An investigation that used satellite images to identify solar panels in 48 states gave a result higher than expected, about 1.47 million installations, according to the Stanford University of California.
The study analyzed high resolution images using an algorithm to locate the photovoltaic panels and obtained valuable information for the management of the US electric power supply system.
"We were able to use recent advances in automated information to find out where all these assets are," said Ram Rajagopal, an associate professor of civil and environmental engineering at Stanford and project supervisor.
Rajagopal stressed that the analysis allows "generate ideas about where the network is going (solar energy) and how we can help it reach a more beneficial place."
The information obtained is important for energy service companies, regulatory entities and solar panel marketing companies, among others, the researcher noted.
The results were obtained through the automated learning program DeepSolar showed that the economic income is a key factor for the installation of solar panels, "although only to a certain extent".
Thus, when the annual family income exceeds $ 150,000, the decision to install natural energy is no longer important, the report said.
On the other hand, it was found that in "low and medium income households, the installations of solar systems are not frequent, despite being in areas where doing so is profitable in the long term".
In areas with abundant sunshine hours and high electricity rates, the monthly energy savings exceed the monthly cost of the equipment and, however, low and middle income families do not acquire the technology, presumably because of the high initial cost, the analysis explains. .
If the suppliers of the solar systems have this factor in mind, they could design economic models that fit the possibilities of these households, the report noted.
DeepSolar spent a month analyzing the characteristics – such as color, texture and size – of the nearly 1 billion satellite images and locating the solar installations, which would have taken years without the use of this technology, Zhecheng Wang, a civil engineer, noted. environmental and study member.