Some thoughts. On the technology side, try pvlib written in python:
Regarding data sources — and these are unfortunately oriented towards Germany and documented in German:
-
Zentrale Stelle Hauskoordinaten, Hausumringe und 3D-Gebäudemodelle (ZSHH) (more here)
-
Happ, Marina, Julia Wielgosch, and Alex Kalevi Dieke (December 2022). Open Data bei Katasterdaten: Status quo, gute Beispiele und Herausforderungen [Open Data in cadastral data: status quo, good examples and challenges] (in German). Bad Honnef, Germany: Wissenschaftliches Institut für Infrastruktur und Kommunikationsdienste (WIK) [Scientific Institute for Infrastructure and Communications Services]. (also covers PV potentials)
Somewhere in the middle is the European‑wide SLiCE model which supports building-integrated photovoltaics (BIPV) as well. See this recent openmod lightning talk (which will be uploaded to YouTube as soon as I get the okay from the presenter):
and also this literature:
-
Röck, Martin, Alexander Passer, and Karen Allacker (forthcoming). “SLiCE: an open building data model for scaling high-definition …”. Life Cycle Engineering and Environmental Hotspot Analysis. (incomplete reference sorry)
-
Röck, Martin, Elena Baldereschi, Evelien Verellen, Alexander Passer, Serenella Sala, and Karen Allacker (1 November 2021). “Environmental modelling of building stocks – An integrated review of life cycle-based assessment models to support EU policy making”. Renewable and Sustainable Energy Reviews. 151: 111550. ISSN 1364-0321. doi:10.1016/j.rser.2021.111550.
The plan is to release SLiCE under suitable open licensing sometime this year — although that decision that rests with the European Commission however.