An Observer-based Approach To Robust Fault Detection And Its Application

Observer-based approach is an important one which makes full use of the mathematical model. The benefits of the observer-based fault diagnosis scheme are with high robustness, high reliability and safety. On the basis of reading lots of reference, observer theory is investigated in this paper. The fault detection filter is not only effective, but also robust.In this paper, the first step is to introduce the latest development of fault detection and diagnosis briefly. The basic fault diagnosis approach based on observers for control systems is discussed. The second step is proposing a full-order unknown input observer for uncertain dynamic system with unknown input disturbance. The residual is designed to be insensitive to unknown disturbances, whilst sensitive to faults. Simulation results show the efficiency of this method is good. Then the design of fault detection filter with robust eigenstructure assignment is investigated in a class of residual signal for the unknown input disturbance can not be completely decoupled. Simulation results show this method can be accurate in detecting sensor failure. In the end, the optimal decoupling strategy for robust fault detection filter method and its application for large-scale interconnected systems are also discussed.