Robust inversion based fault estimation for discrete-time LPV systems
Artikel i vetenskaplig tidskrift, 2012

The article presents a state-space based Fault Diagnosis (FD) method for discrete-time, affine Linear Parameter Varying (LPV) systems. The goal of the technical note is to develop a robust and dynamic inversion based technique for systems with parameter varying representations when an additive, exogenous disturbance signal perturbs the system. After applying geometric concepts for explicit fault inversion, a robust strategy is proposed to attenuate the effect of the unknown disturbance input signal on the fault estimation error. The proposed robust observer is derived as a solution of off-line Linear Matrix Inequality (LMI) conditions. The technical note demonstrates the viability of the novel methodology through a numerical example.

dynamic inversion

robust observers

Discrete-time linear parameter varying (LPV) systems

fault detection and isolation

fault diagnosis (FD)


Balázs Adam Kulcsár

Signaler och system, System- och reglerteknik, Reglerteknik

M. Verhaegen

Chalmers University of Technology

IEEE Transactions on Automatic Control

0018-9286 (ISSN)

Vol. 57 1581-1586 6082383