Title: Design of a proportional integral observer based on sliding mode principle for uncertain Takagi-Sugeno fuzzy systems: applications to a turbo-reactor

Authors: Ilyes Elleuch; Atef Khedher; Kamel Ben Othman

Addresses: Research Laboratory of Automatic Signal Processing and Image (LARATSI), National School of Engineers of Monastir (ENIM), University of Monastir, Monastir, Tunisia ' LARA: Laboratory of Research in Automatic, High School of Engineering of Tunis, ENIT, Tunis, Tunisia ' Research Laboratory of Automatic Signal Processing and Image (LARATSI), National School of Engineers of Monastir (ENIM), University of Monastir, Monastir, Tunisia

Abstract: This work deals with the problem of state and fault estimation for systems described by Takagi-Sugeno fuzzy systems. The state and fault estimation is made using a proportional integral observer based on the sliding mode principle. Only sensor faults are considered in this work. In order to estimate this kind of fault, a particulate mathematical transformation is used. The application of this mathematical transformation to the initial system output allows to conceive an augmented system where the initial sensor fault appears as an unknown input. An adaptive mathematical form is used for the sign function to facilitate the determination of the proportional gains of the conceived observer. The observer convergence conditions are formulated in the form of linear matrix inequalities (LMI) allowing computing the observer gains and the Lyapunov theory is used to guarantee the system stability with faults. The proposed proportional integral sliding mode observer is applied to turbo-reactor showing the efficiency of the fault and state estimation.

Keywords: state estimation; Takagi-Sugeno; unknown input; sensor faults; multiple model.

DOI: 10.1504/IJAAC.2018.090809

International Journal of Automation and Control, 2018 Vol.12 No.2, pp.179 - 194

Received: 04 Jun 2016
Accepted: 14 Oct 2016
Published online: 28 Mar 2018 *

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