Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
775079 | Fluid Dynamics Research | 2008 | 32 Pages |
Abstract
In this paper, the geometrical properties of the resolved vorticity vector ϯ derived from large-eddy simulation are investigated using a statistical method. Numerical tests have been performed based on a turbulent Couette channel flow using three different dynamic linear and nonlinear subgrid-scale stress models. The geometrical properties of ϯ have a significant impact on various physical quantities and processes of the flow. To demonstrate, we examined helicity and helical structure, the attitude of ϯ with respect to the eigenframes of the resolved strain rate tensor S¯ij and negative subgrid-scale stress tensor -Ïij, enstrophy generation, and local vortex stretching and compression. It is observed that the presence of the wall has a strong anisotropic influence on the alignment patterns between ϯ and the eigenvectors of S¯ij, and between ϯ and the resolved vortex stretching vector. Some interesting wall-limiting geometrical alignment patterns and probability density distributions in the form of Dirac delta functions associated with these alignment patterns are reported. To quantify the subgrid-scale modelling effects, the attitude of ϯ with respect to the eigenframe of -Ïij is studied, and the geometrical alignment between ϯ and the Euler axis is also investigated. The Euler axis and angle for describing the relative rotation between the eigenframes of -Ïij and S¯ij are natural invariants of the rotation matrix, and are found to be effective for characterizing a subgrid-scale stress model and for quantifying the associated subgrid-scale modelling effects on the geometrical properties of ϯ.
Related Topics
Physical Sciences and Engineering
Engineering
Mechanical Engineering
Authors
Bing-Chen Wang, Eugene Yee, Donald J. Bergstrom,