Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-16T11:19:23.098Z Has data issue: false hasContentIssue false

Compressibility effects on statistics and coherent structures of compressible turbulent mixing layers

Published online by Cambridge University Press:  30 August 2022

Xiaoning Wang
Affiliation:
Department of Mechanics and Aerospace Engineering, Southern University of Science and Technology, Shenzhen 518055, PR ChinaSouthern Marine Science and Engineering Guangdong Laboratory, Guangzhou 511458, P.R. ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Data-Driven Fluid Mechanics and Engineering Applications, Southern University of Science and Technology, Shenzhen 518055, PR China
Jianchun Wang*
Affiliation:
Department of Mechanics and Aerospace Engineering, Southern University of Science and Technology, Shenzhen 518055, PR ChinaSouthern Marine Science and Engineering Guangdong Laboratory, Guangzhou 511458, P.R. ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Data-Driven Fluid Mechanics and Engineering Applications, Southern University of Science and Technology, Shenzhen 518055, PR China
Shiyi Chen
Affiliation:
Department of Mechanics and Aerospace Engineering, Southern University of Science and Technology, Shenzhen 518055, PR ChinaSouthern Marine Science and Engineering Guangdong Laboratory, Guangzhou 511458, P.R. ChinaGuangdong-Hong Kong-Macao Joint Laboratory for Data-Driven Fluid Mechanics and Engineering Applications, Southern University of Science and Technology, Shenzhen 518055, PR China
*
†Email address for correspondence: wangjc@sustech.edu.cn

Abstract

The effects of compressibility on the statistics and coherent structures of a temporally developing mixing layer are studied using numerical simulations at convective Mach numbers ranging from $M_c=0.2$ to $1.8$ and at Taylor Reynolds numbers up to 290. As the convective Mach number increases, the streamwise dissipation becomes more effective to suppress the turbulent kinetic energy. At $M_c=1.8$, the streamwise dissipation increases much faster than the other two components in the transition region, even larger than pressure–strain redistribution, correlating with the streamwise elongated vortical structures at a higher level of compressibility. We confirm the existence of the large-scale high- and low-speed structures in the mixing layers, which accompany the spanwise Kelvin–Helmholtz rollers at low convective Mach number and dominate the mixing layer at higher convective Mach number. Conditional statistics demonstrate that the large-scale low-speed structures are lifted upwards by a pair of counter-rotating quasi-streamwise rollers flanking the structures. The small-scale vortical structures have an apparent preference for clustering into the top of the low-speed regions, which is directly associated with high-shearing motions on top of the low-speed structures. The high-speed structures statistically exhibit central symmetry with the low-speed structures. The statistics and dynamics of large-scale high- and low-speed structures in the compressible mixing layers resemble those in the outer region of the turbulent boundary layers, which reveals the universality of the large-scale structures in free shear and wall-bounded turbulence. A conceptual model is introduced for the large-scale high- and low-speed structures in turbulent mixing layers.

Information

Type
JFM Papers
Copyright
© The Author(s), 2022. Published by Cambridge University Press
Figure 0

Figure 1.Schematic of the temporally evolving shear layer with mean velocity profile. The red dashed rectangles represent the extent of the sponge layers in vertical direction.

Figure 1

Table 1.Simulation parameters at the beginning ($\tau =\tau _0$) and end ($\tau =\tau _f$) of the self-similar period. Parameters $M_t$, $Re_{\lambda }$, $\eta$, $l_x$ and $l_z$ are obtained at $y = 0$.

Figure 2

Figure 2.Comparison of mean streamwise velocity $\bar {u}/\Delta U$ in similarity coordinates (a) $y/\delta _{\theta }$ and (b) $y/\delta _{\omega }$. (c) Time evolution of normalised momentum thickness for different $M_c$, and the vertical line segments mark the self-similar time periods. (d) Comparison of linear growth rates at different $M_c$ with different numerical and experimental results.

