Filtering techniques for the system prognosis

This paper, concerns the model-based prognosis approach. More precisely, it is about the use of the Bayesian filtering on the prognosis problem. The aim of the proposed approach is to identify the damage function parameter by using an Ensemble Kalman Filter (EnKF), then, estimate the Remaining Useful Life (RUL) by a probabilistic analysis of this damage trajectories. The relevance of the proposed approach is illustrated through the helicopter gearbox crack propagation.