Cinlar Subgrid Scale Model for Large Eddy Simulation

TL;DR

This paper introduces a novel subgrid scale model for large eddy simulation of incompressible flows, utilizing a covariance tensor derived from a Cinlar random velocity field based on ocean vortex structures.

Contribution

The paper develops a new SGS stress model based on Cinlar's random velocity field, integrating Clark's cross stress, and validates it against DNS data in turbulent channel flow.

Findings

The new model accurately predicts small-scale effects in LES.

It outperforms traditional Smagorinsky and one-equation models in tests.

The model effectively captures vortex-induced turbulence structures.

Abstract

We construct a new subgrid scale (SGS) stress model for representing the small scale effects in large eddy simulation (LES) of incompressible flows. We use the covariance tensor for representing the Reynolds stress and include Clark's model for the cross stress. The Reynolds stress is obtained analytically from Cinlar random velocity field, which is based on vortex structures observed in the ocean at the subgrid scale. The validity of the model is tested with turbulent channel flow computed in OpenFOAM. It is compared with the most frequently used Smagorinsky and one-equation eddy SGS models through DNS data.