Bayesian method improves rooftop solar capacity estimates in France
Nationally Consistent, Locally Incomplete: A Bayesian Remote-Sensing Audit of Rooftop Photovoltaic Registries
Machine Learning
Summary
Tracking how much solar power is on roofs is tough because many small solar panels are scattered everywhere, and official counts often miss some. The authors created a statistical method that uses satellite images to better estimate the true amount of rooftop solar power, taking into account errors and uncertainties in detection. When they tested this method in France, it gave a nationwide estimate very close to official records but also revealed places where the official counts were missing up to 61% of local solar power. This way, their approach helps get more accurate and reliable solar data.
What this means in practice
- •For energy system planners: Correct national and local rooftop solar capacity estimates to improve grid planning and renewable integration using uncertainty-aware remote sensing data.
- •For solar installation analytics teams: Identify under-reported rooftop solar deployment areas by combining Bayesian analysis with remote sensing to support market assessments and service targeting.$Commercial implications: Enables solar market analysis tools that provide better regional solar penetration insights for businesses and utilities to optimize sales and services.
Tested on one dataset.
Authors
Gabriel Kasmi, Yves-Marie Saint-Drenan, Laurent Dubus, Philippe Blanc
Abstract
Tracking the energy transition requires reliable statistics on renewable deployment. Rooftop photovoltaics (PV) are especially hard to track, owing to their decentralised nature, and the resulting inaccuracies in official statistics are known but not quantified. Remote sensing offers an independent way to identify rooftop PV systems. We introduce a Bayesian framework to estimate the ground-truth rooftop PV capacity from remote sensing detections, turning an imperfect detector into an uncertainty-aware measurement instrument. Applied to France, the corrected detections estimate a capacity of 4.03 GWp [3.96--4.11] (99% credible interval) of rooftop PV below 36 kWp, matching the transmission system operator's connection data within 3.3% nationally, while identifying local under-reports of up to 61% of local capacity. We also document and quantify a significant truncation bias in French rooftop PV open data. Beyond France, the approach paves the way for more reliable estimates of rooftop PV capacity worldwide.