Hostname: page-component-7b7567c8c-tnkh4 Total loading time: 0 Render date: 2026-11-17T12:24:27.993Z Has data issue: false hasContentIssue false

A compressible Reynolds-averaged mixing model considering turbulent composition and heat fluxes

Published online by Cambridge University Press:  23 September 2025

Hansong Xie
Affiliation:
HEDPS, Center for Applied Physics and Technology and School of Mechanics and Engineering Science, Peking University, Beijing 100871, PR China
Tengfei Luo
Affiliation:
Department of Mechanics and Aerospace Engineering, Southern University of Science and Technology, Shenzhen 518055, PR ChinaCenter for Ocean Research in Hong Kong and Macau and Division of Environment and Sustainability, Hong Kong University of Science and Technology, Hong Kong, PR China
Yaomin Zhao*
Affiliation:
HEDPS, Center for Applied Physics and Technology and School of Mechanics and Engineering Science, Peking University, Beijing 100871, PR China
Yousheng Zhang*
Affiliation:
HEDPS, Center for Applied Physics and Technology and School of Mechanics and Engineering Science, Peking University, Beijing 100871, PR ChinaInstitute of Applied Physics and Computational Mathematics, Beijing 100094, PR ChinaNational Key Laboratory of Computational Physics, Beijing 100088, PR China
Jianchun Wang
Affiliation:
Department of Mechanics and Aerospace Engineering, Southern University of Science and Technology, Shenzhen 518055, PR China
*
Corresponding authors: Yaomin Zhao, yaomin.zhao@pku.edu.cn; Yousheng Zhang, zhang_yousheng@iapcm.ac.cn
Corresponding authors: Yaomin Zhao, yaomin.zhao@pku.edu.cn; Yousheng Zhang, zhang_yousheng@iapcm.ac.cn

Abstract

In typical nature and engineering scenarios, such as supernova explosion and inertial confinement fusion, mixing flows induced by hydrodynamic interfacial instabilities are essentially compressible. Despite their significance, accurate predictive tools for these compressible flows remain scarce. For engineering applications, the Reynolds-averaged Navier–Stokes (RANS) simulation stands out as the most practical approach due to its outstanding computational efficiency. However, existing RANS studies focus primarily on cases where the compressible effect plays an insignificant role in mixing development, with quite limited attention given to scenarios with significant compressibility influence. Moreover, most of the existing RANS mixing models demonstrate significantly inaccurate predictions for the latter. This study develops a novel compressible RANS mixing model by incorporating physical compressibility corrections into the $K$–$L$–$\gamma$ mixing transition model recently proposed by Xie et al. (J. Fluid Mech. 1002, 2025, A31). Specifically, taking the density-stratified Rayleigh–Taylor mixing flows as representative compressible cases, we first analyse the limitations of the existing model for compressible flows, based on high-fidelity data and local instability criteria. Subsequently, the equation of state for a perfect gas is employed to derive comprehensive compressibility corrections. The crucial turbulent composition and heat fluxes are integrated into the closure of the key turbulent mass flux term of the turbulent kinetic energy equation. These corrections enable the model to accurately depict compressible mixing flows. Systematic validations confirm the efficacy of the proposed modelling scheme. This study offers a promising strategy for modelling compressible mixing flows, paving the way for more accurate predictions in complex scenarios.

Information

Type
JFM Papers
Copyright
© The Author(s), 2025. Published by Cambridge University Press
Figure 0

Figure 1.Schematic diagrams of the initial density distributions for $(a)$ the incompressible and $(b)$ the density-stratified compressible RT flows.

Figure 1

Figure 2.Initial distributions for density and pressure fields. The symbols and lines represent the initial flow fields of HiFi simulations and RANS simulations, respectively. The lines coloured in black, red, blue and green correspond to cases of $S_r=$ 0.5, 1, 2 and 3, respectively.

Figure 2

Table 1.Model coefficients of the baseline model.

