Project Grant 2154446

This $294,555 Project Grant from the National Science Foundation's Division of Chemical, Bioengineering, Environmental, and Transport Systems, within the Engineering program (CFDA 47.041), will fund research at San Diego State University to develop neural network-based preconditioning methods to improve the computational efficiency of chemical property evaluations in reactive flow simulations. Specifically, the grantee will combine machine learning and adaptive tabulation approaches to more quickly assess chemical properties, allowing for faster and more accurate combustion simulations using complex chemical models. The grantee organization, San Diego State University Research Foundation, will support graduate and undergraduate students involved in the research, providing valuable training. The proposed techniques aim to reduce the computational cost of detailed chemical simulations, which will enable their use in engineering applications such as combustion devices. If successful, this work could significantly accelerate reactive flow simulations.