Abstract
Joint remaining useful life (RUL) prediction and capacity estimation require representations of both gradual degradation and recent battery behavior. This paper presents a cross-expert framework using partial-charging measurements without requiring measured historical full-cycle capacity as an input. The RUL Expert captures long-term degradation from nominal 10-minute segments sampled across a 30-cycle history, while the Capacity Expert characterizes recent battery behavior from statistical descriptors of nominal 40-minute segments over ten consecutive cycles. Their complementary representations are integrated through feature-wise linear modulation for joint RUL and capacity prediction. A three-stage training strategy preserves expert-specific degradation knowledge while improving the balance between the two prediction tasks. On two public battery-aging datasets, the framework achieves the lowest reported RUL root-mean-square error among the compared methods while maintaining competitive capacity-estimation accuracy.
@article{tran2026joint,
title={Joint Remaining Useful Life Prediction and Capacity Estimation of Lithium-Ion Batteries Using Partial-Charging Data},
author={Tran, Khoa and Nguyen, Ho-Si-Hung and Moe, Phone Wai Yan and Trinh, Hung-Cuong and Tran, Thi-Hoang-Giang},
journal={arXiv preprint arXiv:2609.21932},
year={2026},
doi={10.48550/arXiv.2609.21932}
}