Variable structure neural network control

Abstract

A variable structure neural network control (VSNNC) is proposed for a class of uncertain nonlinear SISO systems, in which neural network is used as an estimator for the system unknown nonlinear functions. And variable structure control strategy is improved by adding the continuous function item to control input. It can adjust discontinuous item in control variable adaptively according to the distance between the state point and the sliding mode switching surface. The proposed control method can not only effectively restrain the chattering around the switching surface but also guarantee the dynamic performance and eliminate the static error. Finally, some results of simulation experiments indicate that the proposed control scheme is feasible.