• Journal of computational fluids engineering
• /
• v.17 no.2
• /
• pp.35-41
• /
• 2012

This paper presents incompressible Navier-Stokes solution algorithm for 2D Free-surface flow problems on the Cartesian mesh, which was implemented to run on Graphics Processing Units(GPU). The INS solver utilizes the variable arrangement on the Cartesian mesh, Finite Volume discretization along Constrained Interpolation Profile-Conservative Semi-Lagrangian(CIP-CSL). Solution procedure of incompressible Navier-Stokes equations for free-surface flow takes considerable amount of computation time and memory space even in modern multi-core computing architecture based on Central Processing Units(CPUs). By the recent development of computer architecture technology, Graphics Processing Unit(GPU)'s scientific computing performance outperforms that of CPU's. This paper focus on the utilization of GPU's high performance computing capability, and presents an efficient solution algorithm for free surface flow simulation. The performance of the GPU implementations with double precision accuracy is compared to that of the CPU code using an representative free-surface flow problem, namely. dam-break problem.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2002.11b
• /
• pp.289-294
• /
• 2002

In this study, performance, flow characteristics and noise of a cross-flow-fan system, used in indoor unit of the split-type air conditioner, were predicted by computational simulation. Triangular elements were used to mesh the calculation domain and quadrilateral elements were attached to the blade surfaces and walls to enhance the simulation quality. The unsteady incompressible Wavier-Stokes equations were solved using a sliding mesh technique on the interface between rotating fan region and the outside. Two stripes of velocity stream inside the cross-flow-fan were shown - the one was due to the eccentric vortex and the other was due to the normal entrance flow. As the flow rate increased, the center of the eccentric vortex moved toward the inner blade tip and rear-guide, and the exiting flow still had velocity variation along the stabilizer, which can increase the noise level. The acoustic pressure was calculated by using Lowson's equation. From the calculated acoustic pressure, it was found that the trailing edge is a dominant of acoustic generation.

• Nuclear Engineering and Technology
• /
• v.20 no.3
• /
• pp.145-154
• /
• 1988

This paper is a study on an accurate and computationally efficient method for reconstructing pointwise power distributions from coarse-mesh nodal calculations. The modern nodal codes can calculate global reactor power shapes and criticality very efficiently and accurately. But inherent in the nodal procedures, there is inevitable loss of information on local heterogeneous quantities. In this study, an improved form function method which reflects the exponential transition of the thermal flux near the assembly surface is developed for the reconstruction of the heterogeneous fluxes. Use of the new form function method in several pressurized water reactor (PWR) benchmark problems reduces the maximum errors in the reconstructed thermal flux to those in the reconstructed fast flux. Even for assemblies adjacent to the steel baffle in realistic PWR cores, use of this method also results in improved pointwise power reconstruction.

• The KSFM Journal of Fluid Machinery
• /
• v.14 no.5
• /
• pp.18-23
• /
• 2011

A CFD analysis method is developed and applied for investigating the gas flow and the byproduct particle trajectory in Roots type vacuum pump. The internal fluid flow and thermal fields between the rotors and the housing of vacuum pump are analyzed by using the dynamic mesh, the numerical methods for unsteady 2-D Navier-Stokes equation and the standard k-$\varepsilon$ turbulence model of the Fluent code. Coupled with the flow simulation results, the particle trajectory of the byproduct flowing into the pump with gas stream is analyzed by using discrete phase modeling technique. The CFD analysis results show the pressure, the velocity and the temperature distributions in pump change abruptly due to the rotation of rotors, and back flows are produced due to the strong reverse pressure gradients at rotor/rotor and rotor/housing clearances. The predicted byproduct particle trajectory results also show the particles impinge on the clearance surfaces between the housing and the rotor of pump and then may form the deposit layer causing the failure of pump.

• Proceedings of the KSME Conference
• /
• 2001.11b
• /
• pp.934-939
• /
• 2001

In computational study of the flow in piston engines and the flow through moving valves, the use of moving vertices is essential for modelling flows with moving boundaries. The positions of cell vertices in such cases must be allowed to vary with time. To simulate 3-dimensional port-valve and piston-cylinder of HIMSEN 6H21/32 engine, a commercially available code, STAR-CD, was used. Changes in mesh geometry was specified by PROSTAR commands.(i.e. the Change Grid operation in the EVENTS command module.) Control of the intake flow is expected to play an important role as designers seek to obtain better fuel spray characteristics, fuel mixing and mixture preparation, combustion performance, and emissions reductions to meet national standards. As a result of analysis, velocity fields indicate the presence of a structured flow comprised of one pair of counter-rotating vortices under the intake valve during the early induction process. These flow structures remain visible for most of the intake process. As the piston moves towards BDC, these vortices develops into a larger tumbling motion that dominates the flow structure.

