In case of flow around fixed circular cylinder at Reynolds number Re=3900, transition of vortices occur in the shear layer. My question is how to handle this phenomenon during numerical analysis (computational-setup). I am using Large eddy simulation (LES) model, ICEM-CFD for meshing and FLUENT for analysis.
Set the dynamic modelling in Fluent, I suggest the NITA method. The grid requires to solve the BL over the cylinder, at least 3-4 nodes are required within y+=1.
Consider that you need to integrate the equation for some times before doing the analysis, a part of time is necessary to let the flow develop from the initial numerical conditions, a part is necessary to cumulate the samples for the statistics.
Thank you, Mr Guowei Li & Mr Flippo Maria Denaro.
My case is for flow around 'fixed' cylinder. So the mesh is static. What do you mean by dynamic modeling/meshing?
Yes, In my case, I have 1st node at position to ensure y+=1. Only maintaining y+=1 Is enough to handle transition in shear layer ? Other than that, I have very fine mesh and having 20layers within BL.
Yes, I run simulation for some time, and then start "data sampling for timestatistic' .
Thank you, Mr. Guowei Li,
I was wondering maybe there will some other criteria as well that need to be fulfilled.
Thanks for the clarification.
Regards,
Niaz
- Badges
- Science topic