The research objective is to advance predictive science, specifically prognostic modeling algorithms by integrating physical science with emerging data analytics techniques, to improve the accuracy, robustness, and reliability of probabilistic prediction of a situation occurrence in Dynamical Systems. One of the envisioned applications is predicting system performance degradation as well as remaining useful life (RUL) through stochastic modeling based on variations in sensor measurements. The RUL prediction can subsequently provide the technical basis for intelligent preventive maintenance, which minimizes the machine downtime, maintenance cost, and reliance on human experience for maintenance scheduling.
Specific research tasks include:
- Deriving system health indicators as a function of operating conditions measured by sensors through machine learning techniques (e.g. deep leaning);
- Stochastic modeling, based on multi-mode particle filter equipped with adaptive resampling, to track variations in health indicators for prediction of gradual deterioration with time-varying degradation rates (different coefficients in one degradation function) and modes (different degradation functions);
- Investigating related signal processing and optimization methods to detect transient changes in system health indicators due to abrupt fault occurrence;
- Improving the computational efficiency of particle filtering by further modifying its sampling and resampling strategy and leveraging the parallel computing structure.