Model ensemble-based prognostic framework for fatigue crack growth prediction - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Model ensemble-based prognostic framework for fatigue crack growth prediction

Résumé

The demand for online fatigue crack growth prognosis has recently increased in industry in order to prevent severe unexpected failures in equipment operated in evolving conditions where static models may no longer perform well. To address this issue, a robust prognostic framework is presented in this paper to assess the reliability of deteriorating equipment due to fatigue crack growth. In this framework, a new model ensemble methodology that integrates multiple stochastic crack growth models based on the quadratic best-worst weighted voting (QBWWV) is proposed for predicting the remaining useful life (RUL) of equipment. To validate the effectiveness of the proposed framework, a case study concerning fatigue crack growth is demonstrated. The results indicate that the proposed prognostic framework outperforms single crack growth models in terms of prediction accuracy under evolving operating conditions. Keywords-Prognostics and Health Management (PHM), remaining useful life (RUL), recursive Bayesian, dynamic ensemble, fatigue crack growth.
Fichier principal
Vignette du fichier
ICSRS2017-89.pdf (1.29 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01784274 , version 1 (26-02-2020)

Identifiants

Citer

Hoang-Phuong Nguyen, Enrico Zio, Jie Liu. Model ensemble-based prognostic framework for fatigue crack growth prediction. 2017 2nd International Conference on System Reliability and Safety (ICSRS), Dec 2017, Milan, Italy. ⟨10.1109/ICSRS.2017.8272843⟩. ⟨hal-01784274⟩
85 Consultations
37 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More