CN-116773598-B - Digital method for automatically inspecting and positioning defects of photovoltaic panel by unmanned aerial vehicle

CN116773598BCN 116773598 BCN116773598 BCN 116773598BCN-116773598-B

Abstract

The invention provides a digital method for automatically inspecting and positioning defects of a photovoltaic panel by an unmanned aerial vehicle. The method comprises the steps of determining shooting points of unmanned aerial vehicle inspection and geographic position indexes of each photovoltaic panel in a shooting area where the shooting points are located on a preset photovoltaic digital map, obtaining unmanned aerial vehicle inspection infrared images according to the shooting points, processing the unmanned aerial vehicle inspection infrared images according to a target defect detection algorithm and an infrared relative temperature difference method, determining photovoltaic area states and infrared thermal defect types in the shooting area corresponding to the shooting points, carrying out image segmentation processing on the unmanned aerial vehicle inspection infrared images, determining mapping from image pixel coordinates to geographic position longitude and latitude coordinates of each photovoltaic panel, and mapping the geographic position longitude and latitude coordinates of each photovoltaic panel to the preset photovoltaic digital map to achieve marking of the operation states of each photovoltaic panel on the photovoltaic digital map. Through the technical scheme, the digital management of the photovoltaic power station is realized, and the photovoltaic panel with the defect is accurately positioned.

Inventors

HAN JUN
MENG XIANLI
WAN YONG
LIU ZHE

Assignees

上海深邃智能科技有限公司

Dates

Publication Date: 20260505
Application Date: 20230523

Claims (9)

