Design of Linguistic Hedge Fuzzy Logic Controller with Heuristic-Algorithm Enhancements

• Main Authors: Bo-yu Syu, 許博喻
• Other Authors: Chuen-Yau Chen
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/2w84g6

碩士 === 國立雲林科技大學 === 電機工程系碩士班 === 96 === In this thesis, four heuristic algorithms are applied to design the linguistic-hedge fuzzy logic controller. These algorithms include genetic algorithm, guided simulated annealing algorithm, guided ant algorithm, particle swarm optimization algorithm. For the controller, we apply different partition strategies on the input-variable domains and analyze the effects on the performance of the controller. The input-variable domains are partitioned in the following manners: the uniform sub-intervals, the more uniform sub-intervals, the fewer uniform sub-intervals, and the non-uniform sub-intervals. The simulation results on the truck-backer upper control system show better performance of the linguistic-hedge fuzzy logic controller. Consider the performances of different strategies. The linguistic-hedge fuzzy logic controller having the input-variable domains with 16 uniform sub-intervals associated with guided simulated annealing algorithm has the best performance. The number of iterations is 1028; the docking error is 0.04.