Big data and batteries help move towards clean energy

Professor David Howey and colleagues from Oxford, Sheffield and Newcastle Universities are using ‘big data’ to make battery testing and diagnostics faster, cheaper, and more effective – and sharing the approach with academics and commercial enterprises around the world.

A battery is a portable electrochemical device which can be charged with an electric current and used wherever it’s needed. From mobile phones to laptops and electric cars, batteries have become essential to our lives and are vital to store and enable the clean energy we need to tackle climate change.

Ensuring that batteries are safe and reliable at all temperatures and in all conditions, maximising their performance, and understanding how to recycle their components are all important to the economy and public well-being. But testing batteries can be extremely challenging.

“It’s important that firms get it right as this will boost sales and avoid the cost of product failures or accidents,” explains Professor Howey. “But battery testing is time-consuming, expensive, and requires specialist equipment. Until recently, there’s been no standard open-source way to store and share testing data – a major barrier to research and development.”

“Our aim was to develop an open-source database system which could be the ‘go to’ option for anyone working on this issue,” explains Howey. “To do this we needed to sort out some of the fundamentals of working with a huge number of non-comparable datasets and testing systems which had been developed using different methodologies.”

The Oxford team researched the best tools and software to declutter and organise data and worked with the Oxford Robotics Institute and Oxford Research Software Engineering Group to build a bespoke software package, Galvanalyser, which can extract data and put it into a standard format in a database.

The software was reviewed by international battery research partners to maximise usability, and firms and researchers can now install their own copy of the system and access the database through the web or via an Application Programming Interface.

“We’re delighted that so many researchers around the world have engaged with the tool already,” says Howey. “It means that in future, data can be exchanged and analysed more easily, speeding up battery design and development. But we are really just at the start of the process. Our aim is to facilitate a worldwide knowledge exchange about battery-testing, which can help to rapidly advance the field. This work on improving lab data management complements our wider research, for example on understanding battery ageing from field data.”

“The potential of battery development is huge. Long-distance, long-life, standardised batteries for electric cars – or even light but powerful batteries for airplanes – could transform the way we travel. And being able to store wind and solar power on a massive scale would bring down the cost of renewable energy to practically nothing. These things seem impossible now, but sharing big data on battery development could help us achieve them – and move us much closer towards a clean energy future.”

This work was part of the Pitch-In project, funded until 2021 through Research England’s Connecting Capability Fund.