Robust fault identification of a polytopic LPV system with neural network

Abstract

In this paper, a discrete-time Linear Parameter-Varying (LPV) system identification method using artificial neural network is described. In particular, neural network is transformed to obtain LPV model of the non-linear system. Moreover, a novel robust fault diagnosis scheme is developed, which is based on an observer within H∞ framework for a class of non-linear systems. The effectiveness of the proposed approach is illustrated by the faults estimation in the multi-tank system.