Solar Cell Defect Automatic Detection And Evaluation System Based On Absolute-value EL Imaging

Solar energy is renewable and clean.Increasing the conversion efficiency of solar cells can promote sustainable development in the new energy industry.However,defects existing in solar cells will decrease the efficiency of solar cells,and even affect their overall characteristics.Therefore,it is of great value to carry out exploration of defects to give more feedback for the design,optimization,and manufacture of solar cells.Electroluminescence(EL)imaging technology is effective in studying defects.When a solar cell is forward biased in dark condition,photons emitted by the recombination of electrons and holes are captured by the Charge Coupled Device(CCD)camera,defects will appear darker than normal pixels in the EL images.However,traditional detection methods mainly rely on the naked eye to manually identify defects in EL images.These methods are not only time-consuming,but also subjective.This thesis conducted researches on the automatic defect detection and evaluation system based on absolute-value EL imaging.The main contents of the research are as follows:(1)An automatic defect detection and evaluation system has been proposed.To make up the automatic system,various functional modules have been established,including image acquisition,defect detection,absolute-value EL distribution,defect recovery,and performance derivation.(2)By using Python scripting language combined with OpenCV,NumPy,Matplotlib,and other open-source tool libraries,the proposed functional modules have been realized.The feasibility of the entire defect automatic detection and evaluation system has been verified.The automatic system is not limited by factors such as the materials and size of solar cells,and the operation is uncomplicated and convenient.(3)Experiments were conducted in different injection current densities to test the EL images of GaAs solar cells and the defects of the GaAs solar cell are automatically detected and evaluated by the automatic system.Results indicate that the proposed automatic system is efficient for visualizing the distributions of defects,simulating solar cell defect recovery,and deriving solar cell I-V characteristics.The proposed system also reduces the time,effort,and error of manual defect analysis,providing a potential standard method in the field of solar cell defect detection.