Research On P-GMAW Arc Sensing Welding Seam Tracking Accuracy

Seam tracking technology is widely used,from thin plate to medium plate,from V-shaped groove to narrow gap groove,from straight weld to intersection line weld,all of them are inseparable from the use of seam tracking sensors.When welding important products,such as aerospace vehicles,large pressure vessels,ships,etc.,it is required that the weld shape and quality reach a high level.However,at present,the domestic arc sensors have not been able to meet the requirements of weld tracking accuracy;especially in the face of complex weld seam tracking accuracy is often not ideal.How to improve the accuracy of arc sensor tracker has become a "neck jamming" technology which restricts its application in complex weld tracking.In this paper,the tracking accuracy of swing arc sensor based on P-GMAW is studied.Firstly,the welding seam tracking test system based on P-GMAW arc sensor is designed,and the test platform is built to support the test under various welding conditions,and has the function of collecting test data.Based on MATLAB software,the test data analysis program is compiled,which can easily and quickly extract the different characteristic signals of the arc,which is prepared to study how different welding conditions affect the tracking accuracy.Secondly,according to the function,the welding seam tracking control system is mainly divided into arc sensing module,robot control system and execution module.The factors influencing the tracking accuracy in these three parts are analyzed and tested.The following rules are obtained: in V-groove welding,under the same conditions,the smaller the groove angle is,the better the tracking accuracy is;under the same conditions,the thinner the wire diameter is,the better the tracking accuracy is;in a certain range,the deviation correction coefficient K should increase with the increase of the distance between the welding gun and the weld center.Then,based on the original feature signal extraction method,the extraction method of "n peak value average method" is obtained.After comparing with the original method,it is found that the sensitivity of seam tracking is significantly improved.Then,the feature signals obtained by this extraction method are detected for abnormal data,and a new method of abnormal data detection is obtained by combining the statistical Grubbs criterion with the fuzzy c-means algorithm based on clustering.After the analysis and processing of MATLAB software,it is concluded that the method can obviously detect the abnormal data in the characteristic parameters,which is conducive to the improvement of tracking accuracy.Finally,in order to explore the phenomenon of arc breaking in the welding process,the narrow gap groove welding,which is easy to arc interruption,is studied.It is found that there is a special shape change of the arc before arc breaking,that is,the arc is established between the end of the welding wire and the molten pool,and the arc will be stretched along the side wall in a parabola shape until it is broken.In addition,the welding voltage before arc breaking also has a special change rule,that is,the abnormal fluctuation or even rise of the base value voltage in several pulse cycles before the arc breaking.Through in-depth analysis of these special phenomena,it is found that they are closely related to the special groove form of narrow gap and the deviation of welding torch.In order to reduce the occurrence of arc breaking,the abnormal behavior of the base voltage of several pulse periods before the arc breaking can be predicted,and the artificial interference can be carried out,so as to improve the welding seam tracking accuracy.