CFD Online Discussion Forums

CFD Online Discussion Forums (
-   Main CFD Forum (
-   -   wind turbine simulation problem (

dennis0131 November 16, 2010 13:44

wind turbine simulation problem
Hi, guys. I’m recently using CFD to simulate NREL Phase VI wind turbine with a commercial software named SC/Tetra. Somehow the torque generation disobey the experiment, it goes down when the wind speed increase. However, when I took away the rotation, torque did increase with wind speed. I had checked geometry and analysis condition, no problem was found. I built another turbine using different airfoil and size, almost the same set-up. Then compared computational result with BEM theory and they match pretty much. Can anyone give me suggestions on what might be the problem? Thank you.

simulationman November 19, 2010 13:54

I am not sure about the simulation technique that you are using. But in general it is not necessary that the torque should increase with wind speed. For every turbomachine there will be an optimum fluid speed for maximum torque.

When you take away the rotations, I guess the torque should increase more faithfully with the wind speed. That part is fine.

dennis0131 November 20, 2010 10:48

That's what I'm wondering. I even tried self-design blades and flat plates, the torque all increased with wind speed. However, when it comes to the Phase VI turbine blades, it just gave me ridiculous results.

simulationman November 20, 2010 16:21

Tell me one thing.... Do you set the rotation speed of the blades... or let it rotate at some speed on its own under some given load........ I am asking cause as such I do not know much about wind turbine simulation.

gerritgroot November 22, 2010 05:26

There is no optimum in Torque, at least not necessarily, there's an optimum in power defined as Power=Torque*RotationalSpeed.

I think this turbine is two bladed and stall regulated, right? If so, you should run your solver at low wind speeds first, not higher than 8m/s.

Your mesh should be a segment of 180 with the condition of periodicity on the lateral walls (be careful, some solvers need a conformal mesh to do so).

Running cases without the proper rotational speed doesn't make sense, because your flow will be stalled all over the blade.

All times are GMT -4. The time now is 19:26.