The Fuzzy-H∞ Optimal Control Design

• Main Authors: Chin-Wei Ko, 柯今偉
• Other Authors: Cheng-Neng Hwang
• Format: Others
• Language: zh-TW
• Online Access: http://ndltd.ncl.edu.tw/handle/12790686348546066726

碩士 === 國立成功大學 === 系統及船舶機電工程學系碩博士班 === 94 === Abstract The characteristic of the H∞ controller is maintaining the stability and robustness when the uncertainly disturbances and the parameters of the system changes. The advantage of the fuzzy controller is able to use the rule of expert's knowledge base system to design the fuzzy controller, and do not need clear mathematic equations. This paper combines the advantages of the fuzzy theory and the H∞ control theory to develop a steady and high performance Fuzzy-H∞ controller. The plant that we want to control is the mechanical arm control system of two axles (two-link robot).When giving a reference input model, we use H∞ control theory to track the states of two-link robot. The system is changed into the standard state space model form, and we can use T-S fuzzy inference rules to reduce the error which produced by the process of nonlinear to linear system. In the course of deriving the system equations, we can join the fuzzy theory to the standard state space dynamic equations, to change the expression of the fuzzy state space. And use the Lyapunov stability theorem to prove the stability of the closed-loop system when joining the Fuzzy-H∞ controller and observer. While doing computer simulation, because the two–link mechanical arm system is MIMO system, and undergo continuously disturbances, we select a servo compensator of the purpose to reject the disturbances and tracking accurately, and select a filter of the purpose to filter out the useless signals, and the computer simulation results show that the steady state error maintain in a very small range, and this proves the feasibility and accuracy of the Fuzzy-H∞ controller.