Design of Fuzzy Sliding Mode Controller with Adaptive Fuzzy Rule Insertion

• Main Authors: Feng-Sang Chen, 陳鳳珊
• Other Authors: Prof. Chung-Chun Kung
• Format: Others
• Language: en_US
• Published: 2000
• Online Access: http://ndltd.ncl.edu.tw/handle/51537681044861339600

碩士 === 大同大學 === 電機工程研究所 === 88 === In this thesis, efficient fuzzy rules of an adaptive fuzzy sliding mode controller (FSMC) are developed systematically according to the performance of the controller. The input variables of conventional FSMC are mostly two fuzzy inputs regardless of the complexity of controlled plants. We propose a new simple FSMC called distance-based FSMC (D-FSMC) by using a single fuzzy input variable called the signed distance, which is the distance from the actual state to the sliding line in the -S plane. Hence, the number of fuzzy rules is greatly reduced compared to the case of the conventional FSMC with two fuzzy inputs. The stability of the fuzzy control system is guaranteed using the Lyapunov stability. We provide learning algorithms to design fuzzy rules of D-FSMC from basic two rules to three rules and upward for performance improvement. The new rules we insert into the rule-base structure are initialized with the knowledge of the redundant fuzzy rules. And the parameters of the membership functions of the fuzzy rules are trained on—line to minimize the cost functions. Thus, the D-FSMC can be systematically designed. The performance of D-FSMC is revealed via computer simulations using nonlinear plants.