On the selection of input variables by using the GA for the estimation of multi-dimensional chaotic dynamics based upon the multi-stage fuzzy inference systems

摘要

This report deals with the selection of input variables by using the GA for the estimation of multi-dimensional chaotic dynamics based upon the multi-stage fuzzy inference systems. Inthe desing of multi-stage fuzzy inference systems, we usually use very small number of rules, and the system is applicable to the system identification using many input variables. It is necessary to determine optimal sets of input variable to the fuzzy inference system by which we analyze the structure of the chaotic dynamics. In the report, we utilize the GA to determine the optimal selection of input variables for the multi-stage fuzzy inference systems, as well as the GA application for the optimization of shapes of membership functions.