Smart AI extends electric vehicle battery life by 23% without increasing charging time

Fast charging shortens the lifespan of the vehicle battery, but is necessary when traveling long distances in an electric vehicle. Researchers at Chalmers University of Technology in Sweden have developed a new AI method to adapt fast charging to battery conditions. Their research shows that battery life can be extended by almost 23% without increasing charging time. All you need to do is update your vehicle’s software.

The possibility of fast charging is an important factor when individuals and companies consider purchasing electric vehicles.

“Access to fast charging is very important for taxis and heavy industrial vehicles, for example, but this also applies to passenger cars. Private car drivers typically charge their electric vehicles at home, but the ability to fast charge outside the home is a very important factor to make commuting and long-distance driving easier,” said Zhang-Fu Zou, professor in the Department of Electrical Engineering at Chalmers University.

Electric vehicle batteries currently have a lifespan of approximately 8 to 15 years*, depending on usage and charging conditions. Several studies* of the European EV market show that consumers considering purchasing an EV are concerned about limited battery life.

The requirement for efficient fast charging also conflicts with battery health. This is because such charging puts stress on the battery and shortens its lifespan.

Changfu Zou worked on this challenge with Meng Yuan, an assistant professor at Victoria University of Wellington in New Zealand and a former Chalmers researcher. A recently published study shows that with the help of artificial intelligence, it is possible to extend battery life without significantly increasing charging speed.

Adjust charging to match battery condition

In this study, researchers present an AI-based charging strategy that adapts the current during each fast charge to the battery’s chemistry and “health status.” Adapted charging extends battery life by approximately 23% compared to current standard methods. At the same time, charging time is not affected even if it is only a few seconds.

“We show that we can charge at about the same speed as today, but significantly reduce long-term battery degradation,” says Meng Yuan.

Rapidly charging a battery forces large currents through the various cells, increasing the risk of chemical side reactions, among other things. One of the most problematic is known as lithium plating, where metallic lithium is not properly stored within the cell structure and deposits on the electrodes. This can reduce capacity and can also impact safety, as the heterogeneity of the lithium structure can, in the worst case, lead to short circuits.

“As batteries get older, the risk of lithium plating increases. However, today’s standard charging methods use the same current and voltage regardless of whether the battery is new or has been used for years,” says Meng Yuan.

Short charging time and low consumption

The new AI-based billing strategy is based on reinforcement learning**, where good behavior is rewarded and reinforced. The training environment consisted of a model of the most common electric vehicle battery on the market and a simulation of the parameters that affect both charging time and battery health.

The AI ​​model was trained to adjust charging depending on how much the battery is charged or discharged during charging. We also had to consider the overall health of the battery, as it is important to both capacity and electrochemistry. The result is a charging strategy that keeps charging times short and minimizes harmful reactions.

“Our research shows that by intelligently adapting the current during charging, taking into account changes in the battery’s electrochemical state, both battery performance and lifespan can be maximized,” said Changfu Zou.

Easy to implement – ​​but requires adaptation

Researchers say the new charging strategy is easy and cost-effective to implement, and could in principle be implemented through a software update to a vehicle’s battery management system. However, this method requires some adjustment for general use.

“Right now, there aren’t that many types of batteries, but we need to adjust the way we use them so that everyone can use them. transfer learningWe can leverage what the AI ​​model has already learned, allowing us to quickly adapt the AI ​​model to new batteries,” says Changfu Zou.

The next step is to test the method directly on a physical battery. Researchers hope that AI-based charging strategies will make a significant contribution to the electrification of the transportation sector.

“To reduce emissions and transition to a fossil-free society, it is important that people are ready to switch to electric vehicles. Fast charging possibilities and extended battery life will be key drivers,” says Meng Yuan.

“For the automotive industry, an approximately 23% increase in battery life means reduced warranty costs, increased resale value, and more efficient use of critical raw materials,” said Changfu Zou.

*source:

European Commission/EAFO, 26 March 2026: Consumer Monitor 2025: EU drivers’ views on electric vehicles
PwC, September 10, 2025: eReadiness 2025 – EVs stay ahead in a growing market
McKinsey, August 5, 2024: How European consumers perceive electric vehicles

**Reinforcement learning is a method of machine learning in which an algorithm learns by interacting with its environment and gradually improves its decisions based on the feedback it receives.

Learn more about the research

The study, “Lifetime reinforcement learning for health-conscious rapid charging of lithium-ion batteries,” was published in IEEE Transactions on Transportation Electrification. The authors are Changfu Zou from Chalmers University of Technology, Sweden, and Meng Yuan from Victoria University of Wellington, New Zealand.

This research was supported by the European Union’s Horizon European Research and Innovation Program through the Marie Sklodowska Curie Action Postdoctoral Fellowship, the Swedish Research Council, and the Swedish Foundation for International Cooperation in Research and Higher Education.

Fast charging and battery life details

Current electric vehicle batteries have a lifespan of approximately 8 to 15 years, depending on usage and charging conditions (1). Battery capacity gradually decreases over time. For example, Volvo Cars electric cars come with an eight-year or 160,000-kilometre battery warranty (2).

In this study, researchers measured battery life by equivalent full cycles (EFC), or how many full charge/discharge cycles a battery can endure before its capacity drops to 80 percent of its original value. The battery will still operate under this limit, but it will degrade significantly, have a shorter range, and have less power (3).

An analysis of 22,000 electric vehicles in the United States, Canada, and Europe shows that fast charging typically accounts for up to approximately 10 to 12 percent of total charges (4). Fast charging is often used by long-distance commuters or people who cannot charge at home. In addition, in regions such as southern Europe and China where there are few opportunities to charge at home, the usage rate of public charging, including fast charging, is high (5).

1. How long do electric car batteries last? Vehicle battery life, charging cycles, and cost | Goe
2. https://www.volvocars.com/se/cars/electrification/battery/ (Swedish)
3.https://www.twaice.com/battery-encyclopedia/End of life
4. EV Battery Health Study: New Data on Fast Charging and Degradation | Geotab
5. Global trends in electric vehicle charging demand and infrastructure development