TABLE OF CONTENTS The design of modern control systems is characterised by stringent performance and robustness requirements and therefore relies on model-based design methods. This introduces a strong need for effective modelling techniques. Many systems are not amenable to conventional modelling approaches due to the lack of precise knowledge and strong nonlinear behaviour of the process under study. Nonlinear identification is therefore becoming an important tool which can lead to improved control systems along with considerable time saving and cost reduction. Among the different nonlinear identification techniques, methods based on fuzzy sets are gradually becoming established not only in the academia but also in industrial applications [1] [3].
Fuzzy modelling techniques can be regarded as grey-box methods on the boundary between nonlinear black-box techniques and qualitative models or expert systems. Their main advantage over purely numerical methods such as neural networks is the transparent representation of knowledge in the form of fuzzy if then rules. Linguistic interpretability and transparency are therefore important aspects in fuzzy modelling [4] [8]. The tools for building fuzzy systems are based on algorithms from the fields of fuzzy logic, approximate reasoning, neural networks, pattern recognition, statistics and regression analysis [1] , [2] , [9].
This chapter gives an overview of system identification techniques for fuzzy models and some selected techniques for model-based fuzzy control. It starts with a brief discussion of the position of fuzzy modelling within the general nonlinear identification setting. The two most commonly used fuzzy models are...
