Performance ratio v1

[spaneja]

• Closes #xxxx
• I am familiar with the contributing guidelines
• Tests added
• Updates entries to docs/sphinx/source/api.rst 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 and Milestone are assigned to the Pull Request and linked Issue.

Here is a version 1 of the performance ratio implementation. It utilizes various PVLib functions and uses the poa global-weighted method for determining Tref. Currently we need to filter for clipping and low light. I figured the community should decide on best way to filter clipping out, however, I can put a place holder. Documentation will need updates, along with other components of this request.

[kandersolar]

Hey @PVPro-SPA, great to see you here! I'm happy to help guide this PR, especially with the "extra" bits like code tests and documentation.

Is this pull request intended to be a faithful implementation of the Dierauf et al. Performance Ratio? I ask because, while they might be in the spirit of that performance test and used in industry, I don't think low-light and clipping filters are mentioned in NREL/TP-5200-57991. I'm okay with either way (following the technical report exactly or extending the method to be more useful), but I want it to be a conscious and documented decision.

Also curious if this is a part of a broader pvlib scope expansion into performance testing. I wonder if pvanalytics is a better fit.

[cwhanse]

thanks @kanderso-nrel I am thinking the same that a PR function might fit better in pvanalytics, but I'm open other views.

[spaneja]

Thanks @kanderso-nrel. It is always a pleasure to cross paths and learn from you.

I agree with you on the clipping/low light filtering. If we do want something there, I wonder if we can put it as an optional parameter in the function call (low_light_filter=True) or something like that. This may be more than needed, and it may be best to stay within confines of paper.

I also agree that this seems a better fit for pvanalytics.

[cwhanse]

Merged in pvlib/pvanalytics