Gas metal arc weld pool surface imaging: Modeling and processing

Observing and measuring the weld pool geometry in the gas metal arc welding (GMAW) process is the key to understanding and control of a complex welding process. The harsh environment and highly dynamic characteristic of a weld pool surface made the observation extremely challenging. In this paper, a reflection-based GMA weld pool surface observation system is proposed to monitor the threedimensional weld pool geometry. A low-power five-line laser pattern is projected onto the weld pool and the reflected pattern, which is the result of an intercepted reflected laser beam on an imaging plane, is captured and analyzed. To image the reflected laser pattern, the elements in the imaging system including laser and weld pool surface are analyzed and modeled. The effects of their parameters such as laser density, projection angle, weld pool surface geometry, and weld pool surface fluctuation on the imaging of the laser pattern are examined and analyzed. The effectiveness of the proposed method and parameter ranges are verified through simulation and experiments. As a result, acceptable images of laser pattern reflected from the weld pool surface in pulsed GMAW were acquired. An image processing algorithm has been developed to successfully extract the reflected laser pattern for subsequent weld pool surface reconstruction.