Using Multiple Objectives Artificial Immune System to Find the Optimal Manufacturing Parameters for a Solar Cell Process

• Main Authors: Yi-fang Lain, 連逸芳
• Other Authors: Tung-hsu Hou
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/20745239240862051067

碩士 === 雲林科技大學 === 工業工程與管理研究所碩士班 === 98 === High-temperature phosphorus diffusion process is the core manufacturing process to produce the solar cell. The purpose is to produce the p-n junction, the main structure of the solar cell, Its uniformity will affect the conversion efficiency of photovoltaic effects. The process contains the complex non-linear physical and chemical reaction. To use past experience to adjust the process parameters may not get the best results we expect and the quality of product may be not steady. This study proposes a multi-objectives artificial immune system, MOAIS to find an optimal parameter set for setting the diffusion process. The first phase use back-propagation neural network, BPN to establish the relationship between input factors and quality characteristics. And the second phase, the local search mechanism to join the MOAIS to optimize process parameters to get a set of non-dominated solutions. We expect that the method proposed in this study to find the non-dominated solution can be provide more information for decision-making.