In-Process Data Integration for Laser Powder Bed Fusion Additive Manufacturing

Abstract

Additive manufacturing (AM) is a powerful technology that can create complex metallic parts and has the potential to improve the economic bottom line for various industries. However, due to process instabilities, and the resulting material defects that impact the part quality, AM still isn’t as widely used as it could be. To overcome this situation, it is crucial to develop an environment for easy, in-process monitoring and real-time control to detect process anomalies and predict part defects as quickly as possible. AM in-process monitoring measures various process variables and the sensors generate large volumes of structured or unstructured, 1D, 2D, and 3D data, some of which are acquired at very high frequencies. Integration of such data and their analysis are necessary for effective in-process monitoring and real-time control, but they are facing many challenges due to the characteristics of AM in-process data. This paper provides an overview of different in-process monitoring data sources and their connection methods and addresses the integration issues associated with acquiring and fusing the data for both on-fly control and offline analysis. The paper also presents a guideline to help high-speed data integration. This guideline can help users to decide the best data-integration configuration for a specific use case.