Bayesian neural networks with physics-based priors for robust battery health assessment
5 affiliations
- 1Electronics and Computer Science Department, Mondragon University
- 2Electronic Technology Department, University of the Basque Country (UPV/EHU)
- 3KBR, NASA Ames Research Center
- 4Quantia Ingeniería y Consultoría S.L.
- 5Computer Science and Artificial Intelligence Department, University of the Basque Country (UPV/EHU)
Abstract
Accurate and uncertainty-aware health assessment of lithium-ion batteries is essential for ensuring the reliability, safety, and efficiency of energy storage systems. While data-driven methods provide strong predictive performance, they often lack interpretability and reliable uncertainty quantification, whereas physics-based models rely on accurate parameterization and may exhibit limited generalization. To address these challenges, this paper proposes a Bayesian neural network (Bayesian Neural Network (BNN)) framework incorporating physics-based priors derived from electrochemical battery simulations. These priors embed electrochemical domain knowledge directly into the Bayesian inference process via the Kullback–Leibler divergence term of the variational objective, providing a principled regularization mechanism that balances data fidelity with physical consistency. The proposed framework produces a physics-informed posterior predictive distribution and enables end-to-end uncertainty quantification. Its performance is evaluated on the publicly available NASA battery dataset and the Wenzhou Pack Degradation dataset, covering diverse discharge profiles, operating conditions, and battery scales. Across both datasets, the method consistently achieves improvements in predictive accuracy, probabilistic calibration, and robustness compared to standard BNNs with conventional priors and state-of-the-art physics-guided Bayesian approaches. These results highlight the effectiveness of physics-informed priors for reliable battery health prognostics across different chemistries and cell-to-pack validation settings, under controlled nominal operating conditions.
Related
- Canonical residual physics-informed neural networks and algorithmic perturbation landscapes map cell-fate transitions in longitudinal H3K27M glioma
- Ambient RF energy harvesting using voltage doubler rectifier for battery-free agricultural sensor networks: Design, statistical analysis, and machine learning validation
- Physics-informed neural operators for rapid electric field prediction in transcranial magnetic stimulation of healthy and tumor subjects
- Mixture of Plasticizer Chemicals and Child Intelligence: A Weight-of-Evidence and Meta-Analysis-Informed Bayesian Approach in a Cohort with Improved Exposure Assessment
- CO036 | NEUROCOGNITIVE ASSESSMENT OF IMMUNE-MEDIATED THROMBOTIC THROMBOCYTOPENIC PURPURA PATIENTS USING A COMPREHENSIVE TEST BATTERY IN A CROSS-SECTIONAL MRI-INTEGRATED STUDY
BibTeX
@article{alcibar2026bayesian,
title = {{Bayesian neural networks with physics-based priors for robust battery health assessment}},
author = {Jokin Alcibar and Aitor Aguirre-Ortuzar and Ekhi Zugasti and Chetan S. Kulkarni and Juan Fernandez and Jose I. Aizpurua},
journal = {Scientific Reports},
year = {2026},
month = jul,
doi = {10.1038/s41598-026-63104-y},
url = {https://doi.org/10.1038/s41598-026-63104-y},
note = {CC BY},
}
RIS
TY - JOUR
TI - Bayesian neural networks with physics-based priors for robust battery health assessment
AU - Jokin Alcibar
AU - Aitor Aguirre-Ortuzar
AU - Ekhi Zugasti
AU - Chetan S. Kulkarni
AU - Juan Fernandez
AU - Jose I. Aizpurua
PY - 2026
DA - 2026/07/01
JO - Scientific Reports
DO - 10.1038/s41598-026-63104-y
UR - https://doi.org/10.1038/s41598-026-63104-y
AB - Accurate and uncertainty-aware health assessment of lithium-ion batteries is essential for ensuring the reliability, safety, and efficiency of energy storage systems. While data-driven methods provide strong predictive performance, they often lack interpretability and reliable uncertainty quantification, whereas physics-based models rely on accurate parameterization and may exhibit limited generalization. To address these challenges, this paper proposes a Bayesian neural network (Bayesian Neural Network (BNN)) framework incorporating physics-based priors derived from electrochemical battery simulations. These priors embed electrochemical domain knowledge directly into the Bayesian inference process via the Kullback–Leibler divergence term of the variational objective, providing a principled regularization mechanism that balances data fidelity with physical consistency. The proposed framework produces a physics-informed posterior predictive distribution and enables end-to-end uncertainty quantification. Its performance is evaluated on the publicly available NASA battery dataset and the Wenzhou Pack Degradation dataset, covering diverse discharge profiles, operating conditions, and battery scales. Across both datasets, the method consistently achieves improvements in predictive accuracy, probabilistic calibration, and robustness compared to standard BNNs with conventional priors and state-of-the-art physics-guided Bayesian approaches. These results highlight the effectiveness of physics-informed priors for reliable battery health prognostics across different chemistries and cell-to-pack validation settings, under controlled nominal operating conditions.
SN - 2045-2322
ER -