Correcting onshore/monopile tower definition in yaml and models

[jennirinker]

This PR is addressing the issues with the monopile/onshore tower documented in #147. It is currently a work in progress and should not be merged. I will use this comment as a master tracker for the status of the PR, and subsequent comments will contain discussions on specifics.

Purpose

Correct the tower yaml, model files, Excel file to match the design and add a test or two.

Type of change

What types of change is it?

• [x] Bugfix (non-breaking change which fixes an issue)
• [ ] New feature (non-breaking change which adds functionality)
• [ ] Breaking change (non-backwards-compatible fix or feature)
• [ ] Code style update (formatting, renaming)
• [ ] Refactoring (no functional changes, no API changes)
• [x] Documentation update
• [x] Maintenance update
• [ ] Other (please describe)

Testing

To be described after we discuss what we want to test.

Tasks to complete before merge

• [x] Update yaml file to match original design
• [x] Update the HAWC2 model to match the yaml
• [x] Update the Excel sheet based on the new yaml
• [x] Update OpenFAST based on the new yaml
• [x] Define and add a test (or multiple tests)
• [ ] Document the tower design/interpretation in the wiki (or is there a better place?)

Final checklist

• [x] I have run existing tests which pass locally with my changes
• [x] I have added new tests or examples that prove my fix is effective or that my feature works
• [ ] I have added necessary documentation

[jennirinker]

@gbarter I've finished the yaml and HAWC2 update, so it's time for us to discuss some things before I can add more commits.

Specific discussion topics (outlined in detail in each section below):

• I made a small change in the tower design specification -- let me know if you don't agree with it
• Do you need to regenerate the Excel/OpenFAST files with the new yaml?
• Should we make a wiki page with notes on the tower specifics?
• What do we want to test?

Note on slight change to tower design

I have intentionally changed the tower definition in the yaml file slightly to not match the design document. Specifically, I removed the 15.001 line and changed the 15.0 thickness to 39.496. This was to remove a vertical jump in the structural properties of the tower.

My proposal/implementation is that the yaml content for the monopile looks like this	Station [m]	Thickness [mm]
...	...
10	42.242
10.001	41.058
15	41.058

and the content for the tower looks like this	Station [m]	Thickness [mm]
15.000	39.496
28.000	39.496
28.001	36.456
...	...

Excel/OpenFAST files

I'm not sure if my updates in the yaml definition impacts the OpenFAST or Excel files. Do you need to regenerate those files?

If you do need to remake those files, feel free to commit/push them to the fix_onshore_tower_interpretation branch.

New wiki page?

I think it could be worthwhile to make a wiki page (or documentation somewhere) that explains how the HAWC2 and OpenFAST tower models are created from the yaml, and how the yaml is interpreted. HAWC2 is relatively straightforward and is in the original report, but I can't remember if the OpenFAST procedure (WISDEM) is also written there?

What do we want to test?

There are are few different things that I could add to the testing suite, and I'd like to get your opinion on which are the best for the repo:

1. Verify that the tower definition in the yaml matches some hard-coded arrays
2. Verify that the HAWC2 structural file for the tower matches the yaml definition (I'm 99% I will add this regardless of the other tests)
3. Verify that the HAWC2 structural properties match OpenFAST (only works if OpenFAST takes the structural values directly from the yaml, like HAWC2 does)
4. Don't cross-verify distributed properties of HAWC2/OpenFAST, but verify statistical properties like total mass, moment arm, mean bending stiffness, etc.
5. Something else?

The most rigorous would be 1, 2, and 3, but that depends on how the OpenFAST tower properties are calculated from the yaml. What do you think?

[jennirinker]

Current status of yaml/HAWC2/OpenFAST properties. Note that yaml no longer matches original design at base for reason explained above; HAWC2 matches yaml, OpenFAST matches neither.

[gbarter]

Good stuff @jennirinker !

In answer to your questions and comments:

• I updated the OpenFAST files with the updated tower discretization
• Ditto the Excel sheet
• Do you mean a new Wiki page or just an FAQ entry. An FAQ entry sounds good to me. Probably don't need a whole page of text
• I'm on board with all of your testing checks (1-4). Let me know which ones you want me to take. If it helps, OpenFAST files are generated from the YAML via WISDEM within WEIS.

[jennirinker]

Current status:

[gbarter]

Looks pretty good to me!

[jennirinker]

This is now ready for review/merge, with a few small caveats (see final section below).

Excel has small jump in beginning of tower

Because the Excel is the exact tower definition as written in the report, it has a small bump as the thickness transitions from 15.000 m to 15.0001 m. This is fine, IMO. I have written the tests such that they correct this value before comparison with yaml/OpenFAST/HAWC2.

New test

I added a test that:

1. Checks that the Excel and yaml tower definitions (outer diameter, thickness, etc.) are consistent.
2. Checks that the re-calculated distributed properties from the yaml match the Excel, HAWC2, and OpenFAST.

This is sufficient in my opinion -- let me know if you think something else should be added.

I also rearranged slightly an older test, but no functionality was changed.

Proposed wiki entry

Now that the OpenFAST/HAWC2 towers match almost exactly, I agree a short wiki entry in the FAQs is fine. What about:

Question: Why is are the tower properties for the onshore/monopile model different than in the report? There is a discrete jump in the thickness from 15.000 to 15.001 m in the original tower design, but in the yaml and in OpenFAST/HAWC2, the "tower" body is modelled from 15 m upwards. To prevent these models having a discrete jump in thickness in the bottom part of the tower, the properties at the base are changed slightly. Please see Issue #147 and pull request #149 for more details.

Final pre-merge steps

1. @gbarter reviews what I did, make sure all the tests run on his side
2. @gbarter Do you mind updating the wiki? I would really like to get this merged today, but I also want to go home...
3. Someone (not me) merges