Robust Fault Detection for Singular Stochastic Systems

摘要

This paper deals with the problem of fault detection filter(FDF) design for singular stochastic systems.By using an observer-based FDF as a residual generator, the robust fault detection is formulated as a filtering problem. Based on linear matrix inequalities (LMIS) techniques and stability theory of stochastic differential equations, stochastic Lyapunov function method is adopted to design a FDF such that, the filter residual system is sensitive to the fault but robust to the exogenous disturbance. Sufficient conditions are proposed to guarantee the stochastically mean-square stablility with an H∞ performance for the faulty detection system. The existence of a FDF for the system under consideration is achieved in terms of LMIS. Moreover, the expressions of desired fault detection filter are given.