An Investigation of Developing An Intelligent Controll System

• Main Authors: Lee Yung-Lung, 李永隆
• Other Authors: Ji Ching-Chai
• Format: Others
• Language: zh-TW
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/24956080719377392486

碩士 === 中正理工學院 === 兵器系統工程研究所 === 88 === Fuzzy logic theory and grey theory has been developing rapidly in recent years, and is being used successfully in control of complex processes and uncertain Problems. But there exist some problems in using thesis two theories. In fuzzy Logical Controller (FLC), there exist two main topics for the design of fuzzy control structure. One is to find adequate and appropriate control rules to direct the decision process for the complex system, and the other is to find a good parameter set of membership functions describing the linguistic terms in the fuzzy rules. In grey predictor (GP), the solution of the "grey model (GM)" is an exponentially which behaved by a first-order differential equation. Thus, it will be had a larger predict error for a non-exacting increase or non-exacting decrease response of controlled system. In this study, the simple mechanics of GA will be applied to establish the genetic algorithm-based fuzzy logic controller to search adequate fuzzy parameters, and the grey predictor will be reduced the rise time. Finally, we introduce the step-size fuzzy inference system (SSFIS) and the grey predictor weighted fuzzy inference system (GPWFIS) into FLC to effectively subdue the overshoot of controlled plane and substantially decrease rise time. Thus, we can design an FLC based on GA and GP, that have the ability of self-tuning and prediction.