Calculating interrow shading from pvfactors_timeseries

[blaarseth]

Hi!
Tried running a search on onter-row shading, but couldn't find anything.

I'm trying to model a bafacial ground mounted system using the pvfactors_timeseries model. I understand that the view factor model takes into account some sort of inter-row shading, but from the docs it unclear wether the shading is just averaged over the target module, meaning that if say 10 % of a module is shaded I loose roughly 10 % direct irradiance per m2, or if it is more sophisticated. I would like to add some bypass diode logic when calculating power, but that seems difficult with the pvfactors output.

Would it make more sense to model the direct in-plane irradiance separetely, and based on this apply a direct_martinez approach on the DC power?

[kandersolar]

if say 10 % of a module is shaded I loose roughly 10 % direct irradiance per m2

Yes, this is what the pvfactors model does. It is possible to retrieve the individual irradiance components from within pvfactors, but it's rather "hacky". A cleaner alternative is to discretize the module surface and calculate cell-level irradiance. This cannot be done using the function in pvlib, but you can use pvfactors (solarfactors) directly: https://solarfactors.readthedocs.io/en/latest/tutorials/Create_discretized_pvarray.html

Or, as you suggest, you could use pvfactors for the rear-side irradiance and model the front-side components separately, using shaded_fraction1d to compute shading, and then direct_martinez or some other suitable approach for the loss.

[blaarseth]

Thanks! Thats very clearifying.
I'll check out the cell-level calculations. It may be a bit overkill, as I don't currently have any way to handle the cell level mismatch as of now.