Particle.news
Download on the App Store

AI Charging Strategy Promises 23% Longer EV Battery Life Without Slower Fast Charges

The peer-reviewed, simulation-based method could ship as a software update pending calibration and real-world validation.

Overview

  • Researchers at Chalmers University and Victoria University built a reinforcement-learning charger that adapts current to a battery’s chemistry and health.
  • Tests in simulation showed nearly a 23% increase in battery life with fast-charge times kept within a few seconds of today’s rates.
  • The approach targets lithium plating, a buildup of metallic lithium during hard charging that cuts capacity and raises safety risks as batteries age.
  • Deployment could come through battery management software updates after transfer-learning calibration for different battery types and checks on physical cells.
  • Longer-lasting packs could lower warranty costs, improve resale values, and reduce raw material use for automakers, according to the team.