Development of a digital assistance system for continuous quality assurance in the plastic injection moulding process with a focus on self-learning algorithms

The research group at Schmalkalden University of Applied Sciences is developing a pilot system for the highly complex and discontinuous injection moulding of plastics. The focus is on self-learning algorithms for permanent operating point monitoring and indirect quality evaluation of the manufacturing quality. The approach pursued is to limit the analysis to the injection mould. All influencing factors relevant to component quality converge in this tool. This central consideration of the problem reduces the use of necessary sensors, evaluation hardware and software. The main objective of this approach is to generate a verified database of technical parameters based on statistical planning. The developed digital assistant remains permanently on the tool and supports the machine operator in his work by autonomous classification of the production quality and provision of target-oriented process correction measures in a clear and uncomplicated software interface. The aim of the research is to ensure quality components from the first shot onwards. Increasing the production level of injection moulds by reducing set-up and downtimes as well as rejects to improve the earnings situation in the injection moulding companies.