• 한국전산유체공학회:학술대회논문집
• /
• 2006.10a
• /
• pp.181-184
• /
• 2006

High-speed flight vehicle have various cavities. The supersonic cavity flow is complicated due to vortices, flow separation and reattachment, shock and expansion waves. The general cavity flow phenomena include the formation and dissipation of vortices, which induce oscillation and noise. The oscillation and noise greatly affect flow control, chemical reaction, and heat transfer processes. The supersonic cavity' flow with high Reynolds number is characterized by the pressure oscillation due to turbulent shear layer, cavity geometry, and resonance phenomenon based on external flow conditions, The resonance phenomena can damage the structures around the cavity and negatively affect aerodynamic performance and stability. In the present study, we performed numerical analysis of cavities by applying the unsteady, compressible three dimensional Reynolds-Averaged Navier-Stokes(RANS) equations with the ${\kappa}-{\omega}$ turbulence model. The cavity model used for numerical calculation had a depth(D) of 15mm cavity aspect ratio(L/D) of 3, width to spanwise ratio(W/D) of 1.0 to 5.0. Based on the PSD(Power Spectral Density) and CSD(Cross Spectral Density) analysis of the pressure variation, the dominant frequency was analyized and compared with the results of Rossiter's Eq.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.27 no.2
• /
• pp.206-214
• /
• 2003

Mean flow and turbulence properties of developing turbulent flows in a 90 degree square bend with spanwise rotation were measured by a hot-wire anemometer. A slanted wire is rotated into 6 orientations and the voltage outputs from them are combined to obtain the mean velocity and Reynolds stress components. The combinative effects of the centrifugal and Coriolis forces due to the curvature and the rotation of bend on the mean motion and turbulence structures are investigated experimentally. Results show that the two body forces can either enhance or counteract each other depending on the flow direction in the bend.

• Journal of Mechanical Science and Technology
• /
• v.14 no.4
• /
• pp.441-447
• /
• 2000

Direct numerical simulation has been used to study turbulent boundary layers with convex curvature. A direct numerical simulation program has been developed to solve incompressible Navier-Stokes equations in generalized coordinates with the finite volume method. We considered two boundary layer thicknesses. When the curvature effect is small, mean velocity statistics show little difference with those of a plane channel flow. Turbulent intensity decreases as curvature increases. Contours suggest that streamwise vorticities are strong where large pressure fluctuations exist.

• 한국전산유체공학회:학술대회논문집
• /
• 2008.08a
• /
• pp.76.3-76.3
• /
• 2008

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.1
• /
• pp.41-51
• /
• 2004

A direct numerical simulation is performed to analyze the effects of a localized time-periodic blowing on a turbulent boundary layer flow at R $e_{+}$=300. Main emphasis is placed on the blowing frequency effect on near-wall turbulent flow structures at downstream. Wall-normal velocity on a spanwise slot is varied periodically at different frequencies (0.004$\leq$ $f^{＋}$$\leq$0.080). The amplitude of periodic blowing is $A^{＋}$=0.5 in wall nit, which corresponds to the value of $v_{rms}$ at $y^{＋}$=15 without blowing. The frequency responses are scrutinized by examining the phase or time-averaged turbulent statistics. The optimal frequency ( $f^{＋}$=0.03) is observed, where maximum increase in Reynolds shear stress, streamwise vorticity fluctuations and energy redistribution occurs. The phase-averaged stretching and tilting term are investigated to analyze the increase of streamwise vorticity fluctuations which are closely related to turbulent coherent structures. It is found that the difference between PB and SB at a high blowing frequencies is negligible.e.e.

• Proceedings of the KSME Conference
• /
• 2000.04b
• /
• pp.472-478
• /
• 2000

An experimental study is performed to analyze flow structures behind a local suction/blowing in a flat-plate turbulent boundary layer, The local forcing is given to the boundary layer flow by means of a sinusoidally oscillating jet issuing from a thin spanwise slot at the wall. The Reynolds number based on the momentum thickness is about $Re_{\theta}=1700$. The effects of local forcing are scrutinized by altering the forcing frequency $(0.011{\leq}f^+{\leq}0.044)$. The forcing amplitude is fixed at $A_0=0.4$. It is found that a small local forcing reduces the skin friction, and this reduction increases with the forcing frequency. A phase-averaging technique is employed to capture the coherent structures. Velocity signals are decomposed into a periodic part and a fluctuating part. An organized spanwise vortical structure is generated by the local forcing. The larger reduction of skin friction for the higher forcing frequencies is attributed to the diminished adverse effect of the secondary vortex. An investigation of the random fluctuation components reveals that turbulent energy is concentrated near the center of vortical structures.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.25 no.3
• /
• pp.381-388
• /
• 2001