• Journal of the Korean Society for Precision Engineering
• /
• v.22 no.2
• /
• pp.133-138
• /
• 2005

The deformation of sheet metal due to the residual stress during blanking or piercing process, is numerically simulated by means of a commercial finite element code. Two dimensional plain strain problem is solved and then its result is applied to the deformation analysis of the lead frame. The plain strain element is applied to the 2D problem to observe the Von Mises equivalent stress concentration at the both shearing edges. As the punch penetrates into the sheet material, the stress concentration generated on both edges is getting increased to be the shearing surface. The limits of the punching depth applied to the simulation is 16% and 24% of the sheet thickness for the plain strain element and the hexahedral element, respectively. The hexahedral element and the limit of punching depth were applied to the deformation analysis of the lead frame for the blanking process. The FEM results for the lead deformation were very good agreement with the experimental ones. This paper shows that the coarse mesh has enabled to analyze the lead deformation generated due to the blanking mechanism. This simple approach to save the calculation time will be very effective to the design of the blanking tools in industries.

• 한국신재생에너지학회:학술대회논문집
• /
• 2009.11a
• /
• pp.600-605
• /
• 2009

A general purpose viscous flow solver Ansys CFX is used to study a Savonius type wave energy converter in a 3D numerical viscous wave tank. This paper presents the results of a computational fluid dynamics (CFD) analysis of the effect of blade configuration on the performance of 3 bladed Savonius rotors for wave energy extraction. A piston-type wave generator was incorporated in the computational domain to generate the desired incident waves. A complete OWC system with a 3-bladed Savonius rotor was modeled in a three dimensional numerical wave tank and the hydrodynamic conversion efficiency was estimated. The flow over the rotors is assumed to be two-dimensional (2D), viscous, turbulent and unsteady. The CFX code is used with a solver of the coupled conservation equations of mass, momentum and energy, with an implicit time scheme and with the adoption of the hexahedral mesh and the moving mesh techniques in areas of moving surfaces. Turbulence is modeled with the k.e model. Simulations were carried out simultaneously for the rotor angle and the helical twist. The results indicate that the developed models are suitable to analyze the water flows both in the chamber and in the turbine. For the turbine, the numerical results of torque were compared for all the cases.

• Nuclear Engineering and Technology
• /
• v.52 no.11
• /
• pp.2431-2441
• /
• 2020

Currently, the pin-by-pin homogenized solvers are a very active research field as they can, unlike the nodal codes, directly predict the local power, while requiring significantly less computational resources than the heterogeneous transport codes. This paper presents a recently developed pin-by-pin diffusion/SP3 solver Tortin, its spatial discretization method and the reflector treatment. Regarding the spatial discretization, it was observed that the finite difference method applied on pin-cell size mesh does not properly capture the big flux change between MOX and uranium fuel, while the nodal expansion method is more accurate but too slow. If the finite difference method is used with a finer mesh in the outer two pin rows of the fuel assembly, it increases the required computation time by only 50%, but decreases the pin power errors below 1% with respect to lattice code reference solutions. The paper further describes the coupling of Tortin with a microscopic depletion solver. Several verification tests show that the SP3 pin-by-pin solver can reproduce the heterogeneous transport solvers results with very good accuracy, even for fuel cycle depletion of very heterogeneous core employing MOX fuel or inserted control rods, while being two orders of magnitude faster.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.23 no.4
• /
• pp.363-370
• /
• 2020

In this study, pitching stability derivatives of the conical shell configuration is predicted using commercial CFD code. Unsteady flow analysis with forced harmonic motion of the model is performed using overset mesh. The test is conducted about Basic finner missile configuration. The static and dynamic stability derivatives are good agreement with available experimental data. As the same way, a conical shell is analyzed in Mach number 1.6 and various reduced frequency. The static and dynamic derivatives are obtained from the time-pitching moment coefficient histories in each of four cases of mean angle of attack. The variation of reduced frequency is not affected static and dynamic derivatives. Increasing the mean angle of attack, static derivatives are increased slowly. Comparison of the Cm curves at the steady and unsteady state results shows that the Cm curve including the damping effect is lower than otherwise case, approximately 9-18 %.

• Journal of Ship and Ocean Technology
• /
• v.12 no.2
• /
• pp.41-52
• /
• 2008

The development of a high-lift rudder is needed because low speed full ships such as the VLCC(Very Large Crude oil Carrier) have difficulty for obtaining enough lifting force from a common rudder. The rudder of a ship is generally positioned behind the hull and propeller. Therefore, rudder design should consider the interactions between hull, propeller, and rudder. In the present study, the FLUENT code and body fitted mesh systems generated by the GRIDGEN program are adopted for the numerical simulations of flow characteristics around a rudder that is interacting with hull and propeller. Sliding mesh model(SMM) is adopted to analyze the interaction between propeller rotation and wake flow behind hull. Several numerical simulations are performed to compare the interactions such as hull-rudder, propeller-rudder, and hull-propeller-rudder. Also, we consider relationships between the interactions. The results of present numerical simulations show the variation of flow characteristics by the interaction between hull, propeller, and rudder, and these results are compared with an existing experimental result. The present study demonstrates that numerical simulations can be used effectively in the design of high-lift rudder behind low speed full ship.