Extended Kalman Filter-Based Sensor Fault Detection for Nonlinear Stochastic Systems with Strong Noise

The problem of sensor fault detection is investigated in this paper for nonlinear stochastic systems with strong noise. Firstly, the state estimation of nonlinear systems is realized by the extended Kalman filter. To suppress the influence of strong noise, a new residual generator and residual evaluation function are designed using the weighted moving average method. Based on the stochastic analysis theory and the definition of fault detectability, a sufficient condition is derived to guarantee the fault detectability in probability. At the same time, the quantitative relationships are analyzed among fault amplitude, window length, false alarm rate and missed detection rate. Finally, the effectiveness of the proposed fault detection method is verified through a numerical simulation for the rotary steerable system.