Fault Detection and Isolation Scheme in Continuous Time Nonlinear Stochastic Dynamical Systems

Abstract

In this paper, the design of a fault detection and isolation scheme is addressed. We work on plants modelled via continuous-time nonlinear uncertain systems, where the uncertainty and the fault function are assumed to be random processes. A model-based analytical redundancy approach, which checks on-line the statistical characteristics of a residual through hypotheses tests, is proposed. Also a simulation example is discussed, with promising results,