EMO Hannover 2023 Online Seminar ‘Let’s Talk Science’, Shows How to Make AI Models More Flexible

Marcel Wichmann holding a real hip implant in front of its digital twin

Frankfurt am Main, Germany – ‘Let’s Talk Science’, a monthly online seminar hosted by EMO Hannover 2023, demonstrates how to make AI models more flexible. In the online seminar, Marcel Wichmann, Head of the Production Systems Department, Institute of Production Engineering and Machine Tools, Leibniz University Hannover, talked about how small and medium-sized businesses will benefit more and more from AI in the future.

Artificial intelligence (AI) is a buzzword that is heard everywhere these days, but its reality has yet to impact everyday practice. The reason for this is that complexly computed AI models for specific processes are rigid. If a process changes, such as the use of different materials or tool wear, the model is unable to adapt. This is something that researchers from Hannover want to change, allowing AI to be used in a variety of ways.

To that end, they employ digital twins, which are precisely mapped digital images of processes. Such simulations have a lot of technical and economic potential because they can help with early detection and compensation of quality deviations. “Until now, however, AI models have been unable to respond to changing process conditions. They had to be retrained, and were thereby naturally a huge drain on time, expertise, and money,” Wichmann explains.

Such learning AI models enable what is known as adaptive process control, which is the ongoing adaptation of a process to changing conditions such as tool wear or errors during processing. Wichmann uses concrete examples to demonstrate the use of digital twins. He gives the example of a turbine blade regeneration application in aerospace engineering. This application allows one to predict the outcome of a specific workpiece and, if necessary, modify the process. This, in turn, improves the component’s quality and safety.

Another example demonstrates tool grinding optimization. The virtual tool’s twin’s forecasts are constantly adjusted based on data feedback, and the process is constantly optimized. In practice, this means fewer faulty workpieces and shorter run-in processes. The Hannover team also hopes to facilitate the imaging of entire production processes, from the manufacturing and use of a component all the way to its quality control. “Last but not least,” says Wichmann, “our simulations will improve the productivity of the processes.”

He further adds, “Quite apart from that, in these times of shortages in skilled staff, we are giving companies the opportunity of creating a team with many virtually supported assistants who have access to the experience of older members of staff, and all of this in a group with an experienced employee and a number of assistants.”

Image Source: IFW Hannover


Eplan
  Facebook   Twitter   Linkedin   Subscribe