Automation and Digitalization of the Start-up Processes for Injection Molding Tools

Determining the optimal set of parameters experimentally is resource-intensive and can be carried out using various strategies. In addition to the chosen strategy, the set of parameters depends on the mold to be tooled, the part geometry, the machine, and the material used. Flow simulations, which model the filling behavior in the injection molding process, usually play only a minor role in trial runs due to their deviations from the actual process. Using machine learning methods, knowledge already generated from simulations—combined with process knowledge from previous trial runs—can be leveraged to achieve successful data-driven operating point determination.

Project
Objective The objective of the research project is to determine an optimized set of setup parameters based on simulation and real experimental data by applying machine learning methods. The data-driven prediction of an optimal set of operating parameters is expected to not only improve process understanding but also significantly reduce the time and cost associated with the validation process.

Funded by: Phoenix Contact Foundation

Your contact: Julia Volke, M.Sc.
Email: volke(at)uni-kassel.de
Phone: +49 561 804 - 2867