Hostname: page-component-7b7567c8c-2n9c9 Total loading time: 0 Render date: 2026-11-17T08:05:30.324Z Has data issue: false hasContentIssue false

Asymptotic theory of Mack-mode receptivity in hypersonic boundary layers due to interaction of a heating/cooling source and a freestream sound wave

Published online by Cambridge University Press:  22 May 2023

Lei Zhao
Affiliation:
Department of Mechanics, Tianjin University, Tianjin 300072, PR China
Jianhong He
Affiliation:
Department of Mechanics, Tianjin University, Tianjin 300072, PR China
Ming Dong*
Affiliation:
State Key Laboratory of Nonlinear Mechanics, Institute of Mechanics, Chinese Academy of Sciences, Beijing 100190, PR ChinaSino-Russian Mathematics Center, Peking University, Beijing 100871, PR China
*
†Email address for correspondence: dongming@imech.ac.cn

Abstract

In this paper, we study the local receptivity of the inviscid Mack modes in hypersonic boundary layers induced by the interaction between a surface heating or cooling source (HCS) and a freestream acoustic wave. The asymptotic analysis reveals that among the three distinguished layers, i.e. the main, wall and Stokes layers, the leading-order receptivity is attributed to the interaction of the HCS-induced mean-flow distortion and the acoustic signature in the wall layer; the second-order contribution appears in the Stokes layer; the third-order contribution appears in both the main and wall layers. Interestingly, at a moderate Reynolds number, the third-order contribution to the receptivity efficiency may be quantitatively greater than the second-order one, but this does not lead to breakdown of this asymptotic theory. Assuming the HCS intensity to be sufficiently weak, the asymptotic predictions are made for four representative cases involving different Mach numbers and wall temperatures, which are compared with the results obtained by the finite-Reynolds-number theory based on either the extended compressible Orr–Sommerfeld equations or the harmonic linearised Navier–Stokes (HLNS) calculations. Taking into account the first three orders of the receptivity efficiency, the asymptotic predictions are confirmed to be sufficiently accurate even when the Reynolds number is a few thousands, and the agreement with the finite-Reynolds-number calculations is better when the wall temperature of the base flow approaches the adiabatic wall temperature. The HLNS calculations are also conducted for moderate HCS intensities. It is found that the nonlinearity does not affect the receptivity coefficient much even when the temperature distortion of the HCS reaches $80\,\%$ of the temperature at the wall.

Information

Type
JFM Papers
Copyright
© The Author(s), 2023. Published by Cambridge University Press
Figure 0

Figure 1.Side-view sketch of the physical model, where the red region denotes the HCS. In the spanwise direction, the HCS is distributed periodically, and the two-dimensional HCS for which its spanwise wavelength is infinity is also included.

Figure 1

Figure 2.Sketch of the scaling estimate of the receptivity mechanism: (a) HCS–acoustic receptivity; (b) roughness–acoustic receptivity.

Figure 2

Figure 3.Sketch of the mesh system for the HLNS approach.

Figure 3

Table 1.Parameters for case studies.

Figure 4

Figure 4.Profiles of (a) $U_B$ and (b) $T_B$ of the compressible Blasius solution for the four cases.

Figure 5

Figure 5.Dependence on $\omega$ of (a,c) $c_r$ and (b,d) $-\alpha _i$ of the 2-D Mack modes, where the velocities of the fast and slow acoustic waves, $1\pm 1/M$, are marked in (a,c). Plots for (a,b) case 1, and (c,d) case 3. R1 and R2 are for $R=7800$ and 1560, respectively. The abbreviations ASMP and OS denote the asymptotic predictions and the OS solutions, respectively.

Figure 6

Figure 6.Profiles of the mean-flow distortion in the wall layer for case 1 with $k_x=k_z=0.028$: (a) $\hat V_0$ and $\hat T_0$ profiles; (b) $\hat U_0$ and $\hat W_0$ profiles. The dashed lines denote the asymptotic estimations.

Figure 7

Figure 7.Normalised profiles of the real parts of (a) $\hat V_{w1}$ and (b) $\hat P_{w1}$ in the main layer for case 1 with $k_x=0.028$ and $k_z=0$. The inset in (a) shows a zoom-in plot in the near-wall region.

