EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Enhanced Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM) attention model with adaptive loss for lithium-ion battery state of health estimation

Guanfu Liao, Mohd Nashrul Mohd Zubir, Hwa Jen Yap, Mohammed A. H. Ali, Mohammad Alkhedher

PeerJ Computer Science · 07.11.2025 · 11 · e3006

DOI: 10.7717/peerj-cs.3006 ↗

Kurz gefasst

Fachartikel in der Zeitschrift PeerJ Computer Science, online erschienen am 07.11.2025. Verfasst von Guanfu Liao und 4 weiteren Autor:innen. Die Arbeit wurde bisher 5-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study proposes an enhanced deep learning framework for accurately estimating the state of health (SOH) of lithium-ion batteries (LIBs), leveraging a refined Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM)-Attention architecture. To improve prediction accuracy and robustness, a novel Multi-Scale Adaptive Wasserstein-Huber Loss (MSAWH Loss) is introduced, which combines the strengths of Huber loss and multi-scale Wasserstein distance to effectively handle outliers and capture complex degradation patterns. Furthermore, an Uncertainty-Guided Adaptive Bayesian Optimization (UGABO) algorithm is employed to optimize model hyperparameters, achieving efficient convergence in high-dimensional and noisy parameter spaces while maintaining a balance between exploration and exploitation. To enhance generalization, four complementary data augmentation techniques—linear interpolation, data slicing, sequence flipping, and cross-sample mixing—are applied to the National Aeronautics and Space Administration (NASA) charge-discharge dataset, significantly enriching the training data. Additionally, a Model Predictive Control (MPC) mechanism is integrated to correct long-horizon prediction errors, dynamically refining outputs based on recent deviations and maintaining consistency with real-world battery constraints. Experimental results demonstrate that the proposed framework outperforms baseline models, achieving a 20.6% reduction in mean absolute error (MAE), an 18.6% reduction in root mean squared error (RMSE), and a 0.1% improvement in R-squared (R 2 ). These findings highlight the effectiveness of integrating adaptive loss design, automated hyperparameter optimization, strategic data augmentation, and feedback-based error correction in advancing SOH prediction performance for lithium-ion batteries.

Zitieren

APA

Liao, G., Zubir, M. N. M., Yap, H. J., Ali, M. A. H., & Alkhedher, M. (2025). Enhanced Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM) attention model with adaptive loss for lithium-ion battery state of health estimation. PeerJ Computer Science, 11, e3006. https://doi.org/10.7717/peerj-cs.3006

Vancouver

Liao G, Zubir MNM, Yap HJ, Ali MAH, Alkhedher M. Enhanced Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM) attention model with adaptive loss for lithium-ion battery state of health estimation. PeerJ Computer Science. 2025;11:e3006. doi:10.7717/peerj-cs.3006

BibTeX
@article{Liao2025,
  author = {Liao, Guanfu and Zubir, Mohd Nashrul Mohd and Yap, Hwa Jen and Ali, Mohammed A. H. and Alkhedher, Mohammad},
  title = {Enhanced Convolutional Neural Network (CNN)-Long Short-Term Memory (LSTM) attention model with adaptive loss for lithium-ion battery state of health estimation},
  journal = {PeerJ Computer Science},
  year = {2025},
  volume = {11},
  pages = {e3006},
  doi = {10.7717/peerj-cs.3006}
}
Kostenlos starten →