Rooftop solar panels are more energy efficient when installed correctly
Mehmet Aslani, a geoinformatics researcher at Euler University, has developed artificial intelligence methods that optimize the placement of solar panels on a multi-level roof.
The amount of electricity generated by rooftop solar panels is difficult to calculate. For individual homeowners, solar panels can be a good investment in their homes. The challenge is to determine how many panels and how to place them in order to optimize the amount of electricity generated on the roof.
"Many rooftop solar panels produce more electricity, but if they sell for a high price, they don't pay the owners much money. Too few solar panels don't produce enough electricity from the total. Investment cost. Solar" panels are expensive. Therefore, you should Investments should be as profitable as possible for the user. And we must find a balance to achieve this,” says Muhammad Aslani.
As part of Mohamed Aslani's doctoral dissertation, Aslani and computer science professor Stefan Seibel developed artificial intelligence methods that included several steps in the process of evaluating the potential of rooftop solar panels. These methods can be useful both at the micro level for individual homeowners and at the macro level for designers and real estate companies.
“This study is useful as a basis for decision making. If a designer wants to assess the potential of solar energy for an entire area, it is important for economic and environmental reasons that the assessment is as accurate as possible.” Aslani says.
Using public sites, aerial photographs, and point target data, the AI engine can first identify all surfaces in an image of the surrounding area. The next step is to evaluate the slope and orientation of the surfaces. Third, the design of each roof is considered, including chimneys and corners. Finally, it suggests how many solar panels should be installed on each roof and how to arrange them to be as efficient as possible. There are six different AI engines in total: two in the three steps of the process described above.
In two case studies, one in Gothenburg and one in Uppsala, the new AI methods were found to be more accurate than previous ones.
"Existing methods are very optimistic. They are not as detailed as our method, so they overestimate energy efficiency. They do not account for spatial details that affect or overlap with solar panel placement, as our methods do." Seibel says.
The next step in Mehmet Aslanis's research is to develop artificial intelligence to estimate a building's energy demand by counting the number of solar panels.
“I also want to change our thinking and explore how we build buildings and roofs to get the most out of solar panels,” concludes Mohammad Aslani.
The study was published as part of the Proceedings of the 8th International Conference on Theory, Application and Management of Geographic Information Systems .
Courtesy of Yavel University
Quote: AI will make rooftop solar panels more efficient (December 13, 2022). Retrieved December 15, 2022 from https://techxplore.com/news/2022-12-ai-rooftop-solar-panels-efficient.html.
This document is protected by copyright. No part may be reproduced without written permission for personal study or research, except for accuracy. Content is provided for informational purposes only.
