Magdalena Pöttinger and Matthias Schmid
Introduction to cutting-edge research currently being conducted at the JKU.
Our new LIT Breakfast Briefings not only provide a brief introduction to cutting-edge research currently being conducted at the JKU, but also take a look at the potential to apply key research developments to address real-world problems. Two topics – 30 minutes. And coffee is included!
October 12, 2021, 8:30 – 9:00 AM, Forum - LIT Open Innovation Center, Stufenforum.
Coffee & croissants provided.
- Magdalena Pöttinger (Institut für Polymer Injection Moulding and Process Automation):
Titel: Simulation-based selection of the machine settings of an injection molding machine with the use case of a ski boot
In order to produce a plastic component on an injection molding machine, many machine settings are necessary to determine. The manual selection of these machine settings involves a lot of material consumption, time, and skilled personnel. To support the machine operator in choosing an optimized machine setting, simulations and material characterization can be carried out before starting the injection molding machine. With the help of this assistance system, production on the injection molding machine can be more resource-efficient and sustainable. In addition, data can be recorded along the value chain and reproducibility can be guaranteed. In this project the ski boot is used as use case.
- Matthias Schmid (Institut für Polymer Injection Moulding and Process Automation):
Titel: A Simulation-Data-Based Machine Learning Model for Predicting Quality Parameters of the Plasticizing Process in Injection Molding
The optimal machine settings in polymer processing are usually the result of time-consuming and expensive trials. Oftentimes, the operator cannot be sure whether the operating point is efficient or not. This research presents a workflow that allows meaningful insights for the plasticizing process in injection molding to be determined with the help of a simulation-driven machine learning model. Given the material, screw geometry and relevant process parameters the model is able to predict quality parameters like melting- and pressure curves along the screw, melt temperature, mean residence time, mass flow rate and power consumption. Additionally, the trained model was implemented into the infrastructure of the LIT-Factory. This enables to obtain real time insights about the process on a specific injection molding machine.