Study On Underwater Welding Seam Automatic Tracking System Of High Speed Rotating Arc Sensor

With the exploitation of deep ocean resource, offshore engineering construction become a large scale. As one of the key technologies in offshore engineering construction and maintenance, underwater welding is required absolutely necessary. Due to the saturation restrict of diving welder in deep water, underwater robot, instead of diving welder, is in emergent need to achieve underwater welding automatization and intelligentization. The robot seam tracking is among the important research area.The robot underwater seam tracking system based on the rotating arc sensor, directly use the welding current to track the seam. When the rotational arc scans the groove, causing the distance between the tip and workpiece (CTWD) to change regularly, the welding current change in accordance. Therefore the offset between the center of the welding torch and the midline of the seam can be detected by the current variation. By analyzing the variation of welding current, offset information obtained in order to achieve underwater seam tracking. With high stability, sensitivity and spatial accessibility, the seam tracking system based on the rotating arc sensor is one of the most effective methods currently.Based on previous research results, we improved of the rotating arc sensor firstly, hence the hardware of the robot underwater seam tracking system based on the rotational arc sensor is set up, which was the research basis for seam tracking. The hardware largely include the rotating arc sensor, the current sensor, singlechip, photoelectric sensor, data capture card, IPC and the robot. The system is composed of welding subsystem, signal acquisition subsystem, filter and offset identification subsystem, and the robot tracking control subsystem.Aimed at the DC motor used in high speed rotational arc sensor of underwater welding, a system of speed feedback controlled by closed-loop has been designed, in which singlechip is used as the core control chip, the control mode of PWM and the control algorithm of PID are adopted. With simple circuit, easy to control and work stable and reliable, it provided a guarantee for offset identification of underwater welding.As the fast cooling speed and water pressure in the underwater welding process, due to the strong arc, splash, short circuit transfer and so forth, the current signal and speed signal have been interfered. After the wavelet filter and median filter in the LabVIEW8.5 software development platform, current waveform have improved significantly. Eliminated the interference of arc on the speed signal, mergered the current signal and speed signal to determine the current signal cycle, provided a more accurate information for offset identification. Finally, left-right integral method was adopted to identify the welding torch offset. Obtained quite high recognition accuracy, builted a foundation for underwater welding seam tracking.The four main factors, which were rotating frequency, the radius of the torch, torch height and welding current, affect the welding forming and the underwater current waveform characteristics. Orthogonal tests were carried through, the three test indicators, which were the tensile strength, weld width and pool depth, were analyzed and compared respectively, to determine the optimal parameters for underwater welding.Finally, set up the communication between the IPC and the robots. The underwater seam tracking experiments were carried out on the steel plates with the bias seam and broken seam, achieved a good tracking results.