Please use this identifier to cite or link to this item:
http://bura.brunel.ac.uk/handle/2438/29859
Title: | PFFN: A Parallel Feature Fusion Network for Remaining Useful Life Early Prediction of Lithium-ion Battery |
Authors: | Dong, Z Yang, M Wang, J Wang, H Lai, CS Ji, X |
Keywords: | remaining useful life (RUL);early prediction;lithium-ion battery;Bayesian optimization |
Issue Date: | 12-Jul-2024 |
Publisher: | Institute of Electrical and Electronics Engineers (IEEE) |
Citation: | Dong, Z. et al. (2024) 'PFFN: A Parallel Feature Fusion Network for Remaining Useful Life Early Prediction of Lithium-ion Battery', IEEE Transactions on Transportation Electrification, 0 (early access), pp. 1 - 11. doi: 10.1109/TTE.2024.3427334. |
Abstract: | Remaining useful life (RUL) early prediction of lithium-ion battery is crucial to develop advanced battery health management and complete security assessment. However, most of existing methods still suffer from two limitations, i.e., the inadaptability to the different data distribution and the inability to capture the relationship between input series and RUL, which always make the RUL early prediction difficult and challengeable. To address these issues, this paper proposes a parallel feature fusion network (PFFN) for RUL early prediction of lithium-ion battery. Firstly, a feature selection strategy is designed to filter the optimal feature sets (containing cycle statistical features and domain knowledge-based features) that are most related to RUL of lithium-ion battery. Secondly, two specific Transformer encoders connected in parallel configuration are developed to integrate the cycle statistical features and domain knowledge-based features, respectively, achieving original RUL early prediction results. Furthermore, the Bayesian optimization is applied for global iterative optimization, aiming to enhance the prediction accuracy and generalization capability. A series of experiments are conducted with different data distributions. Experimental results demonstrate that the proposed PFFN outperforms the state-of-the-art (SOTA) methods, achieving 6.00%~27.61%, 0.58%~6.49%, and 5.95%~7.03% reduction in Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE), and Score respectively. |
URI: | https://bura.brunel.ac.uk/handle/2438/29859 |
DOI: | https://doi.org/10.1109/TTE.2024.3427334 |
Other Identifiers: | ORCiD: Zhekang Dong https://orcid.org/0000-0003-4639-3834 ORCiD: Junfan Wang https://orcid.org/0000-0001-8403-2875 ORCiD: Hao Wang https://orcid.org/0000-0002-4280-554X ORCiD: Chun Sing Lai https://orcid.org/0000-0002-4169-4438 ORCiD: Xiaoyue Ji https://orcid.org/0000-0002-3526-5215 |
Appears in Collections: | Dept of Electronic and Electrical Engineering Research Papers |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
FullText.pdf | Copyright © 2024 Institute of Electrical and Electronics Engineers (IEEE). Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works (see: https://journals.ieeeauthorcenter.ieee.org/become-an-ieee-journal-author/publishing-ethics/guidelines-and-policies/post-publicationpolicies/). | 3.5 MB | Adobe PDF | View/Open |
Items in BURA are protected by copyright, with all rights reserved, unless otherwise indicated.