Physics-Informed Explainable Machine Learning for Hybrid State of Health and Remaining Useful Life Estimation of Lithium-Ion Batteries
- 1CoE, “National Center of Mechatronics and Clean Technologies”, 1000 Sofia, Bulgaria
- 2Department of Intelligent Technology in Industry, Faculty of Computer Systems and Technologies, Technical University of Sofia, 1000 Sofia, Bulgaria
- 3Department of Computer Systems, Faculty of Computer Systems and Technologies, Technical University of Sofia, 1000 Sofia, Bulgaria
Abstract
Accurate health assessment is essential for the reliable and cost-effective operation of lithium-ion battery energy storage systems. This study presents a physics-informed explainable machine-learning framework for hybrid State-of-Health (SoH) estimation and Remaining Useful Life (RUL) prediction under heterogeneous simulated operating and degradation conditions. A multi-cell degradation model represents capacity fade, resistance growth, thermal and operational stress, and nonlinear aging acceleration. Hybrid SoH combines capacity- and resistance-related health information, while end of life (EOL) and RUL are defined exclusively by an 80% capacity-retention criterion. Generalization is evaluated using cell-wise and leave-one-cell-out validation, with right-censored cells excluded from exact RUL error calculation. In the primary evaluation, Gradient Boosting achieves a SoH RMSE of 0.340 percentage points and R2 = 0.9979, while Random Forest provides the lower RUL RMSE of 239.5 cycles with R2 = 0.9278. Whole-cell bootstrap analysis, simulated distribution-shift testing, and expanded feature ablation further assess uncertainty and robustness. Explainability analysis indicates strong model associations with cumulative electrothermal stress, charging-stage behavior, and pulse resistance. The results demonstrate strong simulation-based predictive capability while revealing substantial inter-cell variability in RUL generalization and uncertainty. Independent experimental validation remains necessary to establish transferability to physical battery systems.
Related
- A Physics-Informed TCN-LSTM Approach for Interpretable Internal Temperature Estimation of NMC811 Lithium-Ion Batteries
- A Physics-Informed Fourier Neural Operator Framework for Zero-Plating Fast Charging of Lithium-Ion Batteries via Real-Time Electrochemical State Estimation and Barrier-Riding Optimal Control
- Fast State-of-Health Estimation of Lithium-Ion Batteries Using Short-Term Relaxation-Voltage Features Integrated with a Differential Evolution-Optimized Extreme Learning Machine Algorithm
- Remaining Useful Life Prediction of Lithium-Ion Batteries Based on Artificial Neural Network-Supported Wiener Process with Random Effects
- A Study on Lithium-Ion Battery Health Estimation and Remaining Life Prediction Based on Real-World Operational Data
BibTeX
@article{stanchev2026physicsinformed,
title = {{Physics-Informed Explainable Machine Learning for Hybrid State of Health and Remaining Useful Life Estimation of Lithium-Ion Batteries}},
author = {Plamen Stanchev and Nikolay Hinov},
journal = {Batteries},
year = {2026},
month = oct,
doi = {10.3390/batteries12100401},
url = {https://www.mdpi.com/2313-0105/12/10/401},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Physics-Informed Explainable Machine Learning for Hybrid State of Health and Remaining Useful Life Estimation of Lithium-Ion Batteries
AU - Plamen Stanchev
AU - Nikolay Hinov
PY - 2026
DA - 2026/10/06
JO - Batteries
DO - 10.3390/batteries12100401
UR - https://www.mdpi.com/2313-0105/12/10/401
AB - Accurate health assessment is essential for the reliable and cost-effective operation of lithium-ion battery energy storage systems. This study presents a physics-informed explainable machine-learning framework for hybrid State-of-Health (SoH) estimation and Remaining Useful Life (RUL) prediction under heterogeneous simulated operating and degradation conditions. A multi-cell degradation model represents capacity fade, resistance growth, thermal and operational stress, and nonlinear aging acceleration. Hybrid SoH combines capacity- and resistance-related health information, while end of life (EOL) and RUL are defined exclusively by an 80% capacity-retention criterion. Generalization is evaluated using cell-wise and leave-one-cell-out validation, with right-censored cells excluded from exact RUL error calculation. In the primary evaluation, Gradient Boosting achieves a SoH RMSE of 0.340 percentage points and R2 = 0.9979, while Random Forest provides the lower RUL RMSE of 239.5 cycles with R2 = 0.9278. Whole-cell bootstrap analysis, simulated distribution-shift testing, and expanded feature ablation further assess uncertainty and robustness. Explainability analysis indicates strong model associations with cumulative electrothermal stress, charging-stage behavior, and pulse resistance. The results demonstrate strong simulation-based predictive capability while revealing substantial inter-cell variability in RUL generalization and uncertainty. Independent experimental validation remains necessary to establish transferability to physical battery systems.
SN - 2313-0105
ER -