Although deep learning has been successfully implemented in processing images and video data, it still has limited applications in investigating physical processes for smart process calibration and control. Collaborating with HP research scientists, this study demonstrates how engineering-informed deep learning can assist to understand the powder bed fusion in 3D printing processes. The studies show promising model performance in both prediction quality (e.g., accuracy) and the computational cost.

As an indispensable link of the life-cycle of AM, end-part quality control in Cyber-Physical Additive Manufacturing Systems (CPAMS) is made difficult by enormous differences in product designs/varieties. Statistical monitoring of additive manufacturing (AM) processes faces major challenge due to the nature of one-of-a-kind manufacturing. This posters puts forth a prescriptive SPC scheme to monitor shape deformation from shape to shape. Only a limited number of test shapes are required to establish control limits.

A significant challenge in dimensional accuracy control of cyber-physical additive manufacturing systems (CPAMS) is the specification of geometric shape deviation models. The current practice of constructing tailor-made deviation models for each combination of computer- aided design model, additive manufacturing (AM) process, and process setting is impractical and inefficient for general application in CPAMS. We present a new framework and class of  Bayesian neural networks for automated and efficient deviation model building in CPAMS.