A Hybrid Ensemble Deep Learning Approach for Early Prediction of Battery Remaining Useful Life
Abstract
Accurate estimation of the remaining useful life (RUL) of lithium-ion batteries is critical for their large-scale deployment as energy storage devices in electric vehicles and stationary storage. A fundamental understanding of the factors affecting RUL is crucial for accelerating battery technology development. However, it is very challenging to predict RUL accurately because of complex degradation mechanisms occurring within the batteries, as well as dynamic operating conditions in practical applications. Moreover, due to insignificant capacity degradation in early stages, early prediction of battery life with early cycle data can be more difficult. In this paper, we propose a hybrid deep learning model for early prediction of battery RUL. The proposed method can effectively combine handcrafted features with domain knowledge and latent features learned by deep networks to boost the performance of RUL early prediction. We also design a non-linear correlation-based method to select effective domain knowledge-based features. Moreover, a novel snapshot ensemble learning strategy is proposed to further enhance model generalization ability without increasing any additional training cost. Our experimental results show that the proposed method not only outperforms other approaches in the primary test set having a similar distribution as the training set, but also generalizes well to the secondary test set having a clearly different distribution with the training set. The PyTorch implementation of our proposed approach is available at https://github.com/batteryrullbattery_rul_early_prediction.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Hybrid models in battery SOH estimation improve estimation accuracy and reliability by integrating different types of models.
“By integrating different types of models, these methods can handle the diverse characteristics of battery data, improving both estimation accuracy and reliability.59”
From Battery state of health estimation under fast charging via deep transfer learning (2025) · cites this paper as [59]
Cites
Modeling of Galvanostatic Charge and Discharge of the Lithium/Polymer/Insertion Cell
Doyle, M., Fuller, T. F., & Newman, J. (1993) · Journal of The Electrochemical Society · doi:10.1149/1.2221597 · Cited by 4,328 · cited 1× · see citing works
“Physics-based approaches such as the single particle model (SPM) [8], [9] and pseudo-two-dimensional (P2D) model [10], [11], [12], [13], [14] are based on the electrochemical principles underlying LIBs and can simulate a battery’s current and voltage characteristics from kinetics and transport equations.”
MethodData-driven prediction of battery cycle life before capacity degradation
Severson, K. A., Attia, P. M., Jin, N., Perkins, N., Jiang, B., Yang, Z., Chen, M. H., Aykol, M., Herring, P. K., Fraggedakis, D., et al. (2019) · Nature Energy · doi:10.1038/s41560-019-0356-8 · Cited by 3,228 · see citing works
Simulation and Optimization of the Dual Lithium Ion Insertion Cell
Fuller, T. F., Doyle, M., & Newman, J. (1994) · Journal of The Electrochemical Society · doi:10.1149/1.2054684 · Cited by 2,057 · see citing works
Temporal Ensembling for Semi-Supervised LearningPreprint
Laine, S., & Aila, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1610.02242 · Cited by 1,910 · see citing works
Zhang, Y., Xiong, R., He, H., & Pecht, M. G. (2018) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2018.2805189 · Cited by 1,326 · see citing works
Liu, K., Shang, Y., Ouyang, Q., & Widanage, W. D. (2020) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2020.2973876 · Cited by 838 · see citing works
