In-situ monitoring saves time and cost
Even while a process is running, sensors in the systems continuously monitor various parameters such as pressures, temperatures and energy consumption. »Deviations from the ideal parameters indicate that something is going wrong in the process,« explains Philipp Lelidis, research scientist in the Production Machines and System Management department. Such indicators also point to where the problem lies.
One use case is additive manufacturing, for example in metal 3D printing, which is mainly used in the aerospace and medical industries. Such processes often take several hours. If the employees only realize afterwards that there has been an error and the component is unusable, time and raw materials have been wasted.
»We are researching how the machine can use sensors to automatically detect where the problem may be, what impact this has on the component and what countermeasures we can take to still end up with a perfect component,« says Lelidis. This in-situ monitoring can also reduce the extent of required quality assurance in the end.
AI for pattern recognition and process optimization
Another example of the benefits of in-situ monitoring is laser machining. »A laser beam can cut, drill or remove fine layers from the surface of the material,« explains Luiz Guilherme De Souza Schweitzer, who leads the Process Technologies department in the Ultra- and High-Precision Technology division. Particularly precise work is carried out using laser pulses that are only a few femtoseconds long. When they hit a material, it sublimates – i.e. it changes from the solid to the gaseous state. The acoustic spectrum of this vapor can be measured using optical microphones.
There is an ideal acoustic spectrum for every process. Deviations from it indicate that errors have occurred. At present, it is still very difficult to read out these acoustic spectrums. Schweitzer is therefore working on training an artificial intelligence to recognize faulty patterns and carry out optimization procedures on the process. »We only have this expertise in-house at the moment, but testing with customers is already underway,« explains Schweitzer.
Werk 4.0 increases resilience in production
What works on a small scale now should also function on a large scale in the future. As part of the Werk 4.0 project, Fraunhofer IPK is working on equipping not just individual systems, but an entire plant with digital assistance systems. This should also reduce the workload for employees as well as help to make manufacturing jobs more accessible and fill them despite the shortage of skilled workers.