The rhythms of real-world driving enable EV batteries to live far beyond the predictions of laboratory tests, according to a new study from Stanford University. Thereport, published in December inNature Energy, suggests that EV batteries could last 38 percent longer than previous lab-based estimates. That means drivers could get as much as 314,000 kilometers (195,000 miles) more out of their EVs than academic researchers believed.
The longer lifespan could alleviate the concerns of prospective buyers who are comparing the financial benefits of an EV to that of an internal combustion engine vehicle, says Simona Onori, an associate professor of energy science and engineering at Stanford and a coauthor of the paper. It could also reduce the amount ofnatural resourcesneeded to meet the growing demand for EVs. All of this could lead to greater acceptance of EVs, and a stronger secondhand market, Onori says.
Academic researchers’ previous estimates underestimated battery life because the testing protocols didn’t simulate real-world driving. Typically, when researchers test battery-cell aging in a lab, they use a constant-current cycling technique, which involves continuously charging the cells at a constant rate, fully discharging them, and then continuously charging them again. But this approach looks nothing like the frequent acceleration, braking, and parking that EV batteries actually endure.
“The more realistic the discharge, the greater the gain in lifetime,” Onori and her coauthors concluded in their paper. “These results confirm that constant current cycling is not representative of realistic conditions of use.”
Car companies typically offer 100,000-mile (160,000-km) warranties for EV batteries. The batteries may last longer than this, but their state-of-health declines. Researchers are looking for ways to improve EV battery duration and range with chemical coatings, wireless monitoring systems, and other strategies.
Real-World Testing Shows Longer EV Battery Life
Researchers have to find shortcuts when testing the lifespan of products that can last upwards of 18 years. Academic researchers have typically employed a constant-current cycling approach to battery testing, which compresses the process and mimics some EV uses such as public transportation and industrial vehicles that operate continuously and rest for shorter periods of time compared to personal vehicles. But techniques that simulate personal vehicle driving and get the job done relatively quickly are needed, the Stanford researchers say.
“We need to bring into the lab some other ways of looking at the life of the battery,”Onori says. She and coauthor William Chueh, a materials scientist at Stanford, had been comparing notes on altering testing protocols when they decided to try something new, she says.
They developed three protocols, and compared them with the standard practice of constant-current cycling. One protocol tested cells in the lab with rest periods and short pulses of charge and discharge. The charge pulses were meant to mimic the regenerative braking that gives a little energy back to the battery when the driver brings the car to a stop.
A second protocol created a synthetic dataset based on real-world driving information, and a third used data the researchers gathered from autonomous EVs that operated in a mix of urban and highway settings. The researchers assessed wear on the cells by periodically measuring various forms of resistance in battery cells.
New Testing Methods Challenge Old Models
To evaluate their data, the researchers plotted cell degradation over time for each protocol, at three different rates of discharge, or C-rates. They found that degradation was less severe in each of the new protocols, compared with the standard one, meaning the batteries had more life in them.
What’s more, an optimum rate of discharge emerged from the data. Battery lifespan increasedthe most when a battery cell discharges its capacity between 4 and 5 hours of continuous use—a C-rate between C/4 and C/5. Design factors such as battery chemistry and materials affect a battery’s rate of discharge, as does the specific power needed to move a vehicle based on its weight and engine horsepower. Further research could give EV makers a path to optimizing a battery’s discharge rate for a particular model of vehicle, Onori says.
Chris Rahn, a professor of mechanical engineering at Pennsylvania State University, who was not involved in the study, called the results “exciting” and “very surprising.” Battery experts would expect the intermittent pulses of charge, which have a higher C-rate than continuous charging models, to age batteries faster than standard procedures do. The fact that the data didn’t reflect that kind of aging is a big opportunity for further research, he says. “That’s the real takeaway,” Rahn says. “We really don’t understand how these batteries age.”
