Improved Diffuse Boundary Condition for the DSBGK Method to Eliminate the Unphysical Density Drift

TL;DR

This paper introduces an improved diffuse boundary condition for the DSBGK method that effectively eliminates unphysical density drift in simulations, enhancing accuracy for both closed and open micro-scale flow problems.

Contribution

The paper proposes a new boundary condition calculating molecular fluxes using molecular variables, reducing density drift in DSBGK simulations.

Findings

The improved boundary condition removes unphysical density drift.

The original boundary condition is preferable for tiny perturbations due to less stochastic noise.

The method is effective for both closed and open flow problems.

Abstract

An improved diffuse boundary condition, where the number flux of the incoming real molecules on the wall surface is calculated using the molecular variables rather than the cell's macroscopic variables, is proposed to eliminate the unphysical density drift, which was observed in the previous DSBGK simulation of the lid-driven problem but disappears in the channel flow problem because of the density constraint imposed at the open boundaries. Consequently, the efficient time-average process is valid for sampling all quantities of interest in closed as well as open problems. When the driven velocity of the lid-driven problem is only one micrometer per second that is realistic in the micro-electro-mechanical systems or pore-scale flows of the shale gas, we suggest to use the original boundary condition because the density drift during a very long time interval becomes unperceivable when the perturbation is tiny. Thus, the original diffuse boundary condition, which contains much less stochastic noise than the improved one, is still an advisable choice in closed problems with tiny perturbation and in open problems unless the density drift occurs and fails the goal of the simulations making the use of the improved boundary condition necessary.