Fault diagnosis for uncertain stochastic nonlinear systems via LMI approach

Abstract

The problem of fault diagnosis observer design is researched for a class of stochastic nonlinear systems with uncertainties. The stochastic nonlinear systems with uncertainties model considered in this paper possesses the disturbances, lock-in-place (LIP) and loss of effectiveness (LOE). The proposed model is a more general LOE-LIP stochastic nonlinear model and covers several common actuator fault tolerant situations. Based on the theory of Lyapunov stability theory, Itô stochastic differential theory and linear matrix inequality approach a robust fault diagnosis observer is designed. When faults occur, the fault diagnosis observer can estimate the fault quickly and accurately. Finally, numerical simulation results show that the fault diagnosis observer can diagnosis the fault rapidly and accurately.