With several partners from research and industry, the Chair of PEM of RWTH Aachen University has started the three-year TRAICELL project, aiming to increase the efficiency of battery cell production through the use of artificial intelligence. The three-year project is funded by the German Federal Ministry of Education and Research. Its goals is to make data from various production steps completely traceable and, on this basis, to use machine learning for quality predictions.
Three near-series prototypes
Within the framework of the project, three prototypes are to be developed in different scaling stages from pilot production and research production to near-series implementation. PEMʼs partners include Fraunhofer Research Institution for Battery Cell Production FFB, as cell manufacturer UniverCell, quality assurance specialist BST, and AI developer Merantix Momentum.
Early AI detection of rejects
A key development goal is a system that records production and quality data down to the level of individual electrode layers. At the same time, the process steps of mixing and coating are to be optimized. At the end of the research project, predictive models are expected to be able to assess cell quality at an early stage of production to reduce waste and improve material yield.Industrial implementation of the technology is planned for the end of 2027. The consortium aims to transfer the results to existing infrastructures to enable rapid application in mass production.