Figure 3

Figure 3.Comparison of normalised components of the Reynolds stress for case M02: (a) $\sqrt {{\mathsf{R}}_{11}}/\Delta U$, (b) $\sqrt {{\mathsf{R}}_{22}}/\Delta U$, (c) $\sqrt {{\mathsf{R}}_{33}}/\Delta U$ and (d) $\sqrt {-{\mathsf{R}}_{12}}/\Delta U$. The symbols denote experimental results for incompressible shear layers, and the curves represent numerical results. The legend is the same for all plots.

Figure 4

Figure 4.TKE transport equation terms in (2.10): $\bar {P}$ (green), $\bar {\epsilon }$ (red) and $\bar {T}$ (blue) for cases (a) M02, (b) M08 and (c) M18. The budget terms are normalised by $\Delta U^3/\delta _\theta$. The solid lines correspond to the present DNS. The dashed lines, open and filled symbols represent the results of Rogers & Moser (1994), Pantano & Sarkar (2002) and Vaghefi (2014), respectively.

Figure 5

Figure 5.Visualisation of normalised velocity divergence $\varTheta \delta _{\theta }/\Delta U$ in middle $x$–$y$ plane at two different time instants: (a) $M_c=0.8$, $\tau =250$, (b) $M_c=0.8$, $\tau =1000$, (c) $M_c=1.8$, $\tau =500$ and (d) $M_c=1.8$, $\tau =1250$. The red solid lines correspond to contours of $\omega =0.01\Delta U/\delta _{\theta }^0$, showing the TNTI.

Figure 6

Figure 6.Temporal evolution of the centreline (a) turbulent Mach number $M_{t,y0}$ and (b) r.m.s. velocity divergence $\varTheta _{rms,y0}/(\Delta U/\delta _{\theta _0})$ at the centreplane for simulations with $M_c=0.2$, 0.8 and 1.8. The dashed vertical lines mark the self-similar time periods.

Figure 7

Figure 7.PDF of the normalised velocity divergence $\varTheta /(\Delta U/\delta _{\theta _0})$ at five different time instants (a),(c) in the turbulent region and (b),(d) outside the turbulent region for cases at (a),(b) $M_c=0.8$ and (c),(d) $M_c=1.8$.

Figure 8

Figure 8.Instantaneous fields of (a) $u_s^{\prime \prime }$, (b) $u_d^{\prime \prime }$, (c) $v_s^{\prime \prime }$, (d) $v_d^{\prime \prime }$, (e) $w_s^{\prime \prime }$ and ( f) $w_d^{\prime \prime }$ in the central $x$–$y$ plane for case M08 at $\tau =1000$.

Figure 9

Figure 9.Instantaneous fields of (a) $u_s^{\prime \prime }$, (b) $u_d^{\prime \prime }$, (c) $v_s^{\prime \prime }$, (d) $v_d^{\prime \prime }$, (e) $w_s^{\prime \prime }$ and ( f) $w_d^{\prime \prime }$ in the central $x$–$y$ plane for the case M18 at $\tau =1700$.

Figure 10

Figure 10.Instantaneous fields of (a) $u_s^{\prime \prime }$, (b) $u_d^{\prime \prime }$, (c) $v_s^{\prime \prime }$, (d) $v_d^{\prime \prime }$, (e) $w_s^{\prime \prime }$ and ( f) $w_d^{\prime \prime }$ in the central $x$–$z$ plane for case M08 at $\tau =1000$.

Figure 11

Figure 11.Instantaneous fields of (a) $u_s^{\prime \prime }$, (b) $u_d^{\prime \prime }$, (c) $v_s^{\prime \prime }$, (d) $v_d^{\prime \prime }$, (e) $w_s^{\prime \prime }$ and ( f) $w_d^{\prime \prime }$ in the central $x$–$z$ plane for case M18 at $\tau =1700$.

Figure 12

Figure 12.Joint PDF of (a),(b) $u_s^{\prime \prime }/u_{s,rms}^{\prime \prime }$ and $v_s^{\prime \prime }/v_{s,rms}^{\prime \prime }$ and (c),(d) $u_d^{\prime \prime }/u_{d,rms}^{\prime \prime }$ and $v_d^{\prime \prime }/v_{d,rms}^{\prime \prime }$ for case with $M_c=0.8$ at $\tau =1000$ and $M_c=1.8$ at $\tau =1250$: (a),(c) in the turbulent region and (b),(d) in the non-turbulent region. The thick dashed auxiliary line indicates the long axis of the oval shape. The thin dashed-dotted line represents the symmetry line of the $u_s^{\prime \prime }$–$v_s^{\prime \prime }$ plane. The contour levels 0.001, 0.01 and 0.1 are shown.

