Using a revised gravitational search algorithm to find the optimal parameters for a nano-particle milling process

• Main Authors: Yen-Chuan Tsai, 蔡嬿娟
• Other Authors: Tung-Hsu Hou
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/96617741639835419431

碩士 === 國立雲林科技大學 === 工業工程與管理系 === 102 === In the early 21st century, the nanotechnology were gradually mature, many industries began using nanotechnology in their product. Thus, nanotechnology becomes an indispensable technology in many industries. In general, many researches commonly used Taguchi methods and algorithms to make the optimal decision. However, there would have errors between actual results and predictions which are calculated by Taguchi method. Therefore, many heuristic algorithms have been developed and used in industry. In all manufacturing process, the producers expect to offer the best quality products to customers and they would get good reputation. The aim of this study is to explore the optimal parameters of the manufacturing process. Due to heuristic algorithms still have some problems included the convergence rate and falling into local optimal solution, the problem of convergence into a local optimum need to be improved. Therefore, this study uses Gravitational Search Algorithm whch through the slow convergence and escape local optimum to find the optimal solution. Additionally, this study also uses the fuzzy and mutation methods to improve the gravitational search algorithm. We expected to get the optimal parameters to minimize the mean and variances of grain size in the nano-particle milling process. Finally, this study also compared the results of the Revised Gravitational Search Algorithm and other algorithms. The results of this study could provide researchers or engineers to make parameter settings as reference. These results also could help the industries to improve their manufacturing process.