Applications of Fuzzy Brain Emotional Cerebellar Model Articulation Controller for SIMO Nonlinear System Decoupling Control

• Main Authors: Geng-Lin Guo, 郭畊麟
• Other Authors: Chih-Min Lin
• Format: Others
• Language: en_US
• Published: 2018
• Online Access: http://ndltd.ncl.edu.tw/handle/p946yp

碩士 === 元智大學 === 電機工程學系 === 106 === In this thesis, a Fuzzy Brain Emotional Cerebellar Model Articulation Controller (FBECMAC) is designed for the decoupling nonlinear systems. For the underactuated systems, the decoupling sliding-mode technique is used to achieve the decoupling control performance. The FBECMAC combines cerebellar model with a brain model to introduce a new neural network model, which imitates the judgment of a cerebellar and the emotion of a brain. The brain has an amygdala cortex and a prefrontal cortex, so that the FBECMAC contains two neural networks; the amygdala cortex is a sensory neural network and the prefrontal cortex is an emotional neural network. The proposed FBECMAC has learning ability and can adjust parameters to achieve efficient control performance. The proposed control system is comprised of a FBECMAC and an auxiliary compensation controller. The FBECMAC is used as the main controller to approximate an ideal controller to achieve desired control performance and the auxiliary compensation controller is utilized to attenuate the residual approximation error to achieve system stability. Finally, the FBECMAC control system is used to control a ball-beam system, a Translational Oscillations with a Rotational Actuator(TORA) system, a bridge-crane, an aeroelastic structure system and a double-inverted pendulum system to confirm the validity of the proposed approach.