Dynamic Petri Cerebellar Model Articulation Controller for Magnetic Levitation System

• Main Authors: Yen-Jui Peng, 彭彥瑞
• Other Authors: Chih-Min Lin
• Format: Others
• Language: en_US
• Published: 2011
• Online Access: http://ndltd.ncl.edu.tw/handle/48571506368131012507

碩士 === 元智大學 === 電機工程學系 === 99 === In this thesis, the control of the magnetic levitation system is achieved by the proposed intelligent controller, which is based on the cerebellar model articulation controller (CMAC). The magnetic levitation system is a complicated and tough problem for the conventional controller, because the system is a highly nonlinear system. Since the magnetic levitation system is without mechanical contact, friction and noise, it can be used for precise positioning. These advantages make it a wide range of applications on the maglev train, magnetic bearing, wind tunnel, and conveyor system, etc. The proposed intelligent controller is composed of a novel dynamic Petri CMAC controller which is utilized to approximate an ideal controller. The dynamic Petri is used to determine the passing of the Gaussian function value. Finally, the proposed intelligent control system is applied to the magnetic levitation system, and its performance is verified through simulation and experiments based on the field programmable gate array (FPGA) chip. From the simulation and experimental results for the magnetic levitation system, the system stability and desired control performance can be achieved by the proposed intelligent controller.