Lindholm temperature model for floating PV

[IoannisSifnaios]

• [X] Part of #2068
• [X] I am familiar with the contributing guidelines
• [X] Tests added
• [X] 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`).
• [X] New code is fully documented. Includes numpydoc compliant docstrings, examples, and comments where necessary.
• [ ] Pull request is nearly complete and ready for detailed review.
• [x] Maintainer: Appropriate GitHub Labels (including remote-data) and Milestone are assigned to the Pull Request and linked Issue.

[kandersolar]

Taking a closer look at the references now, I'm having trouble understanding the calculations. I think my main source of confusion is that the references use temp_module_front and temp_module_back as inputs to the calculations, i.e. they must be known ahead of time. Is that true? How is a user supposed to come up with those values -- measured with thermocouples on the module surfaces, or can they be modeled somehow? I would have expected front and rear temperature to be outputs of a temperature model, not inputs to one.

Surely if the front and rear temperatures are known, the cell temperature could be estimated much more simply (tcell = (tfront + tback) / 2??) than with such a complex model as this... What am I misunderstanding?

[IoannisSifnaios]

You are correct, @kandersolar! The model takes the temp_module_front and temp_module_back as input to calculate the radiative heat loss toward the sky and water, respectively. In the paper, the authors measured these temperatures using thermocouples on some panels.

From a modeling perspective, this definitely makes things harder, as it is not easy to know these temperatures in advance. I am not sure what a solution could be; maybe use some other model to estimate these temperatures (e.g., sapm_module)? I am not an expert on this, so maybe someone else has a better idea?

At least for the case study they validated their model with, the cell temperature was not the average of the front and back module temperatures (though I guess that could be module-dependent...).

[IoannisSifnaios]

After communicating with the model's author, the model could be used to estimate the module temperature using an iterative procedure. However, since this is not documented in any peer-reviewed journal, it was decided not to include this model in pvlib.