On the alignment of in-process and post-process measurement datasets acquired for precision enhancement of laser powder bed fusion of metals

Despite laser powder bed fusion technology is spreading widely among several industrial sectors, the process can be affected by low repeatability and manufacturing issues leading to defects formation. Therefore, the development of in-process monitoring systems aimed at detecting possible defects during the fabrication is of increasing interest. In this context, process events must be correlated to actual defects, which implies the need for an accurate comparison between datasets acquired by in-process and post-process measurements. This work is focused on the development of a sample geometry specifically designed for the accurate registration of datasets, which was demonstrated to be fundamental to improve data correlation, but is complicated for example by part distortions occurring after fabrication. The proposed method was verified through a case study, in which it was successfully implemented to correlate in-process hot spot detections with the induced pores measured by post-process X-ray computed tomography.