Calculation of unsteady turbulent flow around obstacles using the large eddy simulation turbulence model /
The premise of the work presented here is to use a common analytical tool, Computational Fluid Dynamics (CFD), along with a prevalent turbulence model, Large Eddy Simulation (LES), to study the flow past rectangular cylinders. In an attempt to use CFD simulations to model the cylinder flow phenomen...
| Main Author: | |
|---|---|
| Format: | Thesis eBook |
| Language: | English |
| Published: |
[Place of publication not identified] :
[publisher not identified] ;
2002.
|
| Subjects: | |
| Online Access: | Link to OAKTrust copy |
| Summary: | The premise of the work presented here is to use a common analytical tool, Computational Fluid Dynamics (CFD), along with a prevalent turbulence model, Large Eddy Simulation (LES), to study the flow past rectangular cylinders. In an attempt to use CFD simulations to model the cylinder flow phenomena, a suitable CFD code (Trio_U) was selected, and implemented. A validation test of flow around a cylinder in an open channel was studied, and grid / model sensitivities were explored. Trio's LES capabilities were then extended to study obstacle flow in a closed channel. Mesh resolution tests indicated that an instability in the solution procedure disallows a grid independent solution. However, it was determined that fine and moderate grid resolutions were able to reproduce experimental results. Through further testing, it was determined that the 4th order central difference scheme, the Smagorinsky sub-grid scale model and the standard law of the wall model were most suitable for their respective applications. In addition, inlet condition tests were run which indicated that a constant inlet velocity condition is the most suitable boundary condition for these simulations. Lastly, extension to a larger grid in the z-direction resulted in increased turbulence dissipation. When the same obstacle is placed at the center of a closed channel, results were very similar to the open channel case, with slight changes in the near-wake velocities, and a slight increase of the drag coefficient. Interaction between the obstacle and the wall was negligible. However, at distances greater than x = 5D behind the obstacle, cylinder-generated vortices were observed to perturb the boundary layer slightly. When the obstacle was moved to a distance of 1D from the lower wall, significant changes in the results were observed. The drag coefficient increased by 17%, and the near-wake turbulent velocity was also shown to have increased. Vortex formation behind the cylinder was observed to occur at an angle to the channel centerline. Furthermore, substantial interaction between the obstacle and the wall was noted, both near the obstacle and further downstream. The boundary layer downstream of the obstacle was dramatically affected by the obstacle's presence. The cumulative result is a flow field which contains physical behavior that is much more complex than that of the freely-stationed obstacle. |
|---|---|
| Item Description: | "Major subject: Nuclear Engineering". Vita. |
| Physical Description: | x, 85 leaves : illustrations ; 28 cm. Also available online. Issued also on microfiche from Lange Micrographics. |
| Bibliography: | Includes bibliographical references (leaves 82-84). |