Integrating next generation technologies to boost the performance of EVs, University of Cambridge researchers have developed a machine learning (ML) algorithm that could help reduce charging times while also prolonging battery life in electric vehicles.
According to the Cambridge researchers in England, they have developed a new machine learning algorithm that predicts how different driving patterns affect battery performance, improve safety, and reliability.
The researchers believe that their algorithm could help drivers, manufacturers, and businesses get the most out of their electric vehicles by suggesting routes and driving patterns that can help to minimise battery degradation and charging times.
In addition to the machine learning algorithm, the Cambridge researchers developed a non-invasive way to probe batteries to get a holistic view of the battery health, data which is then fed into the algorithm to help predict how different driving patterns will affect the future health of the battery.
The Cambridge researchers believe that, if their algorithm is developed commercially, it could provide, for example, recommended routes for drivers that get from point-to-point in the shortest time without degrading the battery, or recommend the fastest way to charge the battery without causing it to degrade.
“Battery health, like human health, is a multi-dimensional thing, and it can degrade in lots of different ways,” said Penelope Jones, from Cambridge’s Cavendish Laboratory and first author of the report.
“Most methods of monitoring battery health assume that a battery is always used in the same way. But that’s not how we use batteries in real life. If I’m streaming a TV show on my phone, it’s going to run down the battery a whole lot faster than if I’m using it for messaging. It’s the same with electric cars – how you drive will affect how the battery degrades.”
“Most of us will replace our phones well before the battery degrades to the point that it’s unusable, but for cars, the batteries need to last for five, ten years or more,” said Alpha Lee, who led the research.
“Battery capacity can change drastically over that time, so we wanted to come up with a better way of checking battery health.”
The non-invasive probe sends high-dimensional electrical pulses into a battery and measures the response, resulting in a series of ‘biomarkers’ of battery health that is nevertheless gentle on the battery and doesn’t cause any unnecessary degradation.
The returning electrical signals from the battery are then converted into a representation of the state of the battery’s health. These signals allowed the machine learning algorithm to predict how the battery would respond in the next charge-discharge cycle depending on how quickly the battery was charged and how fast the car would be going the next time it was on the road.
Tests with 88 commercial batteries, focusing on lithium cobalt oxide (LCO) cells, showed that the algorithm did not need any information about previous usage of the battery to create an accurate prediction.
“This method could unlock value in so many parts of the supply chain, whether you’re a manufacturer, an end user, or a recycler, because it allows us to capture the health of the battery beyond a single number, and because it’s predictive,” added Lee.
“It could reduce the time it takes to develop new types of batteries, because we’ll be able to predict how they will degrade under different operating conditions.”
In addition to the potential benefit to EV drivers and manufacturers, the Cambridge researchers believe that their algorithm could be helpful for businesses operating large fleets of EVs.
“The framework we’ve developed could help companies optimise how they use their vehicles to improve the overall battery life of the fleet,” said Lee.
