Degradation modeling and remaining useful life prediction for dependent competing failure processes

Remaining useful life (RUL) prediction is critical for ensuring the safe and efficient operation of machinery. Due to the existence of multiple influencing factors, the degradation of machinery is often described as dependent competing failure processes (DCFPs). Extensive studies have been conducted on the degradation modeling and RUL prediction for DCFPs. However, they suffer from two limitations: 1) no analytical expression is available for RUL prediction under the first passage time (FPT) concept, and 2) the offline estimation and online update of parameters have not been jointly addressed. Faced with these limitations, this paper investigates the degradation modeling and RUL prediction for DCFPs. The considered DCFPs comprise of soft failure processes subject to gradual degradation and random shocks, and hard failure processes induced by random shocks. First, degradation models for both soft and hard failure processes are formulated, and the FPT-based analytical expression of RUL is correspondingly derived. Second, the offline estimation and online update of parameters are jointly addressed. A sequential estimation scheme is developed for offline estimation, then the estimated results are updated using a specifically designed total variation multiple model particle filter. Finally, a numerical example and an experimental study are provided for demonstration.

Automated summary

This paper investigates the degradation modeling and remaining useful life (RUL) prediction for dependent competing failure processes (DCFPs). By formulating models for both soft and hard failure processes and addressing the offline estimation and online update of parameters, a new approach is proposed to tackle the challenges in RUL prediction.