Abstract
The assessment of rooftop photovoltaic potential has become increasingly accurate due to the expanding availability of satellite imagery and improvements in computer vision methods. However, the analysis of satellite imagery is impeded by a lack of transparency, reproducibility, and standardized description of the methods employed. Studies are heterogeneous, target different types of potential with redundant efforts, and are mostly not open source or use private datasets for training. With respect to the estimation of photovoltaic potential, this study proposes a conceptual frame of reference for clearly identifying tasks, their relationships, and their data. Additionally, the open-source workflow ETHOS.PASSION is introduced, which integrates the assessment of geographical, technical and economic potentials of regions under consideration along with the calculation of surface areas, orientations and slopes of individual rooftop sections. ETHOS.PASSION also includes the detection of superstructures, i.e., obstacles such as windows or existing photovoltaic installations. The novel two-look approach combines two deep learning models identifying rooftops and sections, and an additional model for the identification of superstructures. The three models show a mean Intersection Over Union between classes of 0.8478, 0.7531 and 0.4927 respectively, and more importantly display consistent results amongst randomly sampled real world images. The final results are evaluated for multiple datasets and compared against other studies, with a case study in the Aachen region of Germany being presented.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have