0
SPECIAL SECTION ON RANS/LES/DES/DNS: THE FUTURE PROSPECTS OF TURBULENCE MODELING

The Use of Solution Adaptive Grid for Modeling Small Scale Turbulent Structures

[+] Author and Article Information
G. de With1

 Lafarge Roofing Technical Centers, Sussex Manor Business Park, Gatwick Road, Crawley RH12 4LN, UKg.de.with@lafarge-roofing.com

A. E. Holdø

CFD Research Group, University of Hertfordshire, Dept. Aerospace Automotive and Design Eng., Hatfield Campus, College Lane, Hatfield AL10 9AB, UK

Δtnorm is normalized with cylinder diameter and inlet velocity Δtnorm=ΔtuD.

The drag coefficient is the sum of the streamwise pressure forces acting on the cylinder relative to the density and inlet velocity C¯D=2F¯Dρu2.

Strouhal number is the frequency of the periodic wave relative to cylinder diameter and inlet velocity St=fDu.

The micro scale is the normalized filter width based on the smallest element in the mesh and calculated as Ωmin3D.

1

Correspondence to: Govert de With, Lafarge Roofing Technical Centers. Telephone: +44(0)1293 596039; Fax: +44(0)1293 596427.

J. Fluids Eng 127(5), 936-944 (May 12, 2005) (9 pages) doi:10.1115/1.1989359 History: Received July 23, 2004; Revised May 12, 2005

The use of large eddy simulation (LES) is computationally intensive and various studies demonstrated the considerable range of vortex scales to be resolved in an LES type of simulation. The purpose of this study is to investigate the use of a dynamic grid adaptation (DGA) algorithm. Despite many developments related to adaptive methods and adaptive grid strategies, the use of DGA in the context of turbulence modeling is still not well understood, and various profound problems with DGA in relation to turbulence modeling are still present. The work presented in this paper focuses on the numerical modeling of flow around a circular cylinder in the sub-critical flow regime at a Reynolds number of 3.9103. LES simulations with conventional mesh and DGA have been performed with various mesh sizes, refinement criteria and re-meshing frequency, to investigate the effects of re-meshing on the flow field prediction. The results indicate that the turbulent flow field is sensitive to modifications in the mesh and re-meshing frequency, and it is suggested that the re-meshing in the unsteady flow region is affecting the onset of small scale flow motions in the free shear layer.

FIGURES IN THIS ARTICLE
<>
Copyright © 2005 by American Society of Mechanical Engineers
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Figure 1

Dimensions of the numerical domain

Grahic Jump Location
Figure 2

A plot of the (a) mesh and (b) filter width for NO DGA ∣2.4∙106∣

Grahic Jump Location
Figure 3

A plot of the (a) mesh, (b) cell squish, (c) filter width, and (d) velocity magnitude for ∇U∣20∣1.8∙106

Grahic Jump Location
Figure 4

A plot of the (a) mesh, (b) cell squish, (c) filter width, and (d) velocity magnitude for ∇U1∕3∣20∣3.0∙106

Grahic Jump Location
Figure 5

Mesh resolution in the spanwise direction with DGA and a mesh size of 3.0∙106

Grahic Jump Location
Figure 6

Time integrated normalized velocity along streamwise axis at y∕D=0, for simulations using conventional mesh and DGA

Grahic Jump Location
Figure 7

Time integrated normalized pressure coefficient Cp on the surface of the cylinder

Grahic Jump Location
Figure 8

Time integrated normalized Reynolds stress uu axis at x∕D=1.54, for simulations using conventional mesh and DGA

Grahic Jump Location
Figure 9

Time integrated normalized Reynolds stress vv axis at x∕D=1.54, for simulations using conventional mesh and DGA

Grahic Jump Location
Figure 10

Time integrated normalized velocity along streamwise axis at x∕D=7, for simulations using conventional mesh and DGA

Grahic Jump Location
Figure 11

Time integrated normalized Reynolds stress uu axis at x∕D=7, for simulations using conventional mesh and DGA

Grahic Jump Location
Figure 12

Time integrated normalized Reynolds stress vv axis at x∕D=7, for simulations using conventional mesh and DGA

Grahic Jump Location
Figure 13

One-dimensional power spectrum of the y velocity component compared with the Kolmogorov similarity hypothesis at (a) x∕D=0.5, y∕D=0.5, (b) x∕D=1.54, y∕D=0, and (c) x∕D=7, y∕D=0

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In