AI in practice RWTH start-up makes machining smart

Related Vendors

FAULHABER Drive Systems

Wuxi InfiMotion Technology Co., Ltd.

Datamatters and Fraunhofer-IPT are working on how to optimize various processes in the machining world. Artificial intelligence plays a major role.

Together with the Fraunhofer Institute for Production Technology (IPT) from Aachen, Germany  (project coordinator) and the partners Gemineers, Innoclamp, and Kaitos, the Cologne RWTH start-up Datamatters is working on "intelligent" machining. The goal is to better and more cost-effectively meet the high quality requirements in the machining industry through the use of artificial intelligence (AI). Datamatters founder Dr. Daniel Trauth calls his business concept "Real World AI." He explains: "While the general public is rushing towards generative AI to create texts and images, we focus on linking AI with the real world—from the Smart Factory through Smart Buildings to the Smart City."

The path through the parameter jungle of machining

Machining, in which materials are shaped and sized by turning, drilling, milling, or grinding, forms an essential basis of manufacturing technology. The industries range from automobile production to the manufacture of medical instruments. Errors in the machining process can have serious consequences, ranging from product failures to safety issues, it is said. Strict quality controls are therefore essential, but also time-consuming and expensive. Trauth: "The automated monitoring and analysis of production processes using AI can significantly reduce inspection times and the cost of quality assurance, while improving the accuracy of quality assessment." As part of the research project "FL.IN.NRW," a learning platform is being developed for decentralized training of predictive AI models, as explained by the IPT. As the first application case, the project team is investigating the complex process of machining. The multitude of tool and process parameters in machining poses significant challenges for quality control, which can usually only be managed through time-consuming manual inspections of the components.

AI models detect component defects in machining

By training the models with process data directly from the production machine, experts can enable AI to detect quality issues during machining. This includes deviations in the desired component profile due to tool wear, which are identified by fluctuations in spindle load and clamping pressure, they emphasize. The AI model immediately detects this tool behavior as a dimensional deviation from previously defined tolerance ranges. As a result, time-consuming quality checks can be conducted as needed and significantly reduced, making quality assurance and manufacturing more efficient.