[Field85]

Hi all,

I'm new here, I'm trying to simulate a wake behind a wind turbine, using STAR CCM+ for the simulations, I'm currently going through the tutorials, but I'm just wondering if any of you who may have prior experience may have some pointers or recommendations on what to concentrate on?

Any info much appreciated,

Thanks.

[Field85]

Can anyone tell if the overset meshing method would be the best way of simulating my problem?

[me3840]

I would not use overset for a simple problem like this. Your best method of mesh motion is using the sliding mesh model.

[Field85]

Originally Posted by me3840

I would not use overset for a simple problem like this. Your best method of mesh motion is using the sliding mesh model.

Hi Me3840, thanks for the input, I only thought about using the overset method due to a recommendation from a previous work that suggested the moving reference frame method was not good at producing the wake for their simulations.

When you say sliding mesh model, would that mean a rotating a mesh about a stationary blade?

[me3840]

Overset is not conservative, so I doubt it would be very good at producing wakes either.

MRF is not sliding mesh, those are different things.

For sliding mesh, the blade is stationary relative to the local mesh, but the local mesh moves relative to the domain. This is also just called rigid body motion.

[Field85]

Hi me3840,

Thanks for your feedback above, I've come up with this method for my set-up, could you glance through it and tell if there's anything that could be improved or changed?

Motion option - Rigid Body Motion
Mesh - Surface Wrapper - whilst applying per-region meshing
Surface Remesher - improve the quality to get it ready for volume mesh model
Retain geometric features and aligned meshing also to be activated
Unstructured, Polyhedral and Prism Layer cells for the volume mesh

Cheers

[me3840]

Sure, that looks okay. I would avoid using the surface wrapper if you can, especially for a simple model like a turbine.

[Field85]

What benefit does it give by not using the surface wrapper? Also, not sure if it matters but I'm using Star-works for model transfer purposes to save time on repairing geometry after import. Would there be any difference in setting up my mesh & simulations from within Star-works first or would it be better to just use it to transfer my CAD model and do everything from within Star ccm? I know it may be difficult answer as not many people appear to be using any CAD clients

[me3840]

In your case if you have a parameterized model the CAD client can provide a nice connection to move your designs up and downstream.

The surface wrapper takes time and can misrepresent your geometry. If you don't need it, don't use it. It's a fantastic tool for assemblies of thousands or tens of thousands of parts, but I can't imagine you have anywhere near that here.

[Field85]

Sorry to bother you again me3840, just so I have a clear picture of the setup.

-My wind turbine model will be within a specified wind tunnel with the fluid domain created by extracting the turbine model from the wind tunnel.
-Then a circular region created around the blades, which will rotate as the blades themselves will be stationary?.
-Two more circular regions created downwind of the turbine to resolve the wake of the turbine.
-Is it normal that the spindle between my blade assembly and tower should disappear? coming up as a blank region?

Thanks

[me3840]

-My wind turbine model will be within a specified wind tunnel with the fluid domain created by extracting the turbine model from the wind tunnel.

Alright, good.

-Then a circular region created around the blades, which will rotate as the blades themselves will be stationary?.

The blades are stationary relative to the rotating region. The rotating region is rotating relative to the stationary region. Therefore the blades rotate relative to the stationary region.

-Two more circular regions created downwind of the turbine to resolve the wake of the turbine.

Okay, good.

-Is it normal that the spindle between my blade assembly and tower should disappear? coming up as a blank region?

I'm not sure what you mean.

[Field85]

Thanks for the input,

On the last issue, I have a 3d model of a wind turbine with the three blades attached to the tower via a short rotating spindle, but when the meshing is complete this spindle disappears leaving a gap and I'm just wondering what the reason might be.

Also, once I have ran a steady model, and then use the results as boundary conditions for an unsteady model, am I right that the blade geometry will change from being fixed with the mesh rotating around it to the blade geometry then moving deforming the mesh around it?

[Field85]

Can anyone tell me if the gap behind my wind turbine blade in the attached .png file is a consequence of a mistake I have made or is it common? I've created a Cylinder around my blades as you can see but it appears to cut off the nacelle structure... Can anyone explain how to fix it?

[Field85]

Thought I would give an update, Images attached.

[dai549592484]

Originally Posted by me3840

Overset is not conservative, so I doubt it would be very good at producing wakes either.

MRF is not sliding mesh, those are different things.

For sliding mesh, the blade is stationary relative to the local mesh, but the local mesh moves relative to the domain. This is also just called rigid body motion.

Hi me3840

Could you please give some more details on "overset is not conservative"?
Thanks in advance

[Rab Bo]

Hello, may I know how did you model that wind turbine? Having a hard time modelling it. If you import it, how did you build the model on first hand?