Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/32028
Title: | A self-data-driven approach for online remaining useful life prediction of machinery using a recursive update strategy |
Authors: | Xu, P Lei, Y Wang, Z Li, N Cai, X Feng, K |
Keywords: | self-data-driven method;RUL prediction;latest monitoring data;recursive update strategy;online scenarios |
Issue Date: | 4-Mar-2025 |
Publisher: | Elsevier |
Citation: | Xu, P. et al. (2025) 'A self-data-driven approach for online remaining useful life prediction of machinery using a recursive update strategy', Mechanical Systems and Signal Processing, 229, 112541, pp. 1 - 15. doi: 10.1016/j.ymssp.2025.112541. |
Abstract: | Self-data-driven remaining useful life (RUL) prediction of machinery has gained significant attention in engineering due to its potential to reduce dependency on extensive training data during the prediction process. However, existing self-data-driven methods typically necessitate the storage and reuse of entire historical degradation data for updating prediction models, which limits their applicability in online and real-time scenarios. To overcome this limitation, this paper proposes a self-data-driven method for online RUL prediction using a recursive update strategy. Unlike conventional methods, the proposed recursive update strategy updates model parameters and selects the optimal model based on the latest monitoring data. A model base consisting of various degradation functions is initially established. During online prediction, model parameters are continuously updated in real-time using a sequential Bayesian algorithm, while the optimal degradation model is automatically identified using the proposed recursive Bayesian information criterion (RBIC). The effectiveness of this approach is validated through both simulation and experimental case studies. The results demonstrate that the proposed method not only achieves higher prediction accuracy but also requires less computation time compared to existing methods in online applications. |
Description: | Data availability: Data will be made available on request. |
URI: | https://bura.brunel.ac.uk/handle/2438/32028 |
DOI: | https://doi.org/10.1016/j.ymssp.2025.112541 |
ISSN: | 0888-3270 |
Other Identifiers: | ORCiD: Pengcheng Xu https://orcid.org/0000-0002-4393-4511 ORCiD: Zidong Wang https://orcid.org/0000-0002-9576-7401 ORCiD: Xiao Cai https://orcid.org/0000-0003-1494-9578 Article number: 112541 |
Appears in Collections: | Dept of Computer Science Embargoed Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
FullText.pdf | Embargoed until 4 March 2026. Copyright © 2025 Elsevier Ltd. All rights reserved. This manuscript version is made available under the CC-BY-NC-ND 4.0 license https://creativecommons.org/licenses/by-nc-nd/4.0/ (see: https://www.elsevier.com/about/policies/sharing ). | 3.95 MB | Adobe PDF | View/Open |
This item is licensed under a Creative Commons License