• International Journal of Aeronautical and Space Sciences
• /
• v.16 no.2
• /
• pp.254-263
• /
• 2015

In this study, numerical simulations are carried out to investigate the dynamic SGS model effects in a Kerosene-LOx coaxial swirl injector under high pressure conditions. The turbulent model is based on large-eddy simulation (LES) with real-fluid transport and thermodynamics. To assess the effect of the dynamic subgrid-scale (SGS) model, the dynamic SGS model is compared with that of the algebraic SGS model. In a swirl injector under supercritical pressure, the characteristics of temporal pressure fluctuation and power spectral density (PSD) present comparable discrepancies dependant on the SGS models, which affect the mixing characteristics. Mixing efficiency and the probability density (PDF) function are conducted for a statistical description of the turbulent flow fields according to the SGS models. The back-scattering of turbulent kinetic energy is estimated in terms of the film thickness of the swirl injector.

• Journal of Advanced Marine Engineering and Technology
• /
• v.34 no.8
• /
• pp.1084-1093
• /
• 2010

Large eddy simulation of turbulent premixed flame stabilized by the bluff body is performed by using sub-grid scale combustion model based on the G-equation describing the flame front propagation. The basic idea of LES modeling is to evaluate the filtered-front speed, which should be enhanced in the grid scale by the scale fluctuations. The dynamic subgrid scale models newly introduced into the G-equation are validated by the premixed combustion flow behind the triangle flame holder. The calculated results can predict the velocity and temperature of the combustion flow in good agreement with the experiment data.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.28 no.10
• /
• pp.1255-1263
• /
• 2004

An a priori test has been conducted for the dynamic mixed model which was generalized for the prediction of passive scalar field in a turbulent channel flow The results from a priori tests indicated that dynamic mixed model is capable of predicting both subgrid-scale heat flux and dissipation rather accurately. The success is attributed to the explicitly calculated resolved term incorporated into the model. The actual test of the model in a LES a posteriori showed that dynamic mixed model is superior to the widely used dynamic Smagorinsky model in the prediction of temperature statistics.

• Journal of computational fluids engineering
• /
• v.16 no.4
• /
• pp.28-38
• /
• 2011

Large eddy simulation(LES) of fully developed turbulent pipe flow has been performed to investigate the effect of Reynolds number on the flow field at $Re_{\tau}$=180, 395, 590 based on friction velocity and pipe radius. A dynamic subgrid-scale model for the turbulent subgrid-scale stresses was employed to close the governing equations. The mean flow properties, mean velocity profiles and turbulent intensities obtained from the present LES are in good agreement with the previous numerical and experimental results currently available. The Reynolds number effects were observed in the mean velocity profile, root-mean-square of velocity fluctuations, Reynolds shear stress and turbulent viscosity.

• Journal of computational fluids engineering
• /
• v.16 no.4
• /
• pp.100-109
• /
• 2011

Large eddy simulation(LES) of fully developed turbulent pipe flow has been performed to investigate the effect of Reynolds number on the flow field at $Re_{\tau}$=180, 395, 590 based on friction velocity and pipe radius. A dynamic subgrid-scale model for the turbulent subgrid-scale stresses was employed to close the governing equations. The mean flow properties, mean velocity profiles and turbulent intensities obtained from the present LES are in good agreement with the previous numerical and experimental results currently available. The Reynolds number effects were observed in the higher-order statistics(Skewness and Flatness factor). Furthermore, the budgets of the Reynolds stresses and turbulent kinetic energy were computed and analyzed to elucidate the effect of Reynolds number on the turbulent structures.

