A new AI-based charging method can adapt fast charging to a lithium-ion battery, extending its battery life without increasing charging times.

A key barrier to electric vehicle (EV) adoption is access to fast charging to enable commuting and driving over longer distances. However, fast charging can be stressful for batteries, limiting their lifespan. 

Researchers at Chalmers University of Technology in Sweden have developed an AI-based charging strategy for EVs that adapts the current during each fast charge to the battery’s chemistry and ‘state of health’. As a result, battery life is extended by around 23% compared to standard methods, while charging time remains largely unaffected.

“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,” said Meng Yuan, assistant professor at Victoria University of Wellington, New Zealand, and a former researcher at Chalmers.

One of the most problematic issues with fast charging lithium-ion batteries is a chemical reaction known as lithium plating. Here lithium ions accumulate on the surface of the anode, instead of being properly absorbed into it, forming metallic lithium deposits. Over time, this significantly degrades battery capacity and can even lead to short circuits.

“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,” said Yuan.

The AI-based charging strategy is based on what is known as reinforcement learning, where the algorithm learns by interacting with its environment, gradually improving as the right actions are reinforced. 

To train their AI model, the researchers used a conventional EV battery type widely available on the market and a simulation of the parameters that have an impact on both charging time and overall battery health.

“Our study shows that smart adaptation of the current during charging, taking into account the changing electrochemical state of the battery, can maximise both the performance and the life of the battery,” said Changfu Zou, professor at the Department of Electrical Engineering at Chalmers.

According to the researchers, their new charging strategy is both easy and cost-effective to implement through software updates in the vehicle’s battery management systems. However, they say that some adaptation will be needed for the method to be used generally.

“There are not so many different battery types today, but the method needs to be calibrated for it to be used by everyone. Using transfer learning, we can take advantage of what our AI model has already learned, and thus adapt the AI model to new batteries more quickly,” said Zou.

The next steps is to test the method directly on physical batteries. The hope is that it could bring about reduced costs and prevent a drain on natural resources. 

“An almost 23% increase in battery life can mean lower warranty costs, better resale value and more efficient use of critical raw materials,” said Zou.

Their study – Lifelong reinforcement learning for health-aware fast charging of lithium-ion batteries – has been published in the journal IEEE Transactions on Transportation Electrification.