Evolved Fuzzy NN Control for Discrete-Time Nonlinear Systems

Abstract

To guarantee the asymptotic stability of discrete-time nonlinear systems, this paper proposes an Evolved Bat Algorithm (EBA) fuzzy neural network (NN) controller. In the evolved fuzzy NN modeling, an NN model and linear differential inclusion (LDI) representation are established for arbitrary nonlinear dynamics. This representation is constructed by the use of sector nonlinearity to convert a nonlinear model to the multiple rule base of the linear model, and a new sufficiency condition to guarantee asymptotic stability in the Lyapunov function is implemented in terms of linear matrix inequalities. The proposed method is an enhancement of existing methods which produces good results.