Researchers at Uppsala University, Sweden, have developed a pioneering AI model that could dramatically extend the lifespan and enhance the safety of electric vehicle (EV) batteries, addressing a critical barrier in the electrification of the transport sector.
Battery degradation is a significant challenge for the EV industry, with batteries often becoming the first component to age and fail. This rapid deterioration not only results in resource waste but also hinders the transition to a more sustainable transportation system.
In a bid to mitigate this issue, researchers have focused on developing advanced software, frequently leveraging artificial intelligence to optimize battery management and control.
The new AI model from Uppsala University stands out by significantly increasing the accuracy of battery aging predictions.
Daniel Brandell, a professor of materials chemistry in the Department of Chemistry and director of the Ångström Advanced Battery Centre at Uppsala University, led the study.
"Being able to learn more about the life and ageing of batteries will benefit future control systems in electric vehicles. It also shows how important it is to understand what happens inside the batteries. If we stop looking at them as black boxes that are simply expected to provide power, and instead acquire a detailed picture of the processes, we can manage them so that they stay in good condition longer," Brandell said in a news release.
The research involved extensive battery testing over several years, in collaboration with Aalborg University in Denmark.
The researchers built an extensive database by collecting data from numerous very short charging segments.
This large dataset was then coupled with a detailed model that mapped the various chemical reactions occurring within the battery, providing an exceptionally accurate picture of both power generation and aging processes.
"Altogether, this gives us a very precise picture of the various chemical reactions that result in the battery generating power, but also of how it ages during use," added Wendi Guo, a postdoctoral fellow in Brandell's group, who conducted the study.
An additional benefit of this AI model is its potential to reduce the dependency on sensitive and extensive vehicle data.
"The fact that we only use short charging segments is probably an added advantage. Battery data from electric vehicles is sensitive, both for the industry and from an anonymization point of view for users. This research shows how far you can get without needing complete datasets," Brandell added.
The implications of this research, published in the journal Energy & Environmental Science, could extend far beyond longevity. By better understanding battery chemistry and behavior during charging cycles, the AI model promises to enhance safety by predicting and mitigating potential design flaws and side reactions that often lead to safety concerns.