EUVIMEDObjevte EUVIMED
Open access · CC BY-NC-ND 4.0

Transferable Generative Pretrained Transformer Using Low-Rank Adaptation (TransLoRA-GPT): A Framework for Predicting Battery Aging Trajectories from Charging Fragments

Zhenyu Wang, Renquan Lv, Jiahong Liao, Lecheng Lei, Shaoyun Hao iD, Xingwang Zhang iD

Chem & Bio Engineering · 2026-02-26 · 3 (2) · 294-307

DOI: 10.1021/cbe.5c00101 ↗

At a glance

Journal article in Chem & Bio Engineering, published online on 2025-11-26. Written by Zhenyu Wang and 5 co-authors. Institution: Key Laboratory of Biomass Chemical Engineering of Ministry of Education, Department of Chemical and Biological Engineering, Zhejiang University, Hang…. Cited 7 times so far (Crossref); 53 references. The full text is freely available (open access, CC BY-NC-ND 4.0). Funded by: Quzhou Municipal Science and Technology Bureau, Research Funds of the Institute of Zhejiang University-Quzhou.

AbstraktSubjectsBibliografické údajeCiteRelated studies

Abstract (original language)

Accurate battery lifetime prediction is critical for operational safety and battery management. However, limited data and poor model transferability make this task challenging. This paper proposes the transferable generative pretrained transformer using low-rank adaptation (TransLoRA-GPT) to address the challenges. Specifically, depending on whether the source and target domains are the same, a base model and a transferred model are fine-tuned from the large language model (LLM), respectively. In the evaluation phase, by applying feature engineering from the charging data of the first 20 cycles, the model can predict the state of health (SOH) aging trajectory for the subsequent 100 cycles. The results indicate that by fine-tuning only 3.33% of the total parameters, the model outperforms traditional models with an average root-mean-square error (RMSE) of 0.0128. Moreover, it maintains robustness against data scarcity and delivers reliable predictions across the entire battery life cycle. Furthermore, it demonstrates strong transferability across different charging/discharging strategies (average RMSE of 0.0171) and battery types (average RMSE of 0.0594). Overall, this work highlights the potential of LLM-based transferable predictive models.

Subjects

Author keywords

feature engineeringlarge language modellithium-ion batterystate of healthtransfer learning

Authors and institutions

  • Key Laboratory of Biomass Chemical Engineering of Ministry of Education, Department of Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, China.

ORCID profiles: Shaoyun Hao, Xingwang Zhang

Funding

  • Quzhou Municipal Science and Technology Bureau
  • Research Funds of the Institute of Zhejiang University-Quzhou

Cite

APA

Wang, Z., Lv, R., Liao, J., Lei, L., Hao, S., & Zhang, X. (2026). Transferable Generative Pretrained Transformer Using Low-Rank Adaptation (TransLoRA-GPT): A Framework for Predicting Battery Aging Trajectories from Charging Fragments. Chem & Bio Engineering, 3(2), 294-307. https://doi.org/10.1021/cbe.5c00101

Vancouver

Wang Z, Lv R, Liao J, Lei L, Hao S, Zhang X. Transferable Generative Pretrained Transformer Using Low-Rank Adaptation (TransLoRA-GPT): A Framework for Predicting Battery Aging Trajectories from Charging Fragments. Chem & Bio Engineering. 2026;3(2):294-307. doi:10.1021/cbe.5c00101

BibTeX
@article{Wang2026,
  author = {Wang, Zhenyu and Lv, Renquan and Liao, Jiahong and Lei, Lecheng and Hao, Shaoyun and Zhang, Xingwang},
  title = {Transferable Generative Pretrained Transformer Using Low-Rank Adaptation (TransLoRA-GPT): A Framework for Predicting Battery Aging Trajectories from Charging Fragments},
  journal = {Chem & Bio Engineering},
  year = {2026},
  volume = {3},
  number = {2},
  pages = {294--307},
  doi = {10.1021/cbe.5c00101}
}
Start for free →