Abstract
The integration of solar technology in the built environment is realized mainly through rooftop-installed panels. In this paper, we leverage state-of-the-art Machine Learning and computer vision techniques applied on overhead images to provide a geo-localization of the available rooftop surfaces for solar panel installation. We further exploit a 3D building database to associate them to the corresponding roof geometries by means of a geospatial post-processing approach. The stand-alone Convolutional Neural Network used to segment suitable rooftop areas reaches an intersection over union of 64% and an accuracy of 93%, while a post-processing step using building database improves the rejection of false positives. The model is applied to a case study area in the canton of Geneva and the results are compared with another recent method used in the literature to derive the realistic available area.
Highlights
The built environment has the highest energy demand due to its intense human activity: buildings are the main drivers, with their continuous need for energy supply to perform routine operations
In this work we present a novel method to estimate the available rooftop surfaces for solar panel installation by combining state-of-the-art Machine Learning (ML) and computer vision techniques based on aerial images with 3D building data
The available rooftop areas are detected with an IoU of 64% and an accuracy of 93%
Summary
The built environment has the highest energy demand due to its intense human activity: buildings are the main drivers, with their continuous need for energy supply to perform routine operations. For a strategic integration of RPV in the built environment, an accurate assessment of their potential electricity generation is essential While this assessment can be very precise locally, at larger scale (regional and national) it poses a major challenge. Automatic panel fitting algorithms based on roof shape data have been developed to estimate the available area for RPV installations at regional and national scale [1,2]. While these algorithms are scalable, they depend on the availability and accuracy of the roof shape data.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have