Research On Key Techniques Of Solar Cell Intelligent Slicing System

In the production process of solar cells,it is usually necessary to divide and cut the standard cells into small cells of different specifications.this paper studies a solar cell intelligent dicing system which realized the online intelligent nesting of the standard original cells which may be exist damage,and achieved efficient dicing.When using machine vision technology for intelligent nesting,if the image is directly processed in the image space,there is a problem that the final patch size has a low precision.Therefore,the two-dimensional virtual space is created and the image of the cell is mapped from the image space to the virtual space.The method solves the problem of large size error caused by low image resolution.Based on the idea of greedy algorithm,an intelligent nesting algorithm for solar cells was proposed.First,the effective area of the cell is identified and extracted from the original image.Then in the effective area of the solar cell,according to the order of the rectangle size of the given specifications from the largest to the smallest,according to the current best principles for the nesting.The designed algorithm can achieve a better layout result in a shorter period of time,so as to adapt to the situation of the damaged original cell that may appear during the online layout.After the nesting is completed,it is necessary to use a laser to cut the small size rectangle.A nearest neighbor method is proposed to optimize the cutting path to improve the cutting efficiency.By selecting the coordinate point nearest to the current position of the laser as the starting point of the next cutting line each time,the cutting sequence of the cutting line is determined.Thereby obtaining a better cutting path.The experiment shows that the proposed algorithm can get a better nesting result and cutting path for any type of solar cell.The intelligent dicing system designed has successfully applied in the solar cell industry and has achieved good results.Instead of the previous manual processing,saving manpower and time as much as possible,and improving production efficiency.It achieves the work process intelligent.At the same time,the designed algorithm also has certain reference value in the approximate application field.