[mmd_knr]

Hi all,

I am doing a CFD simulation on vertical axis wind turbines in Fluent. To reduce the computational time I'm only modelling one blade of a two balded wind turbine and apply a periodic boundary along the plane of symmetry of the flow & geometry. In order to simulate the rotation I am using a moving reference frame which applies rotational forces to the fluid domain but not the blade itself.
My directional convention is:
- Y is the direction of wind (perpendicular to rotational plane)
- Z is the radial direction of the blade
- X is the span wise direction of the blade

To ensure that my boundary conditions are correct I first simulated an empty air space which rotates at my desired rotational speed. When analysing the converged results I noticed that I get back flow from my pressure outlet. This I believe is due to the rotation pushing fluid particles that come from the velocity inlet in the radial direction. As the flow now progresses downstream more and more particles are being pushed outwards which eventually requires back flow from my pressure outlet along the rotational axis in order the preserve the mass flow rate across the Z-X plane. This back flow now creates a velocity gradient in the Y direction. It therefore severely affects the performance of my blade (if it had been modelled), depending on where I place it in the Y direction. This however does not seem logical to me.

Could some one please advice me on how to simulate flow around a wind turbine blade using a moving reference frame?
How should I specify the settings in Fluent?

Thanks in advance,
Best regards,
Mohammad

[monaya flower]

hey mohamed
i simulate HAWT .i used also MRF . for the b.c ,i used inlet velocity for the inlet and oulet flow for the outlet and periodic (rotational) ,and make the domain around the blade moving with rotational speed , and the blade is moving wall , rotational to adjacent zone with zero velocity