AI RESEARCH
A graph neural network based chemical mechanism reduction method for combustion applications
arXiv CS.LG
•
ArXi:2603.22318v1 Announce Type: new Direct numerical simulations of turbulent reacting flows involving millions of grid points and detailed chemical mechanisms with hundreds of species and thousands of reactions are computationally prohibitive. To address this challenge, we present two data-driven chemical mechanism reduction formulations based on graph neural networks (GNNs) with message-passing transformer layers that learn nonlinear dependencies among species and reactions.