STABILITY ANALYSIS OF DYNAMIC FUZZY SYSTEM

Abstract

Fuzzy control system can be described as a real-time expert system, implementing a part of a human operator's or process engineer’s expertise. A representation theorem mainly due to Kosko, states that any continuous nonlinear function can be approximated as exactly as needed with a finite set of fuzzy variables, values, and rules. In general the Lyapunov approach reveals only the existence of stable points, not their number or nature [Kosko'92].There are a few stability test methods in the linguistic fuzzy dynamic models represented by ‘if-then’ rules that can be used for modeling real plants. Tanaka and Sugeno [Tanaka'92] suggested an effective method for determining stability. Another method, which has been suggested by Kim et al. [Kim'95] treats the fuzzy model as a linear system having modeling uncertainties. It suggests a new Lyapunov method for determining stability that is less restrictive than the previous one.. The second chapter is followed by theorems giving a su...