Presented are heat data which describe the effect of interaction between bulk flow pulsations and a vortex embedded in a turbulent boundary layer. The pulsation frequencies are 3 Hz, 15 Hz and 30 Hz. A half delta wing with the same height as the boundary layer thickness is used to generate the vortex flow. The convection heat transfer coefficients on a constant heat-flux surface are measured by embedded 77 T-type thermocouples. Spanwise profiles of convection heat transfer coefficients show that upwash region of vortex flow is influenced by bulk flow pulsations. The local heat transfer coefficient increases approximately by 7 percent. The increase in the local change of convection heat transfer coefficient is attributed to the spanwise oscillatory motion of vortex flow especially at the low Strouhal number and to the periodic change of vortex size.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.1
• /
• pp.31-40
• /
• 2004

Direct numerical simulations were performed to analyze the effects of time-periodical blowing through a spanwise slot on a turbulent boundary layer. The blowing velocity was varied in a cyclic manner from 0 to 2A$^{+}$(A$^{+}$ =0.25, 0.50 and 1.00) at a fixed blowing frequency of f$^{+}$=0.017. The effect of steady blowing (SB) was also examined, and the SB results were compared with those for periodic blowing (PB). PB reduced the skin friction near the slot, although to a slightly lesser extent than SB. PB was found to generate a spanwise vortical structure in the downstream of the slot. This vortex generates a reverse flow near the wall, thereby reducing the wall shear stress. The wall-normal and spanwise turbulence intensities under PB are increased as compared to those under SB, whereas the streamwise turbulent intensity under PB is weaker than that under SB. PB enhances more energy redistribution than SB. The periodic response of the streamwise turbulence intensity to PB is propagated to a lesser extent than that of the other components of the turbulence intensities and the Reynolds shear stress.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.9
• /
• pp.1255-1263
• /
• 2000

In this study, the modification of turbulent boundary layer flow by local wall vibration is investigated. The wall is locally vibrated using a wall deformation actuator, which moves up and down at the frequencies of 100Hz and 50Hz. Simultaneous measurements of the streamwise velocities in the spanwise direction are performed at several wall-normal and streamwise locations using an in-house multi-channel hot wire anemometer and a spanwise hot-wire-probe rake. The mean velocity is reduced in most places due to the wall vibration and its reduced amount becomes small as flow goes downstream. Interestingly, the mean velocity is found to increase very near the wall and near the actuator. This is due to the motion induced by the streamwise vortices which are generated by the downward motion of the actuator. In case of the streamwise velocity fluctuations, their magnitude increases as compared to the unperturbed turbulent boundary layer, and the increased amount becomes small as the flow moves downstream. The modified flow field at the forcing frequency of 50Hz is not much different from that of 100Hz, except the reduced amount of modification.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.9
• /
• pp.2283-2296
• /
• 1995

Experiments of normal shock wave/turbulent boundary layer interaction were conducted in a supersonic diffuser. The flow Mach number just upstream of the normal shock wave was in the range of 1.10 to 1.70 and Reynolds number based upon the turbulent boundary layer thickness was varied in the range of 2.2*10$^{[-994]}$ -4.4*10$^{[-994]}$ . The wall pressures in streamwise and spanwise directions were measured for two test cases, in which the turbulent boundary layer thickness incoming into the supersonic diffuser was changed. The results show that the interactions of normal shock wave with turbulent boundary layer in the supersonic diffuser can be divided into three patterns, i.e., transonic interaction, weak interaction and strong interaction, depending on Mach number. The weak interactions generate the post-shock expansion which its strength is strong as the Mach number increases and the strong interactions form the pseudo-shock waves. From the spanwise measurements of wall pressure, it is known that if the flow Mach number is low, the interacting flow fields essentially appear two-dimensional, but they have an apparent 3-dimensionality for the higher Mach numbers.