Manufacturing Process Parameters Predicting and Analyzing by Soft-Computing and Chaos Theory

• Main Authors: Chun-ying Lin, 林俊穎
• Other Authors: Tung-Hsu Hou
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/11452165994341147628

碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 92 === The chaos theory can be used to analyze chaos and unordered data. The phase state reconstruct (PSR) is one of the most important technologies used to analyze chaos data. The PSR can transfer the one-dimension chaos data into multi-dimension ordered data. Traditionally, the phase state local approximate (PSLA) is then used to predict the chaotic data time series by using the local information. In this research, a new system is proposed to predict the chaotic time series. The PSR is used to transfer the one-dimension data into multi-dimension phase state data. The fuzzy clustering is applied to group the phase state data into several clusters which are than used to create the fuzzy set and its rough membership functions. The genetic algorithm (GA) is then used to find the optimal membership functions. Finally, the fuzzy set inference is used to predict the chaotic series. Two examples are used to demonstrate the effectiveness of the proposed system. The first example uses two famous functions to generate chaotic series data which are then used to test the PSLA and the proposed system. The second example uses manufacturing process parameters of an industrial electric cable to test the PSLA and the proposed system. uses comparison results show that the proposed system has better prediction accuracy than the PSLA