DAFODIL
PEM Develops New Techniques to Optimize Lithium-Ion Cell Production
The chair "Production Engineering of E-Mobility Components" (PEM) of RWTH Aachen University, together with several partners from industry and research, is involved in the joint project "DAFODIL" with the development of new methods and techniques for the overall optimization of lithium-ion cell production through the use of machine learning and inline analytics. The project is part of the German government's Battery 2020 research funding initiative.
Investigation of different sensor systems
In the DAFODIL project, the global optimization of production is to be worked out via the intermediate step of investigating new sensor systems. To this end, the collaborative partners are investigating various sensor systems for their suitability for data acquisition in cell production. For all measured values to be investigated, DAFODIL will check whether and at which position in the production plant the respective sensor can be used. In addition, it will be investigated whether this information allows direct conclusions to be drawn about the quality of the cells or the intermediate products. At the same time, indirect conclusions about the cell quality will also be tested and complex as well as hidden parameter correlations will be determined from the holistically recorded environmental and process data. For this purpose, algorithms and methods are to be developed that initially identify these complex indirect correlations. In later steps, a sensitivity analysis will be used to quantitatively investigate how the various identified influencing variables affect each other.
Product and method recommendations
As a result of all project phases, product and method recommendations are to be developed in which the use of individual sensors for the optimization of directly evaluable sensor signals is suggested – as well as the use of several sensors, coupled with a suitable evaluation algorithm. A special focus of the final economic feasibility study is on the actual feasibility and usability.
The project
- "DAFODIL": Data-based manufacturing optimization of battery cells based on end-of-line data through massive use of machine-learning algorithms and inline analytics
Research objectives
- Development of methods and tools to reduce the reject rate in the production of lithium-ion battery cells
- Investigation of further innovative sensor technology with added information value for battery production
- Investigation of direct and indirect effect chains and detection of hidden cause-effect patterns
- Evaluation of the costs and benefits of sensor technology and inline analytics in cell production
Research and project partners
SIEMENS AG
(consortium leader)
Institute for Power Electronics and Electrical Drives (iSEA) (RWTH Aachen University)
PEM of RWTH Aachen University
Safion GmbH
Customcells Tübingen GmbH
SURAGUS GmbH
Münster Electrochemical Energy Technology (MEET) (WWU Münster)
Duration
- 10/01/2021 through 09/30/2024
Project sponsor
Funding code
- 03XP0389E
Grantor