The effects of turbulence modeling on dynamic stall

TL;DR

This study compares various turbulence models in simulating dynamic stall on a pitching airfoil, highlighting their strengths and limitations through detailed numerical analysis and hybrid RANS/LES strategies.

Contribution

It evaluates the impact of different turbulence modeling approaches, including hybrid RANS/LES, on accurately capturing dynamic stall phenomena.

Findings

Hybrid RANS/LES models better capture dynamic stall vortices.

Spectral analysis reveals low-frequency contributions to force coefficients.

Comparison shows hybrid models outperform traditional RANS in dynamic stall simulation.

Abstract

A numerical investigation of the flow evolution over a pitching NACA 0012 airfoil incurring in deep dynamic stall phenomena is presented. The experimental data at Reynolds number Re = 135 000 and reduced frequency k = 0.1, provided by Lee and Gerontakos, are compared to numerical simulations using different turbulence models. After a preliminary space and time convergence study, two- and three-dimensional URANS with different turbulence models are explored, highlighting the advantages and the drawbacks. Then, the turbulence-resolving capabilities of hybrid RANS/LES strategies are exploited to recover and better represent the dynamic stall vortex. In detail, Scale-Adaptive Simulations (SAS) and Stress-Blended Eddy Simulations (SBES) are adopted. Furthermore, the LES resolved portion allows a spectral analysis of the force and moment coefficients to investigate the contribution of frequency lower than the pitching one. Finally, a comparison of the proposed approaches with other numerical simulations is given.