Data-driven construction of a generalized kinetic collision operator from molecular dynamics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhao, Yue, Burby, Joshua W., Christlieb, Andrew, Lei, Huan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909566402297856
author Zhao, Yue
Burby, Joshua W.
Christlieb, Andrew
Lei, Huan
author_facet Zhao, Yue
Burby, Joshua W.
Christlieb, Andrew
Lei, Huan
contents We introduce a data-driven approach to learn a generalized kinetic collision operator directly from molecular dynamics. Unlike the conventional (e.g., Landau) models, the present operator takes an anisotropic form that accounts for a second energy transfer arising from the collective interactions between the pair of collision particles and the environment. Numerical results show that preserving the broadly overlooked anisotropic nature of the collision energy transfer is crucial for predicting the plasma kinetics with non-negligible correlations, where the Landau model shows limitations.
format Preprint
id arxiv_https___arxiv_org_abs_2503_24208
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Data-driven construction of a generalized kinetic collision operator from molecular dynamics
Zhao, Yue
Burby, Joshua W.
Christlieb, Andrew
Lei, Huan
Computational Physics
Machine Learning
Numerical Analysis
Plasma Physics
We introduce a data-driven approach to learn a generalized kinetic collision operator directly from molecular dynamics. Unlike the conventional (e.g., Landau) models, the present operator takes an anisotropic form that accounts for a second energy transfer arising from the collective interactions between the pair of collision particles and the environment. Numerical results show that preserving the broadly overlooked anisotropic nature of the collision energy transfer is crucial for predicting the plasma kinetics with non-negligible correlations, where the Landau model shows limitations.
title Data-driven construction of a generalized kinetic collision operator from molecular dynamics
topic Computational Physics
Machine Learning
Numerical Analysis
Plasma Physics
url https://arxiv.org/abs/2503.24208