Multiscale simulation of interacting turbulent and rarefied gas flows in the DSMC framework Journal: Journal of Computational Physics
Liyan Luo, Songyan Tian, Lei Wu · 2025
Read Multiscale simulation of interacting turbulent and rarefied gas flows in the DSMC framework Journal: Journal of Computational Physics by Liyan Luo, Songyan Tian, Lei Wu (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-10478.txt.
Introduction
Liyan Luo, Songyan Tian, and Lei Wu fuse two simulation worlds that rarely meet: turbulent gas flow and rarefied gas flow. Thin, fast gases near spacecraft behave by particle rules, while turbulence follows fluid equations. Real problems, like a jet punching into thin upper air, mix both at once.
Their method couples a particle simulation with synthetic fluid equations and a standard turbulence model. Particle data feed the fluid solve; the fluid answer then resets the particles. Tests show the blend slides smoothly into whichever single method fits the local regime. Applied to an opposing jet in hypersonic thin gas, it exposed strong swings in surface behavior from the interplay.
Catalog record
| CREATOR | Liyan Luo, Songyan Tian, Lei Wu |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 54 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2503-10478
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Multiscale simulation of interacting turbulent and rarefied gas flows in the DSMC framework Journal: Journal of Computational Physics Liyan Luo, Songyan Tian, Lei Wu Abstract A multiscale stochastic-deterministic coupling method is proposed to investigate the complex interactions between turbulent and rarefied gas flows within a unified framework. This method intermittently integrates the general synthetic iterative scheme with the shear stress transport turbulence model into the direct simulation Monte Carlo (DSMC) approach, enabling the simulation of gas flows across the free-molecular, transition, slip, and turbulent regimes. First, the macroscopic synthetic equations, derived directly from DSMC, are coupled with the turbulence model to establish a constitutive relation that incorporates not only turbulent and laminar transport coefficients but also higher-order terms accounting for rarefaction effects. Second, the macroscopic properties, statistically sampled over specific time intervals in DSMC, along with the turbulent properties provided by the turbulence model, serve as initial conditions for solving the macroscopic synthetic equations. Finally, the simulation particles in DSMC are updated based on the macroscopic properties obtained from the synthetic equations. Numerical simulations demonstrate that the proposed method asymptotically converges to either the turbulence model or DSMC results, adaptively adjusting to different flow regimes. Then, this coupling method is applied to simulate an opposing jet surrounded by hypersonic rarefied gas flows, revealing significant variations in surface properties due to the interplay of turbulent and rarefied effects. This study presents an efficient methodology for simulating the complex interplay between rarefied and turbulent flows, establishing a foundational framework for investigating the coupled effects of turbulence, hypersonic conditions, and chemical reactions in rarefied gas dynamics in the future. 1 Introduction With the rapid progress of modern industry, rarefied gas flows have become increasingly common in critical applications, particularly in the development of near-space hypersonic vehicles with jet flows operating at altitudes of 20 to 100 kilometers. These vehicles encounter not only conventional aerodynamic processes, but also complex rarefied gas dynamics and turbulence. Due to the substantial differences in flow characteristics and applicable conditions, turbulence and rarefied flows are typically studied independently, with turbulence described by the Navier-Stokes (NS) equations and rarefied flows governed by the Boltzmann equation. From a physical perspective, however, the dynamics of dilute gases, whether in rarefied or turbulent flow regimes, are governed by the Boltzmann equation. Only when the Knudsen number (defined as the ratio of the molecular mean free path \lambda to the characteristic system length L ) is small can the NS equations be derived from the Boltzmann equation chapman-1990 . In the simulation of turbulent flows, the NS equations are typically solved directly or in conjunction with turbulence models. However, since the turbulent flow has many length scales, if the characteristic length L is chosen as the smallest Kolmogorov length scale, the local Knudsen number can be large, which might cause the NS equations to fail in capturing rarefaction effects. Although Tennekes and Lumley estimated that this