Today we continue the series of posts by around the new OPEN ACCESS PUBLICATIONS from project PULSELiON. We present the journal article “Time-Dependent Deep Learning Manufacturing Process Model for Battery Electrode Microstructure Prediction”, which has been published in Advanced Energy Materials (Wiley) on 5 March 2024.
Physics-based manufacturing simulations can give physical insights to understand the calendering process in battery manufacturing at the micrometre level, thus allowing to optimize it. However, they might be computationally costly in some scenarios. Deep Learning models, once properly trained, can perform spectacular predictions with little computional cost. Here we combine the two approaches to describe the process evolution with 3D μm-level resolution quickly and accurately. The next generation of electrodes manufacturing simulations is already here.
Big congratulations to the authors Diego E. Galvez-Aranda (CNRS), Tan Le Dinh (CNRS), Utkarsh Vijay (CNRS), Franco M. Zanotto (CNRS), Alejandro A. Franco (CNRS)
Please feel free to download and read the work and to download the interal cover of the Advanced Energy Materials journal about this work?