Visual sensing and quality control in plasma MIG welding

Journal of Manufacturing Processes(2023)

引用 4|浏览10
暂无评分
摘要
By generating a complementary effect of each weld in the welding process, a hybrid welding can improve the welding efficiency while ensuring the welding quality. The combination of plasma arc welding (PAW) with high current density and metal inert gas (MIG) welding utilizing consumable electrodes is available to perform one-pass welding for the thick plate. The plasma torch works for keyhole welding with deep penetration, and the MIG torch works for a build-up welding to form a surface bead. However, due to the complexity and particularity of the two-electrode system, it is still difficult to achieve a high-quality automatic welding. In this study, the authors have observed the weld zone with a complementary metal oxide semiconductor (CMOS) camera, and analyzed the regulation of image features changing with the torch position. Control strategy of the welding torch and image processing to realize the seam tracking and standoff have been proposed. Utilizing a high-speed video camera, system configuration and welding conditions have been determined with the welding images of the arc behavior and droplet transfer. The authors have investigated the corresponding welding conditions when the welding defects occurred. In addition, we have modified the power source with an invertor control and designed a characteristic curve for the power source to stabilize welding conditions and enhance the MIG arc length self-regulation. The performance of welding system for the seam tracking and standoff control is effective in the real-time welding experiment. The power source with an invertor control and the designed characteristic curve can optimize the welding conditions and acquire good welding results. The study has achieved a high-quality welding under the condition of automatic plasma MIG welding.
更多
查看译文
关键词
Plasma MIG welding,Image processing,Seam tracking,Quality control,Intelligent manufacturing
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要