Remaining useful life prediction of turbofan engine using global health degradation representation in federated learning

Xi Chen, Hui Wang, Siliang Lu, Jiawen Xu and Ruqiang Yan

Reliability Engineering and System Safety, 2023, vol. 239, issue C

Abstract: In recent years, deep neural networks have been widely applied in remaining useful life (RUL) prediction, and good prognostic performance has been achieved. However, existing centralized learning methods often ignore data privacy, modeling efficiency, and common feature of learning tasks. This paper presents a new RUL prediction method using global health degradation representation (GHDR) in federated learning (FL) framework named GHDR-FL, which aims to extract GHDR from distributed datasets and build personalized models for multiple clients. Specifically, GHDR is an aggregation of shallow features learned by the FL server and clients jointly. The head of each model is the unique superstructure, which is adopted to extract high-level features from the GHDR and local data on the client side. With the GHDR and unique superstructures, RUL prediction models customized for different operating conditions and fault modes can be built simultaneously in the FL. A degradation dataset of turbofan engines is used to evaluate the proposed method. The experimental results show that the GHDR-FL has high accuracy than the centralized learning methods, and the ready-made GHDR has strong versatility.

Keywords: Convolutional neural network; Federated learning; Global health degradation representation; Gated recurrent unit; Remaining useful life prediction (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0951832023004258
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:reensy:v:239:y:2023:i:c:s0951832023004258

DOI: 10.1016/j.ress.2023.109511

Access Statistics for this article

Reliability Engineering and System Safety is currently edited by Carlos Guedes Soares

More articles in Reliability Engineering and System Safety from Elsevier
Bibliographic data for series maintained by Catherine Liu ().