Modeling_options.yaml inputs not making their way to simulation?

[mlidGWT]

Hey other WEIS users. Has anyone else experienced defining an input in modeling_options.yaml (based on it being available in the modeling_schema.yaml), and yet not had that defined input make its way to the simulation?

I have several examples of this from various 'layers' within the modeling_options.yaml, but most recently I was trying to run with a particular wind file to take a look in more detail than from a previous run. Going forward I'll skip using WEIS for this type of simulation and run directly in OpenFAST, but I still feel like WEIS should have behaved differently than it did.

Within my DLC definition block, I added a line for wind_file: as provided on line 3065 of the schema, and pointed it to the full path of the desired .bts file.

And yet, when WEIS wrote the simulation files, the case_matrix shows a completely different wind seed and wind filename. I have no idea why. It ran everything else how I had specified (wind speed, wind direction, etc.). Are there certain conflicts or prerequisites for some of the inputs in the modeling_schema?

[gbarter]

Thank you for trying to improve the WEIS codebase! What you are chasing should definitely work, however we have all encountered the same issue where it is a silly mistake in the yaml syntax or the python syntax that references the nested options file. There is also the possibility that the correct OpenFAST flag wasn't set to use the bts-file you provided. For debugging, carefully check the OpenFAST input files and don't be afraid to set breakpoints in the code and use the python debugger to inspect variables directly.

[mlidGWT]

So in this case, I do think it was because of my OpenFAST input InflowWind.dat file (I'm using an openfast directory to run WEIS), but that raises a new question. I thought that WEIS would only use the specified OpenFAST .dat files as a starting point, and would overwrite inputs that are defined in the modeling_options.yaml - for instance, it did successfully overwrite the input InflowWind.dat file's WindType from 1 to 3 for the simulation in question, but it did not then overwrite FileName_BTS. Is that a bug? How am I supposed to know which inputs will and won't be overwritten by WEIS?

For more detail, it seems like WEIS did grab the right filepath as specified, and part of the filename, but the seed number is the same as from my input InflowWind file:

• Input InflowWind.dat filename: (different path)/IEA15_NTM_U4.000000_Seed1501552846.0.bts
• modeling_options.yaml defined filename: (correct path)/DLC62_1EWM50_U70.000000_Seed249069577.0.bts
• WEIS-written InflowWind filename: (correct path)/DLC62_1EWM50_U70.000000_Seed1501552846.0.bts

Do I need to specify both the wind_file and wind_seed? (I'll try this now; If so, could the schema file parameter descriptions be updated to reflect this dependency?)

[mlidGWT]

The wind seed seemed to take when I defined both the wind_file and wind_seed (it may have been sufficient to only specify wind_seed), though as I mentioned in Issue #245 it didn't seem to work for the HydroDyn wave seeds.