Recent progress of sensing and machine learning technologies for process monitoring and defects detection in wire arc additive manufacturing

Wire Arc Additive Manufacturing possesses advantages of high deposition rate and low cost compared with other metal additive manufacturing processes. However, potential defects may occur during the process, such as pores, cracks, lack of fusion, inclusions, delamination, and geometrical deviations. These defects are undesirable and have negative effects. To optimize the performance of the as-built components, and to reduce the potential defects, a feasible solution is to conduct in-process sensing and provide feedback to the control system. This article aims to give a comprehensive review of recent progress on sensing technologies, such as optical, acoustic, vision, thermal, and multiple signals-based sensing technologies, and the application of machine learning to enhance the ability to extract the needed feedback from the in-process monitoring raw data. Effective monitoring of different types of defects typically requires different sensing technologies, focus points, and attentions. Multi-sensor-based sensing systems may thus be needed to provide full-scale information. These necessities include the need for in-time data fusion and more complex data processing. This review analyzes recently explored sensing technologies for their principles and remaining challenges to provide directions for future invention, exploration, and investigation.