Fault detection and isolation based on the combination of a bank of interval observers and invariant sets

Feng Xu, Florin Stoican, Vicenc Puig, Carlos Ocampo-Martinez, Sorin Olaru
2013 21st Mediterranean Conference on Control and Automation  
In this paper, a fault detection and isolation (FDI) approach using a bank of interval observers is developed. From the methodological point of view, a bank of interval observers is designed according to different dynamical models of the system under different modes (healthy or faulty). Each interval observer matches one system mode while all the interval observers monitor the system simultaneously. In order to guarantee FDI, a set of FDI conditions based on invariant set notions are
more » ... ions are established. These conditions ensure that the considered faults can be accurately isolated after a period of monitoring time. Finally, simulation results are used to present the effectiveness of the approach. I. INTRODUCTION According to the recent literature [2], [3], [5], interval observers have being successfully used for fault detection (FD) purposes, but not for fault isolation (FI) purposes. The aim of this present paper is to extend the use of interval observers to fault isolation (FI). The interval observer-based FD approach consists in propagating the effect of uncertainties by means of the system mathematical models to generate adaptive thresholds for residuals. Then, the FD is performed by testing the consistency between model predictions and current measurements of the corresponding residuals [2], [5] . Whatever mode the system is under, an interval observer consistent with the current mode model of the system always predict state or output interval vectors that confine the current system states or outputs at each time instant. This mechanism provides useful information to detect and isolate faults when a bank of interval observers is used. An FDI approach based on a bank of set-valued observersdifferent from interval observers-is proposed in [6] . Under a set of assumptions regarding the system, that proposed approach can implement FDI scheme. However, comparatively, the approach proposed in this paper provides a set of definite and pre-checkable FDI conditions that allow to know a priori whether faults are detectable and isolable. Considering the good balance among expressional compactness, computational precision and complexity offered by zonotopes [1], this paper focuses on the representation
doi:10.1109/med.2013.6608816 dblp:conf/med/0006SPOO13 fatcat:62iztshsojbrjcc573rqy3mxpe