An asymptotic-preserving Monte Carlo method for the Boltzmann equation

Abstract

In this work, we propose an asymptotic-preserving Monte Carlo method for the Boltzmann equation that is more efficient than the currently available Monte Carlo methods in the fluid dynamic regime. This method is based on the successive penalty method [39], which is an improved BGK-penalization method originally proposed by Filbet and Jin [16]. Here we introduce the Monte Carlo implementation of the method, which, despite its lower order accuracy, is very efficient in higher dimensions or simulating some complicated chemical processes. This method allows the time step independent of the mean free time which is prohibitively small in the fluid dynamic regime. We study some basic properties of this method, and compare it with some other asymptotic-preserving Monte Carlo methods in terms of numerical performance in different regimes, from rarefied to fluid dynamic regimes, and their computational efficiency.