• 한국전산유체공학회:학술대회논문집
• /
• 2009.04a
• /
• pp.104-111
• /
• 2009

The existing code which solves two-dimensional RANS(Reynolds Averaged Navier-Stokes) equations and 2-equation turbulence model equations was modified to enable numerical simulation of various supersonic flows. For this, various boundary conditions have been implemented to the code. Bleed boundary condition was incorporated into the code for calculating wall mean flow quantities. Furthermore, the boundary conditions for the turbulence quantities along rough surfaces as well as porous walls were applied to the code. The code was verified and validated by comparing the computational results against the experimental data for the supersonic flows over bleed region on a flat plate. Using the newly modified code, numerical simulations were performed and compared with other computational results as well as the experimental data for the supersonic flows over an oblique shock with a bleed region.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.36 no.11
• /
• pp.1055-1063
• /
• 2012

Fluctuating wall pressures were measured using an array of 16 piezoelectric transducers beneath a turbulent boundary layer. The coating used in this experiment was an open-cell, urethane-type foam with a porosity of approximately 50 ppi. The ultimate objective of the coating is to provide a mechanical filter to reduce the wall pressure fluctuations. The boundary layer on the flat plate was measured by using a hot wire probe, and the CPM method was used to determine the skin friction coefficient. The wall pressure autospectra and streamwise wavenumber-frequency spectra were compared to assess the attenuation of the wall pressure field by the coating. The coating is shown to attenuate the convective wall pressure energy. However, the relatively rough surface of the coating in this investigation resulted in a higher mean wall shear stress, thicker boundary layer, and higher low-frequency wall pressure spectral levels compared to a smooth wall.

• Journal of computational fluids engineering
• /
• v.14 no.4
• /
• pp.23-30
• /
• 2009

The existing code which solves two-dimensional RANS(Reynolds Averaged Navier-Stokes) equations and 2-equation turbulence model equations was modified to enable numerical simulation of various supersonic flows. For this, various boundary conditions have been implemented to the code. Bleed boundary condition was incorporated into the code for calculating wall mean flow quantities. Furthermore, boundary conditions for the turbulence quantities along rough surfaces as well as porous walls were applied to the code. The code was verified and validated by comparing the computational results against the experimental data for the supersonic flows over bleed region on a flat plate. Furthermore, numerical simulations for supersonic shock boundary layer interaction with a bleed region were performed and their results were compared with the existing computational results.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 1999.04a
• /
• pp.7-7
• /
• 1999

충돌/유출냉각은 연소실 벽면 등과 같이 열부하가 많이 걸리는 영역에서 고온의 연소가스로부터 표면을 보호하는 막냉각과 더불어 내부에 다공성 판을 설치하여 충돌제트에 의해 내부표면을 냉각시켜 가스터빈엔진의 고온요소의 냉각성능을 극대화시키는 방법 중의 하나이다. 이와 같은 냉각방법을 적용하면 벽면을 충돌제트에 의한 냉각과 함께 냉각유채를 막냉각에 활용함으로써 냉각효율을 극대화할 수 있다. 본 연구에서는 국소적인 값들을 획득하기 용이한 물질전달실험방법의 하나인 나프탈렌 승화법을 이용하여 수직으로 분사되는 충돌제트에 의한 유출판 내면에서의 열/물질전달특성을 분사판(injection plate)과 유출판(effusion plate) 사이의 높이, 분사제트의 속도, 분사홀간의 배열을 변화시켜가며 유출판만이 있는 경우와 비교, 분석하였다. 분사홀과 유출홀의 관 사이의 간격은 0.33d에서 10d까지 변화시켜가며 그 효과를 관찰하였으며, 홀배열 효과를 보기 위하여 2가지 홀배열(staggered array, shifted array)에 대하여 실험하였다. 또한 분사제트의 속도효과를 고찰하기 위하여 분사제트의 Rc$_{d}$=5,000-12,000까지 변화시켜가며 실험을 수행하였다.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.36 no.4
• /
• pp.425-430
• /
• 2012

In this study, a preconverter of an MCFC for an emergency electric power supplier is numerically simulated to increase the hydrogen production from natural gas (methane). A commercial code is used to simulate a porous catalyst with a user subroutine to model three dominant chemical reactions-steam reforming, water-gas shift, and direct steam reforming. To achieve a fuel conversion rate of 10% in the preconverter, the required external heat flux is supplied from the outer wall of the preconverter. The calculated results show that the temperature distribution and chemical reaction are extremely nonuniform near the wall of the preconverter. These phenomena can be explained by the low heat conductivity of the porous catalyst and the endothermic reforming reaction. The calculated results indicate that the use of a compact-size preconverter makes the chemical reaction more uniform and provides many advantages for catalyst maintenance.

