Speciale_EMO_Hannover

14 SPECIAL EMO 2023 vices and workpieces - explains Markus Michelberger, Head of Sales Clamping Technology at the company’s site in Men- gen -. At EMO Hannover, we will be show- casing further automation and digitaliza- tion options in machining, as well as new developments in the field of sustainable, resource-saving clamping equipment”. AI meets Digital Twins Although feed drives rarely fail, they can cause high costs in production when they do. In its Kidz project, iwb is conduct- ing research into how predictive mainte- nance can be implemented by combining algorithms and traditional machine tool models, even in cases where only a few failures have been observed. The back- ground is that feed drives have a con- siderable influence on the productivity of machine tools: On the one hand, such failures are time-consuming and costly to repair. On the other hand, they signif- icantly influence the machine dynamics and thus the actual metal removal rate. Switching to a predictive maintenance strategy therefore offers great economic potential. However, there is currently no comprehensive system on the market that only uses existing sensors to moni- tor and predict the condition of feed drive components of producing machine tools. The Kidz project is therefore carrying out research into a hybrid system for predic- tive maintenance. It combines a digital twin with modern AI methods. “In con- trast to purely data-driven approaches, the hybrid system can work with much less training data in the form of failure ob- servations, which are difficult to obtain in the case of machine tools - says Michael Zäh, Professor of Machine Tools and The main topics addressed at the EMO by the machine tool manufacturer Hermle digitalization in conjunction with automation. For Hermle, EMO 2023 Hannover is ‘the’ international engineering trade fair of the year. Constant wear of the feed drives impairs the dynamic properties of the machine tool. (Photo_ Institute for Machine Tools and Industrial Management).

RkJQdWJsaXNoZXIy Mzg4NjYz