2-D Transformer-Based Approach for Process Monitoring of Metal 3-D Printing via Coaxial High-Speed Imaging

Defects in the metal 3-D printing process exhibit randomness and low frequency, making them difficult to predict and control. This severely hinders the application of this technology in critical industrial fields. Extracting useful features from massive process monitoring data to ensure forming quality has become a popular research direction for intelligent additive manufacturing practitioners. In this study, a coaxial machine vision monitoring system is utilized to monitor the entire forming process of the melt track. More importantly, this study constructed a high-speed video dataset of typical metal 3-D printing working conditions by changing the powder layer thickness, which can serve actual industrial production. To promptly recognize unhealthy melt tracks from massive process monitoring data, a 2-D transformer-based framework named super frame feature pyramid transformer (SFFPT) is designed for video classification. This framework transforms the video understanding task into a 2-D feature map processing task, allowing the video classification task to be completed directly using only an image classifier. In comparison to state-of-the-art methods, SFFPT achieves the best classification accuracy in this study.

I. Introduction

Metal 3-D printing (M3DP), also known as metal additive manufacturing, has broad application prospects in fields such as automotive manufacturing, medical equipment, and aerospace [1]. However, the forming process of this technology is often interfered by defects such as balling, warping, porosity, and cracks. These defects exhibit randomness and low frequency characteristics, making them difficult to predict and control, thus severely hindering the application of this technology in crucial industrial fields. Randomness refers to the uncertainty of the time, location, and type of defects that occur during the forming process; low frequency refers to the very small frequency of defect occurrence, or even no occurrence. Process monitoring involves capturing characteristic information about defect generation from a large amount of process monitoring data. Effectively detecting defects during part forming has become a crucial research issue in the field of intelligent process monitoring [2].