• TUHH (IM)
• Fraunhofer IST, Braunschweig
During the first three-year funding years, greybox models will be developed and qualified to predict tool wear during the turning process of high-alloy stainless steel with TiAlN-coated tools. In addition to the development and commissioning of an automatic wear test rig to generate mass data, the main focus is on methods that combine machine learning and domain-specific knowledge to build models that can extrapolate the tool wear beyond the training limits. The mandatory machining of stainless steels, e.g. for maritime applications, is considered difficult due to the high strength and low thermal conductivity of the material. With the help of the basic research carried out as part of this project, both the production machines (reduction in energy consumption) and the resource tool (tungsten and cobalt) can be better utilized by accurate prediction of tool life.
The multidisciplinary team consists of Dr. Jan Dege and M.Sc Sebastian Schibsdat (Institute of Production Management and Technology (IPMT), TUHH), Dr. Daniel Höche and M.Sc Ya-Jing Wu (Helmholtz-Zentrum Hereon), Dr. Christoph Herrmann and Dr. Martin Keunecke and M.Sc. Sarah Baron (Fraunhofer IST, Braunschweig), as well as Dr. Sebastian Götschel and Dr. Jens-Peter M. Zemke (Institute of Mathematics, TUHH).