Figure 3

Figure 3.Temporal evolution of the mixing width predicted by $(a)$ the $K$–$L$ model and $(b)$ the $K$–$L$–$\gamma$ baseline model. The colours used to distinguish different $S_r$ cases are consistent with figure 2.

Figure 4

Figure 4.Budgets for terms $I\text{-} V$ of the TKE equation (2.16). Panels $(a){-}(c)$ and $(d){-} (f)$ correspond to moments of $t/\tau$ = 1, 2.5 and 4 for cases of $S_r$ = 0.5 and 3, respectively.

Figure 5

Figure 5.Identification of unstable regions based on the local instability criteria for incompressible flow $Q_1=-({\partial \bar {\rho }}/{\partial x})( {\partial \bar {p}}/{\partial x})$ and for compressible flow $Q_2=\bar {c}^2({\partial \bar {\rho }}/{\partial x})-( {\partial \bar {p}}/{\partial x})$. The $Q_2$ is equivalent to the expression (2.18) for the RT cases considered here, owing to the negative pressure gradient in the gravitational field. The ordinate has been normalised by their maximum amplitudes. Panels $(a){-} (c)$ and $(d){-} (f)$ correspond to moments of $t/\tau$ = 1, 2.5 and 4 for the cases of $S_r$ = 0.5 and 3, respectively.

Figure 6

Figure 6.Spatial distributions of $-\overline {p^{\prime }u_i^{\prime }}/\bar {p}$ and $\overline {e^{\prime }u_i^{\prime }}/\bar {e}$. Panels $(a){-} (c)$ and $(d){-} (f)$ correspond to moments of $t/\tau$ = 1, 2.5 and 4 for the cases of $S_r$ = 0.5 and 3, respectively.

Figure 7

Figure 7.Contours of instantaneous temperature in the central $x{-}z$ plane for cases of $(a)$$S_r$ = 0.5 and ($b$) $S_r$ = 3 at $t/\tau$ = 4. The instantaneous turbulent heat flux is also displayed with the vector arrows.

Figure 8

Figure 8.Spatial distributions of the ratio $C_4/C_3$ across the mixing regions (marked by the arrows) at $t/\tau$ = 1, 2.5 and 4.

Figure 9

Figure 9.Temporal evolutions of the mixing widths for the case of $S_r$ = 3, based on different $C_4$.

Figure 10

Figure 10.Temporal evolutions of the mixing widths. Panels $(a){-} (d)$ correspond to the cases of $S_r$ = 0.5, 1, 2 and 3, respectively.

Figure 11

Figure 11.Spatial profiles of the mean species at three different moments: $t/\tau =1$ (coloured in black), $t/\tau =2.5$ (coloured in blue) and $t/\tau =4$ (coloured in red). Panels $(a){-} (d)$ correspond to the cases with $S_r=$ 0.5, 1, 2 and 3, respectively.

Figure 12

Figure 12.Spatial profiles of the mean density at three different moments: $t/\tau =1$ (coloured in black), $t/\tau =2.5$ (coloured in blue) and $t/\tau =4$ (coloured in red). Panels $(a){-} (d)$ correspond to the cases with $S_r=$ 0.5, 1, 2 and 3, respectively.

Figure 13

Figure 13.Temporal evolution of the mixing quantity for the inverse chevron case predicted by the $K$–$L$–$\gamma$ model. The red and blue lines represent the RANS results with and without compressibility corrections, respectively.

Figure 14

Figure 14.Temporal evolutions of the mixing widths predicted by the $K$–$L$–$a$–$b$ model, whose physical model and model coefficients can be found from Zhao et al. (2025). Panels $(a){-} (d)$ correspond to the cases of $S_r$ = 0.5, 1, 2 and 3, respectively. The used numerical schemes are consistent with those in § 2.2: the initial TKE $\tilde {K}_0=0.0001A_tg\tilde {L}_0$, turbulent mass flux velocity $a_0=\sqrt {\tilde {K}_0}/4$, density-specific volume covariance $b=1\times 10^{-8}$. Particularly, the initial turbulent length scale $\tilde {L}_0$ is adjusted to obtain optimal results. All the cases use the same RANS initialisations.