Stability analysis of a general class of singularly perturbed linear hybrid systems - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Automatica Année : 2018

Stability analysis of a general class of singularly perturbed linear hybrid systems

Résumé

We introduce and analyze a general class of singularly perturbed linear hybrid systems with both switches and impulses, in which the slow or fast nature of the variables can be mode-dependent. This means that, at switching instants, some of the slow variables can become fast and vice-versa. Firstly, we show that using a mode-dependent variable reordering we can rewrite this class of systems in a form in which the variables preserve their slow or fast nature over time. Secondly, we establish, through singular perturbation techniques, an upper bound on the minimum dwell-time ensuring the overall system's stability. Remarkably, this bound is the sum of two terms. The first term, which can be equal to zero, only depends on the matrices of the reduced order linear hybrid system describing the slow dynamics and corresponds to an upper bound on the minimum dwell time ensuring the stability of that system. The order of magnitude of the second term is determined by that of the parameter defining the ratio between the two timescales of the singularly perturbed system. We show that the proposed framework can also take into account the change of dimension of the state vector at switching instants. Numerical illustrations complete our study.

Domaines

Automatique
Fichier principal
Vignette du fichier
SwitchedSingularPerturbation_Auto.pdf (389.21 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01653073 , version 1 (01-12-2017)

Identifiants

Citer

Jihene Ben Rejeb, Irinel-Constantin Morarescu, Antoine Girard, Jamal Daafouz. Stability analysis of a general class of singularly perturbed linear hybrid systems. Automatica, 2018, 90, pp.98-108. ⟨10.1016/j.automatica.2017.12.019⟩. ⟨hal-01653073⟩
225 Consultations
142 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More