| Summary: | 碩士 === 南臺科技大學 === 機械工程系 === 103 === Multi-crystalline silicon (mc-Si) solar cells will be the future mainstream because it can significantly reduce the manufacture cost. However, Electroluminescence (EL) technology cannot effectively inspect the quality of mc-Si solar cells due to their high density of dislocation and grain boundary. The programmable logic controller (PLC) can load the control command into the memory and execute it. By using the input and output modules to control the machine motion and manufacturing procedures, PLC is an important tool for industrial automation.
This study is divided into two parts. The first part is the colorization processing of grayscale EL images. Because human eyes are more sensitive to colorful than black and white images, we developed programs to colorize the EL grayscale images in order to highlight the defect positions. The second part is the application of the PLC control to construct a solar cells transport system and the integration of these two features on a man-machine interface.
The results show that both block algorithms and texture algorithms can clearly show the high efficiency and middle efficiency regions by colors so that the interpretation of EL images of the mc-Si solar cells can be improved. In addition, the texture algorithm can perform more delicate colorization of the structure around a defect. Although the colorization effect of the block algorithm for defect structure is rougher and hazier, but it can colorize a specimen in less than 4 seconds while the colorization time of the texture algorithm is less than 8 seconds. The integrated system developed in this study can complete the following procedures: solar cells transportation, EL image capture, colorization, cells defects inspection, and cells classification. It can be used as the foundation for future development of an automatic inspection system.
|
|---|
