Input variables selection of fuzzy dynamic models by using genetic algorithm

In this paper, an input selection technique for fuzzy-based dynamic modelling is presented. When a system is composed of a large number of input-output variables with strong coupled dynamics, a systematic method is required in order to minimise the modelling error through the selection of proper variables. The selection approach is based on a Genetic algorithm optimisation to obtain solutions as binary strings that effectively determine a proper input field structure.