Back
Full-text
Journal article

Machine Learning for Observables: Reactant to Product State Distributions for Atom-Diatom Collisions.

  • Arnold J Department of Chemistry, University of Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland.
  • Koner D Department of Chemistry, University of Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland.
  • Käser S Department of Chemistry, University of Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland.
  • Singh N Department of Mechanical Engineering, Stanford University Stanford, California 94305, United States.
  • Bemish RJ Air Force Research Laboratory, Space Vehicles Directorate, Kirtland AFB, New Mexico 87117, United States.
  • Meuwly M Department of Chemistry, University of Basel, Klingelbergstrasse 80, CH-4056 Basel, Switzerland.
Show more…
  • 2020-07-24
Published in:
  • The journal of physical chemistry. A. - 2020
Physical and Theoretical Chemistry
English Machine learning based models to predict product state distributions from a distribution of reactant conditions for atom-diatom collisions are presented and quantitatively tested. The models are based on function-, kernel-, and grid-based representations of the reactant and product state distributions. All three methods predict final state distributions from explicit quasi-classical trajectory simulations with R2 > 0.998. Although a function-based approach is found to be more than two times better in computational performance, the grid-based approach is preferred in terms of prediction accuracy, practicability, and generality. For the function-based approach, the choice of parametrized functions is crucial and this aspect is explicitly probed for final vibrational state distributions. Applications of the grid-based approach to nonequilibrium, multitemperature initial state distributions are presented, a situation common to energy and state distributions in hypersonic flows. The role of such models in direct simulation Monte Carlo and computational fluid dynamics simulations is also discussed.
Language
  • English
Open access status
green
Identifiers
Persistent URL
https://sonar.ch/global/documents/92600
Statistics
Document views: 38 File downloads:
  • Full-text: 0