Comparison of stochastic BGK and FP methods for the simulation of non-equilibrium multi-species molecular gas flows

TL;DR

This paper compares stochastic BGK and FP methods for simulating non-equilibrium multi-species gas flows, highlighting their implementation and verification against test cases to evaluate computational efficiency and accuracy.

Contribution

It introduces implementations of stochastic BGK and FP methods in an open-source code and compares their performance for multi-species gas flow simulations.

Findings

Stochastic BGK and FP methods are implemented in PICLas.

Comparison of methods shows differences in accuracy and computational efficiency.

Verification against test cases demonstrates applicability for non-equilibrium flows.

Abstract

Due to limited possibilities of experimental investigations for non-equilibrium gas flows, numerical results are of highest interest. Although the well-established Direct Simulation Monte Carlo (DSMC) method achieves highly accurate solutions, the computational requirements increase excessively for lower Knudsen regimes. Computationally more efficient simulations can be achieved with stochastic continuum-based methods using either the Bhatnagar-Gross-Krook (BGK) or the Fokker-Planck (FP) approximations where, instead of particle collisions, particle relaxation processes are considered. This paper explains the implementation of different stochastic BGK and FP methods in the open-source particle code PICLas for multi-species molecular gas flows. For verification, the results of different test cases are compared.