Low torque values in high wind speeds for a wind turbine

Seervan · July 20, 2019, 14:11

Greetings all

I am simulating a NREL Phase VI wind turbine in different wind speeds. everything seems fine in wind speeds of 5~7 m/s so torque and power values agree very well with experimental results. Errors are less than 3%. But it gets totally rogue in wind speeds of 8 m/s and above which I think is because of separation. I get values at best the half of experimental results and tried almost all turbulence models which some get results slightly better than the others. My Y+ values are less than 1 in most parts of the blade surface and its average value is around 4~5 throughout the blade surface. Have used 16 prism layers with a trimmer mesh. Nothing changed with polyhedral and tetrahedral meshes either. I also tried "Turbulence suppression" and chose the blade surface as the pertinent boundaries and got a little bit better results but the torque values are still way lower than the experimental values in the literature. It is getting really annoying. Any advice will be highly appreciated.

Regards

me3840 · July 28, 2019, 16:56

Can you post pictures of your mesh?

Seervan · July 29, 2019, 08:15

Hi, thanks for the reply, here is my mesh:

me3840 · July 29, 2019, 22:56

That mesh has a pretty large growth rate, the cells jump in size by like an order of magnitude in much less than the chord. It would probably pay to smooth it way out. The prism mesh to core transition ratio is not that great past what appears to be the leading edge too.

Also your domain seems very small for something with a big external influence like a wind turbine. Are you sure those boundary conditions are well posed?

Seervan · August 5, 2019, 05:13

Thanks for the reply. You're right, the transition from prism layer to core mesh is not that smooth. I prepared this mesh using Advancing Layer Mesher so there is no volumetric control to smooth out the mesh. I have generated it again using prism and polyhedral mesher to use volumetric controls so it is getting a little bit better. Do you have any other suggestions about cell count over the wind turbine blade and transition ratio, etc?

As for the domain I think that It's not the cause of the problem since I have chosen the dimensions based on the literature.

me3840 · August 5, 2019, 20:36

I'm not sure why you can't use volume controls with the advancing layer mesher, that's certainly possible to do. Anyway if you have fixed the mesh go ahead and give it a re-run.

Are those boundary conditions supposed to be freestream conditions or is this emulating some sort of wind tunnel? If I saw conditions like that in literature I'm not sure I'd trust the paper.

July 20, 2019, 14:11	Low torque values in high wind speeds for a wind turbine	#1
Seervan Member Soroush Kargar Join Date: Apr 2017 Posts: 45 Rep Power: 9	Greetings all I am simulating a NREL Phase VI wind turbine in different wind speeds. everything seems fine in wind speeds of 5~7 m/s so torque and power values agree very well with experimental results. Errors are less than 3%. But it gets totally rogue in wind speeds of 8 m/s and above which I think is because of separation. I get values at best the half of experimental results and tried almost all turbulence models which some get results slightly better than the others. My Y+ values are less than 1 in most parts of the blade surface and its average value is around 4~5 throughout the blade surface. Have used 16 prism layers with a trimmer mesh. Nothing changed with polyhedral and tetrahedral meshes either. I also tried "Turbulence suppression" and chose the blade surface as the pertinent boundaries and got a little bit better results but the torque values are still way lower than the experimental values in the literature. It is getting really annoying. Any advice will be highly appreciated. Regards

Thread Tools	Search this Thread
Show Printable Version Email this Page	Search this Thread: Advanced Search
Display Modes
Linear Mode Switch to Hybrid Mode Switch to Threaded Mode

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Problem with torque output from a turbine simulation.	pa-dundas	FLUENT	5	July 17, 2022 05:27
How to calculate torque in wind turbine using fluent?	dty08395	Fluent UDF and Scheme Programming	1	March 1, 2017 05:33
Instabilities with too low yplus values?	BendikS	OpenFOAM Running, Solving & CFD	0	February 27, 2017 05:37
Air Flow Analysis around a Wind Turbine	NightHawkRaven	FLUENT	4	August 12, 2014 17:18
Torque calculation of 2D VAWT Wind Turbine	Anonymized_JL1	FLUENT	14	September 26, 2013 03:35

July 28, 2019, 16:56		#2
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	Can you post pictures of your mesh?

July 29, 2019, 22:56		#4
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	That mesh has a pretty large growth rate, the cells jump in size by like an order of magnitude in much less than the chord. It would probably pay to smooth it way out. The prism mesh to core transition ratio is not that great past what appears to be the leading edge too. Also your domain seems very small for something with a big external influence like a wind turbine. Are you sure those boundary conditions are well posed?

August 5, 2019, 05:13		#5
Seervan Member Soroush Kargar Join Date: Apr 2017 Posts: 45 Rep Power: 9	Thanks for the reply. You're right, the transition from prism layer to core mesh is not that smooth. I prepared this mesh using Advancing Layer Mesher so there is no volumetric control to smooth out the mesh. I have generated it again using prism and polyhedral mesher to use volumetric controls so it is getting a little bit better. Do you have any other suggestions about cell count over the wind turbine blade and transition ratio, etc? As for the domain I think that It's not the cause of the problem since I have chosen the dimensions based on the literature.

August 5, 2019, 20:36		#6
me3840 Senior Member Join Date: Nov 2010 Location: USA Posts: 1,232 Rep Power: 24	I'm not sure why you can't use volume controls with the advancing layer mesher, that's certainly possible to do. Anyway if you have fixed the mesh go ahead and give it a re-run. Are those boundary conditions supposed to be freestream conditions or is this emulating some sort of wind tunnel? If I saw conditions like that in literature I'm not sure I'd trust the paper.