A new technique for fast detection of progressive faults

In this paper, we propose a new technique for enhancing the residual signal in a way that reduces the rate of false alarms without introducing detection delay. This is achieved by eliminating most of the noise from the residual signal by performing an autoregressive modeling, where the model parameters are estimated in real time by an ordinary Kalman filter. Computer experiments using a Boeing 747 model are used to demonstrate the implementation of the proposed technique.