In-Process Condition Monitoring in Laser Powder Bed Fusion (LPBF)

The goal of this work is to monitor the laser powder bed fusion (LPBF) process using an array of heterogeneous sensors. This goal is termed as build status monitoring. The overarching aim is to usher a qualify-as-you-build quality assurance paradigm in LPBF whereby incipient build defects are detected through analysis of data acquired from in-process sensors. This will allow opportune preventive action to avert propagation of build defects across multiple layers. In pursuit of this goal, a commercial LPBF machine at the National Institute of Standards and Technology (NIST) was integrated with three types of sensors, namely, a photodetector, high-speed video camera, and SWIR thermal camera with the following objective: to develop and apply a spectral graph theoretic approach to monitor the LPBF build status from the data acquired by the three sensors. This objective will lead to early identification of incipient defects that afflict LPBF despite extensive process automation. The proposed approach is illustrated with experimental sensor data acquired during LPBF of a part having a steep overhang feature of ~ 40.5o . Parts with such steep overhang features may exacerbate deleterious consequences such as poor surface finish, porosity, and distortion. Hence, close monitoring of the signal patterns during scanning of overhang areas is consequential for early detection of build defects. The proposed approach detected differences between overhang and non-overhang build status for different sensors with the statistical fidelity (F-score) of 95% from thermal camera signatures to 79% with the photodetector. In comparison, conventional signal analysis techniques - e.g., neural networks, support vector machines, linear discriminant analysis, etc., are evaluated with F-score in the range of 40% to 60%. As part of our forthcoming work, this study will be further expanded to include more build defects, e.g., due to material contamination.