Publication
Actuator and sensor fault estimation based on a proportional-integral quasi-LPV observer with inexact scheduling parameters
Conference Article
Conference
IFAC Workshop on Linear Parameter-Varying Systems (LPVS)
Edition
3rd
Pages
100-105
Doc link
https://doi.org/10.1016/j.ifacol.2019.12.355
File
Authors
-
Gomez Penate, Samuel
-
Lopez Estrada, Francisco Ronay
-
Valencia Palomo, Guillermo
-
Rotondo, Damiano
-
Enriquez Zarate, Josue
Projects associated
Abstract
This paper presents a method for actuator and sensor fault estimation based on a proportional-integral observer (PIO) for a class of nonlinear system described by a polytopic quasi-linear parameter varying (qLPV) mathematical model. Contrarily to the traditional approach, which considers measurable or unmeasurable scheduling parameters, this work proposes a methodology that considers inexact scheduling parameters. This condition is present in many physical systems where the scheduling parameters can be affected by noise, offsets, calibration errors, and other factors that have a negative impact on the measurements. A H∞ performance criterion is considered in the design in order to guarantee robustness against sensor noise, disturbance, and inexact scheduling parameters. Then, a set of linear matrix inequalities (LMIs) is derived by the use of a quadratic Lyapunov function. The solution of the LMI guarantees asymptotic stability of the PIO. Finally, the performance and applicability of the proposed method are illustrated through a numerical experiment in a nonlinear system.
Categories
control theory.
Author keywords
qLPV system, Inexact scheduling parameters, PI observer, fault diagnosis
Scientific reference
S. Gomez, F.R. Lopez, G. Valencia, D. Rotondo and J. Enriquez. Actuator and sensor fault estimation based on a proportional-integral quasi-LPV observer with inexact scheduling parameters, 3rd IFAC Workshop on Linear Parameter-Varying Systems, 2019, Eindhoven, The Netherlands, Vol 52:28 of IFAC Papers Online, pp. 100-105.
Follow us!