Probabilistic Hysteresis Factor Prediction for Electric Vehicle Batteries with Graphite Anodes Containing Silicon

Runyao Yu1,3, Viviana Kleine1,2, Philipp Gromotka2, Thomas Rudolf2, Adrian Eisenmann2,
Gautham Ram Chandra Mouli1, Peter Palensky1, Jochen L. Cremer1,3,
1 Delft University of Technology
2 Porsche Engineering Group
3 Austrian Institute of Technology
Accepted to IEEE Transactions on Transportation Electrification
IEEE arXiv
Teaser Image

Illustration of OCV-SoC curves. a The OCV-SoC relationship provided by the cell manufacturer. b The OCV-SoC relationship with voltage hysteresis, which is particularly significant for silicon-graphite anode-based batteries.

Abstract

Accurate and computationally efficient state-of-charge (SoC) estimation is essential for battery management systems (BMS) in electric vehicles (EVs). However, batteries with silicon-graphite based anodes, which offer higher energy density and improved charging performance, introduce pronounced voltage hysteresis, making SoC estimation particularly challenging. This work introduces a data-driven approach for probabilistic hysteresis factor prediction, with a particular emphasis on applications where silicon-graphite anode-based batteries are involved. A data harmonization framework is proposed to standardize heterogeneous driving cycles across varying operating conditions. Statistical learning and deep learning models are applied to assess performance in predicting the hysteresis factor with uncertainties while considering computational efficiency. Extensive experiments are conducted to evaluate the generalizability of the optimal model configuration in unseen vehicle models through retraining, zero-shot prediction, fine-tuning, and joint training. By addressing key challenges in SoC estimation, this research facilitates the adoption of advanced battery technologies.

Disclaimer

Due to commercial reasons, we are unable to disclose further details regarding the dataset and labeling algorithm provided by Porsche Engineering Group. If the dataset becomes commercially non-sensitive in the future, we will update this website to release additional information.

BibTeX

@article{HysteresisYu2026,
  title   = {Probabilistic Hysteresis Factor Prediction for Electric Vehicle Batteries with Graphite Anodes Containing Silicon},
  author  = {Yu, Runyao and Kleine, Viviana and Gromotka, Philipp and Rudolf, Thomas and Eisenmann, Adrian and Mouli, Gautham Ram Chandra and Palensky, Peter and Cremer, Jochen L.},
  journal = {IEEE Transactions on Transportation Electrification},
  year    = {2026},
}