Title: Robust diagnosis for hybrid noise systems

Authors: Asma Takrouni; Nadia Zanzouri; Islem Labidi

Addresses: Department of Electrical Engineering, Laboratory ACS, Box 37, Belvedere, 1002 Tunis, Tunisia ' Department of Electrical Engineering, Laboratory ACS, Box 37, Belvedere, 1002 Tunis, Tunisia ' Department of Electrical Engineering, Laboratory ACS, Box 37, Belvedere, 1002 Tunis, Tunisia

Abstract: This paper investigates the synthesis of robust hybrid observer dedicated to identification mode and fault detection and isolation (FDI) for MIMO hybrid dynamical system (HDS) modelled via hybrid automata. The observer design is divided into two modules. The first is used to identify the current mode and the second is synthesised around a dedicated observer scheme (DOS) to detect faults. In order to highlight the efficiency of linear matrix inequality (LMI) approach in synthesis of robust hybrid observer, a comparison between this technique and pole placement method is proposed. The performance required by the LMI approach is manifested by ensuring a minimum convergence time.

Keywords: identification; fault detection; fault isolation; FDI; MIMO HDS; hybrid dynamical systems; modelling; hybrid automata; observer design; hybrid observers; dedicated observer scheme; DOS; pole placement; linear matrix inequalities; LMI.

DOI: 10.1504/IJMIC.2016.075818

International Journal of Modelling, Identification and Control, 2016 Vol.25 No.3, pp.177 - 189

Received: 19 Dec 2014
Accepted: 21 Jul 2015

Published online: 06 Apr 2016 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article