Utkarsh Vijay
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Utkarsh Vijay has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Alejandro A. Franco
UMR CNRS, Université de Picardie Jules Verne, 15, rue Baudelocque, Amiens Cedex, France, 80039
2 shared publications - Diego E. Galvez-Aranda
Texas A&M University, USA
1 shared publication - Diego Eduardo Galvez-Aranda
Laboratoire de Réactivité et Chimie des Solides (LRCS), Université de Picardie Jules Verne, Hub de l’Energie, UMR CNRS 7314,15 rue Baudelocque, 80039 Amiens, France
1 shared publication - Franco Martín Zanotto
Laboratoire de Réactivité et Chimie des Solides (LRCS)
1 shared publication - Hartmut Schlenz
Forschungszentrum Juelich, Institute of Energy and Climate Research (IEK), IEK-1: Materials Synthesis and Processing, Wilhelm-Johnen-Strasse, D-52425 Juelich, Germany
1 shared publication - Norbert Heribert Menzler
Forschungszentrum Jülich GmbH
1 shared publication
Latest Publications
- Open Access
- 14 June 2026
- Published by Elsevier in Journal of Power Sources
- Vol. 688
Deep learning-enhanced physical modelling for tape-casting slurry microstructures of solid oxide cell substrates
- 14 June 2026
- Published by Elsevier in Journal of Power Sources
- Vol. 688
Open AccessDeep learning-enhanced physical modelling for tape-casting slurry microstructures of solid oxide cell substrates
- Open Access
- 5 February 2024
- Published by American Chemical Society (ACS) in ChemRxiv
Time-dependent deep learning manufacturing process model for battery electrode microstructure prediction
The manufacturing process of Lithium-ion battery electrodes directly affects the practical properties of the cells, such as their performance, durability, and safety. While computational physics-based modeling has been proved as a useful method to produce insights on the manufacturing properties interdependencies as well as the formation of electrode microstructures, their high computational costs avoid their direct utilization in electrode optimization loops. In this work, we report a novel time-dependent deep learning (DL) model of battery electrodes manufacturing process, demonstrated for calendering of NMC111 electrodes, and trained with time-series data arising from physics-based Discrete Element Method (DEM) simulations. The DL model predictions are validated by comparing evaluation metrics (e.g. MSE and R2 score) and electrode functional metrics (contact surface area, porosity, diffusivity and tortuosity factor), showing very good accuracy with respect to the DEM simulations. Our DL model can remarkably capture the elastic recovery of the electrode upon compression (spring-back phenomenon) and the main 3D electrode microstructure features without using the functional descriptors for its training. Furthermore, our DL model has a significantly less computational cost that the DEM simulations, paving the way towards quasi-real time optimization loops of the 3D electrode architecture predicting the calendering conditions to adopt in order to obtain the desired electrode performance.
- 5 February 2024
- Published by American Chemical Society (ACS) in ChemRxiv
Open AccessTime-dependent deep learning manufacturing process model for battery electrode microstructure prediction
The manufacturing process of Lithium-ion battery electrodes directly affects the practical properties of the cells, such as their performance, durability, and safety. While computational physics-based modeling has been proved as a useful method to produce insights on the manufacturing properties interdependencies as well as the formation of electrode microstructures, their high computational costs avoid their direct utilization in electrode optimization loops. In this work, we report a novel time-dependent deep learning (DL) model of battery electrodes manufacturing process, demonstrated for calendering of NMC111 electrodes, and trained with time-series data arising from physics-based Discrete Element Method (DEM) simulations. The DL model predictions are validated by comparing evaluation metrics (e.g. MSE and R2 score) and electrode functional metrics (contact surface area, porosity, diffusivity and tortuosity factor), showing very good accuracy with respect to the DEM simulations. Our DL model can remarkably capture the elastic recovery of the electrode upon compression (spring-back phenomenon) and the main 3D electrode microstructure features without using the functional descriptors for its training. Furthermore, our DL model has a significantly less computational cost that the DEM simulations, paving the way towards quasi-real time optimization loops of the 3D electrode architecture predicting the calendering conditions to adopt in order to obtain the desired electrode performance.