EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Physics‐Based Inverse Modeling of Battery Degradation with Bayesian Methods

Micha C J Philipp iD, Yannick Kuhn iD, Arnulf Latz iD, Birger Horstmann iD

ChemSusChem · 27.07.2025 · 18 (15)

DOI: 10.1002/cssc.202402336 ↗

Kurz gefasst

Fachartikel in der Zeitschrift ChemSusChem, online erschienen am 09.07.2025. Verfasst von Micha C J Philipp und 3 weiteren Autor:innen. Beteiligte Einrichtung: Institute for Engineering Thermodynamics, German Aerospace Center (DLR), Wilhelm-Runge-Straße 10, 89081, Ulm, Germany. Die Arbeit wurde bisher 4-mal zitiert (Crossref) und nennt 52 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: European Union's Horizon 2020 research and innovation program, German Research Foundation, Helmholtz Association.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

To further improve lithium‐ion batteries, a profound understanding of complex battery processes is crucial. Physical models offer understanding but are difficult to validate and parameterize. Therefore, automated machine‐learning methods are necessary to evaluate models with experimental data. Bayesian methods, e.g., Expectation Propagation + Bayesian Optimization for Likelihood‐Free Inference (EP‐BOLFI), stand out as they capture uncertainties in models and data while granting meaningful parameterization. An important topic is prolonging battery lifetime, which is limited by degradation, such as the solid‐electrolyte interphase (SEI) growth. As a case study, EP‐BOLFI is applied to parametrize SEI growth models with synthetic and real degradation data. EP‐BOLFI allows for incorporating human expertise in the form of suitable feature selection, which improves the parametrization. It is shown that even under impeded conditions, correct parameterization is achieved with reasonable uncertainty quantification, needing less computational effort than standard Markov Chain Monte Carlo methods. Additionally, the physically reliable summary statistics show if parameters are strongly correlated and not unambiguously identifiable. Further, Bayesian Alternately Subsampled Quadrature (BASQ) is investigated, which calculates model probabilities, to confirm electron diffusion as the best theoretical model to describe SEI growth during battery storage.

Schlagwörter

Stichwörter der Autor:innen

Bayesian methodsinverse modelinglithium‐ion batteriesmachine learningmodel selectionparameterizationsuncertainty quantifications

Autor:innen und Institutionen

  • Institute for Engineering Thermodynamics, German Aerospace Center (DLR), Wilhelm-Runge-Straße 10, 89081, Ulm, Germany.
  • Theory of Electrochemical Materials, Helmholtz Institute Ulm (HIU), Helmholtzstraße 11, 89081, Ulm, Germany.
  • Faculty of Natural Sciences, Ulm University, Albert-Einstein-Allee 11, 89081, Ulm, Germany.

ORCID-Profile: Micha C J Philipp, Yannick Kuhn, Arnulf Latz, Birger Horstmann

Förderung

  • European Union's Horizon 2020 research and innovation program
  • German Research Foundation
  • Helmholtz Association

Zitieren

APA

Philipp, M. C. J., Kuhn, Y., Latz, A., & Horstmann, B. (2025). Physics‐Based Inverse Modeling of Battery Degradation with Bayesian Methods. ChemSusChem, 18(15). https://doi.org/10.1002/cssc.202402336

Vancouver

Philipp MCJ, Kuhn Y, Latz A, Horstmann B. Physics‐Based Inverse Modeling of Battery Degradation with Bayesian Methods. ChemSusChem. 2025;18(15). doi:10.1002/cssc.202402336

BibTeX
@article{Philipp2025,
  author = {Philipp, Micha C J and Kuhn, Yannick and Latz, Arnulf and Horstmann, Birger},
  title = {Physics‐Based Inverse Modeling of Battery Degradation with Bayesian Methods},
  journal = {ChemSusChem},
  year = {2025},
  volume = {18},
  number = {15},
  doi = {10.1002/cssc.202402336}
}
Kostenlos starten →