Researchers Develop Tool to Identify Rooftop’s Potential for Solar Systems

‘DeepRoof’ approach uses satellite imagery to accurately determine roof geometry, nearby structures, and trees that affect the solar potential of a roof

September 9, 2019

thumbnail

A research team from the University of Massachusetts Amherst, U.S., has proposed a new, data-driven approach that uses machine learning techniques and widely available satellite images to identify roofs that have the maximum potential to produce cost-effective solar power.

Led by Prashant Shenoy and Subhransu Maji, the faculty from the University of Massachusetts Amherst, College of Information and Computer Sciences (CICS), the team of scientists state that the new data-driven ‘DeepRoof’ approach takes advantage of recent advances in computer vision techniques and uses satellite imagery to accurately determine roof geometry, nearby structures and trees that affect the solar potential of the roof.

The researchers cite industry data showing that the global rate of solar energy installations grew by 30% in one year, and the average cost of installing solar has fallen from $7 (~₹501)/W to $2.8 (~₹200.4)/ W, making rooftop solar attractive to many more homeowners. According to the scientists, “The progress of rooftop installations is often slowed by a shortage of trained professionals who must use expensive tools to conduct labor-intensive structure assessments one by one.”

Shenoy and his colleagues presented their new ‘DeepRoof’ tool during the 25th Association for Computing Machinery’s Special Interest Group on Knowledge Discovery and Data Mining (ACM SIGKDD) conference held recently in Anchorage, Alaska.

In an email interview with Mercom, Stephen Lee, a Ph.D. student at CICS and the lead author, said, “Depending on the location, there may be different factors that affect the growth of solar installations; for example, government efforts, cheap and efficient solar, and renewable targets to fight climate change. In the U.S., rooftop solar has increased primarily because it is now very easy and cost-effective to install solar. In California, they follow a zero-money down lease model, where homeowners have no upfront costs, making it easy for them to install solar. Also, for most homeowners, installing solar cut the electric bill by 40% to 50%.”

He pointed out that the solar potential estimation of a roof can substantially benefit homeowners deciding to adopt solar, “but current automated tools work only for cities and towns where LIDAR data is available, thereby limiting their reach to just a few places in the world.”

LIDAR (light detection and ranging), is a remote sensing method that uses light in the form of a pulsed laser to measure ranges at variable distances to the Earth.

“DeepRoof estimates can be used to identify ideal locations on the roof for installing solar panels,” Lee adds.

The team experimented with DeepRoof using different roof shapes and sizes from six different cities to recognize and extract planar roof segments, Lee says. Results show that DeepRoof can identify the solar potential of roofs with 91% accuracy. Further, the tool can be scaled to automatically analyze satellite images of an entire city to identify all building roofs with the most solar potential.

“Trained professionals are required to assess the potential of your roof and to analyze whether there is any benefit of installing solar. If your rooftop does not generate enough solar energy, there will be no economic benefits in the long run. Instead of having trained professionals, the initial step is to use artificial intelligence to estimate the suitability of a roof for installing solar,” said Lee adding that this was one of the main motivations behind their research work.

Asked how they came up with the idea of ‘DeepRoof’ tool, Lee said, “The main idea of DeepRoof came about from the fact that current automated approaches do not work for large parts of the world. Solar estimates are mostly available for big cities. If you look at Google Sunroof or other tools, many smaller towns do not have coverage. We wanted to find a solution that can work in other parts of the world.”

Further, Lee claims that the USP of the tool is that it has an accurate way of determining rooftops which can provide better solar estimation or the amount of sunlight a roof can receive.

Along with Lee, Shenoy, and Maji, the research team also includes CICS alumni, Srinivasan Iyengar, and Menghong Feng.

Meanwhile, the researchers at the University of Waterloo have developed a method to utilize better the amount of energy that’s collected by solar panels. Read more about this innovation here.

Recently, the Ministry of New and Renewable Energy (MNRE) invited proposals for research and development (R&D) and demonstration in the field of solar energy. The proposals have been sent to start-ups, industries, and R&D laboratories. This move will also facilitate the expansion of solar capacity making use of the latest innovations and trends in the field.

Anjana is a news editor at Mercom India. Before joining Mercom, she held roles of senior editor, district correspondent, and sub-editor for The Times of India, Biospectrum and The Sunday Guardian. Before that, she worked at the Deccan Herald and the Asianlite as chief sub-editor and news editor. She has also contributed to The Quint, Hindustan Times, The New Indian Express, Reader’s Digest (UK edition), IndiaSe (Singapore-based magazine) and Asiaville. Anjana holds a Master’s degree in Geography from North Bengal University, and a diploma in mass communication and journalism from Guru Ghasidas University, Bhopal.

RELATED POSTS