Application Of Machine Vision In Automatic Solar Cell Laminated Welding Machine

Photovoltaic power generation is the main way to use solar emergy and plays an important role in the utilization of new energy and its development.A large part of the photovoltaic power generated by solar modules,in order to further improve the efficiency of power generation,resulting in the new process of solar modules invented.The traditional cell string welding technology has reached a bottleneck of the photoelectric conversion rate,it is necessary to invent a new technology of the solar modules process.In the new process,grid lines of solar cells will be more refined,and the positioning of solar cells will be further improved.Ensuring its positioning accuracy and at the same time realizing the real-time testing is one of the important factors affecting the development of the new technology.Therefore,it is necessary to study the machine vision system in the new process equipment.Based on the traditional machine vision of stringing machine,this paper analyzes and designs the hardware system through the application scenario of the actual solar cell,and improves the camera calibration system.The algorithm for positioning the solar cells is proposed and this paper verifies the accuracy of the algorithm.In the end,this paper verifies results of the defect testFirstly,according to the research background of battery components,this paper introduces the automatic stringing machine and the machine vision system and its current research status.Technical difficulties of the new process of solar modules and the significance and main content of the research are proposed.And this paper introduces the structure and working progress of the machine vision system in the new automatic laminating stringing machine,analyzes the key technologies involved in the system.Combined with the characteristics of the solar cell sheet and the actual working conditions of the laminator,the hardware system in the machine vision system is designed and selected,and the software flow is confirmed.Then,according to the camera model theory,this paper analyzes the influence of the camera's internal and external parameters and camera distortion on the collected image information,and expounds the importance of camera calibration.According to the nine-point manual calibration principle,a self-calibration method for solar cell slice is proposed and the calibration results obtained by the method are verified.Next,this paper introduces the commonly used edge detection operators and selects the pixel-level edge detection operator used in this paper.In this chapter,the advantages of sub-pixel algorithms and pixel-level edge detection operators are contrasted.Several methods of the sub-pixel edge detection operators are compared and selected.A sub-pixel edge detection algorithm for solar cell slice is proposed.Then,according to the algorithm,the solar cell slice is positioned,the positioning results are analyzed,and its accuracy and reliability is verified.This paper introduces the common types of defects in solar cells,classifies defects in the machine vision system of the laminator,and designs the defect detection process.In order to meet the later demand changes,the parameter setting interface is determined.At the same time,this paper expounds the method of the defect detection,and its defect detection results are analyzed.At last,this paper summarizes the main research work carried out in the full text,and puts forward further research directions and improvement methods for the future direction of development.