Fault Detection for Stochastic Time-Varying Systems Based on Adaptive Disturbance Rejection Observer

The paper studies the state estimation problem for a class of time-varying stochastic systems with unknown disturbance. A type of adaptive disturbance rejection observer is proposed, which can achieve the unknown disturbance decoupling with the state estimation error. The observer design procedure is given to realize the minimum variance estimation. The sufficient and necessary condition for the existence of the observer is presented, which can provide the state estimation error with disturbance decoupling and minimum variance properties. Finally, a fault diagnosis algorithm based on the designed observer is investigated and simulation results illustrate the effectiveness of the proposed approach.