State of Health Estimation in EV Batteries Using AI-enhanced BMS
Keywords:
BMS, EV, LSTM, Machine learning, SoHAbstract
The accurate estimation of battery State of Health (SoH) is critical for enhancing the safety, reliability, and lifespan of Electric Vehicle (EV) battery systems. Traditional Battery Management Systems (BMS) rely on model-based approaches, which often struggle to capture nonlinear degradation patterns under diverse operating conditions. This study presents an AI-powered BMS framework that integrates data-driven machine learning algorithms, specifically a hybrid Long Short-Term Memory (LSTM) neural network combined with gradient boosting regression to predict SoH in real time. The method leverages historical voltage, current, temperature, and charge–discharge cycle data from EV lithium-ion batteries, performing feature extraction through time-frequency analysis and dimensionality reduction via Principal Component Analysis (PCA). Model training and validation are conducted using both laboratory cycling datasets and in-field driving profiles. The proposed AI-enhanced BMS demonstrates a prediction accuracy exceeding 97% on test datasets, with an average Root Mean Square Error (RMSE) below 1.5% in SoH estimation. Results also indicate a 20–25% improvement in prediction stability under variable load conditions compared to conventional algorithms. The findings highlight the potential of integrating advanced AI models into EV BMS to enable predictive maintenance, extend battery life, and optimize performance throughout the vehicle lifecycle.
References
J. Lu, R. Xiong, J. Tian, C. Wang, and F. Sun, “Deep learning to estimate lithium-ion battery state of health without additional degradation experiments,” Nature Communications, vol. 14, no. 1, p. 2760, May 2023, doi: https://doi.org/10.1038/s41467-023-38458-w
G. R. Sylvestrinet al., “State of the art in electric batteries’ state of health (SoH) estimation with machine learning: A review,” Energies, vol. 18, no. 3, p. 746, Feb. 2025, doi: https://doi.org/10.3390/en18030746
A. Lanubile, P. Bosoni, G. Pozzato, A. Allam, M. Acquarone, and S. Onori, “Domain knowledge-guided machine learning framework for state of health estimation in lithium-ion batteries,” Communications Engineering, vol. 3, no. 1, Nov. 2024, doi: https://doi.org/10.1038/s44172-024-00304-2
G. Xu, J. Xu, and Y. Zhu, “LSTM-based estimation of lithium-ion battery SoH using data characteristics and spatio-temporal attention,” PLoS ONE, vol. 19, no. 12, pp. e0312856–e0312856, Dec. 2024, doi: https://doi.org/10.1371/journal.pone.0312856
Z. Xiao, B. Jiang, J. Zhu, X. Wei, and H. Dai, “State of health estimation for lithium-ion batteries using an explainable XGBoost model with parameter optimization,” Batteries, vol. 10, no. 11, pp. 394–394, Nov. 2024, doi: https://doi.org/10.3390/batteries10110394
Y. Wang et al., “A comprehensive review of machine learning-based state of health estimation for lithium-ion batteries: data, features, algorithms, and future challenges,” Renewable and Sustainable Energy Reviews, vol. 224, p. 116125, Jul. 2025, doi: https://doi.org/10.1016/j.rser.2025.116125
M. Li et al., “State of health estimation and battery management: A review of health indicators, models and machine learning,” Materials, vol. 18, no. 1, pp. 145–145, Jan. 2025, doi: https://doi.org/10.3390/ma18010145
L. Zhang et al., “Improved LSTM based state of health estimation using random segments of the charging curves for lithium-ion batteries,” Journal of Energy Storage, vol. 74, pp. 109370–109370, Oct. 2023, doi: https://doi.org/10.1016/j.est.2023.109370
W. Taparhudee, R. Jongjaraunsuk, S. Nimitkul, P. Suwannasing, and W. Mathurossuwan, “Optimizing convolutional neural networks, XGBoost, and hybrid CNN-XGBoost for precise Red tilapia (Oreochromis niloticus Linn.) weight estimation in river cage culture with aerial imagery,” AgriEngineering, vol. 6, no. 2, pp. 1235–1251, May 2024, doi: https://doi.org/10.3390/agriengineering6020070
X. Qiu, W. Yan, S. Wang, and K. Chen, “A general multi-source ensemble transfer learning framework for health prognostic of lithium-ion batteries,” Applied Energy, vol. 376, p. 124245, Dec. 2024, doi: https://doi.org/10.1016/j.apenergy.2024.124245
Y. Feng, G. Hu, and Z. Zhang, “GPT4Battery: An LLM-driven framework for adaptive state of health estimation of raw Li-ion batteries,” arXiv (Cornell University), Jan. 2024, doi: https://doi.org/10.48550/arxiv.2402.00068
J. Zhao, D. Li, Y. Li, D. Shi, J. Nan, and A. F. Burke, “Battery state of health estimation under fast charging via deep transfer learning,” iScience, vol. 28, no. 5, p. 112235, May 2025, doi: https://doi.org/10.1016/j.isci.2025.112235
I. T. Jolliffe and J. Cadima, “Principal component analysis: A review and recent developments,” Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences, vol. 374, no. 2065, p. 20150202, Apr. 2016, doi: https://doi.org/10.1098/rsta.2015.0202
S. Hochreiter and J. Schmidhuber, “Long short-term memory,” Neural Computation, vol. 9, no. 8, pp. 1735–1780, Nov. 1997, doi: https://doi.org/10.1162/neco.1997.9.8.1735
T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining - KDD ’16, vol. 1, no. 1, pp. 785–794, Aug. 2016, doi: https://doi.org/10.1145/2939672.2939785
Y. Gal and Z. Ghahramani, “Dropout as a Bayesian approximation: Representing model uncertainty in deep learning,” proceedings.mlr.press, Jun. 11, 2016. https://proceedings.mlr.press/v48/gal16.html
