Multi-layered and hierarchical fuzzy modelling using evolutionary algorithms

In this presentation we examine issues in the construction of a fuzzy logic system to model a complex (nonlinear) system associated with the decomposition into hierarchical/multi-layered fuzzy logic sub-systems and the learning of fuzzy rules and internal parameters. The decomposition into hierarchical/multi-layered fuzzy logic sub-systems reduces greatly the number of fuzzy rules to be defined and to be learnt but such decomposition is not unique and may give rise to variables with no physical significance. This can raise then major difficulties in obtaining a complete class of rules from experts even when the number of variables is small. We will examine the learning of fuzzy rules in such systems using evolutionary algorithms. Application areas considered are: the prediction of interest rate, hierarchical control of the inverted pendulum, robot control, feedback boundary control for a distributed optimal control system and image processing.