• Journal of the Korean Institute of Gas
• /
• v.18 no.4
• /
• pp.56-62
• /
• 2014

The present study investigated numerically heat and mass transfer characteristics of a fixed bed reactor by using a computational fluid dynamics (CFD) code of Fluent (ver. 13.0). The temperature and species fraction were estimated for different porosities. For modeling of the catalyst in a fixed bed tube, catalysts were regarded as the porous material, and the empirical correlation of pressure drop based on the modified Eugun equation was used for simulation. In addition, the averaged porosities were taken as 0.545, 0.409, and 0.443 and compared with non-porous state. The predicted results showed that the temperature at the tube wall became higher than that estimated along the center line of tube, leading to higher hydrogen generation by the endothermic reaction and heat transfer. As the mean porosity increases, the hydrogen yield and the outlet temperature decreased because of the pressure drop inside the reformer tube.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.13 no.5
• /
• pp.979-990
• /
• 1989

Natural convection heat transfer has been investigated numerically in the vertical annulus filled withsaturated porous material for the aspect ratio less than unity. The inner wall of the annulus is exposed to constant heat flux condition and the outer wall is cooled to keep isothermal condition. The upper and the lower horizontal wall are assumed to be insulated. Under conditions ranging 50 .leq. Ra .leq. 10000, 1 .leq. RD .leq. 12, the characteristics of flow and heat transfer have been investigated. The results show that average Nusselt numbers increase when the radius ratio increases and the multicellular flows are not detected under the present conditions. Isothermal lines are plotted within the porous media. Temperatures of the inner wall with constant heat flux conditions and the local heat flux rate of the cooled outer wall with constant temperature are also obtained.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.24 no.12
• /
• pp.1635-1643
• /
• 2000

The present work investigates a heat transfer phenomenon at the interface between a porous medium and an impermeable wall. In an effort to appropriately describe the heat transfer phenomenon at the interface, the heat transfer at the interface between the microchannel heat sink, which is an ideally organized porous medium, and the finite-thickness substrate is examined. From the examination, it is clarified that the he heat flux distribution at the interface is not uniform for the impermeable wall with finite thickness. On the other hand, the first approach, based on the energy balance for the representative elementary volume in the porous medium, is physically reason able. When the first approach is applied to the thermal boundary condition, and additional boundary condition based on the local thermal equilibrium assumption at the interface is used. This additional boundary condition is applicable except for the very th in impermeable wall. Hence, for practical situations, the first approach in combination with the local thermal equilibrium assumption at the interface is suggested as an appropriate thermal boundary condition. In order to confirm our suggestion, convective flows both in a microchannel heat sink and in a sintered porous channel subject to a constant heat flux condition are analyzed. The analytically obtained thermal resistance of the microchannel heat sink and the numerically obtained overall Nusselt number for the sintered porous channel are shown to be in close agreement with available experimental results when our suggestion for the thermal boundary conditions is applied.

• Journal of Biomedical Engineering Research
• /
• v.24 no.1
• /
• pp.37-44
• /
• 2003

This experimental study is aimed at evaluating the hydrodynamic performance of newly designed self-expandable graft stents under steady flow condition. Two graft stents with different coating materials and a bare TiNi metallic stent for comparison test were used in the experiment. Pressure variation and velocity distribution at the upstream and downstream of the stents were measured at flow rates of 5, 10, and 15 l/min, respectively. Pressure loss due to insertion of the stent increased with increasing flow rate exponentially as expected. At a flow rate of 15 l/min, pressure loss of Polyure-thane(PU)-coated graft stent was 6 times higher than that of TiNi metallic stent, while the pressure loss of a porous Polytetrafluoroethylene(PTFE)-coated graft stent was comparable to a bare TiNi metallic stent. Velocity profiles of the porous PTFE-coated graft stent were similar to those of a bare TiNi metallic stent regardless of flow rate. Furthermore, the velocity profile of PU-coated graft stent revealed an asymmetrical and relatively low central velocity at a higher flow rate than 10 1/min, expecially, where the effects resulted in increases of wall shear stress and normal stress. The worse hydrodynamic behavior of PU-coated graft stent than the other two stents might be attributed to formation of folds due to poor flexibility of coated material when inserting the graft stent into the pipe with a more smaller size, which later gave rise non-symmetry of flow area, increase of surface roughness and jet flow via the crevice between the stent and cylinder wall.

• Journal of computational fluids engineering
• /
• v.18 no.4
• /
• pp.82-89
• /
• 2013

Database of a bleed model has been corrected and numerical simulations have been performed to control buzz using the corrected bleed model. The existing bleed model, which was developed as a part of a boundary condition model for porous bleed walls, underestimates bleed flow rate because flow accelerations near the bleed regions are ignored. Also, it overpredicts the sonic flow coefficient when the bleed plenum pressure ratio is high. To correct these problems, and to enhance the performance of the bleed model, the database has been corrected using CFD simulations to compensate for the flow acceleration near the bleed region. Futhermore, the database of the bleed model is extended with the second order extrapolation. The corrected bleed model is validated with numerical simulations of a shock-boundary layer interaction problem over a solid wall with a bleed region. Using the corrected bleed model, numerical simulations of supersonic inlet buzz are performed to find the deterrent effects of bleed on buzz. The results reveal that bleed is effective to prevent buzz and to enhance the inlet performance.