Figure 13

Figure 13.The Reynolds shear stress and its decomposed components for cases with (a) $M_c=0.8$ and (b) $M_c=1.8$. The legend is the same for both plots.

Figure 14

Figure 14.Time evolution of $\hat {K}/\Delta U^2$, $\hat {K}_i/\Delta U^2$, $0.5\hat {R}_{12}/\Delta U^2$ at (a) $M_c=0.2$, (b) $M_c=0.8$ and (c) $M_c=1.8$. The legend is the same for all plots. The dashed vertical lines mark the self-similar time periods.

Figure 15

Figure 15.Time evolution of $\hat {P}/\Delta U^3$, $\hat {\epsilon }/\Delta U^3$, $\hat {\varPhi }/\Delta U^3$ at (a),(b) $M_c=0.2$, (c) $M_c=0.8$ and (d) $M_c=1.8$. The legend is the same for (a), (c) and (d). The black solid lines correspond to $\hat {\epsilon }/\Delta U^3 \propto -\tau ^2$. The dashed vertical lines mark the self-similar time periods.

Figure 16

Figure 16.Time evolution of $\hat {P}_i/\Delta U^3$, $\hat {\epsilon }_i/\Delta U^3$, $\hat {\varPhi }_i/\Delta U^3$ at (a) $M_c=0.2$, (b) $M_c=0.8$ and (c) $M_c=1.8$. The legend is the same for all plots. The black solid lines correspond to $\hat {\epsilon }_i/\Delta U^3 \propto -\tau ^2$.

Figure 17

Figure 17.PDF of viscous dissipation components ${\epsilon }_i/\Delta U^3$ at $\tau =250$ for cases with (a) $M_c=0.2$ and (b$M_c=1.8$.

Figure 18

Figure 18.Time evolution of $\hat {P}_{12}/\Delta U^3$, $\hat {\epsilon }_{12}/\Delta U^3$, $\hat {\varPhi }_{12}/\Delta U^3$ at (a) $M_c=0.2$, (b) $M_c=0.8$ and (c) $M_c=1.8$. The legend is the same for all plots. The dashed vertical lines mark the self-similar time periods.

Figure 19

Figure 19.(a) Time evolution of the centreline Kolmogorov length scale $\eta _{y0}/\Delta x$ and (b) mean profile of $\eta /\Delta x$ at $M_c=0.2, 0.8$ and $1.8$. The dashed vertical lines mark the self-similar time periods in (a).

Figure 20

Figure 20.Small-scale vortical structures are visualised by the iso-surfaces of the second invariant $Q$ of the velocity gradient tensor and coloured based on local values of $u^{\prime \prime }$: (a) $Q=20$, $\tau =125$ and $M_c=0.2$; (b) $Q=2$, $\tau =250$ and $M_c=1.8$.

Figure 21

Figure 21.Contours of instantaneous velocity fluctuations $u_i^{\prime \prime }$ along with stronger events of viscous dissipation indicated by black solid lines: (a) ${\epsilon _1}/\Delta U^3=0.03$, $-0.3< u^{\prime \prime }<0.3$ at $M_c=0.2$ and $\tau =125$; (b) ${\epsilon _1}/\Delta U^3=0.01$, $-0.3< u^{\prime \prime }<0.3$, (c) ${\epsilon _2}/\Delta U^3=0.005$, $-0.2< v^{\prime \prime }<0.2$, and (d) ${\epsilon _3}/\Delta U^3=0.005$, $-0.2< w^{\prime \prime }<0.2$ at $M_c=1.8$ and $\tau =250$. (e) Enlarged view of the region in red-lined rectangular box in plot (b), showing the velocity vectors of $v$ and $w$.

