Possible Bug in BeamDyn -- Root Pitching Moment Doesn't Match Pitching Moment from First Node?

[jjonkman]

In a topic on the forum related to BeamDyn, Ryan Walker has raised concerns that the pitching moment (Mz) at the root doesn't match the pitching moment at the first node, when the first node is located at the root. I have not confirmed this result myself. Perhaps the issue has been fixed in the yet-to-be merged pull request #163?

Ryan has also reported that Figures 6.7, 6.8, and 6.9 do not show up in the readthedocs user documentation for BeamDyn.

See the following topic on the forum for more information: https://wind.nrel.gov/forum/wind/viewtopic.php?f=4&t=2215.

[rafmudaf]

@jjonkman a somewhat similar problem was fixed here: #114. However, that was merged into OpenFAST prior to the commit where these problems are reported: v1.0.0-198-gb31cfbad-dirty.

[jjonkman]

@rafmudaf -- Yes, I agree, which is why I was surprised Ryan Walker raised the issue again (you'll notice in the forum that my first question to Ryan was which OpenFAST version was being used).

[rafmudaf]

It was also mentioned that he started with v2.0.0-25-g50c88c2f-dirty then at one point modified some BeamDyn files from FAST 8 to OpenFAST (no version given) while the attached output files are at v1.0.0-198-gb31cfbad-dirty. I'm not implying that any of this is the cause of the problem, but just documenting for future reference. Both of the OpenFAST commits here are after the pull request which fixed a similar issue.

[vparmar58]

Hi @jjonkman @rafmudaf ,
I work with Ryan and conducted a test with NREL input files. I ran simulation using Test18 from the input files provided with FAST 8 package. I had to make some changes in the input files to make them compatible with the openfast. I have attached the input and out files that I ran using "openfast v1.0.0-198-gb31cfbad-dirty"
The simulation used an 8 m/s steady wind, no controllers used, and no grid connection. All the blades had same beamdyn input file. From the output file, I noticed two issues:

1. The Mz loads are different for B1RootMzr and B1N1Mzl
2. The Mz and My loads are also different for blade 1 compared to blade 2 and 3 at node 1. Please see the files attached.
   It would be really helpful to understand disparity between My and Mz loads at node 1 between the blades.
   Thanks.

Best,
Vismay.

BeamDyn_Node_1_testing.zip

[vparmar58]

Hello @jjonkman @rafmudaf ,

The issues addressed in the the previous post were resolved after I simulated with compiled version commit ed9daba-dirty obtained from (git clone https://github.com/bjonkman/openfast.git --branch b/BD-loads)

However, while calculating blade spanwise DELs (using NREL 5MW model), there is bump towards the end of the tip. We see a similar bump from our input files as well. It would be really helpful to know if there is a way to validate the spanwise loads distribution towards the tip.

Zooming towards the tip.

Note: the DEL calculations are done using only 1 sim at 12 mps windspeed. However, the bump towards the tip is consistent with our input files as well.

Thank you.

Best,
Vismay.

[rafmudaf]

@jjonkman pull request #262 addressed the documentation issues described here.