Degradation model based remaining useful life (RUL) prediction is often used in prognostics and health management (PHM). In practice, three-source of variability, i.e., the temporal variability, unit-to-unit variability, and measurement variability, is often encountered in degradation modeling, leading to complex degradation models and great challenge for the parameter estimation and RUL prediction. Commonly, off-line methods are used, which, however, cannot fulfill the real-time requirement of decision-making in the PHM. In this extended abstract, a generic degradation model is introduced, which can characterize the three-source variability and provide a flexible for on-line parameter and RUL updating. An integrated simulation-based filtering method is introduced by feeding the output of a Markov chain Monte Carlo simulation into an extended particle filter, which can fuse the historical trajectories and condition monitoring observations and update the parameter and RUL simultaneously. Critical aspects of the generic model and the filtering method are presented.
Liu, J., West M., (2001). Combined parameter and state estimation in simulation-based filtering. Sequential Monte Carlo Methods in Practice, New York: Springer.
Si, X.-S., Wang, W., Hu C.-H., Zhou, D.-H., (2011). Remaining useful life estimation – A review on the statistical data driven approaches. European Journal of Operational Research, vol. 213, pp. 1-14.
Si, X.-S., Wang, W., Hu, C.-H., Zhou, D.-H., (2014). Estimating remaining useful life with three-source variability in degradation modeling. IEEE Transactions on Reliability, vol. 63, pp. 167-189.
Ye, Z.-S., Xie, M., (2014). Stochastic modelling and analysis of degradation for highly reliable products. Applied Stochastic Models in Business and Industry, vol. 31, pp. 16-32.