To improve the prediction of turbulent flows, a two-scale, non-linear Reynolds stress turbulence model is proposed in this study. It is known that for the near-wall low-Reynolds number turbulent flows, the Kolmogorov turbulence scale, based on the fluid kinematic viscosity and dissipation rate of turbulent kinetic energy (ν,ε), is the dominant turbulence scale, hence it is adopted to address the viscous effects and the rapid increase of dissipation rate in the near wall region. As a wall is approached, the turbulence scale transits smoothly from turbulent kinetic energy based (k, ε) scale to (ν,ε) scale. The damping functions of the low-Reynolds number models can thus be simplified and the near-wall turbulence characteristics, such as the ε distribution, are correctly reproduced. Furthermore, to improve the prediction of the anisotropic Reynolds stresses for complex flows, a nonlinear algebraic Reynolds stress model is incorporated. The same turbulence scales are adopted in the nonlinear algebraic Reynolds stress model. The developed two-scale non-linear Reynolds stress model is first calibrated with the DNS budgets of two-dimensional channel flows, and then applied to predict the separation flow behind a backward facing step. It is found that the proposed two-scale nonlinear Reynolds stress turbulence model is capable of providing satisfactory results without increasing much computation efforts or causing numerical stability problems.
A reynolds-stress model for near-wall regions.
