Abstract

The shape of welding arc is closely related to the quality of welding. For micro plasma arc welding (MPAW), however, the change of arc shape is not obvious enough to evaluate the quality of welding, due to the small size of welding arc. Thus, it needs to extract the specific features of arc shape through image processing algorithm to judge its change quantitatively. Furthermore, for workpiece with bright surface, it is difficult to extract the arc shape directly from the arc images because of the strong arc reflection on the workpiece surface. In this paper, a new method based on vertical Sobel operator was proposed to overcome the above-mentioned challenges, and to extract the clear arc shapes from the arc images of pulse micro plasma arc welding (P-MPAW). The arc length, arc width and the height difference between two sides of arc were extracted as characteristic parameters to reveal the influence of different welding defects on the arc shape. The changes of arc characteristic parameters have been studied in depth, when there are different defects such as burn through, misalignment, undercut, concave deformation, excessive gap and overlap. The results show that, excessive welding gap and misalignment, caused by the assembly and machining accuracy, have obvious characteristics in the instantaneous arc shape. The defects caused by welding thermal process like burn through, concave deformation, overlap and undercut have their own distinctive characteristics in the changing process of the arc shape.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call