Large-eddy simulation of a developing turbulent boundary layer over a wall-mounted hemisphere under the influence of a surrounding ditch: Flow characteristics and aerodynamic forces

Abstract

In this study, we have investigated the interaction between a wall-mounted hemisphere and a developing turbulent boundary layer. The Reynolds number based on the hemisphere diameter and the freestream velocity is 50,000. Two configurations are considered, namely a plain hemisphere and a hemisphere with a surrounding ditch. The latter is imposed to eliminate the horseshoe vortex formed in front of the plain hemisphere. The incoming boundary layer is generated through a precursor simulation to have realistic turbulence in the upstream. Wall-resolved Large-eddy Simulation is carried out to characterize the flow features in the wake of the hemispheres. At first, the plain hemisphere mean flow and velocity fluctuations are comprehensively validated against a well-documented experimental study. In the case of the hemisphere with the ditch, it is shown that the latter produce coherent structures in the wake, termed as “ditch vortex.” The effects of the ditch vortex on the mean flow and velocity fluctuations are studied. The ditch is shown to have a substantial effect on both flow and turbulent statistics, particularly on the spanwise fluctuations. Further, the aerodynamic forces and the lateral force oscillations are compared and analyzed between the plain hemisphere and the hemisphere with the ditch. Spectral analysis is utilized to identify the dominant frequencies of the wake structures in both flow configurations. To examine the flow features in greater details, Proper Orthogonal Decomposition is performed on the symmetry plane to identify the energetic structures and their associated frequencies. The results suggest that the ditch shifts the separated shear layer and also increases the turbulent kinetic energy at higher elevations but, at the same time, greatly alleviates the spanwise velocity fluctuations near the bottom wall.