The Fraunhofer Institute for Material and Beam Technology (IWS) in Dresden has been working on laser powder build-up welding for decades and plans to use this method to produce very strong and heat-resistant materials by means of 3D printing. Among other materials, there are plans to process nickel, which requires the precise setting of numerous parameters such as temperature, type of powder and feed rate. This nevertheless generates a hardly manageable volume of high frequency sensor data.

As part of the futureAM – Next Generation Additive Manufacturing project , researchers in Dresden are now using artificial intelligence (AI) and machine learning to determine the optimum process parameters. For example, the sensor values are linked to the institute’s powder database. The objective is to develop effective methods for the construction of aircraft engines, for example, which have to withstand temperatures of 1,200 degrees and higher. Molding complete components using a high-performance alloy would cost too much. Additive manufacturing allows the expensive material to be used only where it is actually needed and the use of cheaper alloys where the stresses on the component are lower.