While the race to develop fast-charging, better-performing batteries for electric vehicles (EVs) is ongoing, various challenges have made the process a slow one. A portion of a collaborative research effort among Stanford University, MIT and the Toyota Research Institute has developed an artificial intelligence (AI) method to tackle one of those challenges: testing evaluation time.
Led by Stanford researchers, their discovery resulted in slashing testing times by 98 percent. While the team focused on battery charge time, the machine-based learning technology could be used for other areas of battery development.
“In battery testing, you have to try a massive number of things, because the performance you get will vary drastically,” said Stefano Ermon, an assistant professor of computer science. “With AI, we’re able to quickly identify the most promising approaches and cut out a lot of unnecessary experiments.”
Their work, recently published in Nature, involved developing a program that could predict a battery’s response to different charging methods within a few charging cycles. The software was able to determine optimal charging methods in real time and discover data patterns to determine battery life, which significantly boosted the time-consuming trial-and-error methods used by humans.
“Machine learning is trial-and-error, but in a smarter way,” said Aditya Grover, a graduate student in computer science who co-led the study. “Computers are far better than us at figuring out when to explore—try new and different approaches—and when to exploit, or zero in, on the most promising ones.”
Instead of traditional methods of continuously charging until failure occurs, their software learned how to reduce the number of testing methods. This allowed the researchers to determine the optimal quick-charging solution for the batteries tested, which happened to use the highest current mid-charge instead of at the beginning of the charging process, as was the previous method.
The team’s work not only assists with charging times, but it also could impact every stage of battery development, as well as types of energy storage.
“This is a new way of doing battery development,” said Patrick Herring, coauthor of the study and a scientist at the Toyota Research Institute. “Having data that you can share among a large number of people in academia and industry, and that is automatically analyzed, enables much faster innovation.”
The research will be available for free use to usher in the next innovations in batteries. According to the team, it has the potential to be used in other industries that use machine learning for data testing.
Interested in learning more about EV innovations? Check out New Device Enables Two-Way Electric Vehicle Charging and Nanoparticles Pave the Way for a Million-Mile EV Battery.