Steady state vs. mean transient analysis of flow behind a circular cylinder
I'm looking at the current reduction behind a circular cylinder in a steady flow, at Re = 2100. I have solved it with a steady state solver and k-eps turbulence model, using Fluent. But is this a good enough approach, if one needs a detailed "image" of the flow reduction behind the cylinder? My guess is that a statistical mean velocity of a transient analysis, would show a broder flow reduction path close to the cylinder, than what would be predicted from the steady state solution, due to the detail of the vortex street. Any comments about this proposal, or does anby one know of articles regarding this subject?
I have some experience at higher Reynolds numbers (form 1e05 to 3.6e06) in simulating the two-dimensional flow around circular cylinders. I can tell you that accounting for the vortex shedding makes a significant difference in terms, for instance, of the predicted pressure drag coefficient: applying the RNG k-epsilon turbulence model, my time-averaged URANS results were far more accurate than the steady-state RANS results.
|All times are GMT -4. The time now is 10:39.|