Error using curl model

[MYMahfouz]

Dear Jason,

I am receiving the following error when trying to use the curl model.

I am using the same models that already worked with the polar model just setting Mod_Wake = 2 instead of 1 for the polar implementation.

Am I overlooking something? Should I update other inputs for the curl model?

 Time: 19 of 4200 seconds.  Estimated final completion at 01:34:25.

FARM_CalcOutput:AWAE_CalcOutput:LowResGridCalcOutput:The center of wake plane #1 for turbine #1
has passed the lowest-most X boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #1 for turbine #1 has passed the lowest-most Y
boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #1 for turbine #1 has passed the upper-most Z
boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #2 for turbine #1 has passed the lowest-most X
boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #2 for turbine #1 has passed the lowest-most Y
boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #2 for turbine #1 has passed the upper-most Z
boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #3 for turbine #1 has passed the lowest-most X
boundary of the low-resolution domain.
LowResGridCalcOutput:The center of wake plane #3 for turbine #1 has pa

                                                                                                  
 Time: 23 of 4200 seconds.  Estimated final completion at 01:16:51.
Warning: SkewedWakeCorrection encountered a large value of chi (-174.7 deg), so the yaw
correction will be limited. This warning will not be repeated though the condition may persist.
See the AD15 chi output channels, and consider turning off the Pitt/Peters skew model (set
SkewMod=1) if this condition persists.

The BEM solution is being turned off due to low TSR.  (TSR = -2.30740E-05). This warning will not
be repeated though the condition may persist. (See GeomPhi output channel.)

T9:FARM_UpdateStates:FWrap_Increment:FAST_Solution:CalcOutputs_And_SolveForInputs:SolveOption2:AD_
CalcOutput:RotCalcOutput:BEMT_CalcOutput(node 37, blade 1):UA_CalcOutput:Mach number exceeds 0.3.
Theory is invalid. This warning will not be repeated though the condition may persist.
BEMT_CalcOutput(node 38, blade 1):UA_CalcOutput:Mach number exceeds 1.0. Equations cannot be
evaluated.

Best regards

[jjonkman]

Dear @MYMahfouz,

If all you've changed in your FAST.Farm input is Mod_Wake, from 1 to 2, my guess is that you are not following the modeling guidance for curled wake. In particular, when using Mod_Wake = 2, DT_Low is tied to dr, see: https://openfast.readthedocs.io/en/main/source/user/fast.farm/ModelGuidance.html#low-resolution-domain. (Using a larger value of DT_Low will result in numerically instability of the curled wake model.) Please note, though, that dr can likely be bigger than you are using, e.g., dr = D/10 or D/15, as documented here: https://openfast.readthedocs.io/en/main/source/user/fast.farm/ModelGuidance.html#wake-dynamics-parameters. (This same guidance on dr should actually apply to the polar model as well, but we still have to correct this section of the FAST.Farm modeling guidance.)

Best regards,

[MYMahfouz]

I also sometimes get this error and restarting the simulation fixes it. The error appear right after the Mann box is read and before all time steps.

Farm_Initialize:AWAE_Init:AWAE_UpdateStates:InflowWind_CalcOutput:InflowWind_GetSpatialAverage:Cal
cOutput:IfW_FFWind_CalcOutput [position=(NaN, NaN, NaN) in wind-file coordinates]: FF wind array
boundaries violated: Grid too small in Y direction. Y=NaN; Y boundaries = [-2557.5, 2557.5]

[jjonkman]

I would not expect that running multiple simulations with the same input files would produce different results. Are you running FAST.Farm with OpenMP parallelization?

@ebranlard -- Any ideas?

[MYMahfouz]

I am not using OpenMP.

Just a quick explanation on why i started playing with NumRadii values. As the figure here this is at the beginning of simulation. There is this unstable wake behind the turbine. The width of this instability is the NumRadii or at least that's what i think.

After few seconds this disappears and only a part of it, that i circled remains.

From what i saw this changes a lot with changing NumRadii.

Another observation that makes me think something is weird is that this always happens only behind the first turbine in the .fstf file. At the beginning i thought maybe the wind field is corrupted, but this happens for all wind directions in which i rotate the turbines and not the wind field. Therefore, the first turbine is located at different positions of the wind field and still this happens in all wind directions.

These observations i made with the binary for linux i didnt check if similar observations can be made with the windows executable.

[MYMahfouz]

Sorry for spamming, but one last thing that i forgot to mention cause it was not so important for me but maybe it can help you tell if there is a difference between the difference in compilation.

For the inflowind input file:
In the executable on windows for the values of of nx, ny, and nz, I had floats so for example my nz was 128.0.

In the linux binary this caused an error and i had to use integers so i switched the value to 128.

I thought it is not that important but maybe this means there are more differences in how the inflowind module is compiled?

[ebranlard]

I don't have much idea of what can be going on... I haven't seen these issues before.