PV on Rolling Terrain

[Anton Driesse]

Has anyone ever come across a published or otherwise publicly accessible report or case study on (a) PV system(s) installed on significantly rolling terrain?

Anton

[Frederic Rivollier]

Hi Anton,

Try to contact Nithya (nithya....@canadiansolar.com) from our Japan CSI office.
She performs all the energy modeling for our projects and they are always on rolling terrain or worse on side of mountains.
Check this article she wrote recently. https://www.linkedin.com/pulse/user-experience-pvcase-complex-japanese-solar-context-nithya-susan

[Mark Mikofski]

I can second that. We have been using SolarFarmer to analyze sites in Japan, also China

Check out all these amazing youtube videos of a PV systems in China:

https://youtu.be/TVQxp9ekPe0 
https://youtu.be/lf2tN0oaX8A

here's a google image search: https://www.google.com/search?q=chinese+solar+hills&source=lnms&tbm=isch

you can do the same for Japan.

Here's an article in PVTech: https://www.pv-tech.org/news/pacifico-energy-starts-construction-on-111mw-pv-project-in-wakayama-prefect

And here's a really cool image of a really cool site in China. Don't worry it's a stock photo you can find for free on the internet:

[Alberto Caldas]

Hi! I would like to bring this post back to life :)

Have there been any advancements regarding modelling with PVlib at complex/rolling terrain?

As far as I could find out on similar posts and the pvlib docs, considering the terrain slope is only implemented for modelling SAT and backtracking. I am interested in developing/using a wider solution for fixed and SAT structures that considers (to some extent) the slope and complexity of the terrain.

Any clues, hints on that? 

The way I see it, one should make use of the PVLib feature that allows splitting a full solar PV plant into different arrays. Then, gathering bunches of PV panels according to the terrain slope and azimuth they are placed on and calculate the irradiance on each of these bunches separately. 

Thanks in advance for your time and comments!

Cheers 

[Nicholas Riedel]

Our very own pvlib maintainers Kevin and Adam just published an excellent paper on this topic: https://doi.org/10.1063/5.0202220

They present a new analytical model to adjust the backtracking angle in complex 2D terrain based on an acceptable shaded array fraction. The paper also has a nice literature review on the topic.

Both the paper and code are open access! When will we see the model integrated into pvlib?

-Nicholas

[Alberto Caldas]

Thank you very much Nicholas. I wasnt aware of the publicaiton. 

Regards

[Echedey]

Hi Nicholas and everyone else!

I've got a PR open with the current implementation of that model. You can check it out at PR#1962. I expect it to make it into the next official release.

In any case, there's a tested implementation at http://doi.org/10.5281/zenodo.10513987 that you may find useful.

Best,

Echedey.