Review of models for predicting the cycling performance of lithium ion batteries
Santhanagopalan, S., Guo, Q., Ramadass, P., & White, R. E. (2005) · Journal of Power Sources · doi:10.1016/j.jpowsour.2005.05.070 · Cited by 808 · see citing works
A Data-Driven Auto-CNN-LSTM Prediction Model for Lithium-Ion Battery Remaining Useful Life
Ren, L., Dong, J., Wang, X., Meng, Z., Zhao, L., & Deen, M. J. (2021) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2020.3008223 · Cited by 690 · see citing works
Patil, M. A., Tagade, P., Hariharan, K. S., Kolake, S. M., Song, T., Yeo, T., & Doo, S. (2015) · Applied Energy · doi:10.1016/j.apenergy.2015.08.119 · Cited by 631 · see citing works
Review of simplified Pseudo-two-Dimensional models of lithium-ion batteries
Jokar, A., Rajabloo, B., Désilets, M., & Lacroix, M. (2016) · Journal of Power Sources · doi:10.1016/j.jpowsour.2016.07.036 · Cited by 594 · see citing works
Single-Particle Model for a Lithium-Ion Cell: Thermal Behavior
Guo, M., Sikha, G., & White, R. E. (2010) · Journal of The Electrochemical Society · doi:10.1149/1.3521314 · Cited by 572 · see citing works
Weng, C., Cui, Y., Sun, J., & Peng, H. (2013) · Journal of Power Sources · doi:10.1016/j.jpowsour.2013.02.012 · Cited by 561 · see citing works
Wu, J., Zhang, C., & Chen, Z. (2016) · Applied Energy · doi:10.1016/j.apenergy.2016.04.057 · Cited by 531 · see citing works
Self-citationReview—“Knees” in Lithium-Ion Battery Aging Trajectories
Attia, P. M., Bills, A., Planella, F. B., Dechent, P., dos Reis, G., Dubarry, M., Gasper, P. J., Gilchrist, R., Greenbank, S., Howey, D. A., et al. (2022) · Journal of The Electrochemical Society · doi:10.1149/1945-7111/ac6d13 · Cited by 500 · see citing works
Self-citationReniers, J. M., Mulder, G., & Howey, D. A. (2019) · Journal of The Electrochemical Society · doi:10.1149/2.0281914jes · Cited by 464 · see citing works
Li, X., Zhang, L., Wang, Z., & Dong, P. (2019) · Journal of Energy Storage · doi:10.1016/j.est.2018.12.011 · Cited by 445 · see citing works
Self-citationRemaining Useful Life Prediction for Lithium-Ion Battery: A Deep Learning Approach
Ren, L., Li, Z., Hong, S., Zhao, S., Wang, H. H., & Zhang, L. (2018) · IEEE Access · doi:10.1109/access.2018.2858856 · Cited by 434 · see citing works
Ma, G., Zhang, Y. A., Cheng, C., Zhou, B., Hu, P., & Yuan, Y. (2019) · Applied Energy · doi:10.1016/j.apenergy.2019.113626 · Cited by 343 · see citing works
Zhang, S., Zhai, B., Guo, X., Wang, K., Peng, N., & Zhang, X. (2019) · Journal of Energy Storage · doi:10.1016/j.est.2019.100951 · Cited by 336 · see citing works
Towards Long Lifetime Battery: AI-Based Manufacturing and Management
Liu, K., Wei, Z., Zhang, C., Shang, Y., Teodorescu, R., & Han, Q. (2022) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2022.105599 · Cited by 304 · see citing works
Perspective—Combining Physics and Machine Learning to Predict Battery Lifetime
Aykol, M., Gopal, C. B., Anapolsky, A., Herring, P. K., van Vlijmen, B., Berliner, M. D., Bazant, M. Z., Braatz, R. D., Chueh, W. C., & Storey, B. D. (2021) · Journal of The Electrochemical Society · doi:10.1149/1945-7111/abec55 · Cited by 267 · see citing works
Zhang, L., Mu, Z., & Sun, C. (2018) · IEEE Access · doi:10.1109/access.2018.2816684 · Cited by 252 · see citing works
Zhang, Y., Xiong, R., He, H., & PECHT, M. G. (2018) · IEEE Transactions on Industrial Electronics · doi:10.1109/tie.2018.2808918 · Cited by 243 · see citing works
A new hybrid method for the prediction of the remaining useful life of a lithium-ion battery
Chang, Y. L., Fang, H., & Zhang, Y. (2017) · Applied Energy · doi:10.1016/j.apenergy.2017.09.106 · Cited by 231 · see citing works
Multiphase Porous Electrode Theory
Smith, R. B., & Bazant, M. Z. (2017) · Journal of The Electrochemical Society · doi:10.1149/2.0171711jes · Cited by 230 · see citing works
