Research On Photovoltaic Panel Defect Detection Technology Based On High Frame Rate InGaAs Camer

With the large-scale application of solar energy,the productivity of solar panels is increasing year by year,and the higher request for the accuracy and nondestructive testing of defects in solar panels are also set.The traditional electroluminescence detection method is not competent to the long-time and large-area detection operation due to the needs of changing the original wiring circuit of the solar panel and the possible damage to the solar panel caused by the external high current.Photoluminescence,as a defect detection method that can replace electroluminescence,has drawn more and more attention because of its portability and controllability.However,when using photoluminescence detection,the requirement for ambient light that the conditions of darkroom need to be strictly guaranteed are too high,which limits its development and application range.Therefore,this project focuses on how to carry out photoluminescence detection and how to reduce the dependence on weak ambient light.This thesis introduces the status quo of defect detection technology for solar panels.Comparing the advantages and disadvantages of electroluminescence and photoluminescence defect detection methods,the more portable and efficient one is selected.To solve the problem that photoluminescence can only be detected indoors or in almost dark room,a solution for defect detection under a certain degree of sunlight irradiance is put forward.Firstly,based on solar panel photoluminescence principles,a set of light source system with light source,heat dissipation and focus as three basic components is designed.Secondly,a miniaturized near infrared hardware imaging system is designed rooted in the high quantum efficiency of InGaAs material in near infrared band.Adopting modular design ideas,the whole hardware in this system is divided into six modules:detector transfer,power control,operational amplifier and AD conversion,output interface,field programmer sub-card,TEC sub-card,and through each module.Then,the inter-plate stacking is designed through modularization of each function,minimizing the overall design in which the maximum area of a single circuit board is only50×50mm.Next,in the digital logic of the overall imaging system,the driver of the detector,high frame frequency,high bandwidth data transmission are completed through Register-Transter Level(RTL),then using micro-HDMI or Camera Link display and optional USB3.0digital logic design for transmission.In addition,in order to solve the problems like blind cells,bad points and non-uniformity,several algorithms such as blind cell substitution algorithm,fixed-mode noise removal algorithm under dark and bright conditions,and digital filter are added to improve the image quality.Finally,an experimental detection platform is set up.Comparing three near-infrared cameras,the high frame-frequency InGaAs near-infrared camera is employed as the main detection imaging device.In the experiment,the light source system is modulated,and the background reduction and defect significance enhancement algorithms are designed.The result proved that the irradiance of ambient light which can be effectively detected has been raised from the 50 W/m~2 to around 400 W/m~2,which greatly reduces the dependence on dark environment and achieves the goal of defect detection under a certain degree of sunlight radiation.