SOURCE-LINKED INTELLIGENCE

ML-Enhanced CFD Modelling of Carbothermic Reduction of Cathode Material for Optimized and Environmentally Friendly Recycling of Lithium-Ion Batteries

CORDIS · observation · Publication date unknown

erstood due to the lack of predictive computational models. ENDEAVOR addresses this gap by developing the first high-fidelity computational fluid dynamics (CFD) model for CTR of LiCoO2, enhanced with machine learning (ML) for accelerated simulation. The project will integrate detailed multiphase CFD with a trained ML surrogate model to enable fast and accurate predictions of species source terms. This hybrid CFD–ML framework will then feed into a multi-objective optimization strategy (MOBO) to identify operating conditions that maximize metal recovery, reduce greenhouse gas emissions, and improve overall process efficiency. Hosted at the CRECK Modeling Lab (Politecnico di Milano), ENDEAVOR will benefit from world-class expertise in reactive flow modelling, access to high-performance computing (HPC), and a supportive MSCA training environment. The fellowship will empower the researcher through advanced training, two-way knowledge transfer, and interdisciplinary collaboration, laying the foundations for a long-term career at the interface of chemical engineering, computational science,

Read original source ↗ Open in workspace

recordType

award

status

SIGNED

region

EU

value

209483.28

unit

EUR