Modeling of Multilinear Dynamical Systems from Experimental Data
- 796 Downloads
Since the introduction of fuzzy sets by Zadeh in 1965 , many researchers have shown interest in applying this theory to system identification, which is an essential part of any control system design. Rapid development of intelligent control methodologies such as artificial neural networks, fuzzy logic theory, and rule-based expert systems, have provided alternative tools to tackle the problem of system identification . A large number of fuzzy identification techniques have been developed using neural networks, genetic algorithms, clustering techniques, Kalman filtering and other methods, including ad hoc ones . Consequently, fuzzy identification has become a very important area in fuzzy system theory . The main approaches to fuzzy identification are based on linguistic fuzzy modeling, fuzzy relational equation modeling and Takagi-Sugeno modeling . In this chapter we present a new effective method of modeling continuous multilinear dynamical systems using the Takagi-Sugeno fuzzy expert system.
KeywordsFuzzy Rule Recursive Little Square Control System Design Recursive Little Square Algorithm Fuzzy Identification
Unable to display preview. Download preview PDF.