1. The digital method for automatically inspecting and positioning the defects of the photovoltaic panel by using the unmanned aerial vehicle is characterized by comprising the following steps of: determining shooting points of unmanned aerial vehicle inspection and geographic position indexes of each photovoltaic panel in a shooting area where the shooting points are located according to the preset unmanned aerial vehicle inspection height on a preset photovoltaic digital map; Acquiring an unmanned aerial vehicle inspection infrared image according to the shooting point of the unmanned aerial vehicle inspection; processing the inspection infrared image of the unmanned aerial vehicle according to a target defect detection algorithm and an infrared relative temperature difference method, and determining the state of a photovoltaic area and the type of infrared thermal defects in a shooting area corresponding to the shooting point; Performing image segmentation processing on the unmanned aerial vehicle inspection infrared image, and determining mapping from image pixel coordinates of each photovoltaic panel to longitude and latitude coordinates of a geographic position; Mapping the longitude and latitude coordinates of the geographic position of each photovoltaic panel to the preset photovoltaic digital map, positioning the photovoltaic panel with the infrared thermal defect, determining the running state of each photovoltaic panel, and further marking the running state of each photovoltaic panel on the photovoltaic digital map to digitally manage each photovoltaic panel; the method for determining the geographical position index of each photovoltaic panel in the shooting area where the shooting point is located and the shooting point of unmanned aerial vehicle inspection is determined according to the preset unmanned aerial vehicle inspection height comprises the following steps: continuously shooting images of a target photovoltaic area through an unmanned aerial vehicle, determining an orthographic image of the target photovoltaic area, adopting a map making tool, making a photovoltaic digital map of the target photovoltaic area according to the orthographic image, and establishing a geographic position index for each photovoltaic panel in the photovoltaic digital map; determining a shooting area of the unmanned aerial vehicle under orthographic projection according to the preset unmanned aerial vehicle inspection height; according to the length, the width and the inclination angle of the photovoltaic panel, determining row intervals and column intervals between adjacent shooting points in the inspection of the unmanned aerial vehicle; traversing the photovoltaic digital map according to the row interval and the column interval, and determining longitude and latitude coordinates of all shooting points of the unmanned aerial vehicle inspection; And determining a geographic position index corresponding to each photovoltaic panel in a shooting area where the shooting point is located according to all shooting points of the unmanned aerial vehicle inspection and the photovoltaic digital map.
2. The method according to claim 1, wherein the determining a shooting area of the drone under the orthographic projection according to the preset drone patrol height includes: ; ; Wherein F represents the focal length of the camera, Representing the length of the camera sensing area pixels, The width of the camera sensing image element is represented, H represents the preset unmanned aerial vehicle inspection height, Indicating the length of the photographing region, Representing the width of the photographing region.
3. The method of claim 1, wherein determining row and column spacing between adjacent shots of the drone based on the length, width, and tilt angle of the photovoltaic panel comprises: = ; ; ; ; Wherein, the Representing the length of the photovoltaic panel, Representing the width of the photovoltaic panel, Indicating the inclination angle of the photovoltaic panel installation, Representing the oblique projected length of the photovoltaic panel, Representing the channel spacing from row to row in the photovoltaic panel array, Representing the oblique projected length of the photovoltaic panel forming a repeating row spacing with the channel spacing, Indicating the overlapping ratio of photographing the target photovoltaic region, Representing a line interval between the adjacent photographing points; Representing the column spacing between the adjacent shots.
4. The method according to claim 1, wherein the processing the unmanned aerial vehicle inspection infrared image according to the target defect detection algorithm and the infrared relative temperature difference method, and determining the state of the photovoltaic area and the type of the infrared thermal defect in the shooting area corresponding to the shooting point, comprises: Acquiring unmanned aerial vehicle inspection training images with various heights, establishing a label according to the type of the infrared thermal defect, performing model training on the target defect detection algorithm, and determining an infrared photovoltaic defect prediction model, wherein the type of the infrared thermal defect comprises a blocky hot spot, a linear hot spot and a short circuit; Inputting the inspection infrared image of the unmanned aerial vehicle into the infrared photovoltaic defect prediction model, and determining a predicted photovoltaic defect area and an infrared thermal defect type corresponding to the predicted photovoltaic defect area; Detecting the predicted photovoltaic defect area according to the infrared relative temperature difference method, and determining the state of the predicted photovoltaic defect area, wherein the state of the predicted photovoltaic defect area comprises a photovoltaic defect state, a photovoltaic early warning state and a photovoltaic normal state.
5. The method of claim 4, wherein detecting the predicted photovoltaic defect region from the infrared relative temperature differential method, determining a status of the predicted photovoltaic defect region, comprises: dividing the predicted photovoltaic defect area according to the OTSU of the discipline method, and determining an inner area of the binarization temperature area and an outer area of the binarization temperature area; Determining the highest temperature of the inner area of the binarization temperature area, the average temperature of the inner area of the binarization temperature area and the average temperature of the outer area of the binarization temperature area; if the highest temperature of the inner area of the binarized temperature area is larger than a first preset temperature and the temperature difference is larger than a second preset temperature, determining that the state of the predicted photovoltaic defect area is a photovoltaic defect state; if the temperature difference is larger than a third preset temperature and smaller than a second preset temperature, determining that the state of the predicted photovoltaic defect area is a photovoltaic early warning state; and if the temperature difference is smaller than a third preset temperature, determining that the state of the predicted photovoltaic defect area is a photovoltaic normal state.
6. The method of claim 1, wherein the image pixel coordinates of the photovoltaic panel comprise image center point coordinates of the photovoltaic panel and image position latitude and longitude coordinates of the photovoltaic panel; the image segmentation processing is carried out on the unmanned aerial vehicle inspection infrared image, and the mapping from the image pixel coordinates of each photovoltaic panel to the longitude and latitude coordinates of the geographic position is determined, which comprises the following steps: Performing image segmentation processing on the unmanned aerial vehicle inspection infrared image, determining the area where each photovoltaic panel is located, extracting the outline of the area where each photovoltaic panel is located, and determining the coordinates of the image center point of each photovoltaic panel according to the outline of the area where each photovoltaic panel is located; and determining the longitude and latitude coordinates of the geographic position of each photovoltaic panel according to the coordinates of the image center point of each photovoltaic panel and the longitude and latitude coordinates of the image position of each photovoltaic panel carried on the unmanned aerial vehicle inspection infrared image.
7. The method of claim 6, wherein determining the geographic location longitude and latitude coordinates of each photovoltaic panel based on the image center point coordinates of each photovoltaic panel and the image location longitude and latitude coordinates of each photovoltaic panel carried on the unmanned aerial vehicle inspection infrared image comprises: ; ; ; ; Wherein, the Representing the pixel width of the drone inspection infrared image, The pixel height of the inspection infrared image of the unmanned aerial vehicle is represented, Longitude coefficients representing the longitude and latitude coordinates of the geographic location, Latitude coefficients representing latitude and longitude coordinates of a geographic location, Longitude representing latitude and longitude coordinates of the image position of the photovoltaic panel, Latitude representing latitude and longitude coordinates of the image position of the photovoltaic panel, The abscissa representing the center point of the photovoltaic panel, Representing the ordinate of the center point of the photovoltaic panel, Longitude representing the latitude and longitude coordinates of the geographic location of the photovoltaic panel, And the latitude of the longitude and latitude coordinates of the geographic position of the photovoltaic panel is represented.
8. The method of claim 1, the determining an operational status of each photovoltaic panel comprising: If the overlapping area of the photovoltaic area in the shooting area corresponding to the shooting point and each photovoltaic panel in the unmanned aerial vehicle inspection infrared image after the image segmentation processing is larger than a preset area, the state of the photovoltaic area of the overlapping area represents the state of the photovoltaic panel in the overlapping area, and the infrared thermal defect type of the overlapping area represents the infrared thermal defect type of the photovoltaic panel in the overlapping area.
9. The method according to claim 1, wherein mapping the longitude and latitude coordinates of the geographic location of each photovoltaic panel to the preset photovoltaic digital map, and positioning the photovoltaic panels with infrared thermal defects, determining the operation state of each photovoltaic panel, and further marking the operation state of each photovoltaic panel on the photovoltaic digital map, and digitally managing each photovoltaic panel, comprises: traversing the unmanned aerial vehicle inspection infrared image, and searching the minimum distance according to the space longitude and latitude coordinates of each photovoltaic panel and the longitude and latitude coordinates of the image center point of each photovoltaic panel on the photovoltaic digital map; And if the distance is smaller than or equal to a preset distance, mapping longitude and latitude coordinates of the geographic position of the photovoltaic panel corresponding to the photovoltaic digital map of the unmanned aerial vehicle inspection infrared image to the preset photovoltaic digital map, marking the running state of each photovoltaic panel on the photovoltaic digital map, and carrying out digital management.

