Invalid Output channel - FAST.Farm about openfast HOT 21 CLOSED

Dear @jjonkman,
I checked and it seems that the TwrAeroflag in the AeroDyn file is set to True, so this has to be the reason for the error regarding the high-res domain, thank you so much.

from openfast.

from openfast.

Dear @ILDENIEL,

The FAST.Farm outputs related to radial output nodes are only available for when using the polar wake formulation (Mod_Wake = 1). It appears that you are using curled wake model (Mod_Wake = 2), which is solved on a Cartesian grid. To output wake deficits with the curled wake model, you should use alternative outputs such as the flow visualization outputs or OutAllPlanes.

Best regards,

from openfast.

Dear @jjonkman,
thanks a lot for your help.
I would like to exploit your answer to ask you for a tip: if the main goal of the simulation is the assessment of the combined effect of yaw and distance of the turbines, would you use the polar or the curled wake model?
Thanks in advance

from openfast.

Dear @ILDENIEL,

The curled wake model works better than the polar model when there is sizeable skew of the flow relative to the rotor, e.g., from yaw error associated with wake steering or rotor tilt associated with floater pitch (in floating offshore wind turbines).

Best regards,

from openfast.

Dear @jjonkman,
thanks again for your help.
Best regards
Daniele

from openfast.

Dear @jjonkman,
I'm sorry to bother you again, but I still have some issues regarding the simulation of 2 on-shore turbines with Y offset = 0.

The first one is that when I try to set the z-coordinate origin of the high-resolution grid equal to the one of the low-res grid: the Z0_low is set to 6 (6m above ground) and

• when I set Z0_high to the same value the simulation does not start
• when I set Z0_high to 0 (ground level) or 1m, the simulation starts without problems.
  Why is it so?

I'm still using the curled wake model and, as you suggested, I set the OutAllPlanes flag to true.
Looking from above (in Paraview) everything seems correct, but the visualization of the wake planes does not seem to be correct.
Is there something I could have wrongly set?

Thanks in advance

from openfast.

from openfast.

Dear @ILDENIEL,

Regarding Z0_High, can you clarify what error you receiving (or share the full message written to the command prompt) when executing with Z0_High = 6 m?

Regarding the visualization, I'm not sure what you are illustrating. Can you clarify what you are showing or share the VTK files directly?

Best regards,

from openfast.

Dear @jjonkman,
These are the pictures of the .FSTF file and the message of the error, respectively.
I solved the problem by setting the Z-coordinate of the high-res grid origins at 0.

from openfast.

Regarding the visualization, I've tried to plot on the same Pataview window all the files from:

• the vtk folder
• the vtk_ff folder
  resulting from the simulation.
  The second folder is generated when I set the following flag to "True"
  OutAllPlanes [-] Output all wake planes in VTK at all time steps. Note: this option requires intensive writing to disk and will drastically slow down the simulation. DEFAULT is False.

I don't understand why the wake planes have the center at ground level and distant from the turbines, so it is a little confusing: whether their origin were on the low-res domain reference system or in the global FoR, their location seems to be wrong, or (probably) I'm missing something.

Thanks in advance
Best regards

from openfast.

Dear @ILDENIEL,

Regarding Z0_High, the error you are receiving is caused because an aerodynamic analysis node of the OpenFAST model is trying to access wind data outside of the high-resolution domain. I would guess that you have a tower aerodynamic analysis node in AeroDyn that is at or close to ground level. The solution is to ensure that all aerodynamic analysis nodes in each OpenFAST model stay within their high-resolution domain for all time steps, e.g., by lower Z0_High as you have done or raising the lowest tower aerodynamic analysis node.

Regarding OutAllPlanes, can you share an example VTK file of a plane both at time zero and well into the simulation? Also, which version of FAST.Farm/OpenFAST are you running?

Best regards,

from openfast.

I'm using OpenFAST and FAST.Farm-v3.4.1.
I think I cannot share the simulation with you, because when I try to select the planes from the Paraview pipeline window:

1. Paraview takes a lot of time to process all the files
2. I'm not managing to view the evolution of a single plane in time, since when I select the files from the folder (more than 100K elements...) they automatically appear on the visualization window.

Maybe it is just a matter of Paraview knowledge, so it is only my fault, but

this is what I see when trying to select the files.

from openfast.

Can you share a couple of the VTK files from the vtk_ff directory?

Best regards,

from openfast.

In this case there is a yawed turbine(tilt and cone = 0) upstream and one turbine with yaw, tilt and coning =0 after 3 diameters

from openfast.

in this case, without the vtk_ff_planes, the visualization works without problems

from openfast.

By adding, for example

just one plane group from the vtk_ff_planes folder, the selected plane does not move/evolve even though the simulation keeps running

from openfast.

I mean can you attach a few of the VTK files from the vtk_ff directory directly so that I can review the files rather than the Paraview plots?

Best regards,

from openfast.

only_yaw.zip
sorry I didn't understand because it is not possible to directly attach the vtk files.
Thanks again

from openfast.

Thanks for sharing. The issue is that the Z value of the ORIGIN is not being written properly in the VTK file. I don't see an issue in the latest version of source code, so, I would guess this bug was found and fixed since v3.4.1. I suggest upgrading to the newest version of FAST.Farm (currently within OpenFAST v3.5.2) to see if that solves the issue.

Best regards,

from openfast.

Thanks for your time and your disposability, I'll surely try the newest versions.
Best regards.

from openfast.