There’s still a lot we don’t know about EV batteries. This massive new research project aims to find answers
This article originally appeared on Inside Climate News . It is republished with permission. Sign up for their newsletter here.
How much does an electric vehicle’s battery performance change in hot weather? How about cold?
If someone drives aggressively in an EV, how does that affect the battery life?
How much do variations in battery materials make a difference in how an EV performs in various conditions?
Researchers and manufacturers have partial answers to these questions based on the data they have collected. But they would know much more if they shared their data in formats they all could understand.
This is the premise behind the Battery Data Genome, a new initiative led by Argonne National Laboratory in Illinois and Idaho National Laboratory, among others. The name is a reference to the Human Genome Project, a monumental data-sharing effort launched in 1990 that contributed to innovations in medical science.
“It’s going to take a lot of data, data from a lot of sources,” said George Crabtree, a distinguished fellow at Argonne and director of the Department of Energy’s Joint Center for Energy Storage Research.
Crabtree is one of more than two dozen coauthors of a paper published this month in the journal Joule announcing the project. Regular readers will recognize him as someone I often ask to help translate battery science into plain language.
The Battery Data Genome will collect information from every part of the battery life cycle, including basic data like how batteries respond to different types of charging and discharging, and additional variables like the effects of temperature, driving speed, and differences in the materials within the batteries.
In addition to national labs like Argonne and Idaho, virtually anyone who wants to participate in the initiative can join, including universities, automakers, and other businesses. The partners can choose how much data they want to share.
“I think one of the things that everyone realizes is that some will be reluctant to join, because, you know, it compromises their secrets, trade secrets, and that’s okay,” Crabtree said. “It’s kind of an open decision for anyone who wishes to participate.”
The project aims to create a common set of standards for how battery data is formatted, so everyone is speaking the equivalent of the same language.
The organizers are hoping that once a vast amount of data is in one place, researchers and companies can use artificial intelligence and other sophisticated methods of analysis to unlock ways to make batteries that are more effective.
Sue Babinec, an Argonne battery scientist, said in an email that the announcement of the project follows more than a year of meetings and conversations among researchers about how to standardize their data for better sharing. She was the lead writer of the paper, along with Eric Dufek, a manager at the Idaho lab.
The authors’ work is an attempt to counteract what the paper says is the “current fragmented ecosystem” in the ability of researchers to build on each others’ progress, which is holding back the potential for a renaissance in battery data science.
The paper notes that there are already several data-sharing initiatives in battery science, including the Battery Data Toolkit maintained by Argonne. The new project is building on what the others have done.
Consumers, businesses, and the R&D community would be the beneficiaries of such research, in that it should lead to batteries that are less expensive, more functional, and more durable. This would apply to batteries used in EVs along with stationary battery storage and other applications.
Crabtree sees the potential for the insurance industry to use some of the data to get a better idea of how to insure EVs. Also, consumers may be able to allow their driving habits to be monitored, and drivers who are putting less stress on their batteries may be able to qualify for lower rates.
The most exciting thing, he said, is the idea that sharing data on a large scale can yield insights that are beyond even what researchers know to be looking for, insights that otherwise would not be available.
(31)
