Toward the development of an anisotropy-resolving subgrid-scale model for large eddy simulation

Abstract

To improve the prediction accuracy of large eddy simulation, an anisotropy-resolving subgrid-scale (SGS) model is a promising strategy. Although an anisotropic term in this type of SGS model is known to effectively enhance unsteady motions of vortex structures particularly in the near-wall region, it has not been made clear how well this term reproduces the real SGS-stress components. Therefore, we performed a detailed investigation of the model performance by an a priori test using the direct numerical simulation (DNS) data of a plane channel flow. The anisotropic SGS model is constructed by combining an isotropic linear eddy-viscosity model with an extra anisotropic term that does not produce undesirable energy transfer between the grid-scale and SGS components. This modeling concept contributes to a reasonable prediction, while maintaining computational stability. Comparison of the SGS stresses evaluated by the model expressions with those obtained directly from the DNS data provided several insights useful for further development of this type of SGS models. From the present investigation, this anisotropic SGS model was found generally to produce a reasonable trend for the SGS-stress anisotropy.