• Microsoft releases Surface battery data to standardize fragmented testing environments
• Battery dataset reveals inconsistencies across lithium-ion testing methods and tools
• Open format aims to reduce repeated engineering work across battery research teams

Microsoft has contributed a standardized battery dataset through its Surface Battery Development team to the Linux Foundation initiative known as LF Energy’s Battery Data Alliance.

The release coincides with the introduction of the Battery Data Format, an open specification designed to improve consistency and interoperability in battery data workflows.

The dataset focuses on cell architecture design variations, enabling direct comparison across multiple lithium-ion configurations, including end tab, middle tab, and multi tab designs.

Article continues below

Raw test data made openly accessible

The dataset has been made publicly accessible through a repository, where it appears primarily as time-series measurements of current and voltage collected during controlled test cycles.

The format defines a structured approach for experimental, simulation, and metadata-rich datasets.

It can be shared and reused across laboratories, software tools, and engineering environments without extensive modification.

The Linux Foundation notes the contribution is more than a routine data release, noting that “it reflects more than a standalone dataset release.”

The organization adds that it demonstrates how an emerging standard can be applied in real testing scenarios rather than remaining conceptual.

Battery data has remained fragmented across institutions, vendors, and platforms, often requiring custom handling before analysis can begin.

The Battery Data Format introduces a unified schema supported by ontology-driven definitions derived from initiatives such as BattINFO, enabling machine-readable metadata and compatibility with broader linked data practices.

This structure allows datasets generated under different conditions or by different cycler systems to be combined and analyzed in a consistent manner.

It also supports compatibility between analytical models developed independently, reducing the need for repeated data preparation across research groups.

The dataset contributed by Microsoft focuses on lithium-ion cell architecture variations, including end tab, middle tab, and multi-tab configurations.

It includes initial performance benchmarks and cycle aging measurements, allowing engineers to examine how design differences influence degradation patterns over time.

These comparisons are often difficult when datasets originate from incompatible systems or follow inconsistent naming conventions.

Supporting tools within the Battery Data Format ecosystem include Python libraries for validation and conversion utilities that transform vendor-specific formats into standardized datasets.

The Battery Data Alliance includes a range of research institutions and companies, with participation from groups such as SINTEF, the Faraday Institution, and several university laboratories.

The broader development of the format has also incorporated contributions from projects such as PyProBE and modeling frameworks like PyBaMM, linking experimental data with simulation workflows.

Though the biggest names in the industry are missing, the Linux Foundation argues that the shared datasets are necessary for advanced computational analysis.

“Having universal standards for data management for each segment of the battery community is required for data creation to unlock the power of AI algorithms designed to identify everything from new candidate electrode materials to improved battery pack construction to cell lifetimes,” said Gabe Hege, Chair of the LF Energy Battery Data Alliance.

The dataset released by Microsoft is the inaugural entry in a vendor-neutral datastore, though participation from other large manufacturers remains uncertain.

“This is a call to action,” said Argonne battery scientist Noah Paulson. “We’re trying to energize and organize the battery community to contribute their data…to enable powerful data science methods to catalyze breakthroughs.”

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds. Make sure to click the Follow button!

And of course you can also follow TechRadar on TikTok for news, reviews, unboxings in video form, and get regular updates from us on WhatsApp too.