Published: May 28, 2024
A longstanding project of the chemical kinetics community is to predict reaction rates and behavior reacting systems, even for systems where there are no experimental data. Many important (atmosphere, combustion, pyrolysis, partial oxidations) involve a large number reactions occurring simultaneously, intermediates that have never been observed, making this goal more challenging. Improvements in our ability compute rate coefficients other parameters accurately from first principles, improvements automated kinetic modeling software, partially overcome many challenges. Indeed, some cases quite complicated models constructed which predicted results independent experiments. However, process constructing models, deciding measure or ab initio, relies on accurate estimates (and indeed most numerical estimates.) Machine-learned trained datasets can improve accuracy these estimates, allow better integration quantum chemistry The need continued development shared (perhaps open-source) software databases, directions improvement, highlighted. As we model weaknesses traditional ways doing modeling, testing exposed, identifying several challenges future research by
Language: Английский