the dust part; After converting the dust image of the photovoltaic panel into a binary image, a closing operation is performed to eliminate noise, fill holes in the image, and make the edge of the target part smooth; After performing the image region closing operation, background information irrelevant to the target part is eliminated to obtain the required binary image of the target part; The area occupied by dust on each solar photovoltaic panel is automatically calculated using the formula embedded in the algorithm; Automatically warning or automatically cleaning is performed according to the abnormal results after the analysis and processing, including: When it is determined that the area occupied by the dust exceeds 1 / 4 of the total area of ​​the photovoltaic panel, the intelligent inspection robot is controlled to clean the photovoltaic panel; When an abnormal temperature area is detected or exposed wires or abnormal dial data are identified, the intelligent inspection robot is controlled to trigger an alarm mechanism.

2. The intelligent inspection method based on photovoltaic power station according to claim 1 is characterized in that: The method of obtaining environmental parameter data of the photovoltaic power station and operating status images of the photovoltaic equipment through the data acquisition module and the visual detection module respectively includes: The data acquisition module is arranged outside the photovoltaic power station to collect the temperature, humidity and solar radiation intensity of the photovoltaic power station; High-definition cameras are used to collect dust images of photovoltaic panels, exposed wires and instrument panels in photovoltaic power plants.

3. The intelligent inspection method based on photovoltaic power station according to claim 1 is characterized in that: Transmitting the environmental parameter data to a monitoring center for analysis and processing, including: After filtering, denoising and calibrating the environmental parameter data, the temperature distribution is analyzed to identify abnormal temperature areas; By comparing the temperature difference and change rate parameters of different abnormal temperature areas, it is possible to determine whether there is local overheating or excessive temperature difference.

4. The intelligent inspection method based on photovoltaic power station according to claim 1 is characterized in that: When it is determined that the area occupied by the dust exceeds 1 / 4 of the overall area of ​​the photovoltaic panel, controlling the intelligent inspection robot to clean the photovoltaic panel includes: After determining the target photovoltaic panel to be cleaned, obtaining path environment data to reach the target photovoltaic panel; Determine the current position of the inspection robot and the position of the target photovoltaic panel, and use the A* path search algorithm combined with the path environment data to obtain the optimal path; When the inspection robot is running along the optimal path, a visually assisted navigation method is used to perform deviation correction control on the robot; When the target photovoltaic panel is reached, a cleaning area is determined and cleaning is performed.

5. The intelligent inspection method based on photovoltaic power station according to claim 4 is characterized in that: When the inspection robot is running along the optimal path, a visual-assisted navigation method is used to perform deviation correction control on the robot, including: The road image acquired by the high-definition camera is calibrated and then subjected to distortion correction; The detection range is narrowed by cropping the ROI area, and the lane lines with high saturation are separated by converting to the HSV color space; The road image noise is removed by image filtering, and the lane line coordinates are obtained by using the Canny operator and Hough line detection to draw the lane area; Then, the bird's-eye view of the road image is obtained through inverse perspective transformation, and the coordinates of the corner points in the lane area are obtained to calculate the navigation lateral deviation and directional deviation; Finally, the fuzzy PID controller is imported to perform correction simulation for different initial deviation conditions.

6. An intelligent inspection robot based on a photovoltaic power station, characterized in that: include: A patrol robot body, and an intelligent patrol device, a communication module and a monitoring center electrically connected to the patrol robot body, wherein the intelligent patrol device includes a data acquisition module and a visual detection module, and the environmental parameter data obtained by the data acquisition module and the operating status image of the photovoltaic equipment obtained by the visual detection module are transmitted to the monitoring center through the communication module. The monitoring center executes the method described in any one of claims 1 to 5 to realize intelligent patrol of the photovoltaic power station and performs abnormal processing based on the patrol results.

Citation Information

Patent Citations

  • Transformation method of picture grey scale for use in the production of copying picture without bias colour

    CN1050934A

  • Photovoltaic module state detection method and device, storage medium and electronic equipment

    CN114648708A

  • Comprehensive inspection fault elimination method and system for unattended photovoltaic power station

    CN116388682A