Fault Detection and Estimation for a Nonlinear System Based on Interval Multiobservers

In this paper the fault detection and estimation of a nonlinear system described by a decoupled multimodel subjected to unknown but bounded disturbances and measurement noise is studied. For detection two residual are generated to determine the occurrence of a fault, whereas for fault estimation an augmented system is introduced and an interval multiobserver is proposed. Sufficient conditions are formed under Linear Matrix Inequalities (LMI) to ensure convergence. Simulation results are presented to show the effectiveness of the proposed method.