Testing compiled FAST and ADAMS/AeroDyn environments

Wind & Water Computer-Aided Engineering Software Tools
1

Hi,

I would like to address three question to you concerning the testing of my compiled FAST executable and Adams_forBladedDLL.dll.

  1. I performed CertTest.bat on my compiled FAST executable with the DISCON.dll of the NREL 5 MW wind turbine which gives me following error message
    Error.png
    What might be the reason? The other test cases run fine and agree with those in the TstFiles folder.

  2. Performing Adams_CertTest.bat on the Adams dll gives me the message that file Test01_ADAMS.cru is missing. I searched both batch files but could not find anything creating those files. How can I gain these files?

  3. Beside these testings I performed load cases according to the OC3 report with these environments. The results agree very well with the published ones. However, my power spectra of load case 3.2 are very noisy compared to yours. Here is an example
    Power.png
    Did you processed those power spectra and if so, could you tell me how you did that?

With kind regards,
Markus

2

Hello again,

I realised that regarding to my first question the use of the DISCON.dll for the NREL 5 MW makes no sense. Is it possible to get the control routines which had been used to create the reference results for the WP 1.5 MW and SWRT?

With kind regards,
Markus

3

Markus,

The *.cru files are not generated by anything. The Crunch input files come in the FAST archive.

I do not think there is an external DLL for controlling the WP 1.5 or SWRT. I believe that software is included in the FAST executable. I think PitchControl_ACH is the pitch controller Craig Hansen wrote for the WP 1.5.

Marshall

4

Dear Markus,

I’m not sure I understand your question regarding the controller. Here are the controllers used in the FAST CertTest by the WindPACT 1.5, SWRT, and NREL 5-MW models:

WindPACT 1.5 MW - Simple (built-in) variable-speed torque controller; Craig Hansen’s user-defined pitch controller, PitchCntrl_ACH.f90, for pitch, which requires a “Pitch.ipt” input file
SWRT - Kirk Pierce’s user-defined torque controller, UserVSCont_KP.f90, which requires a" spd_trq.dat" input file; nothing for pitch (fixed pitch)
NREL 5-MW - Bladed-style DISCON.dll for both torque and pitch and control

In OC3 Phase 1, we binned the PSDs into 0.0333-Hz bins to make them “smoother” in appearance.

Best regards,

5

Hi, Markus.

  1. When you run the FAST CertTest, make sure you are using the executable distributed in the FAST v7.02.00 archive. You should not be using an executable that calls the Bladed DLL (i.e., don’t use the executable you use to run the NREL 5MW models). The control input files you need for the 1.5 MW (Pitch.ipt) and the SWRT (Spd_Trq.dat) models are in the CertTest folder. If you need more information about what the code does with these files, you can look at the source code.

  2. If you are trying to run the Adams CertTest, you can get the Crunch input files and the old results from the maintenance version of the FAST v7.02.00d archive. This archive also contains the FAST and Adams time-series output files generated in the CertTest.

  3. I am not familiar with how the OC3 spectra were generated, so I will let someone else answer that one.

6

Dear Markus,

I would like to follow up on what was said about the PSDs. The noisiness you see in the chart you showed is very typical with a small delta frequency. Like Jason said, you would have to do some sort of averaging or other filtering to get a smoother curve. If you are just trying to compare PSDs between simulations (say, between FAST and ADAMS), you can use MCrunch to process the data and ask for the integral of the PSDs. That should give you much smoother curves to aid in comparison.

7

Thank you guys for your quick replies and hints.