Machine Learning Can Help Speed EV Charging, Idaho Lab Researchers Say

September 1, 2022

by Peter Maloney
September 1, 2022

Machine learning has the potential to help bring an electric vehicle battery to a nearly fully charge in 10 minutes, according to researchers at the Department of Energy’s Idaho National Laboratory (INL).

“Currently, we’re seeing batteries charge to over 90 percent in 10 minutes without lithium plating or cathode cracking,” Eric Dufek. manager for INL’s energy storage and electric transportation department, said in a statement. At best, current protocols can fully charge an electric vehicle battery in about half an hour, he said.

When charging, lithium ions migrate from a battery’s cathode to its anode. Fast charging causes the ions to migrate more quickly, but sometimes the lithium ions do not fully move into the anode, which can cause lithium metal to build up and trigger early battery failure. Fast charging can also cause the cathode to wear and crack. Both conditions will reduce battery life and the effective range of an electric vehicle.

To charge a battery with optimal speed and minimum damage requires a huge amount of data about how different charging methods can affect a wide variety of batteries of varying designs and conditions, as well as the feasibility of applying a given charging protocol with the current electric grid infrastructure.

By inputting information about the condition of many lithium-ion batteries during their charging and discharging cycles, Idaho National Laboratory scientists say there were able to train machine learning analysis to predict battery lifetimes and the ways that different battery designs would eventually fail. The INL researchers fed that data back into the analysis to identify and optimize new protocols they then tested on real batteries.

“We’ve significantly increased the amount of energy that can go into a battery cell in a short amount of time,” Dufek said. One advantage of INL’s machine learning model is that it ties the protocols to the physics of what is actually happening in a battery, he said.

The researchers plan to use their model to develop even better methods and to help design new lithium-ion batteries that are optimized to undergo fast charging. The ultimate goal is for electric vehicles to be able to “tell” charging stations how to power up their specific batteries quickly and safely, Dufek said.

The INL scientists presented the results of their research at an Aug. 22 meeting of the American Chemical Society.