The results are computed when you open the tab Results (Enter in a panel field opens it).
No results yet: set the site and the weather (first tab), optionally the obstruction and the panel; the results are computed when you open this tab.
Key figures
Average daily values per month
Orientation comparison
Annual values for the panel's orientation, the best orientations and the ones to compare; azimuth as compass angle (180 = south).
Sky hemisphere
Annual radiation per sky patch (direct normal plus diffuse from the patch), independent of the panel orientation; obstructed patches dimmed, their border outlined in cyan. Tap or hover over a patch for its values; to pan or zoom, switch the tools on in the plot's toolbar.
Average daily profile per month
Radiation on the panel in W/m² over the hour of the day: total (black), direct (orange) and diffuse (blue), each unobstructed (dashed) and obstructed (solid).
Radiation per day
Total radiation on the panel per day: unobstructed (light) behind obstructed (dark).
Carpet plots
Hourly total radiation on the panel over the year: hour of the day across, day of the year downwards; unobstructed and obstructed side by side (on a phone, one below the other).
Panel orientation
Annual radiation on the panel over azimuth and tilt, unobstructed and obstructed; star: each plot's best orientation, circle: the panel's (an optimized panel is the obstructed best, so in the unobstructed plot it differs from the star); dotted lines enclose 95 % and 90 % of the best.
PV yield estimator
Photovoltaic yield from weather data and an obstructed sky description; all computation runs in your browser (Python via Pyodide). Only the site search, the weather download and the panel map contact other services.
Steps: set the site and the weather (Site & weather), optionally describe the obstructed sky from a LiDAR scan, photos or a sky map (Obstruction), set the panel (Panel), then open the Results: they are computed when needed.
Loaded inputs and the form are kept in this browser for your next visit, including the photos (not a LiDAR point cloud, which may be large). To keep everything or move it to another device, save the project (menu ☰: one zip with all inputs and the matching results) and load it there; single files (weather, obstructed sky description, photos, config) can be saved and loaded as well.
Source code and CLI: github.com/laggondo/pv_yield. Sample weather data: PVGIS © European Union, 2001–2026. Downloaded weather data: Open-Meteo (CC BY 4.0). The first visit downloads about 60 MB (Pyodide, pvlib, pandas, bokeh), later visits use the browser cache.