Process Drift Detection in SLM process using in-situ monitoring and machine learning approach

Laser-powder bed fusion (L-PBF) has seen a growing demand in niche industry applications for its complexity free manufacturing capabilities. Besides, all the advantages over traditional manufacturing techniques, reliability and repeatability is a major concern in this domain. In-situ monitoring of the process using high precision sensors such as photodiodes, High-speed IR cameras both in co-axial and off-axis positions have been installed in commercially available machines. But understanding the correlations between the acquired data from sensors and build quality is a challenge and time-consuming. Therefore, a comprehensive analysis of the sensitivity of the sensors and easy detectability of the drift during the process is need of the time. In our work, we would like to present a sensitivity analysis of the the in-situ monitoring systems installed on commercial SLM280^HL machine and linking it with in-house developed bench. We have studied the link between variation in signal captured by various coaxial photodiodes (visible and infrared region) with drift in the process. Later, we make use machine learning (ML) approaches to treat the data in an automated fashion. Use of ML for treating the enormous data and detection of the drift will reduce the time and data storage problem linked to the current system.