Big Data Aims to Ease Electric Vehicle Driver Anxiety
October 21, 2014Researchers from North Carolina State University have developed software that estimates how much farther electric vehicles can drive before needing to recharge. The technique requires drivers to plug in their destination and pulls in data on a host of variables to predict energy use for the vehicle.
“Electric cars already have range-estimation software, but we believe our approach is more accurate,” says Dr. Habiballah Rahimi-Eichi, a postdoctoral researcher at NC State and lead author of a paper on the work.
“Existing technologies estimate remaining range based on average energy consumption of the past 5 miles, 15 miles, etc.,” Rahimi-Eichi says. By plugging in the destination, the software looks at traffic data, whether the driver will be on the highway or in the city, weather, road grade and other variables. This predictive, big-data approach reduces the range estimation error to a couple of miles. “In some case studies, we were able to get 95% range estimation accuracy,” Rahimi-Eichi says.
The software takes all of the data related to the route between starting point and destination and uses big data techniques to determine which pieces of information are important and extract features that can be plugged into an algorithm to estimate how far the vehicle can go before recharging.
Two other variables are also plugged into the algorithm: the performance characteristics of the vehicle and its battery; and the amount of charge remaining in the battery. The state of charge is estimated using a technique developed and patented by Rahimi-Eichi and Dr. Mo-Yuen Chow in 2012. Chow is a professor of electrical and computer engineering at NC State and a co-author of the paper.
“People have a lot of ‘range anxiety’ in regard to electric vehicles – they’re afraid they’ll get stuck on the side of the road,” Chow says. “Hopefully, our new range estimation software will make people more confident about using electric vehicles.”
The paper, “Big-Data Framework for Electric Vehicle Range Estimation,” will be presented at the 40th Annual Conference of the IEEE Industrial Electronics Society, being held Oct. 29 to Nov. 1 in Dallas.
Additional Resources:
