Model based fault detection using unknown input observer- state dependent H∞ filter

This paper presents an approach for robust fault detection (FD) for discrete-time non-linear systems by employing a modified version of the unknown input observer (UIO) to form a nonlinear stochastic observer using state dependent H∞ filter (SDHF). The new Nonlinear UIO-SDHF structure for detecting faults in the presence of external disturbances and uncertainties are applied to a nonlinear biological process. Simulation results are given to show the effectiveness of the approach to discriminate disturbance and detect sensor and actuator faults.