Figure 8

Figure 8.Comparison of the distortion of the transverse velocity between the asymptotic composite solution (solid lines) and the EOS solutions (dot-dashed lines) for $k_x=0.028$ and $k_z=0$: (a,b) case 1, and (c,d) case 3. Panels (b,d) show zoom-in plots of (a,c) in the near-wall region, respectively.

Figure 9

Figure 9.Asymptotic predictions of the efficiency function at each order at different frequencies for the receptivity to a 2-D freestream acoustic wave with incident angle $\theta =45^\circ$: (a,c,e,g) fast acoustic forcing; (b,d, f,h) slow acoustic forcing. Plots for (a,b) case 1, (c,d) case 2, (e, f) case 3, and (g,h) case 4. For plotting convenience, the blue curves for $|\varLambda _2|$ are multiplied by 0.1.

Figure 10

Figure 10.Dependence of the efficiency function on the incident angle of the 2-D freestream fast acoustic wave. The line types are the same as in figure 9, and the blue curves for $\varLambda _2$ are multiplied by 0.1. Plots for (a) case 1 with $\omega =0.108$; (b) case 2 with $\omega =0.126$; (c) case 3 with $\omega =0.153$; (d) case 4 with $\omega =0.185$.

Figure 11

Figure 11.Dependence of the efficiency function on the spanwise wavenumber of the freestream fast acoustic wave $\beta _a$, where $k_z=0$ and $\theta =45^\circ$. The line types are the same as in figure 9, and the blue curves for $\varLambda _2$ are multiplied by 0.1. Plots for (a) case 1 with $\omega =0.108$; (b) case 2 with $\omega =0.126$; (c) case 3 with $\omega =0.153$; and (d) case 4 with $\omega =0.185$.

Figure 12

Figure 12.Comparison of the receptivity coefficient $\varLambda$ between the EOS solutions and the asymptotic predictions for different $R$. The freestream perturbation is the fast acoustic wave with $\theta =45^\circ$. Leading order is $\epsilon \varLambda _0$; two orders is $\epsilon (\varLambda _0+\epsilon ^{1/2}\varLambda _1)$; three orders is $\epsilon (\varLambda _0+\epsilon ^{1/2}\varLambda _1+\epsilon \varLambda _2)$. Plots for (a) case 1, (b) case 2, (c) case 3, and (d) case 4. For plotting convenience, the curves for $\omega =0.145$ and 0.160 in (b) are multiplied by 0.1 and 0.01, respectively; those for $\omega =0.175$ and 0.190 in (c) are multiplied by 0.1 and 0.01, respectively; those for $\omega =0.224$ in (d) are multiplied by 0.1.

Figure 13

Figure 13.Comparison of the receptivity coefficient $\varLambda$ between the EOS solutions and the asymptotic predictions for different $\omega$: (a) case 1, (b) case 4.

Figure 14

Figure 14.Contours of the mean (a) temperature and (b) pressure distorted by a heating source with $d/\delta _1=0.65$ and $\varTheta _m=0.1$.

Figure 15

Figure 15.Contours of (a i,b i) the perturbation pressure $\breve p$, and (a ii,b ii) the excited perturbation pressure $\breve p_e=\breve p-\breve p_a$, for (a) $\omega =0.178$, and (b) $\omega =0.202$.

Figure 16

Figure 16.Streamwise evolution of the excited pressure perturbation at the wall $\breve p_e(x,0)$, for (a) $\omega =0.178$, and (b) $\omega =0.202$.

Figure 17

Figure 17.Streamwise evolution of (a) the mode-decomposition amplitude $|\breve p_M(x,0)|$, and (b) its normalisation by its linear evolution $|\breve p_M(x,0)/\exp (\textrm {i}\alpha x)|$.

Figure 18

Figure 18.Comparison of the receptivity coefficient $\varLambda$ between the asymptotic predictions and the HLNS calculations: (a) heating source with $\varTheta _m>0$; (b) cooling source with $\varTheta _m<0$.

Figure 19

Figure 19.Dependence of the receptivity coefficient on $\varTheta _m$ for three representative frequencies.