DIAGBAT*: Towards Faster, More Accurate and Smarter Battery Health Diagnostics

With the growth of electric mobility and stationary energy storage, accurately assessing a battery’s State of Health (SOH) and actual performance has become a major challenge for industry.

How can we quickly and accurately assess a battery’s condition, anticipate how it will evolve over time, and determine its potential for continued use or reuse? This is the challenge addressed by the DIAGBAT project, launched on 1 November 2025 and endorsed by the i-Trans competitiveness cluster.

Developing New Battery Diagnostic Methods

Current diagnostic methods can require relatively lengthy testing or provide only a partial picture of a battery’s actual condition.

The DIAGBAT project aims to develop advanced methods for assessing battery State of Health more quickly and accurately, while meeting the requirements and constraints of industrial applications.

The objective is to gain a better understanding of a battery’s actual condition in order to:

• optimise its use and performance;
• anticipate potential failures;
• contribute to improved safety;
• support second-life and battery reuse strategies.

Combining Physics, Artificial Intelligence and Experimental Data

To achieve these objectives, DIAGBAT relies on a multidisciplinary approach combining physics-based models, artificial intelligence and experimental data.

By bringing these approaches together, the project aims to develop diagnostic methods that are fast, reliable and robust, making effective use of available data to accurately characterise a battery’s State of Health.

The project brings together CRITT M2A, a testing and R&D centre specialising in batteries and electric powertrain systems, and LGI2A, a research laboratory with expertise in artificial intelligence.

This collaboration combines expertise in battery testing and characterisation, data analysis, and the development of new diagnostic methods.

The Project Enters Its Experimental Phase

Following an initial phase dedicated to reviewing existing diagnostic methods and the current state of the art, the teams have begun the first experimental tests and data collection activities.

This work will provide the data needed to develop and assess the various methods explored as part of the project.

Through this R&D project, CRITT M2A is continuing to expand its expertise in battery characterisation, ageing and diagnostics, with the aim of developing new solutions tailored to future industrial needs.

*The DIAGBAT project is co-funded by the European Union.