Research progress of laser welding process dynamic monitoring technology based on plasma characteristics signal

Abstract

During the high-power laser welding process, plasmas are induced by the evaporation of metal under laser radiation, which can affect the coupling of laser energy and the workpiece, and ultimately impact on the reliability of laser welding quality and process directly. The research of laser-induced plasma is a focus in high-power deep penetration welding field, which provides a promising research area for realizing the automation of welding process quality inspection. In recent years, the research of laser welding process dynamic monitoring technology based on plasma characteristics is mainly in two aspects, namely the research of plasma signal detection and the research of laser welding process modeling. The laser-induced plasma in the laser welding is introduced, and the related research of laser welding process dynamic monitoring technology based on plasma characteristics at home and abroad is analyzed. The current problems in the field are summarized, and the future development trend is put forward.

Keywords

• pattern recognition
• characteristics of laser-induced metal vapor
• high-power laser welding
• least squares support vector regression
• multiple kernel learning