LF Energy Battery Data Alliance Announces the Battery Data Format (BDF): A New Open Standard for Battery Data Interoperability
Summary
- Announcement: LF Energy’s Battery Data Alliance released the Battery Data Format (BDF), an open, community-driven standard for battery data interoperability.
- Purpose: BDF provides a unified, machine-readable schema and ontology-aligned metadata to make battery data shareable, reproducible, and model-ready across labs, vendors, and software.
- Ecosystem & Validation: Aligned with BattINFO; validated via Faraday Institution tooling (PyProBE/BDX); compatible with PyBaMM and BattMo; supported by open source libraries and converters.
- Adoption & Impact: Backed by major contributions (Microsoft dataset, Ohm converter, largest open BDF dataset from European research labs) and governed openly under LF Energy.
Press Release
SAN FRANCISCO – 22 DECEMBER 2025 – LF Energy, the open source foundation accelerating the energy transition, today announced that its Battery Data Alliance (BDA) project has released the Battery Data Format (BDF), an open, community-driven standard designed to bring order, interoperability, and transparency to the rapidly growing world of battery data. BDF defines a consistent structure for experimental, simulation, and metadata-rich battery datasets making it easier for researchers, industry, and software developers to collaborate and build next-generation tools.
Battery data today is fragmented across institutions, vendors, and platforms. BDF provides a unified schema and ontology-driven approach that enables datasets to be shared, analyzed, and reproduced reliably across the battery ecosystem.
Built on Collaborative Global Research
The development of BDF builds on extensive contributions from leading institutions across academia, industry, and government:
- BattINFO Ontology
BDF is aligned with the terminology and structure defined by the BattINFO Ontology from the Battery2030+ and BIG-MAP projects, ensuring consistent definitions, machine-readable metadata, and compatibility with broader FAIR linked-data practices. - Faraday Institution’s PyProBE and BDX
Python Processing for Battery Experiments (PyProBE https://github.com/ImperialCollegeLondon/PyProBE) is an open source Python package designed to simplify and accelerate the process of analysing data from battery cyclers. It was created at Imperial College London within the Faraday Institution’s ( https://www.faraday.ac.uk/) Multi-scale Modelling project and has been instrumental in validating BDF’s column naming conventions and metadata definitions. Faraday Institution is funding the modification of PyProBE to adopt BDF-aligned naming, enabling interoperability between BDF and BDX (Battery Data eXchange), the optimised, binary format used by PyProBE to provide file size and processing performance benefits. - Microsoft Open Battery Dataset Contribution
Microsoft plans to release a new battery dataset in the BDF format, providing the community with a high-quality reference dataset for benchmarking, tooling development, and educational use. - Ohm BDF Converter Contribution
Ohm (YC W23) to contribute the Battery Data Format (BDF) converter to the community. This web-based tool enables users to upload raw cycler data files and download a BDF-compliant .csv, streamlining adoption of the BDF standard across laboratories and workflows. Ohm’s converter supports major commercial cycler data formats, is freely available, and is designed to accelerate interoperability and reduce friction in transitioning to BDF-aligned data practices. Ohm is a pioneering leader in PhD-level, industrial AI agents purpose-built for battery science. - Largest Open Source Battery Data Contribution in the BDF (August 2025)
Collaboration between Empa, ETH Zurich, EPFL, SINTEF produced dataset from 199 coin cell batteries, featuring both NMC//graphite and LFP//graphite chemistries, each tested for 1,000 cycles under fully automated, precisely controlled workflows in BDF.
These collaborations reinforce BDF as a standard shaped by real-world data workflows, from experimental cycling and materials characterization to physics-based models such as PyBaMM and BattMo.
Designed for Practical Use Across the Battery Lifecycle
The BDF provides a standard structure for data generated in battery labs. It is expected that adoption of the BDF will empower the battery science community to leverage advances in open source battery models.
Developed with input from leading scientists and engineers, the BDF addresses two main challenges:
- Data Consistency
With a common format, labs and cycler brands can eliminate the inconsistencies in data structure that arise with each software update. - Model Compatibility
A unified format means battery model developers can easily adapt their models to accept BDF data, making it possible for scientists to experiment with multiple models without custom coding each time.
The format is defined as open, extensible, and implementation-agnostic, enabling future growth as new chemistries, devices, and data types emerge.
Early Software Support Includes
- BDF Python Library, for reading and writing datasets and validating metadata
- Conversion tools for transforming vendor-specific formats (e.g., Arbin, MACCOR) into BDF
- Reference visualization tools (web and notebook-based) for quick exploration of BDF datasets
- Conversion to and from BDX to access all PyProBE tools
- Compatibility with leading modeling frameworks (PyBaMM, BattMo) and analysis platforms
Initial Scope of the BDF
- The initial scope is intended to facilitate use and comparison of cycler time-series data.
- The BDF provides a fixed table schema for time-series battery data, which is supplemented with a machine-readable application ontology for integration with the Semantic Web.
- The BDF application ontology is defined as an extension of the BattINFO domain ontology, which provides interoperability within the broader field of battery data.
- An immediate next step will be launching a parallel format for storing metadata for the BDF.
- Future development will focus on formats for other types of lab data such as impedance data.
A Shared Foundation for the Battery Data Community
The Battery Data Alliance aims to make BDF a globally adopted standard and invites companies, universities, labs, and software developers to participate.
“BDF provides a common language for battery data, designed in the open and strengthened by contributions from across the global community. We encourage organizations of all sizes to participate and help advance a unified standard for the battery industry.”
– Gabe Hege, Chairperson for the Battery Data Alliance
A public specification, documentation, examples, and reference datasets are available at:
https://batterydataalliance.energy
Organizations interested in participating in the BDF working group or contributing datasets are invited to contact:
info@batterydataalliance.energy
About LF Energy
A first-of-its-kind initiative, LF Energy is the community for technologists to co-develop open, industrial-grade technology, standards, and data to deliver affordable, reliable, safe and clean energy. Strategic Members include Alliander, Google, Hydro-Quebec, Microsoft, RTE and Shell, in addition to over 60 General and Associate Members from across the energy industry, technology, academia, and government. Find further information here: https://www.lfenergy.org.
About The Linux Foundation
The Linux Foundation is the world’s leading home for collaboration on open source software, hardware, standards, and data. Linux Foundation projects are critical to the world’s infrastructure including Linux, Kubernetes, Node.js, ONAP, OpenChain, OpenSSF, OpenStack, PyTorch, RISC-V, SPDX, Zephyr, and more. The Linux Foundation is focused on leveraging best practices and addressing the needs of contributors, users, and solution providers to create sustainable models for open collaboration. For more information, please visit linuxfoundation.org. For a list of trademarks of the Linux Foundation, please see its trademark usage page: www.linuxfoundation.org/trademark-usage. Linux is a registered trademark of Linus Torvalds.
