LAUSR

Abstract In-situ monitoring of metal additive manufacturing (AM) processes is a key issue to determine the quality and stability of the process during the layer-wise production of the part. The quantities that can be measured via in-situ sensing can be referred to as “process signatures”, and can represent the source of information to detect possible defects. Most of the literature on in-situ monitoring of Laser Power Bed Fusion (LPBF) processes focuses on the melt-pool, laser track and layer image as source of information to detect the onset of possible defects. Up to our knowledge, this paper represents a first attempt to investigate the suitability of including spatter-related information to characterize the LPBF process quality. High-speed image acquisition, coupled with image segmentation and feature extraction, is used to estimate different statistical descriptors of the spattering behaviour along the laser scan path. A logistic regression model is developed to determine the ability of spatter-related descriptors to classify different energy density conditions corresponding to different quality states. The results show that by including spatters as process signature driver, a significant increase of the capability to detect under-melting and over-melting conditions is observed. This is why future research on spatter signature analysis and modelling is highly encouraged to improve the effectiveness of in-situ monitoring tools.