• Journal of computational fluids engineering
• /
• v.18 no.3
• /
• pp.79-86
• /
• 2013

Preconditioned compressible Navier-Stokes equations are solved for almost incompressible flows. Unstructured meshes are utilized for spatial discretization of complex flow domain. Effectiveness of the current preconditioning algorithm, with respect to various Reynolds numbers and Mach numbers, is demonstrated by the solution of canonical problems for incompressible flows, e.g. driven cavity flows.

• International Journal of Automotive Technology
• /
• v.7 no.2
• /
• pp.139-143
• /
• 2006

As an automobile tends to be high grade, the needs for more luxurious interior and comfortable HVAC system are emerged. The defrosting ability is another major factor of the performances of HVAC system. The present work is to simulate the flow and the temperature field of cabin interior during the defrost mode. The three-dimensional incompressible Navier-Stokes equations and energy equation were solved on the multi blocked grid system by the iterative time marching method and AF scheme, respectively. The present computations were validated by the comparison of the temperature field of a driven cavity and velocity field of 1/5 model scale of an automobile. Generally good agreements were obtained. By the present computation, the complicated features of flow and temperature within the automotive cabin interior could be well understood.

• Journal of computational fluids engineering
• /
• v.12 no.4
• /
• pp.38-43
• /
• 2007

Owing to the rapid progress in manufacturing technology of microscale devices, there are active research works in developing microscale propulsion systems. In this study, gas flows in nozzles with size of milli and sub-millimeter are investigated by using a CFD code based on the Navier-Stokes equations. The prediction results were compared with theoretical results of quasi-one-dimensional nozzle flow and experiment data. In general, theoretical values agree very well with the CFD results. However, theoretical values begin to deviate from the CFD and experimental data for relatively small Reynolds numbers and the nozzle shape with rectangular cross section. The primary reason for this discrepancy is due to the existence of the thick boundary layer at the wall in low Reynolds flows.

• Journal of computational fluids engineering
• /
• v.1 no.1
• /
• pp.71-80
• /
• 1996

An iterative time marching procedure for solving incompressible turbulent flow has been applied to the flows around a high speed train including cross-wind effects. This procedure solves three-dimensional unsteady incompressible Reynolds-averaged Navier-Stokes equations on a non-orthogonal curvilinear coordinate system using first-order accurate schemes for the time derivatives and third/second-order accurate schemes for the spatial derivatives. Turbulent flows have been modeled by Baldwin-Lomax turbulent model. To validate present procedure, the flow around a high speed train at zero yaw angle was simulated and compared with experimental data. Generally good agreement with experiments was achieved. The flow fields around the high speed train at 9.2°, 16.7°, and 45° of yaw angle were also simulated.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2011.11a
• /
• pp.263-267
• /
• 2011

This paper presents a numerical investigation of the deflagration to detonation transition (DDT) of flame acceleration by a shock wave filled with an ethylene-air mixture in bent tube. A model consisting of the reactive compressible Navier-Stokes equations and the ghost fluid method (GFM) for complex boundary treatment is used. A various intensities of incident shock wave simulations show the generation of hot spots by shock-flame interaction and the accelerated flame propagation due to geometrical effect. Also the first detonation occurs nearly constant chemical heat release rate, 20 MJ/($g{\cdot}s$). Through our simulation's results, we concentrate the complex confinement effects in generating strong shock wave, shock-flame interaction, hot spot and DDT in pipe.

• 한국전산유체공학회:학술대회논문집
• /
• 2006.05a
• /
• pp.185-187
• /
• 2006

• The KSFM Journal of Fluid Machinery
• /
• v.5 no.3 s.16
• /
• pp.40-45
• /
• 2002

Since the previous cut-and-try design algorithm requires much cost and time, the automated design technique with the CFD and optimum design algorithm has recently been concerned. In this work, the Navier-Stokes equation was solved to gain more detailed viscous flow information of cascade with rotor-stator interactions. The H-grid embedded by O-grid was generated to obtain more accurate solution by eliminating the branch cut of H-grid near airfoil surface. To handle the relative motion of the rotor to the stationary stator, the sliding multiblock method was applied and the cubic-spline interpolation was used on the block interface boundary. To validate present procedure, the time-averaged aerodynamic loads were compared with experimeatal data. A good agreement was obtained. The Modified Method of Feasible Direction (MMFD) was used to carry out the sensitivity analysis of the change of aerodynamic performance by the changes of the cascade geometry. The present optimization of the cascade gave a dramatic reduction of the drag while the lift maintains at the value within the user-specified tolerance.

• Journal of computational fluids engineering
• /
• v.2 no.1
• /
• pp.46-53
• /
• 1997

The flow field around a three dimensional minivan-like body has been simulated. This study solves 3-D unsteady incompressible Navier-Stokes equations on a non-orthogonal curvilinear coordinate system using second-order accurate schemes for the time derivatives, and third/second-order scheme for the spatial derivatives. The Marker-and-Cell concept is applied to efficiently solve continuity equation. A H-H type of multi-block grid system is generated around a three dimensional minivan-like body. Turbulent flows have been modeled by the Baldwin-Lomax turbulent model. To validate present procedure, the flows around the Ahmed body with 12.5° of slant angle are simulated. A good agreement with other numerical results is achived. After code validation, the flows around a mimivan-like body are simulated. The simulation shows three dimensional vortex-pair just behind body. The flow separation is also observed on the rear of the body. It has concluded that the results of present study properly agreed with physical flow phenomena.

• Journal of computational fluids engineering
• /
• v.17 no.3
• /
• pp.25-32
• /
• 2012

This work was devoted to compare two different cavitation models to study the dependency of model constants. The cavitation model of Merkle et al.(2006) and Kunz et al.(2000) were used for the present computational study. The cavitation models were coupled with the incompressible unsteady Reynolds-Averaged Navier-Stokes solver to indicate the vaporization and condensation processes. For this purpose, a preconditioning method was added as the pseudo-time term to solve the unsteady stiffness problems. For the validation of the numerical simulation, the computation was performed for the cavitating flow in a converging-diverging channel. The present results show that Merkle's cavitation model is independent to the model constants, and the higher numerical accuracy over Kunz's cavitation model.

• Transactions of the Korean Society of Mechanical Engineers
• /
• v.19 no.10
• /
• pp.2617-2629
• /
• 1995

The flow through a centrifugal compressor rotor was calculated using the quasi-3-dimensional and fully 3-dimensional Navier-Stokes solution methods. The calculated results, obtained during the development of the computer codes for both methods are discussed. In the inviscid quasi 3-dimensional analysis, stream function formulation was used for the blade to blade (B-B) plane calculations, and the streamline curvature method was used for the meridional (H-S) plane calculations. In the viscous 3-dimensional flow analysis, a control volume method based on a general rotating curvilinear coordinate system was used to solve the time-averaged Navier-Stokes equations, and a standard k-.epsilon. model was used to obtain eddy viscosity. The quasi-3-dimensional analysis reasonably predicts the pressure distributions and requires much less computation time in the region where viscous effects are not strong; however, it fails to predict velocity field and loss mechanism through the impeller passage. The viscous 3-dimensional flow analysis shows reasonable pressure distributions and typical jet-wake flow field through the impeller passage. Secondary flow and total pressure distributions on cross-sectional planes explain the loss mechanisms through the impeller.