Title: Arnoldi model reduction for switched linear systems

Authors: Kouki Mohamed; Abbes Mehdi; Mami Abdelkader

Addresses: Laboratoire d'Analyse, de Commande et de Conception des Systemes, École Nationale d'Ingénieurs de Tunis, LR-11-ES20, Université de Tunis El Manar, BP 37, LE BELVEDERE 1002, Tunis, Tunisia ' Laboratoire d'Analyse, de Commande et de Conception des Systemes, École Nationale d'Ingénieurs de Tunis, LR-11-ES20, Université de Tunis El Manar, BP 37, LE BELVEDERE 1002, Tunis, Tunisia ' Laboratoire d'Analyse, de Commande et de Conception des Systemes, École Nationale d'Ingénieurs de Tunis, LR-11-ES20, Université de Tunis El Manar, BP 37, LE BELVEDERE 1002, Tunis, Tunisia

Abstract: In this paper, we present a new method of reduction for large-scale switched systems, which is an important class of hybrid and non-linear systems. In this work, we focused on the linear switched systems. This new method is based on the generation of Krylov subspace by the use of the Arnoldi algorithm, which is numerically efficient and applicable for any order, from which we solve the problems H_∞ and H₂ model reduction. We take three examples to evaluate this method.

Keywords: model order reduction; Krylov subspace; multiple points moment matching; hybrid dynamic systems; Arnoldi model reduction; switched linear systems.

DOI: 10.1504/IJOR.2016.078456

International Journal of Operational Research, 2016 Vol.27 No.1/2, pp.85 - 112

Received: 11 Sep 2013
Accepted: 09 Apr 2014
Published online: 22 Aug 2016 *

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