CFD Online Discussion Forums

CFD Online Discussion Forums (https://www.cfd-online.com/Forums/)
-   Main CFD Forum (https://www.cfd-online.com/Forums/main/)
-   -   Initial Boundary condition of velocity for flow over the cylinder by LES simulation (https://www.cfd-online.com/Forums/main/199901-initial-boundary-condition-velocity-flow-over-cylinder-les-simulation.html)

AM10 March 19, 2018 03:06

Initial Boundary condition of velocity for flow over the cylinder by LES simulation
 
Hi all,

I am simulating flow over the cylinder with LES in OpenFOAM. I am not sure about the initial boundary condition for velocity at the inlet. I ran the simulation with the fixed uniform velocity at the inlet but there is overprediction in Lift and Drag coefficient. For this case, the Turbulent intensity has been kept same as experimental data.

Simulation: Cd=1.48, Clrms=0.808
Experiment: Cd=1.14 Clrms=0.36

Please provide me some suggestion to overcome this problem

FMDenaro March 19, 2018 03:43

Quote:

Originally Posted by AM10 (Post 685664)
Hi all,

I am simulating flow over the cylinder with LES in OpenFOAM. I am not sure about the initial boundary condition for velocity at the inlet. I ran the simulation with the fixed uniform velocity at the inlet but there is overprediction in Lift and Drag coefficient. For this case, the Turbulent intensity has been kept same as experimental data.

Simulation: Cd=1.48, Clrms=0.808
Experiment: Cd=1.14 Clrms=0.36

Please provide me some suggestion to overcome this problem


First, you need to ensure that your grid is enough refined over the cylinder. What about dx+,dy+,dz+?
Then, the inflow in LES should be properly generate being the turbulent field 3D and time-dependent. You can read this issue in the book of Sagaut.
Alternatively, you could assume a laminar inlet with a small fluctuations superimposed and set the inlet far from the cylinder in such a way to let the flow physically correlating.


All times are GMT -4. The time now is 06:40.