A Combined Model-based and Intelligent Method for Small Fault Detection and Isolation of Actuators

This paper proposes a combined method to detect and isolate small faults of actuators in closed-loop control systems. The fault with some tiny magnitude (not larger than the magnitude of disturbances) is mainly considered for the nonlinear system subjected to model uncertainties, disturbances, and noises. The basic idea of our study is to use model-based method to decouple the possible disturbances and isolate faults, and then, appeal to computing intelligence to further reduce the influence of remaining model uncertainties. Specifically, the proposed approach is an extension of the observer-based method with artificial neural networks (ANNs) modeling technique to enhance the performance of the diagnosis system. With the decision logic considered, the small faults of actuators can be detected successfully. This method is applied to a satellite attitude control system and the effectiveness is shown.