Description

Digital method for automatically inspecting and positioning defects of photovoltaic panel by unmanned aerial vehicle Technical Field The invention relates to the technical field of unmanned aerial vehicle inspection photovoltaic power stations, in particular to a digital method for automatically inspecting and positioning defects of a photovoltaic panel by using an unmanned aerial vehicle. Background In recent years, the number of newly installed photovoltaic devices is continuously increased, common photovoltaic power stations comprise mountain hilly photovoltaic power stations, desert gobi photovoltaic power stations, light complementation, agricultural light complementation, water light complementation photovoltaic power stations, distributed photovoltaic power stations, offshore photovoltaic power stations and the like, and for daily maintenance of the photovoltaic devices, photovoltaic maintenance staff can hardly reach the photovoltaic power stations, and the maintenance efficiency of the photovoltaic power stations is low and the cost is high. Along with the maturity of unmanned aerial vehicle technique, photovoltaic inspection is no longer limited to manual operation, but in fact, does not have the standardized photovoltaic power plant scheme of inspecting, trees in the environment and the high voltage transmission line that photovoltaic power plant top passed through can cause the influence to unmanned aerial vehicle's flight to unmanned aerial vehicle inspection photovoltaic area is limited, influences inspection efficiency, along with the increase of photovoltaic service life, has more and more urgent demand to defect state diagnosis and management of photovoltaic board. Disclosure of Invention Aiming at the defects in the prior art, the invention aims to provide a digital method for automatically inspecting and positioning defects of a photovoltaic panel by using an unmanned aerial vehicle. According to one aspect of the invention, a digital method for automatically inspecting and locating defects of a photovoltaic panel by using an unmanned aerial vehicle is provided Optionally, determining a shooting point of unmanned aerial vehicle inspection and a geographic position index of each photovoltaic panel in a shooting area where the shooting point is located according to a preset unmanned aerial vehicle inspection height on a preset photovoltaic digital map; Acquiring an unmanned aerial vehicle inspection infrared image according to the shooting point of the unmanned aerial vehicle inspection; processing the inspection infrared image of the unmanned aerial vehicle according to a target defect detection algorithm and an infrared relative temperature difference method, and determining the state of a photovoltaic area and the type of infrared thermal defects in a shooting area corresponding to the shooting point; Performing image segmentation processing on the unmanned aerial vehicle inspection infrared image, and determining mapping from image pixel coordinates of each photovoltaic panel to longitude and latitude coordinates of a geographic position; Mapping the longitude and latitude coordinates of the geographic position of each photovoltaic panel to the preset photovoltaic digital map, positioning the photovoltaic panel with the infrared thermal defect, determining the running state of each photovoltaic panel, and further marking the running state of each photovoltaic panel on the photovoltaic digital map to digitally manage each photovoltaic panel. Optionally, on a preset photovoltaic digital map, determining, according to a preset unmanned aerial vehicle inspection height, a shooting point of unmanned aerial vehicle inspection and a geographic position index of each photovoltaic panel in a shooting area where the shooting point is located, including: continuously shooting images of a target photovoltaic area through an unmanned aerial vehicle, determining an orthographic image of the target photovoltaic area, adopting a map making tool, making a photovoltaic digital map of the target photovoltaic area according to the orthographic image, and establishing a geographic position index for each photovoltaic panel in the photovoltaic digital map; determining a shooting area of the unmanned aerial vehicle under orthographic projection according to the preset unmanned aerial vehicle inspection height; According to the length, the width and the inclination angle of the photovoltaic panel, determining row intervals and column intervals between shooting points in the unmanned aerial vehicle inspection and shooting points in the shooting area; traversing the photovoltaic digital map according to the row interval and the column interval, and determining longitude and latitude coordinates of all shooting points of the unmanned aerial vehicle inspection; And determining a geographic position index corresponding to each photovoltaic panel in a shooting area where the shooting point is located according to al