窄间隙旋转电弧熔化极活性气体保护焊视觉焊缝偏差检测

发布时间：2018-05-02 03:43

本文选题：窄间隙 + 旋转电弧　；参考：《上海交通大学学报》2015年03期

【摘要】：针对旋转电弧窄间隙熔化极活性气体保护焊多层单道焊焊缝跟踪需要,提出一种基于被动视觉传感的焊缝偏差识别方法.首先通过旋转电弧位置传感器和处于触发工作模式的CCD摄像机获取电弧旋转到坡口左侧和坡口右侧时的焊接图像,然后根据图像灰度直方图特点,构建了自适应双阈值获取算法.大阈值用于获取电弧区域进而得到电弧中心位置;小阈值用于获取坡口工件区域,进而通过计算水平方向一阶差分得到坡口边缘.通过对比一个电弧旋转周期内获取的两幅图像,可计算电弧旋转中心和坡口中心的偏差,得到焊缝偏差.该偏差检测算法高效、可靠且可避免坡口底部改变带来的误差,同时可用于不同偏差算法的比较和融合.
[Abstract]:Aiming at the need of seam tracking in multi-layer single-pass welding with narrow gap of rotating arc, a method of weld deviation identification based on passive vision sensor is proposed. Firstly, the welding images of arc rotating to the left and right of the groove are obtained by rotating the arc position sensor and the CCD camera in the trigger mode, and then according to the characteristics of the gray histogram of the image, the welding image is obtained when the arc rotates to the left and the right of the groove. An adaptive double threshold acquisition algorithm is constructed. A large threshold is used to obtain the arc region and a small threshold is used to obtain the groove workpiece area, and then the edge of the groove is obtained by calculating the first order difference in the horizontal direction. By comparing two images obtained in one arc rotation cycle, the deviation of arc rotation center and groove center can be calculated, and the weld seam deviation can be obtained. The error detection algorithm is efficient, reliable and can avoid the error caused by the change of the bottom of the groove. It can also be used for the comparison and fusion of different deviation algorithms.
【作者单位】：江苏科技大学材料科学与工程学院;江苏现代造船技术有限公司;
【基金】：国家自然科学基金项目(51005107,51475218) 江苏省自然科学基金项目(BK2011509) 江苏省“青蓝工程”科技创新团队优秀青年骨干教师项目江苏高校优势学科建设工程项目资助
【分类号】：TG409

【参考文献】