EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multiview state-of-health estimation for lithium-ion batteries using time–frequency image fusion and attention-based deep learning

Peijian Jin, Shuo Yang iD, Xinwan Xu, Chaoqun Li, Shihao Guo, Wei Yan, Hui Miao, Shimei Sun

PLOS One · 2025

DOI: 10.1371/journal.pone.0335351 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PLOS One, 2025. Verfasst von Peijian Jin und 7 weiteren Autor:innen. Beteiligte Einrichtung: School of Emergency Science and Engineering, Jilin Jianzhu University, Changchun, Jilin, China. Hauptthemen laut MEDLINE: Deep Learning, Electric Power Supplies, Lithium. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Jilin Province.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Lithium-ion batteries are high-performance energy storage devices that have been widely used in a variety of applications. Accurate early-stage prediction of their remaining useful life is essential for preventing failures and mitigating safety risks. This study proposes a novel multiview approach for estimating the State-of-Health (SOH) of lithium-ion batteries by integrating time-domain and time–frequency features. Firstly, time-domain signals are transformed into time–frequency images using a wavelet transform. Three representative features are then selected and converted into grayscale images, which are combined into three-channel color images as inputs for a convolutional neural network (CNN) to extract spatial features. These features are subsequently passed into a long short-term memory (LSTM) network to capture spatial dependencies. In parallel, raw temporal features are processed through a two-stage attention mechanism to explore both temporal and spatial correlations, followed by another LSTM to model temporal dependencies. The outputs from the two branches are fused using weighted integration and passed through a fully connected layer to generate the final SOH estimate. Comparative experiments with four baseline models demonstrate that the proposed time–frequency fusion architecture significantly enhances prediction accuracy, and that each component makes a meaningful contribution to the overall performance.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Deep LearningElectric Power SuppliesLithiumNeural Networks, ComputerHumansWavelet Analysis

Substanzen

Lithium

Autor:innen und Institutionen

  • School of Emergency Science and Engineering, Jilin Jianzhu University, Changchun, Jilin, China.
  • Jilin Provincial Key Laboratory of Fire Risk Prevention and Emergency Rescue for Building, Changchun, PR China.
  • Ji Lin Sinopoly New Energy Technology Co, Ltd., Liaoyuan, Jilin, China.

ORCID-Profile: Shuo Yang

Förderung

  • Natural Science Foundation of Jilin Province

Zitieren

APA

Jin, P., Yang, S., Xu, X., Li, C., Guo, S., Yan, W., Miao, H., & Sun, S. (2025). Multiview state-of-health estimation for lithium-ion batteries using time–frequency image fusion and attention-based deep learning. PLOS One. https://doi.org/10.1371/journal.pone.0335351

Vancouver

Jin P, Yang S, Xu X, Li C, Guo S, Yan W, et al. Multiview state-of-health estimation for lithium-ion batteries using time–frequency image fusion and attention-based deep learning. PLOS One. 2025. doi:10.1371/journal.pone.0335351

BibTeX
@article{Jin2025,
  author = {Jin, Peijian and Yang, Shuo and Xu, Xinwan and Li, Chaoqun and Guo, Shihao and Yan, Wei and Miao, Hui and Sun, Shimei},
  title = {Multiview state-of-health estimation for lithium-ion batteries using time–frequency image fusion and attention-based deep learning},
  journal = {PLOS One},
  year = {2025},
  doi = {10.1371/journal.pone.0335351}
}
Kostenlos starten →