Online inspection technology for weld formation quality of welded structural components

Currently, the non-destructive testing for weld formation quality of welded structural components still predominantly employs visual evaluation with low efficiency and low accuracy. To improve the situation, an online inspection technology for weld formation quality based on structured light vision was proposed. The established 3D coordinate recognition model endowed this technology with the capability to perceive spatial 3D information; the image processing algorithm integrating YOLOv5 and the spatial distance judgment method autonomously identified weld types from structured light images and automatically measured geometrical dimensions such as weld width, weld reinforcement, and weld leg sizes. Particularly, this algorithm was not limited to monitoring weld profiles, but it could also accurately identify, classify, and locate various geometrical defects including excess weld metal, align deviation, incompletely filled sags, undercut, overlap, excessive convexity, and asymmetric weld leg. Moreover, it could autonomously evaluate defect quality grades online according to the ISO 5 817:1992 standard. Experimental results demonstrate that this technology achieves measurement accuracy at the 10⁻² mm level for geometrical dimensions of welds, attains 100% grading accuracy for geometrical defect quality, and fulfills the requirements for online inspection of weld formation quality in automated welding production lines.