FF WAT: change default WAT_Scalebox to True

[andrew-platt]

Almost ready to merge. See list below

Feature or improvement description
This PR sets WAT_Scalebox = .true., and updates the method for setting the WAT_dXdYdZ parameter with WAT = 1.

• The recommendation for wake added turbulence (WAT) is to rescale by default with. However the code and test cases were setup with a default for WAT_Scalebox = .false.. This PR changes this default to .true.. Without the rescaling, cases were occasionally appearing where the velocity from the WAT was causing wakes to propagate in the negative direction.
• When WAT = 1, the discretization in WAT_dXdYdZ is now set to match the high resolution grid if all turbine hi-res domains are identical, otherwise it will fall back to calculating from recommended practices.

Related issue, if one exists
N/A

Impacted areas of the software
This affects the wake added turbulence in FAST.Farm only.

Additional supporting information

Test results, if applicable
Tests will change slightly

Finish before merge

• update results (as needed)
• update docs

[jjonkman]

Thanks!

[andrew-platt]

There are changes in the results as the wakes evolve differently.

TSinflow

At time 90 seconds. XY slice at 90 m.

Before: WAT_dXdYdZ = [5,5,5]

After: WAT: DxDyDz set to: 10.170 10.000 10.000 (using high res grid resolution)

Note that the WAT is causing a larger amount of wake deflection in this case than would be expected indicating the default values for k_Def and k_Grad are not well suited to this case. Also note that there are other artifacts in the wake that should be investigated further.

[andrew-platt]

TSinflow_curl

At time 58 seconds. XY slice at 90 m.

Before: WAT_DxDyDz = 5.0, 5.0, 5.0

After: WAT: DxDyDz set to: 10.170 10.000 10.000 (using high res grid resolution)

Very minor differences in the wakes (almost imperceptible after the wake from T1 arrives at T2).

[andrew-platt]

LESinflow

At 48 seconds. XY slice at 90 m.

Before: WAT_DxDyDz 5.0, 5.0, 5.0

After: WAT_DxDyDz 5.0, 5.0, 5.0 (user defined). ScaleBox = .true. introduces very minor changes to results.

[andrew-platt]

Mod_Amb3

Very minor differences appear in the wake velocities (2nd decimal place, values like W1VDisY).

[andrew-platt]

TSinflowADskSED

At T=90 seconds, 90 m height.

Before: WAT_DxDyDz 5.0, 5.0, 5.0

After: WAT_DxDyDz: 10.170 10.000 10.000 and WAT_ScaleBox: .TRUE.

[andrew-platt]

The py_seastate_1 regression test is failing. This is fixed in #3065, which has not been merged yet.