• Wind and Structures
• /
• v.13 no.6
• /
• pp.557-580
• /
• 2010

A new inflow turbulence generation method and a combined dynamic SGS model recently developed by the authors were applied to evaluate the wind effects on 508 m high Taipei 101 Tower. Unlike the majority of the past studies on large eddy simulation (LES) of wind effects on tall buildings, the present numerical simulations were conducted for the full-scale tall building with Reynolds number greater than $10^8$. The inflow turbulent flow field was generated based on the new method called discretizing and synthesizing of random flow generation technique (DSRFG) with a prominent feature that the generated wind velocity fluctuations satisfy any target spectrum and target profiles of turbulence intensity and turbulence integral length scale. The new dynamic SGS model takes both advantages of one-equation SGS model and a dynamic production term without test-filtering operation, which is particular suitable to relative coarse grid situations and high Reynolds number flows. The results of comparative investigations with and without generation of inflow turbulence show that: (1) proper simulation of an inflow turbulent field is essential in accurate evaluation of dynamic wind loads on a tall building and the prescribed inflow turbulence characteristics can be adequately imposed on the inflow boundary by the DSRFG method; (2) the DSRFG can generate a large number of random vortex-like patterns in oncoming flow, leading to good agreements of both mean and dynamic forces with wind tunnel test results; (3) The dynamic mechanism of the adopted SGS model behaves adequately in the present LES and its integration with the DSRFG technique can provide satisfactory predictions of the wind effects on the super-tall building.

• Proceedings of the KSME Conference
• /
• 2007.05b
• /
• pp.2737-2742
• /
• 2007

Strong wind flow around a building complex was numerically studied by LES. The original motivation of this work stemmed from the efforts to develop a risk assessment technique for windstorm hazards. Lagrangian-averaged scale-invariant dynamic subgrid-scale model was used for turbulence modeling, and a log-law-based wall model was employed on all the solid surfaces including the ground and the surface of buildings to replace the no-slip condition. The shape of buildings was implemented on the Cartesian grid system by an immersed boundary method. Key flow quantities for the risk assessment such as mean and RMS values of pressure on the surface of the selected buildings are presented. In addition, characteristics of the velocity field at some selected locations vital to safety of human beings is also reported.

• Proceedings of the KSME Conference
• /
• 2001.06e
• /
• pp.392-397
• /
• 2001

A numerical study is performed to investigate the interaction between subsonic axial turbine blade boundary layer and periodically oncoming rotor induced wakes. An implicit scheme for solving the compressible Navier-Stokes equation is developed, which adopts a 4th-order compact difference for spatial discretiztion, a 2nd order Crank-Nicolson scheme for temporal discretization and the dynamic eddy viscosity model as the subgrid scale model. The efficiency and the accuracy of the proposed method are verified by applying to some benchmark problems such as laminar cylinder flow, laminar airfoil cascade flow and a transitional flat plate boundary layer flow. Computational results show good agreements with previous experimental and numerical results. Finally, flow through a stator cascade is simulated at $Re = 7.5{\times}10^5$ without free-stream turbulence intensity. The velocity fields and skin friction coefficients in the transitional region show similar trends with previous boundary layer natural transition.

• 유체기계공업학회:학술대회논문집
• /
• 2002.12a
• /
• pp.433-438
• /
• 2002

The flow field around a symmetrical airfoil in a uniform flow under the generation of noise was numerically studied and compared with experimental datum. The numerical simulation was carried out by LES which employs a deductive dynamic model as subgrid-scale model. The result of an attack angle of $6^{\circ}$ indicate that the discrete frequency noise is generated when the separated laminar flow reattaches near the trailing edge of the pressure side and the turbulent boundary layer is formed over the suction side of the airfoil near the trailing edge. The periodic behavior of vortex formation was observed around the trailing edge and it persists further downstream in the wake. The frequency of the vortex formation in the wake was consistent with that of the discrete frequency noise.

• Proceedings of the KSME Conference
• /
• 2002.08a
• /
• pp.819-822
• /
• 2002

It is necessary for the numerical simulation of 3-dimensional incompressible isotropic decaying turbulence to construct 3-dimensional initial velocity field which resembles the fully developed turbulence. Although the previous velocity field generation method proposed by Rogallo(1981) satisfies continuity equation and 3-dimensional energy spectrum, it has limitation, as indicated in his paper, that it does not produce the higher velocity moments(e. g. velocity derivative skewness) characteristic of real turbulence. In this study, a new velocity field generation method which is able to control velocity derivative skewness of initial velocity field is proposed. Brief descriptions of the new method and a few parameters which is used to control velocity derivative skewness are given. A large eddy simulation(LES) of isotropic decaying turbulence using dynamic subgrid-scale model is carried out to evaluate the performance of the initial velocity field generated by the new method. It was shown that the resolved turbulent kinetic energy decay curve and the resolved enstrophy decay curve from the initial field of new method were more realistic than those from the initial field of Rogallo's method. It was found that the dynamic model coefficient from the former was initially half the stationary value and experienced relatively short transition period, though that from the latter was initially zero and experienced relatively longer transition period.