Robust-Neural Observer Design for Discrete-Time Uncertain Non-Affine Nonlinear System

Abstract

This paper proposed a new Nonlinear Discrete-Time Robust-Neural Observer (DTRNO) which capable to give estimation for the states of Discrete-Time Uncertain Non-affine Non-linear Systems in presence of external disturbances. The Neural network is a kind of discrete-time Multi Layered Perceptron (MLP) which Trained with an Extended Kalman-Filter (EKF) based algorithm, which this neural observer is robust in presence of external and internal uncertainties, using a parallel configuration.This work includes the stability proof of the estimation error on the basis of the Lyapunov approach, and for demonstrate observer performance an Uncertain Non-affine Nonlinear Systems have been simulated to formulations validate the theoretical. Simulation results confirm the proficiency of the DTRNO even at the different operating conditions and presence of parameters uncertainties.

DOI:http://dx.doi.org/10.11591/ijece.v4i4.6171