Physics-constrained transfer learning : Open-circuit voltage curve reconstruction and degradation mode estimation of lithium-ion batteries

DOI zum Zitieren der Version auf EPub Bayreuth: https://doi.org/10.15495/EPub_UBT_00009094
URN to cite this document: urn:nbn:de:bvb:703-epub-9094-1

Title data

Hofmann, Tobias ; Hamar, Jacob ; Mager, Bastian ; Erhard, Simon ; Schmidt, Jan Philipp:
Physics-constrained transfer learning : Open-circuit voltage curve reconstruction and degradation mode estimation of lithium-ion batteries.
In: Energy and AI. Vol. 20 (2025) . - 100493.
ISSN 2666-5468
DOI der Verlagsversion: https://doi.org/10.1016/j.egyai.2025.100493

	Format: PDF Name: 1-s2.0-S2666546825000254-main.pdf Version: Published Version Available under License Creative Commons BY 4.0: Attribution
	Download (4MB)

Project information

Project title:	Project's official title Project's id Open Access Publizieren No information

Abstract

Open-circuit voltage (OCV) updates are the key to accurate state of charge (SOC) estimates over lifetime. Degradation modes (DM) are directly coupled to OCV estimation. They offer a more detailed analysis of the battery’s state of health (SOH) and yield optimized usage strategy, and with that, a prolonged lifetime. In this study two data-driven models are coupled with physics-based models and compared in regards of their OCV and DM estimation accuracy: Two temporal convolutional — long short-term memory neural networks (TCN-LSTM) are trained from synthetic NCA-graphite battery data for OCV curve estimation (model 1) and alignment parameter estimation (model 2). Both models are fine-tuned with varying amounts of experimental NMC-graphite battery data during the transfer learning (TL) step. In the subsequent physics-constraining part the DMs are derived via optimization (model 1), i.e., fitting the OCV with half cell open-circuit potentials, or directly via mathematical equations (model 2). Both models prove that fine-tuning data from one aging path suffices, if it includes the maximum appearing DMs of the target domain. For these use cases both models maintain OCV mean absolute errors (MAEs), DM MAEs and SOH mean absolute percentage errors (MAPEs) under 10 mV, 3.10 % and 1.98 %, respectively. The model 2 has less computational complexity and reaches slightly better results but requires labeled target data including alignment parameters for its application. This study shows that synthetic data is eligible for TL, even for varying cell chemistries, and that the mechanistic model helps to physically constrain the output.

Further data

Item Type:	Article in a journal
Keywords:	Lithium-ion battery; State of health estimation; Transfer learning; Degradation modes; Mechanistic model
DDC Subjects:	600 Technology, medicine, applied sciences > 620 Engineering
Institutions of the University:	Faculties > Faculty of Engineering Science > Chair Systems Engineering for Electrical Energy Storage > Chair Systems Engineering for Electrical Energy Storage - Univ.-Prof. Dr.-Ing. Jan Philipp Schmidt Research Institutions > Central research institutes > Bayerisches Zentrum für Batterietechnik - BayBatt Faculties Faculties > Faculty of Engineering Science Faculties > Faculty of Engineering Science > Chair Systems Engineering for Electrical Energy Storage Research Institutions Research Institutions > Central research institutes
Language:	English
Originates at UBT:	Yes
URN:	urn:nbn:de:bvb:703-epub-9094-1
Date Deposited:	09 Apr 2026 10:53
Last Modified:	09 Apr 2026 10:54
URI:	https://epub.uni-bayreuth.de/id/eprint/9094

Download Statistics

Download Statistics

Download Statistics

Downloads

Downloads per month over past year