Model-based Numerical Control (MNC) for the automation of casting processes

点击翻译按钮获取中文摘要

Casting is an elementary molding process for producing metallic products with gravity casting being the most common form of casting. Presently, a full automation of casting processes only takes place for large quantities using special control systems. In (partially) automated systems, the process sequence is programmed by means of teach-in processes. This programming requires, however, expertise by the worker and a direct interaction with the casting process in the danger zone, which results from handling liquid metal. This type of programming cannot be parameterized, therefore, scalability is not given and the motion sequences must be reprogrammed with every change. However, mathematical and physical models for the casting process, which are suitable for calculating the motions in compliance with the relevant process targets such as flow rate and target point of the substance to be cast, are already explored. But this special control system lack the necessary flexibility, so that the casting process can be simple and machine-independent programmed. NC systems have a high flexibility and the potential of easily parameterizing the process. These benefits are to be transferred to the casting process by combining an NC with process models. The objective of the project is the development of an NC architecture that, based on process models, turns gravity casting for any lot sizes into a fully automated process.