A theory for the subfilterscale model in largeeddy simulation
Abstract
A new subfilterscale (SFS) stress model for largeeddy simulation (LES) is proposed using a series expansion to represent the unknown full velocity in terms of the filtered velocity. Using this estimate of the full velocity, SFS models can be derived to an arbitrary order of accuracy in the filter width. Furthermore, the SFS model satisfies the SFS stress evolution equations to the specified order of accuracy. The modeled SFS stress is thus influenced by buoyancy, viscous, pressure, and Coriolis effects just as the velocity field is. The model has no free parameters, and is simple to implement. The series expansion model is of scalesimilarity form, and reduces to the Bardina model at lowest order. Preliminary validation of the model is performed with a priori tests using a direct numerical simulation of sheared, stablystratified homogeneous turbulence. It is found that the model exhibits very high correlations with the exact SFS stress, significantly higher than commonly used eddy viscosity models. A posteriori tests are also performed for the neutrally stratified atmospheric boundary layer (ABL). The results show good agreement with previous numerical studies performed in the neutral ABL.
 Publication:

APS Division of Fluid Dynamics Meeting Abstracts
 Pub Date:
 November 2001
 Bibcode:
 2001APS..DFD.AE005K