Multi-Objective Optimization of Lithium Ion Battery Manufacturing by Using Machine Learning Coupled to Physics-Based Process Modeling

In this lecture I present a digital twin for the accelerated optimization of the manufacturing process of Lithium Ion Batteries (LIBs). This digital twin is developed within the context of our ARTISTIC project¹ and it is supported on a hybrid computational approach encompassing a physics-based process modeling workflow and machine learning models, validated with in house experimental data acquired in our battery pilot line.² This digital twin simulates the different steps along the LIB cells manufacturing process, including the electrode slurry, the coating, the drying, the calendering and the electrolyte infiltration. The physics-based process modeling workflow is supported on the sequential coupling of experimentally-validated Coarse Grained Molecular Dynamics, Discrete Element Method and Lattice Boltzmann simulations which allows predicting the impact of the process parameters on the final electrode microstructures in three dimensions. The predicted electrode microstructures are injected in a continuum performance simulator capturing the influence of the pore networks and spatial location of active material particles and carbon-binder within the electrodes on their electrochemical response. Machine learning models are used to accelerate the physical models’ parameterization and to derive surrogate models capturing with high accuracy the predictions of the physics-based modeling workflow. Such surrogate models are incorporated in a Bayesian Optimization-based algortihmic loop which allows to predict which manufacturing parameters to adopt in order to maximize and minimize several electrode properties simultaneously. The predictive and optimization capabilities of our digital twin are illustrated with results for different electrode formulations and manufacturing parameters. Finally, the free online battery manufacturing simulation services offered by the project³ and designed to support the optimization of battery electrodes manufacturing are illustrated through several examples.<br/><br/><b>1. ERC Consolidator Project ARTISTIC, grant agreement #772873 (</b>https://www.erc-artistic.eu/<b>).</b><br/><b>2. See our publications</b> <b>here:</b> https://www.erc-artistic.eu/scientific-production/publications .<br/><b>3. </b> T. Lombardo, F. Caro, A. C. Ngandjong, J.B. Hoock, M. Duquesnoy, J. C. Delepine, A. Ponchelet, S. Doison, A. A. Franco,. "The ARTISTIC online calculator: exploring the impact of lithium ion battery electrode manufacturing parameters interactively through your browser." <i>Batteries & Supercaps</i> 5, no. 3 (2022): e202100324.