Optimal Fuzzy Controller Design By Genetic Algorithm

• Main Authors: Zhn He Nian, 粘志河
• Other Authors: Jun Feng Lu
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/73966082206548721514

碩士 === 中州技術學院 === 工程技術研究所 === 96 === The purpose of the study is about optimal fuzzy controller design by genetic algorithm. To consider all of the fuzzy control factors to code membership functions, fuzzy rules and scaling factors together, in order to preclude the limit of a single factor. Others also improve the genetic operators and mutation operator especially. To compare with the traditional genetic algorithm, proves the optimal genetic algorithm can avoid premature, and can’t fall in local search. To compare with the PID controller proves its response is fast, and overshoots is less. Finally the inverted pendulum system proves it can get the better control. Its response curve is smooth and steady. The nonlinear control system shows good dynamic performance, steady-state performance, disturbance rejection and robustness.