Data-driven hybrid remaining useful life estimation approach for spacecraft lithium-ion battery
Song, Y., Liu, D., Chen, Y., & Peng, Y. (2017) · Microelectronics Reliability · doi:10.1016/j.microrel.2017.06.045 · Cited by 195 · see citing works
Liu, K., Gao, Y., Zhu, C., Li, K., Fei, M., Chen, P., Zhang, X., & Han, Q. (2022) · Control Engineering Practice · doi:10.1016/j.conengprac.2022.105176 · Cited by 179 · see citing works
Chen, L., Zhang, Y., Zheng, Y., Li, X., & Zheng, X. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.07.081 · Cited by 124 · see citing works
Self-citationIntercalation Kinetics in Multiphase-Layered Materials
Smith, R. B., Khoo, E., & Bazant, M. Z. (2017) · The Journal of Physical Chemistry C · doi:10.1021/acs.jpcc.7b00185 · Cited by 114 · see citing works
Self-citationArdeshiri, R. R., & Ma, C. (2021) · International Journal of Energy Research · doi:10.1002/er.6910 · Cited by 105 · see citing works
Fractional modeling and SOC estimation of lithium-ion battery
Ma, Y., Zhou, X., Li, B., & Chen, H. (2016) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2016.7508803 · Cited by 83 · see citing works
Distributed Cooperative Control of Battery Energy Storage Systems in DC Microgrids
Meng, T., Lin, Z., & Shamash, Y. (2021) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2021.1003874 · Cited by 77 · see citing works
Kemper, P., Li, S. E., & Kum, D. (2015) · Journal of Power Sources · doi:10.1016/j.jpowsour.2015.03.134 · Cited by 70 · see citing works
A nonlinear observer approach of SOC estimation based on hysteresis model for lithium-ion battery
Ma, Y., Li, B., Li, G., et al. (2017) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2017.7510502 · Cited by 48 · see citing works
Pang, Z., Si, X., Hu, C., & Zhang, Z. (2021) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2021.1003859 · Cited by 21 · see citing works
Lin, J. C., Zhang, Y., & Khoo, E. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.13661 · Cited by 7 · see citing works
Self-citation
Flagged first, then by citations
Cited by
Sorted by citations
Ma, G., Wang, Z., Liu, W., Fang, J., Zhang, Y., Ding, H., & Yuan, Y. (2023) · IEEE/CAA Journal of Automatica Sinica · doi:10.1109/jas.2023.123531 · Cited by 70 · see citing works
Battery state of health estimation under fast charging via deep transfer learning
Zhao, J., Li, D., Li, Y., Shi, D., Nan, J., & Burke, A. F. (2025) · iScience · doi:10.1016/j.isci.2025.112235 · Cited by 53 · see citing works
“59 60–62 Compared to single models, hybrid models are more effective at processing complex battery data, particularly excelling in handling multimodal and time-series data.”
- Authors
- Xu, Q., Wu, M., Khoo, E., Chen, Z., & Li, X.
- Affiliations
- Agency for Science, Technology and Research (SG)
- Venue
- IEEE/CAA Journal of Automatica Sinica
- Published
- 2023-01
- Volume
- 10
- Issue
- 1
- Pages
- 177-187
- ISSN
- 2329-9266 (ISSN-L)
- 2329-9274
- OpenAlex
- W4313598793
- Cited by
- 119
- References
- 36
- Citations by year
- 2026: 31 · 2025: 43 · 2024: 36 · 2023: 8 · 2018: 1
- FWCI
- 8.58
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE/CAA Journal of Automatica Sinica (publishedVersion)
- Topics
- Physical Sciences › Engineering › Automotive Engineering › Advanced Battery Technologies Research
- Physical Sciences › Engineering › Electrical and Electronic Engineering › Advancements in Battery Materials
- Physical Sciences › Engineering › Control and Systems Engineering › Machine Fault Diagnosis Techniques
- Keywords
- lithium-ion batteries, hybrid deep learning model, ensemble learning, snapshot ensemble, feature selection, capacity fade, distribution shift, battery degradation modeling
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
