Development of a large scale rooftop PV potential assessment tool

Development of a large scale rooftop PV potential assessment tool

Etxebarria Azanza, Josu (TU Delft Electrical Engineering, Mathematics and Computer Science)

Ziar, Hesan (mentor)
Isabella, Olindo (mentor)

Delft University of Technology

2019-12-12

Recently, climate action has become a priority in the agendas of every government, and the Netherlands was no exception. Almost 6 months ago, the so-called Klimaatakkord was released. This climate agreement sets ambitious goals to reduce greenhouse gas emissions, such as ensuring that 70% of the electricity comes from renewable sources by 2030 (only 13% does today). Considering that the Netherlands is the 2nd most densely populated country in the EU and that by 2050 95% of the population will live in urban areas it can be predicted that rooftop PV will play a major role, since it occupies no extra space and generates energy next to the consumption points. This project aims to develop a large scale rooftop PV potential assessment tool which can help urban planners, grid operators and homeowners stimulate the uptake of clean solar energy. In order to build such a tool, the first step was to extract the information of the rooftop surfaces. To do so, a model that uses openly available height point cloud data (AHN3) and cadastral data (BAG) was developed in-house. This model can extract the tilt, orientation, and area of every roof surface on a building. It can do it with an average estimation deviation of -0.02° for tilts and -0.66° for orientations. The area estimation percentage is 80.5% of the total area of the roof, mainly due to data quality issues, while the model takes an average of 5 seconds to extract the roof information of a building. The second step was to develop a PV yield calculation tool for estimating the potential of the extracted roof surfaces. As computational time is a constraint for large scale PV potential assessment, the Simplified skyline-based method was used. This method was improved by implementing two correction factors, optical airmass and angle of incidence, which can now be applied to surfaces with any tilt and orientation. Consequently, the PV yield calculation tool estimates the energy potential with an average estimation deviation of -6.64% and takes 2.7 seconds on average per building. In an effort to prove the application of the developed tool a PV potential assessment was done for the municipality of Delft. The results show that if every rooftop in the city would be covered with the state of the art PV module technology (SunPOWER MAXEON 3 | 400 W), 437.38 GWh could be generated, which represents 81% of the total electricity consumption in 2018. In the case of only suitable rooftops being covered, those with a yield of over 650 kWh/kW and payback time of less than 10 years, 384.81 GWh could be yielded, which could cover up to the 72% of the electricity demand in 2018. The research made in this project together with the positive findings hope to support in accelerating the transition towards a more sustainable world.

Solar energy
urban energy transition
rooftop PV

http://resolver.tudelft.nl/uuid:f0c41e9c-df95-4841-b088-1ffbdfb4ead1

Student theses

master thesis

© 2019 Josu Etxebarria Azanza