 |
Center for Nonlinear Studies |
In this talk our experience in applying the Navier-Stokes-alpha model of flow turbulence to wall-bounded flows will be discussed. Turbulence models are necessary for many practical problems because the full range of active scales usually cannot be resolved on an affordable computational domain. Many approaches parameterize the effects of the small, unresolved scales on the large, resolved ones using a dissipative term. In contrast, the Navier-Stokes-alpha model has a modified non-linearity, where the velocity that carries a material line is smoothed. The model has been applied successfully to a number of cases, but has seen little application to wall-bounded flows of interest to engineers. In this talk we will investigate a subgrid model based on the anisotropic Navier-Stokes-alpha equations. Results from applying the model to two test cases, turbulent channel flow and a lid-driven cavity flow, will be discussed.
For both test cases the model plays an important role in tilting vorticity into the streamwise direction. The positive and negative impacts of this will be demonstrated. For the lid-driven cavity, this is problematic when the model parameter, alpha2 was based on the
mesh spacing. An alternative definition of the model parameter, alpha2, based on velocity gradients, was more effective. The results can be seen as a step towards tackling the difficult problem of wall-bounded flows; bringing the model to a broader audience.
Host: Mark Petersen, CCS-2