is unlikely under most practical conditions tennekes-1972 , the advancement of computational capabilities has spurred efforts to identify rarefaction effects in turbulent flows. The most reliable large scale simulation is conducted by McMullen et al. who, however, demonstrated that the energy spectrum obtained from the direct simulation Monte Carlo (DSMC 1 1 1 Since DSMC is primarily designed for simulating laminar rarefied flows, its computational cost is prohibitive in near-continuum regimes due to its stringent constraints on grid size and time step; in this case the DSMC simulation costed over 500 hours on half a million CPU cores. , a numerical solver for the Boltzmann equation) bird-1994 and NS simulations agree well with each other, except that discrepancy occurs at large wave numbers, where the thermal fluctuations (in DSMC, as well as in real gas) cause the spectrum to scale with the square of the wave number mcmullen-2022 . Nevertheless, they emphasized that “thermal fluctuations have little impact on the large-scale evolution of the flow". Recently, a GSIS-SST coupling solver has been developed to investigate the possibility of coexistence and interaction of turbulent and rarefied flows tian-2024 . This approach integrates the general synthetic iterative scheme (GSIS) for solving the Boltzmann equation su-2020 , su-2020-can with the k - \omega shear stress transport (SST) model for turbulent flows menter-1994 . By combining these methods, the constitutive relations incorporate both turbulent and laminar viscosities, as well as higher-order terms that capture rarefaction effects. Consequently, the GSIS-SST method is capable of describing gas flows from highly rarefied conditions to fully turbulent continuum flows. Interestingly, in studies of supersonic rarefied gas flows interacting with turbulent jets, the GSIS-SST solver predicts surface heat flux and pitch moment that differ significantly from those predicted by the pure SST model or the pure Boltzmann equation tian-2024 , tian-2025 . This highlights the critical importance of simultaneously accounting for turbulence and rarefied gas dynamics in such scenarios. However, since the GSIS-SST is a deterministic solver for the Boltzmann equation where both physical and velocity space are discretized, its application to high Mach number cases requires large computational memory. To address this challenge, we aim to develop a particle version of the GSIS-SST within the standard DSMC framework. This approach will build on our recent advances in the Direct Intermittent GSIS-DSMC coupling method (DIG) luo-2024 , which solves the Boltzmann equation for laminar flows across multiscale regimes, with fast convergence and less numerical dissipation. Inspired by GSIS-SST, the DIG method will be extended to couple with the k - \omega SST model to simulate the high-density jet flows in higher-altitude and high-Mach-number rarefied flows, in the DSMC framework. The remainder of this paper is organized as follows. Section 2 provides a detailed description of the DSMC method for polyatomic gases and the DIG method. In Section 3 , the DIG and SST models are coupled to enable the DSMC method to accurately describe rarefied and turbulent flows using coarse grids. The algorithm is validated under both turbulent and rarefied conditions through simulations of hypersonic flow over a wedge and a blunt body in Section 4 . Section 5 investigates the interaction between turbulent and rarefied flows based on simulations of an opposing jet. Finally, Section 6 presents concluding remarks and outlines future research directions. 2 The DSMC method for polyatomic gas flow and its acceleration by DIG Throughout this paper, all macroscopic variables are non-dimensionalized using the reference length L_{0} , reference density \rho_{0} , reference temperature T_{0} , and most probable speed c_{0}=\sqrt{k_{B}T_{0}/m} , with k_{B} and m being the Boltzmann constant and molecular mass, respectively. In this section, we first review the governing equations and the DSMC method for simulating rarefied polyatomic gases. Then we introduce the DIG method, which enables efficient simulation of polyatomic gas flows across a wide range of Knudsen numbers, from the laminar continuum to the rarefied regime, but without turbulence. 2.1 Kinetics of polyatomic gas According to quantum mechanics, polyatomic gases possess internal energy levels. In the gas kinetic theory, the state of polyatomic gas flows at the i -th level of internal energy is determined by the probability distribution function f_{i}(t,\bm{x},\bm{v}) , where \bm{v} and \bm{x} represent the molecular velocity and position, respectively, while t denotes time. The evolution of the distribution function is governed by the Wang-Chang-Uhlenbeck