TY - JOUR
T1 - Unsteady simulations of the flow around a short surface-mounted cylinder
AU - Pattenden, R. J.
AU - Bressloff, N. W.
AU - Turnock, S. R.
AU - Zhang, X.
PY - 2007/2/28
Y1 - 2007/2/28
N2 - The flow around a surface-mounted circular cylinder, of height/diameter ratio 1 with a free end, is simulated using large-eddy simulation (LES) and detached-eddy simulation (DES) at a Reynolds number based on diameter of 200 000. A comparison is made between the abilities of the two models to capture flow features observed in particle image velocimetry (PIV) experiments carried out by the authors. The flow contains three interacting features formed from the junction flow between the cylinder and the ground, separation from the cylinder wall and resultant turbulent wake, and the flow over the free-end of the cylinder. Both LES and DES overpredict the length of the recirculation region by 30%, but the turbulence quantities are close to the measured values. The topology of the flow over the free-end is confirmed as consisting of an arch or 'mushroom' vortex. Due to the high Reynolds number the grid resolution is insufficient to resolve the approaching ground-plane boundary layer flow with LES, leading to inaccuracies in the horseshoe vortex system. The DES model improves this area, though still has grid induced separation effects.
AB - The flow around a surface-mounted circular cylinder, of height/diameter ratio 1 with a free end, is simulated using large-eddy simulation (LES) and detached-eddy simulation (DES) at a Reynolds number based on diameter of 200 000. A comparison is made between the abilities of the two models to capture flow features observed in particle image velocimetry (PIV) experiments carried out by the authors. The flow contains three interacting features formed from the junction flow between the cylinder and the ground, separation from the cylinder wall and resultant turbulent wake, and the flow over the free-end of the cylinder. Both LES and DES overpredict the length of the recirculation region by 30%, but the turbulence quantities are close to the measured values. The topology of the flow over the free-end is confirmed as consisting of an arch or 'mushroom' vortex. Due to the high Reynolds number the grid resolution is insufficient to resolve the approaching ground-plane boundary layer flow with LES, leading to inaccuracies in the horseshoe vortex system. The DES model improves this area, though still has grid induced separation effects.
KW - Detached-eddy simulation
KW - Finite cylinder
KW - Large-eddy simulation
UR - https://www.webofscience.com/wos/woscc/full-record/WOS:000244108200001
UR - https://openalex.org/W2132981756
UR - https://www.scopus.com/pages/publications/33846673356
U2 - 10.1002/fld.1309
DO - 10.1002/fld.1309
M3 - Journal Article
SN - 0271-2091
VL - 53
SP - 895
EP - 914
JO - International Journal for Numerical Methods in Fluids
JF - International Journal for Numerical Methods in Fluids
IS - 6
ER -