As the use of renewable resources becomes more important, the development of tools to identify areas that could host productive PV systems parallels in importance. This project used data attained from the city of Louisville, KY, and developed an automated model to identify rooftop areas within the city that are best suited for hosting PV systems. The model identified that 30% of residential rooftops in a suburban area were suitable to host productive PV systems, whereas, the urban downtown region had only 9.7% residential rooftop suitability. This automated tool can be applied to larger areas, for example, the entirety of Louisville, KY but also has the potential to be used in other cities. While this study faced several challenges, including missing and low-quality data, it was overall considered a successful starting point for automating the identification of suitable rooftops across a large area. Further exploration, adjustment, and improvements to input data will be essential as the field progresses. As renewable energy use becomes a more prominent part of the larger energy-production picture, methods like the one proposed have the potential to contribute hugely to the field.
The team would like to thank Kirby Calvert, Adam Bonnycastle and John Lindsay for their assistance and guidance throughout this project, and also to our teaching assistant Melanie Chabot for her valuable feedback.