Methodik zur Arbeitsvorbereitung mittels Maschinensteuerungs- und Sensordaten für extrusionsbasierte additive Verfahren

The plastics processing industry is undergoing a paradigm shift from the mass production of short-lived, non-degradable products to a sustainable production method aimed at the economical and ecological manufacturing as well as the longest possible recycling of plastics. Additive manufacturing processes, especially extrusion-based technologies, are a promising technology for enabling more sustainable plastics production. Thanks to tool-free production, they offer the potential to make the manufacturing of consumer goods and technical products in small to medium quantities more economical. Various forward-looking industrial application scenarios can be derived for large-format, extrusion-based additive manufacturing. These scenarios position large-format, extrusion-based additive manufacturing as an equal plastics processing method with future prospects for OEMs and suppliers. However, extrusion-based additive manufacturing processes face the challenge of achieving the necessary level of technological maturity for these industrial application scenarios. The technologies are complex, as a large number of process parameters, their combinations and interactions play a significant role in the successful manufacturing process. Due to the comparatively young technologies, these are less researched than in long-established conventional processes. The work preparation, with its steps of calibrating the plant technology and the empirical process development prior to the manufacturing of multiple components or large, complex components, requires deep expert knowledge. The goal of producing structurally flawless components with all the required properties can only be achieved with the help of elaborate work preparation, which often requires lengthy commissioning or the use of service providers. In order to keep pace with rapid technological development, automated work preparation is required that can process and analyse large amounts of data from process and production parameters. The aim of this dissertation is therefore to enable system manufacturers and users of extrusion-based additive manufacturing processes to quickly set optimal process parameters with the help of machine learning methods with a systemic approach. To this end, a methodology is being developed that comprises the four methodological analysis of internal process data, integration of sensor technology, process analysis and process optimisation starting from and feeding back into the work preparation of extrusion-based additive manufacturing processes. To ensure that the methodology for work preparation can be adequately used in practice, the steps build on each other, but also offer added value and knowledge gain for the work preparation of extrusion-based additive plant technology even if the methodology is not completed. Validation of the methodology using two practical examples enables industrial practice to use the methodology to optimise the process of work preparation for extrusion-based additive manufacturing processes.