aoi definition in variables_style_rules.csv

[RDaxini]

• Closes #xxxx
• I am familiar with the contributing guidelines
• Tests added
• Updates entries in docs/sphinx/source/reference for API changes.
• Adds description and name entries in the appropriate "what's new" file in docs/sphinx/source/whatsnew for all changes. Includes link to the GitHub Issue with :issue:`num` or this Pull Request with :pull:`num`. Includes contributor name and/or GitHub username (link with :ghuser:`user`).
• New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
• Pull request is nearly complete and ready for detailed review.
• Maintainer: Appropriate GitHub Labels (including remote-data) and Milestone are assigned to the Pull Request and linked Issue.

Original definition says angles "between 90deg and 90deg", but this should be plus/minus 90deg (right?)

[cwhanse]

I read "angle...between" as going to tell me the two rays that define the angle, rather than the limits. What about "angle between the normal vector and the vector pointing to the sun's center"? I also think we can drop the limits +/-90 since pvlib doesn't enforce that.

[echedey-ls]

In addition to @cwhanse note, I would add normal vector of the module. To be very clear.

[kandersolar]

Just a general comment: AOI goes from 0 to 180. Negative AOI doesn't make sense.

[RDaxini]

Just a general comment: AOI goes from 0 to 180. Negative AOI doesn't make sense.

I was wondering about that...

I was trying to make sense of this line:

pvlib-python/pvlib/iam.py

Line 273 in cbe4cc5

-90 <= aoi <= 90, therefore `iam` is constrained to 0.0 outside this

Maybe in this case, the normal to the surface is defined as the 0, and then +/- 90 either side? (not saying that's a correct way to do it, just trying to interpret what was written originally)
As far as I can tell the only iam=0 is set in line 228 for aoi>=90 (no mention of any negatives) so does this function docstring need revising?

Anyway, I'll remove the range reference as suggested by @cwhanse

[RDaxini]

I read "angle...between" as going to tell me the two rays that define the angle, rather than the limits.

Makes sense

"the vector pointing to the sun's center"

Would "sun's rays" be sufficient here or is it not precise enough? (I looked here on sciencedirect and here on sandia for lexical inspiration 👀)

[cwhanse]

@RDaxini that text probably originated in PVLib for Matlab, as a warning to the user when AOI<0 was input. When AOI<0, the Martin-Ruiz function would give IAM<0. At some point in transfer to pvlib-python or creating of iam.py from functions in pvsystem.py, the message was confused.

iam.martin_ruiz handles aoi<0 by setting the IAM to zero.

To be clear, I am suggesting deleting the range from the description of AOI in the Glossary. I would correct, not delete, the statement in the martin_ruiz docstring that IAM= for AOI outside of 0 <= aoi <=90.

Let's not mix editing docstrings with creating a glossary in this PR, it won't ever get finished.

It would be very helpful if you can create an issue to accumulate docstring edits as a checklist. An entry in that checklist: this line is a cut/paste error and should be removed.

[RDaxini]

To be clear, I am suggesting deleting the range from the description of AOI in the Glossary. I would correct, not delete, the statement in the martin_ruiz docstring that IAM= for AOI outside of 0 <= aoi <=90.

Understood, we are on the same page here

Let's not mix editing docstrings with creating a glossary in this PR, it won't ever get finished.

Agreed

It would be very helpful if you can create an issue to accumulate docstring edits as a checklist. An entry in that checklist: this line is a cut/paste error and should be removed.

Okay I will do this 👍🏾

[RDaxini]

this line is a cut/paste error and should be removed.

@cwhanse is the same line on 381 and 471 correct?

Yes, but it belongs in those other two functions martin_ruiz_diffuse and interp

[RDaxini]

"the vector pointing to the sun's center"

Would "sun's rays" be sufficient here or is it not precise enough? (I looked here on sciencedirect and here on sandia for lexical inspiration 👀)

Any thoughts on this? Just an idea, not a definite suggestion.

[kandersolar]

The sun's rays originate from a disc with an angular diameter of a degree or so. It is the center of this disc which defines aoi, in the sense that pvlib uses it.

Seems like nobody objects to the definition as it currently stands in this PR, so I'll go ahead and merge it. Thanks @RDaxini!