EUVIMEDOpdag EUVIMED
Open access · CC BY 4.0

Unified Deep Learning Framework for Interpretable Remaining Useful Life Prediction of Lithium–Ion Batteries under Stochastic Group‐Behavioral Usage Scenarios

Yong Jin Jeong iD, Hyunjae Kim, Sangwon Jung, Seongha An iD, Yeonwoo Chae, Gahyun Lee, Inwoo Kim iD, Jun Hee Lee, Chun Yong Kang, Jang Wook Choi iD, Taesup Moon iD

Advanced Science · 2026-11-15

DOI: 10.1002/advs.78745 ↗

At a glance

Journal article in Advanced Science, published online on 2026-11-15. Written by Yong Jin Jeong and 10 co-authors. Institution: Interdisciplinary Program in Artificial Intelligence, Seoul National University, Seoul, Republic of Korea. 74 references. The full text is freely available (open access, CC BY 4.0). Funded by: National Research Foundation of Korea, National Research Council of Science and Technology, Institute for Information and Communications Technology Planning and Evaluation.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

With the widespread use of lithium–ion batteries (LIBs) across various industries, accurate remaining useful life (RUL) prediction is essential for end‐users. However, stochastic degradation patterns under unpredictable real‐world operations pose a fundamental challenge to conventional prognostic models. This study develops a unified deep learning‐based framework for RUL prediction under group‐behavioral usage scenarios. A group‐behavioral battery dataset is constructed using synthesized charging protocols characterized by a unique Markov transition matrix designed to capture distinct group usage patterns. Systematic feature engineering identifies 13 health indicators (HIs) with strong predictive relevance to RUL. By utilizing these HIs across multiple learning models, the framework offers a flexible paradigm where lightweight models provide high‐fidelity predictions or integrate into a weighted ensemble scheme to enhance accuracy and robustness. Across three heterogeneous datasets with varying cell chemistries and group behavioral patterns, the framework consistently yielded normalized root mean square errors (NRMSEs) 6.20%. Benchmarking against state‐of‐the‐art (SOTA) architectures showed that systematic feature selection paired with lightweight learning models remained highly competitive with the tested complex architectures. By elucidating the underlying mechanisms of the HIs, the proposed framework enables interpretable and specialized RUL prediction tailored to group‐specific driving patterns, highlighting its potential for real‐world applications.

Subjects

Author keywords

deep learningfeature engineeringgroup‐behavioral usage scenarioslithium–ion batterymarkov matrixremaining useful life

Authors and institutions

  • Interdisciplinary Program in Artificial Intelligence, Seoul National University, Seoul, Republic of Korea.
  • School of Chemical and Biological Engineering and Institute of Chemical Process, Seoul National University, Seoul, Republic of Korea.
  • Hyundai Motor Group-Seoul National University (HMG-SNU) Joint Battery Research Center (JBRC), Seoul National University, Seoul, Republic of Korea.
  • Department of Electrical and Computer Engineering, Seoul National University, Seoul, Republic of Korea.
  • Department of Industrial Engineering, Seoul National University, Seoul, Republic of Korea.
  • Department of Applied Biology and Chemistry, Seoul National University, Seoul, Republic of Korea.

ORCID profiles: Yong Jin Jeong, Seongha An, Inwoo Kim, Jang Wook Choi, Taesup Moon

Funding

  • National Research Foundation of Korea
  • National Research Council of Science and Technology
  • Institute for Information and Communications Technology Planning and Evaluation

Cite

APA

Jeong, Y. J., Kim, H., Jung, S., An, S., Chae, Y., Lee, G., Kim, I., Lee, J. H., Kang, C. Y., Choi, J. W., & Moon, T. (2026). Unified Deep Learning Framework for Interpretable Remaining Useful Life Prediction of Lithium–Ion Batteries under Stochastic Group‐Behavioral Usage Scenarios. Advanced Science. https://doi.org/10.1002/advs.78745

Vancouver

Jeong YJ, Kim H, Jung S, An S, Chae Y, Lee G, et al. Unified Deep Learning Framework for Interpretable Remaining Useful Life Prediction of Lithium–Ion Batteries under Stochastic Group‐Behavioral Usage Scenarios. Advanced Science. 2026. doi:10.1002/advs.78745

BibTeX
@article{Jeong2026,
  author = {Jeong, Yong Jin and Kim, Hyunjae and Jung, Sangwon and An, Seongha and Chae, Yeonwoo and Lee, Gahyun and Kim, Inwoo and Lee, Jun Hee and Kang, Chun Yong and Choi, Jang Wook and Moon, Taesup},
  title = {Unified Deep Learning Framework for Interpretable Remaining Useful Life Prediction of Lithium–Ion Batteries under Stochastic Group‐Behavioral Usage Scenarios},
  journal = {Advanced Science},
  year = {2026},
  doi = {10.1002/advs.78745}
}
Start for free →