Photovoltaic Module Cleaning Robot Control System And Infrared Hot Spot Detection Device Design

With the increasing global energy crisis,people pay more and more attention to clean solar energy resources.The photovoltaic power generation industry is developing rapidly as the country has set the goal and vision of carbon peaking by 2030 and carbon neutrality by 2060.However,photovoltaic modules work in harsh environment for a long time and are susceptible to debris such as dust and bird droppings.If not removed in time,the photoelectric conversion efficiency of the photovoltaic module will be reduced,and even permanent irreversible damage will occur.In order to solve the above problems,the control system of photovoltaic module cleaning robot and infrared hot spot detection device are designed in the thesis.The main work contents include the following aspects:Firstly,the power generation principle and output characteristics of photovoltaic modules were analyzed.And the influence of different dust accumulation on photovoltaic module photoelectric conversion efficiency was described.Based on the infrared thermal camera,the feasibility of photovoltaic module hot spot detection was analyzed.Secondly,according to the cleaning demand of photovoltaic module ash accumulation,the design scheme of cleaning robot control system was given.The control system was composed of master STC15W4 K,two slave controllers STC8H3 K,Raspberry PI 4B,LoRa,infrared camera and other modules.The master controller was responsible for data communication between the upper and lower machines and control of function modules.The two-way slave controller realized BLDC motor speed control and ADIS dip Angle detection,while the infrared camera on the Raspberry PI 4B was responsible for photovoltaic hot spot detection.Then,in order to effectively remove the noise caused by camera jitter and transmission of photovoltaic module images,the thesis used an improved filtering algorithm for preprocessing operations.For the detection of hot spots,the foreground and background were separated by adaptive threshold segmentation,the isolated noise points were removed by morphological operation,the influence of component bridge was removed by image mask processing,and the initial detection of hot spots was completed by improved Canny algorithm and hot spot contour screening.The area of hot spots,fitting corner points and the minimum aspect ratio of external rectangle in the initial detection image were further discriminated to realized the precise detection of hot spots.Finally,based on Qt software,the upper computer function interface of PC side of the cleaning robot and the hot spot detection interface of raspberry PI side were designed.Through LoRa module and the main controller STC15W4K communication,cleaning robot implemented the remote automatic mode control,manual mode control and hot spot detection.When the photovoltaic module hot spots were detected,the relevant information was stored in the MySQL database.The experimental show that the system achieves the expected design goal of the subject.