深入理解熟练焊工经验和操作技能对实现复杂环境下的机器人智能化及高精度焊接具有重要理论意义和工程应用价值。然而，如何检测传感焊工与熔池交互过程中熔池动态变化特征信息是一个难题。针对该问题基于激光视觉传感原理建立动态熔池和焊工实时调控特征信息同步采集试验系统，研究熔透连续变化时焊工对熔池形态及流态的调控行为，并获得了相应的调控参数（焊接速度、焊接电流、焊接弧长）和焊枪姿态与熔池特征参数之间的关系。通过对比分析焊接规范参数与焊枪姿态调整时的熔池形态和流态变化发现：焊接速度、焊接电流和焊接弧长的调整主要是基于熔池输入能量的调控，表现出熔池形态变化剧烈，易出现焊塌缺陷，焊工调控过程呈现出单一、盲目性。焊枪姿态的调整表现出焊工对熔池形态和液态金属流动状态的多元化和指向性调控，可避免焊接缺陷的产生。
To understand welding experience and operation skills of skilled human welders which has theoretical and practical significance for realization intelligence and the high-precision of robot welding in complex situation. There is a big problem that to detect the weld pool dynamic change and characteristic information reflecting the welding experience and operation skills. Aiming at this problem, a synchronous experiment system is setup with laser vision and a dynamic tilt sensor to precisely obtain real-time controlling characteristic information. The correlation between control parameters in welding speed, welding current, arc length and characteristic information of weld pool is obtained. It is found that the energy input in weld pool determines the change of shape of weld pool and flow pattern by adjusting welding speed, welding current, arc length. As a result, the regulation of welder is blindly and simply, the weld pool is unsteady and welding defect occurs. The adjustment of welding torch posture shows the diversity and directional control of welding pool shape and liquid metal flow pattern, overcoming the welding defect.
[1] 张宗郁,高洪明,韩庆璘,等. 飞机导管机器人焊接手眼关系标定[J]. 上海交通大学学报,2015,49(3):392-401. ZHANG Zongyu,GAO Hongming,HAN Qinglin,et al. Hand-eye calibration in robot welding of aero tube[J]. Journal of Shanghai Jiao Tong University,2015,49(3):392-401.
[2] LIU Y K,ZHANG W J,ZHANG Y M. Dynamic neuro-fuzzy based human intelligence modeling and control in GTAW[J]. IEEE Transactions on Automation Science and Engineering,2015,12(1):324-335.
[3] LIU Y K,ZHANG Y M. Control of human arm movement in machine-human cooperative welding process[J]. Control Engineering Practice,2014,32:161-171.
[4] LIU Y K,ZHANG Y M. Iterative local ANFIS based human welder intelligence modeling and control in pipe GTAW process:a data-driven approach[J]. IEEE/ASME Transactions on Mechatronics,2015,20(3):1079-1088.
[5] ZHANG G,SHI Y,GU Y F. Welding torch attitude based study of human welder interactive behavior with weld pool in GTAW[J]. Robotics and Computer-Integrated Manufacturing,2017,48:145-156.
[6] ZHANG W J. Modeling of human welder behavior in gas tungsten arc welding of stainless steel tubes[J]. Welding in the World,2014,58:601-617.
[7] LIU Y K,ZHANG Y M,KVIDAHL L. Skilled human welder intelligence modeling and control:Part Ⅱ-Analysis and control applications[J]. Welding Journal,2014,93:162-170.
[8] LIU Y K,ZHANG Y M,KVIDAHL L. Skilled human welder intelligence modeling and control:Part I-Modeling[J]. Welding Journal,2014,93:46-52.
[9] LIU Y K,SHAO Z,ZHANG Y M. Learning human welder movement in pipe GTAW:A virtualized welding approach[J]. Welding Journal,2014,93:388-398.
[10] HASHIMOTO N,NAKAMURA A,OISHI S. Training system for manual arc welding using artificial reality measurement of manual welding motion for skill evaluation[C]//JSPE Semestrial Meeting,2010,93:388-398.
[11] ERDEN M S,TOMIYAMA T. Identifying welding skills for training and assistance with robot[J]. Science and Technology of Welding and Joining,2009,14(6):523-532.
[12] ERDEN M S,TOMIYAMA T. Human-intent detection and physically interactive control of a robot without force sensors[J]. IEEE Transactions on Robotics,2010,26(2):370-382.
[13] 石玗,张刚,马晓骥,等. 基于激光视觉的脉冲GTAW熔池振荡检测与分析[J]. 机械工程学报,2012,48(24):28-32. SHI Yu,ZHANG Gang,MA Xiaoji. Laser vision based detection and analysis of weld pool oscillation for pulsed GTAW[J]. Journal of Mechanical Engineering,2012,48(24):28-32.