Figure 22

Figure 22.Instantaneous fields of (a),(c),(e) $u^{\prime \prime }$ and (b),(d),(f) $\rho^{\prime}$ in the central $x$–$z$ plane for case M02: (a),(b) $\tau =100$; (c),(d) $\tau =375$; and (e),(f) $\tau =750$.

Figure 23

Figure 23.Instantaneous fields of (a),(c),(e) $u^{\prime \prime }$ and (b),(d),(f) $\rho ^{\prime }$ in the central $x$–$z$ plane for case M18: (a),(b) $\tau =125$; (c),(d) $\tau =500$; and (e),(f) $\tau =2000$.

Figure 24

Figure 24.Two-point correlations of the streamwise velocity fluctuation $u^{\prime \prime }$ in the (a),(b) $x$–$y$, (c),(d) $x$–$z$ and (e),(f) $y$–$z$ planes at three different time instants: (a),(c),(e) $M_c=0.2$ and (b),(d),(f) $M_c=1.8$. The contour levels range from 0.1 to 1 with increments of 0.1.

Figure 25

Figure 25.Evolutions of the normalised integral length scales in the streamwise direction $l_x/\delta _\omega$ and in the spanwise direction $l_z/\delta _\omega$ for (a) $M_c=0.2$ and (b) $M_c=1.8$. The dashed vertical lines mark the self-similar time periods.

Figure 26

Figure 26.Comparisons of the two-point correlation of $u^{\prime \prime }$ in the mixing shear layer with that in the boundary layers (Pirozzoli & Bernardini 2011; Bross et al.2021). The red solid lines indicate the wall of the boundary layer. The red + symbols show the centre points located with a distance of (a) $0.2\delta$ and (b) $0.3\delta$ away from the wall. The levels of contour lines are 0.2, 0.4, 0.6 and 0.8.

Figure 27

Figure 27.Top view of coherent structures in the upper half-domain $y>0$ for $M_c=1.8$ at $\tau =1250$. (a) Small-scale vortical structures visualised by the iso-surfaces of the second invariant of the velocity gradient tensor $Q =2$ and coloured based on local values of $u^{\prime \prime }$. (b) The high- and low-speed large-scale structures presented by the iso-surfaces of $u^{\prime \prime } = 0.3$ (blue) and $-$0.3 (red), respectively.

Figure 28

Figure 28.Conditional average of flow field conditioned to the low-speed structures, $u^{\prime \prime } < 0$, in the (a),(c) $x$–$y$ and (b),(d) $y$–$z$ planes: (a),(b) $M_c=0.2$ and (c),(d) $M_c=1.8$. The colour contours display $\langle \omega \mid u^{\prime \prime }<0 \rangle$ and $\langle u^{\prime \prime }\mid u^{\prime \prime } < 0 \rangle$ is plotted with solid black contour lines representing levels range from $-$0.2 to $-$0.05 (from inside to outside) with increments of 0.05. Vectors represent conditional-averaged velocity components with the longest arrow measuring 0.2. The $+$ (red) symbols show the centre location of roll modes.

Figure 29

Figure 29.Schematic of large-scale coherent structures and its interaction across the mixing shear layer for higher convective Mach number $M_c \ge 0.8$. The red and blue regions indicate the high- and low-speed large-scale structures, respectively. The black circles is the counter-rotating roll modes. The gray contours show the clustering of small-scale vortical structures. The red and black dashed lines represent centreplane and boundaries of mixing layer, respectively.

Figure 30

Table 2.Parameters for the additional simulations: $L_b$ denotes the buffer layer width; $L_x$, $L_y$, $L_z$ and $L_b$ are measured in terms of initial momentum thickness $\delta _\theta ^0$.

Figure 31

Figure 30.Temporal evolution of normalised momentum thickness at $M_c=0.8$ for (a) different computational grid sizes and (b) different buffer layer width.

Figure 32

Figure 31.The Reynolds normal stresses and their decomposed components for case with $M_c=1.8$: (a) ${\mathsf{R}}_{11}$; (b) $\boldsymbol{\mathsf{R}}_{22}$; and (c) ${\mathsf{R}}_{33}$.