SOURCE-LINKED INTELLIGENCE

Closed-loop machine learning platform for the optimisation of battery materials

CORDIS · observation · Publication date unknown

Closed-loop machine learning platform for the optimisation of battery materials BATMAT will deliver a proof-of-concept, closed-loop machine-learning platform that makes high-fidelity, first-principles screening of battery materials practical for industrial decision cycles. Battery R&D is increasingly constrained by late discovery of instability, slow ion transport, or reactive interfaces, i.e. failure modes that often require expensive electronic-structure simulations under realistic conditions (defects, diffusion pathways, interphases). Building on algorithms and software foundations developed in the ERC Advanced Grant FITMOL, BATMAT will integrate next-generation machine-learning force fields with an accuracy-anchored reference layer based on advanced electronic-structure methods (e.g., hybrid DFT with many-body dispersion, PBE0+MBD-NL. The platform will automate (i) exploration of configurational

Read original source ↗ Open in workspace

recordType

award

status

SIGNED

region

EU

value

150000

unit

EUR