APS Logo

Near-wall modeling for LES of turbulent channel flows with rough walls

ORAL

Abstract

The interaction between turbulent boundary layers and surface roughness poses significant challenges in various propulsion and turbomachinery devices, as well as in the presence of ice on wings, impacting their aerodynamic performance and operability. However, the high-fidelity simulations of turbulent flows over rough walls using predictive computational tools remain limited. Advances in the development of roughness functions have shown promising results, revealing the necessity of considering at least two or more geometric roughness parameters to characterize near-wall velocity deficit accurately. In this study, we incorporate these geometric roughness functions to model the near-wall region to capture the effect of roughness using LES. The simulation setup consists of a turbulent flow in a channel with rough walls generated from laser scans of sand-blasted roughness (Busse et al., Comp. & Fluids, 2015). Two different simulation strategies are explored: (1) partially grid-resolved roughness using a classical algebraic wall model, and (2) fully subgrid roughness using a modified wall model that incorporates roughness functions. The focus of this work is to evaluate these modeling strategies by comparing them with the existing direct numerical simulations of Thakker et al. (JFM, 2017). The best practices for wall-modeled LES of turbulent flow over rough walls will be discussed.

Presenters

  • Brett Bornhoft

    Center for Turbulence Research, Stanford University

Authors

  • Brett Bornhoft

    Center for Turbulence Research, Stanford University

  • Suhas S Jain

    Stanford University, Center for Turbulence Research, Stanford University, CA, 94305, Center for Turbulence Research, Stanford University

  • Sanjeeb Bose

    Stanford University, Cadence Design Systems

  • Parviz Moin

    Stanford University, Center for Turbulence Research, Stanford University, Stanford Univ