Image: ChalmersWhile recent studies have shown that battery wear within the first 100,000 kilometres is often lower than previously expected, the State of Health (SoH) of traction batteries inevitably declines as electric vehicles age and accumulate mileage. This results in a gradual reduction in usable battery capacity compared to the manufacturer’s specifications.As declining battery capacity is a major concern for potential buyers of electric cars and commercial electric vehicles, researchers at Chalmers University of Technology in Gothenburg, Sweden, have developed a method to improve battery health. Their focus was on DC fast charging, which is known to accelerate battery degradation more quickly than slower AC chargers.“For taxis or heavy vehicles in industry, for example, access to fast charging means a lot, but this is also true for passenger cars, ” says Changfu Zou, Professor at the Department of Electrical Engineering at Chalmers University. “Although private motorists usually charge their electric cars at home, the availability of fast charging outside the home is a crucial factor, as it facilitates commuting and driving over longer distances.”Extending lifespan with AIIn a recently published study, Zou, alongside Meng Yuan, Assistant Professor at Victoria University of Wellington in New Zealand and former researcher at Chalmers University, demonstrated that artificial intelligence can extend battery lifespan without significantly increasing charging time.This AI-powered charging method extends battery lifespan by approximately 23 per cent compared to current standard procedures, while charging time remains almost unchanged. “We show that it is possible to charge more or less as fast as today, but with significantly less long-term degradation of the battery,” says Meng Yuan.Fast charging is a stress factor for batteries, as it forces high-current electricity through individual cells, increasing the risk of chemical side reactions. A particularly problematic issue is lithium plating. Here, metallic lithium deposits on the electrode instead of being properly stored within the battery structure. This can reduce capacity and compromise safety, as uneven lithium distribution may, in the worst case, cause a short circuit.Preventing lithium plating“The risk of lithium plating increases with the age of the battery. However, the standard methods of charging today use the same current and voltage regardless of whether the battery is new or has been used for years,” says Meng Yuan.The new AI-based charging strategy relies on reinforcement learning, where correct actions are rewarded and reinforced. The AI model was trained to adapt the charging process to the battery’s state of charge (SoC) while taking its overall condition into account, as this is critical for both capacity and electrochemical performance. The result is a charging strategy that reduces charging time and minimises harmful reactions.According to the researchers, the new charging strategy is both simple and cost-effective to implement, as it could potentially be deployed through software updates in vehicle battery management systems. However, some adjustments would be required for widespread adoption.cision.com