AI-based quality control for sustainable battery production
The Fraunhofer Institute for Solar Energy Systems (ISE) is conducting research into methods for conserving critical resources as part of a project.
Dr Moritz Kroll, team leader of Battery Data Science at Fraunhofer ISE, will establish the independent junior research group NADINE (Battery Data Science) at the University of Freiburg as part of the BattFutur – Junior Research Groups in Battery Research programme. The interdisciplinary project will start in February 2026 at INATECH in the Faculty of Engineering and will run for five years. The focus is on data-based methods for non-destructive quality control of battery cells that reduce the consumption of critical resources in manufacturing.
Battery cell production faces considerable challenges: new production facilities and even well-established factories generate significant amounts of waste. Current electrical testing methods can record basic functional parameters but provide only limited information about safety-critical defects inside the cells.
This is where the independent junior research group NADINE comes in: “We will further develop non-destructive methods for analysing battery cells. Innovative quality assurance methods are a basic prerequisite for competitive European battery cell production and for the development of safe, long-lasting batteries,” explains group leader Dr Moritz Kroll, who has secured funding of around 2.3 million euros for the research group from the Federal Ministry of Education and Research (BMBF).
More precise quality control would have a significant ecological and economic impact: according to calculations by Fraunhofer ISE, reducing the reject rate by just one percentage point in a 10 GWh production plant would cut annual CO₂ emissions by around 5,000 tonnes and reduce the consumption of critical raw materials such as lithium and cobalt. The decisive factor here is not only the detection of defects itself, but also their spatial localisation and classification within the cell. This location information makes it possible to draw conclusions about individual production steps and parameters and to adapt production in an agile and timely manner.
Interdisciplinary approach at the interface of battery research and data science
The research project combines expertise from battery research with methods from data science and artificial intelligence and collaborates with researchers from TU Dresden and the University of Auckland, New Zealand. The focus is on promising approaches such as ultrasound tomography and the characterisation of the expansion behaviour of battery cells. Through the consistent digitalisation of research and development processes, data can be linked across multiple stages of the battery value chain. Machine learning methods will help to identify new correlations between manufacturing parameters, operational management, health status and safety. In line with the FAIR principle for data management (findable, accessible, interoperable, reusable), data sets are made available in an open science approach.
Another focus is on the evaluation of batteries for secondary use: after being used in electric vehicles, batteries can be reused in stationary energy storage systems – provided that their condition can be reliably and efficiently assessed. The development of appropriate methods can make an important contribution to the circular economy and to reducing the carbon footprint of energy storage systems. “The project thus ideally combines basic research and concrete applications for industry,” explains Prof Andreas Bett, Director of Fraunhofer ISE.
The independent five-member junior research group will be based at the Fraunhofer ISE's Centre for Electrical Energy Storage in Freiburg and will be merged with the existing group. The ISE will grant access to its research infrastructure, and the group will benefit from the opportunities for cell production and characterisation in a safe test environment. An industry advisory board will accompany the project and ensure continuous application relevance and timely exploitation of the results in German industry. “We are very pleased to further strengthen the successful cooperation between the University of Freiburg and Fraunhofer ISE with the NADINE junior research group,” says Prof Dr-Ing Frank Balle, Dean of the Faculty of Engineering. “The combination of battery research and data science at INATECH is an important building block for research into future-proof and sustainable energy storage. Dr Kroll brings with him outstanding experience in developing application-oriented solutions at this interface between basic research and industrial application.”
About the funding programme
Funding is provided as part of BattFutur – Junior Research Groups in Battery Research, which is part of the BMFTR's Battery Research umbrella concept. With this programme, the ministry supports young scientists in establishing independent research groups that work at the interface between academic basic research and application-oriented research. The aim is to strengthen innovation in the field of battery research and contribute to the establishment of a sustainable and competitive battery value chain in Germany.
Caption: The junior research group uses ultrasound microscopy to determine the health status of battery cells (image: Fraunhofer ISE).
Source: Fraunhofer ISE, quotes are translated from German by the editorial team