Research And Simulations Of System Identification

Traditional controlling and identifying theories are good used for linear system identifications, as for non-linear systems, they are difficult to identify or even impractical. From the period of 1980s, with the rise of human intelligence, fuzzy theory and its application is getting more and more widely developed. In 1990s, with syncretic development of artificial neural networks and fuzzy logics, the application fuzzy reasoning systems is getting into identification field from controlling field. This thesis mainly focuses on the application of fuzzy logics for non-linear, SISO and MIMO system identification. Main contents include: Firstly, the evaluation function necessarily used in identification is discussed and details about MME and its application in non-linear, uncertain systems are given. And it is also given that how to choose an optimal value of positive matrix under satisfying conditions. Secondly, fuzzy theory is introduced, including its development, application and extrapolated...