AI holds key to limiting risk of EV battery fires, says UK firm

LONDON: UK-based company Eatron Technologies says it is using Artificial Intelligence (AI) in its battery management software to help vehicle manufacturers reduce the risk of electric vehicle (EV) battery fires.

The recent focus on EV fires has raised concerns among consumers, and the automotive industry is now working to regain their trust. Eatron Technologies aims to eliminate catastrophic battery failures by using intelligent and connected battery management software that incorporates AI.

Battery failure in EVs can be caused by various factors, with one common cause being lithium plating.

This occurs when metallic lithium deposits form around the anode, usually during fast charging at low temperatures. Over time, these deposits can erode battery performance and lead to the growth of dendrites, which can pierce through the separator between the anode and cathode, causing a short circuit and potentially initiating thermal runaway.

Detecting lithium plating without opening the battery cell is a challenge, but Eatron Technologies said it has demonstrated that AI can effectively detect and predict when it might occur.

By using feature extraction to transform raw health data from the battery into a format that makes anomalies easier to identify, combined with their proprietary AI pipeline, Eatron claimed its AI diagnostics can predict cell failures with up to 90% accuracy and zero false positives.

Detecting failures before they happen allows for more effective and convenient management. This could involve altering the way the battery is managed to minimise further damage in the short term and scheduling a service visit for rectification at the driver's convenience.

Eatron Technologies CEO Dr Umut Genc said by using AI in its battery management software, Eatron Technologies aims to stop catastrophic battery failures.