Remaining useful life prediction of lithium-ion batteries based on IBKA-VMD and TCN-AM-ESN
Physica Scripta · 12.06.2026 · 101 (23) · 236001
DOI: 10.1088/1402-4896/ae73f8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Physica Scripta, online erschienen am 05.06.2026. Verfasst von Shuyi Xue und 1 weiteren Autor:innen. Die Arbeit nennt 41 Literaturangaben.
Abstract (Originalsprache)
The remaining useful life of lithium-ion batteries predicted accurately is essential for safety and reliability in various applications, from electric vehicles to portable electronics. However, conventional model-based methods and data-driven methods often struggle with capturing the complex nonlinear degradation dynamics of lithium-ion batteries, suffer from high sensitivity to noise and parameter initialization, and exhibit limited generalization capability when applied across different datasets or operating conditions. To overcome these challenges, this paper proposes a hybrid framework that integrates an improved black-winged kite algorithm (IBKA)-variational mode decomposition (VMD) with a temporal convolutional network (TCN) enhanced by the attention mechanism (AM) and echo state network (ESN). IBKA-VMD adaptively decomposes the raw capacity sequences into informative intrinsic mode functions, where the chaotic initialization and enhanced migration strategies of IBKA optimize the quality of the decomposition. Subsequently, the TCN-AM-ESN leverages residual modules and AM to capture both local degradation patterns and long-term dependencies, thus improving predictive stability. Comprehensive experimental validation involving NASA and CALCE public datasets, as well as real-world electric vehicle operational data, demonstrates that the proposed IBKA-VMD-TCN-AM-ESN model achieves superior accuracy and robustness compared to baseline methods. The results confirm the model’s effectiveness in reducing noise sensitivity and maintaining high generalization performance even under complex, dynamic real-world driving conditions.
Zitieren
APA
Xue, S., & Li, L. (2026). Remaining useful life prediction of lithium-ion batteries based on IBKA-VMD and TCN-AM-ESN. Physica Scripta, 101(23), 236001. https://doi.org/10.1088/1402-4896/ae73f8
Vancouver
Xue S, Li L. Remaining useful life prediction of lithium-ion batteries based on IBKA-VMD and TCN-AM-ESN. Physica Scripta. 2026;101(23):236001. doi:10.1088/1402-4896/ae73f8
BibTeX
@article{Xue2026,
author = {Xue, Shuyi and Li, Lei},
title = {Remaining useful life prediction of lithium-ion batteries based on IBKA-VMD and TCN-AM-ESN},
journal = {Physica Scripta},
year = {2026},
volume = {101},
number = {23},
pages = {236001},
doi = {10.1088/1402-4896/ae73f8}
}
EUVIMED