Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Cloud integration of electrode manufacturing via automated tracking and analysis of machine and quality data

Presentation held at International Battery Production Conference, 14 to 16 November 2018, Braunschweig
 
: Schmauder, Martin; Boonen, Laura; Frommknecht, Andreas; Glanz, Carsten; Schulz, Fabian; Yesilyurt, Ozan

:
Präsentation urn:nbn:de:0011-n-5256598 (680 KByte PDF)
MD5 Fingerprint: 664b3027a0f92bbe47ce9e92976e1e69
Erstellt am: 8.1.2019


2018, 15 Folien
International Battery Production Conference <2018, Braunschweig>
Englisch
Vortrag, Elektronische Publikation
Fraunhofer IPA ()
Batterieproduktion; Elektrode; Fertigung; digitale Produktion; Qualitätsprüfung

Abstract
Electrode manufacturing is one of the most critical processes within the energy storage production chain. The process step of electrode coating is influenced by the process parameters during coating itself, but also by the mixing process of the electrode slurry and the quality of the raw materials. Within the research project DigiBattPro the automated and cloud-based collection of machine and quality data of these processes was developed.
Recorded parameters are e.g. the particle size or the specific surface area of the raw materials; the energy input while mixing the slurry; and coating and drying machine parameters like web speed, web tension and drying temperature. A visual camera and a 3D sensor detect resp. measure quality features of the coating, such as its thickness, pinholes in the surface, and the gradient of the edge. For easy later identification, the electrode foil is marked with equidistant unique codes using an ink jet printing head.
With the usage of the Experiment Managing System developed at Fraunhofer IPA all experiments can be fully digitally planned, managed, documented and reviewed. Algorithms have been developed that evaluate the datasets collected in the executed experiments. The recorded parameters are combined using machine learning methods.

: http://publica.fraunhofer.de/dokumente/N-525659.html