Closed-loop Actuator-fault Detection and Isolation using Invariant Sets and Tubes

Abstract

Abstract This paper proposes a new actuator-fault detection and isolation (FDI) strategy for closed-loop discrete time-invariant systems by using invariant sets and tubes. In this approach, invariant sets are used for fault detection (FD) and the establishment of FDI conditions, while the tubes are generated for fault isolation (FI) at transient state. Comparing with the existing set-theoretic FDI techniques, the advantage of this approach consists in that it exhibits a balance between the conservativeness of FDI conditions, the fast FI response and the computational complexity. The effectiveness of the proposed technique is illustrated by a numerical example.