Publication: Physics-Assisted Machine Learning for the Simulation of the Slurry Drying in the Manufacturing Process of Battery Electrodes

In this publication with the full title "Physics-Assisted Machine Learning for the Simulation of the Slurry Drying in the Manufacturing Process of Battery Electrodes: A Hybrid Time-Dependent VGG16-DEM Model", DigiCell partner CNRS/Université de Picardie Jules Verne presents a hybrid Physics-Assisted Machine Learning (PAML) model that integrates Deep Learning (DL) techniques with the classical Discrete Element Method (DEM) to simulate slurry drying during a lithium-ion battery electrode manufacturing process. This model predicts the microstructure evolution leading to the formation of the electrode as a time-series along the drying process. The hybrid approach consists in performing a certain amount of DEM simulation steps, nDEM, after every DL prediction, mitigating the risk of unphysical predictions, like overlapping particles. The PAML model was rigorously tested by evaluating different functional metrics of the predicted electrodes, including density, porosity, tortuosity factor, and radial distribution function. They conducted an in-depth analysis of performance versus accuracy, particularly focusing on the impact of the nDEM hyperparameter, which represents the number of DEM steps executed between two subsequent DL predictions.

The full abstract and publication are available here.

Image from the publication's abstract.