Research Of Narrow Groove Pipe Welding Seam Tracking Technology Based On The Arc-Based Sensing

All position welding robot for pipeline laying is an important pipeline laying equipment in deepwater pipeline laying system. The stable working performance and high welding efficiency are the first factor to determine the efficiency of the pipe laying efficiency and the economic benefit of construction. The narrow groove pipe welding based on arc sensing welding is a key technology in the automatic welding of submarine pipeline laying. Pipeline welding belongs to all position welding with narrow groove. The steep sidewall of the groove, small current droplet transfer and all position welding process parameters'changing increase the difficulty of traditional welding seam tracking method based on arc sensing.With sufficient researches of the domestic and the foreign pipeline welding robot arc sensor technology research, this paper analyzes arc sensor technical difficulties of the narrow groove pipe welding and proposes a research program for the arc sensing technique of submarine pipeline laying welding robot. This essay also put out the key technologies in the program. The main research work is as follows:(1) Based on the swing arc sensor a welding test system platform with functions of data acquisition, error calculation and deviation adjustment is built. Effect of welding torch swing frequency to the arc sensor signal sensitivity was studied. Through the melting characteristics of different side wall distances, the best distance to the side wall of the groove and the stable welding electrical signals were found.(2) Studying the narrow groove arc sensing mechanism under the different droplet transition the numerical models of arc sensor in narrow groove welding is established. The model can realize free conversions from one of the three kinds of droplet transfer mode to another. A torch swing model in the narrow groove is added to the numerical models of welding transfer modes, forming the narrow groove welding models, by using which the electrical signal waveform of different oscillation frequency and swing wide conditions.(3) In view of the influence of the droplet transfer to the arc signal, the coif5 wavelet is used to filter the arc signal, which can remove the noise signal in the arc sensor signal and obtain the stable arc deviation signal.(4) Through analysis of welding arc signals a deviation extraction algorithm using integral interval is proposed, and the dead zone PID control strategy is used to compiled the weld tracking algorithm on LabVIEW, to obtain a better stability.The above research result provide the basis for the prototype manufacturing of the submarine pipeline laying welding robot, laid the foundation for the further study on the key technologies of intelligent and will effectively promote submarine pipeline laying welding robot in the actual application process.