A group of scientists from the University of Central Florida said their new research in artificial intelligence could be used to create spray-on solar cells. This is a special liquid - perovskite solution.
Perovskite solar cells (PSCs) can convert sunlight into electricity just like conventional silicon solar panels. They were developed more than 10 years ago, but until now scientists have not been able to find a cheap and energy-efficient recipe for obtaining a perovskite solution.
Now machine learning technologies have come to the aid of researchers. A team from the University of Florida trained AI using more than 2, 000 published research papers on perovskite, and now it can predict which solution recipe will be optimal.
Scientists hope that the new AI will help to produce sufficiently flexible, efficient and, most importantly, cost-effective perovskite. But more research is still needed to achieve this goal.