A physics-informed machine learning framework for predictive maintenance applied to turbomachinery assets
Journal of the Global Power and Propulsion Society · 25.05.2021 · (May) · 1-15
DOI: 10.33737/jgpps/134845 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of the Global Power and Propulsion Society, online erschienen am 25.05.2021. Verfasst von Marzia Sepe und 6 weiteren Autor:innen. Die Arbeit wurde bisher 29-mal zitiert (Crossref) und nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0).
Abstract (Originalsprache)
The paper presents an overview of Baker Hughes digital framework for a predictive maintenance service boosted by Machine Learning and asset knowledge, applied to turbomachinery assets. Optimization of the maintenance scenario is performed through a risk model that assesses online health status and probability of failure, by detecting functional anomalies and aging phenomena and evaluating their impact on asset serviceability. Turbomachinery domain knowledge is used to create physics-based models, to configure a severity assessment layer and to properly map maintenance actions to anomaly types. The implemented analytics framework is able also to forecast engine behaviour over the future in order to optimize asset operation and maintenance, minimizing downtime and residual risk. Predictive capabilities are optimized thanks to the hybrid approach, where physics-based knowledge empowers long term prediction accuracy while data-driven analytics ensure fast-events prognostics. Accuracy of the hybrid approach is a differentiator for maintenance optimization, allowing activities to be planned properly and in early advance with respect to outage execution.
Zitieren
APA
Sepe, M., Graziano, A., Badora, M., Stazio, A. D., Bellani, L., Compare, M., & Zio, E. (2021). A physics-informed machine learning framework for predictive maintenance applied to turbomachinery assets. Journal of the Global Power and Propulsion Society, 1-15. https://doi.org/10.33737/jgpps/134845
Vancouver
Sepe M, Graziano A, Badora M, Stazio AD, Bellani L, Compare M, et al. A physics-informed machine learning framework for predictive maintenance applied to turbomachinery assets. Journal of the Global Power and Propulsion Society. 2021:1-15. doi:10.33737/jgpps/134845
BibTeX
@article{Sepe2021,
author = {Sepe, Marzia and Graziano, Antonino and Badora, Maciej and Stazio, Alessandro Di and Bellani, Luca and Compare, Michele and Zio, Enrico},
title = {A physics-informed machine learning framework for predictive maintenance applied to turbomachinery assets},
journal = {Journal of the Global Power and Propulsion Society},
year = {2021},
number = {May},
pages = {1--15},
doi = {10.33737/jgpps/134845}
}
EUVIMED