In-Situ Process Monitoring in Metal Powder Bed Fusion Processes by Means of Multi-Sensor Data Mining Methods

Thanks to rapid technological advances, metal additive manufacturing (AM) technologies enable the production of complex shapes, topologically optimized and lightweight structures that are of industrial interest for advanced applications in different sectors, like aerospace and health-care. However, stringent quality standards and aerospace process qualification requirements impose defect-free and first-time-right capabilities that are still challenging to achieve with state-of-the-art AM systems. This paper reviews different methods to gather and make sense of in-situ data from different sensors during powder bed fusion processes. These methods aim to enhance the embedded intelligence of the AM system by integrating the capability of automatically detecting and localizing process defects since their onset stage.