Title page for etd-0712100-123711


[Back to Results | New Search]

URN etd-0712100-123711
Author Ing-Hao Chen
Author's Email Address m8732667@student.nsysu.edu.tw
Statistics This thesis had been viewed 5364 times. Download 2223 times.
Department Mechanical Engineering
Year 1999
Semester 2
Degree Master
Type of Document
Language English
Title System Identification for Transmission Mechanism by Using Genetic Algorithms
Date of Defense 2000-06-30
Page Count 97
Keyword
  • Genetic algorithm
  • PID controller
  • DC servomotor
  • System identification
  • Harmonic drive
  • Transmission system
  • Minimum variance controller
  • Abstract In this study, the use of modified genetic algorithms (MGA) in the parameterization of the Transmission Mechanisms is facilitated. The new algorithm is proposed from the genetic algorithm with some additional strategies, and yields a faster convergence and a more accurate search. Firstly, this near-optimum search technique, MGA-based ID method, is used to identify the parameters of a system described by an ARMAX model in the presence of white noise and to compare with the LMS (Least mean-squares) method and GA method. Then, this proposed algorithm is applied to the identification of the Transmission Mechanisms of DC motor. The parameters of the friction force and DC motor are estimated in a single identification experiment. It is also shown that this technique is capable of identifying the whole transmission system. Finally, the Minimum Variance Controller (MVC) is taken to track the desired speed trajectory and then a comparison to the conventional digital PID controller is shown. Experiment results are included to demonstrate the excellent performance of the MVC.
    Advisory Committee
  • Yih-Tun Tseng - co-chair
  • Huey-Yang Horng - co-chair
  • Ing-Rong Horng - advisor
  • Files
  • 論文_完稿.pdf
  • indicate access worldwide
    Date of Submission 2000-07-12

    [Back to Results | New Search]


    Browse | Search All Available ETDs

    If you have more questions or technical problems, please contact eThesys