System Identification for Car-Pendulum by Using Genetic Algorithm

• Main Authors: Bi-wei Guo, 郭璧瑋
• Other Authors: Huey-Yang Horng
• Format: Others
• Language: zh-TW
• Published: 2008
• Online Access: http://ndltd.ncl.edu.tw/handle/85802366073794800985

碩士 === 義守大學 === 電子工程學系碩士班 === 96 === In this study, the main purpose is to identified the car-pendulum’s system by using genetic algorithm. For model based fussy control, we should obtain the system model prior to design a stable control. The car-pendulum system is unstable. Hence, we design the stable controller by pole-placement method. And when, we offer chirp signal to the process in order to measure the input and output data adequately. Moreover, identify the parameters of Takagi-Sugeno (T-S) model by the genetic algorithm form the measured data. In order to enable the parameter fast to receive in the search process also has better essence, we considered develops in the genetic algorithm to join some strategies, for example: migration, elite count and so on. The advantages of the proposed methods are illustrated through simulations.