Research On Measuring Method Of Arc Length Of Pulsed TIG Welding In Annular Seam Of Tube Plate

The welding of the annular seam of tube plate is the key link in the manufacturing process of tube and shell heat exchanger.At present,pulsed TIG welding equipment is used.Due to the low processing and assembly accuracy of the pipe and tube plate,and the number of ring cracks in the tube plate,it is easy to appear in the welding process of tube plate deformation or tungsten electrode melting caused by the change of welding arc length,if not timely detection and adjustment,it will lead to the heat exchange tube wall burning or the ring joint is not melted and other welding defects.Therefore,this paper proposes an arc length detection method based on welding loop voltage oriented to the welding process of tube plate ring seam,and carries out the following work:(1)Based on the TIG welding platform of automatic tube plate ring seam,the arc voltage acquisition circuit board is designed to solve the problem of signal interference during arc voltage sampling.The arc voltage acquisition circuit board collects signals from the electric control cabinet through the circuit board to avoid the influence of high temperature caused by extreme arc voltage signal collection from tungsten.The arc voltage acquisition circuit board has step-down multistage filtering processing,which can effectively eliminate the high-frequency noise in the signal.(2)The peak arc voltage signal that can reflect the arc length change characteristics of pulse signals is the peak arc voltage signal.The pulse square wave arc voltage signal is analyzed,and the moving window method is used to extract the peak arc voltage data that can reflect the arc length change characteristics,and the sampling points of arc pressure base value,rising edge and falling edge are interpolated to compensate,so as to reduce the influence of peak data sample extraction on arc length prediction.(3)Aiming at the high frequency noise interference in the extracted peak arc voltage data,discrete Symlet4 wavelet transform,sliding window mean filtering and sliding window median filtering are used to filter the peak data samples,and the filtering effects of the three filters are compared.Finally,it is determined that the discrete Symlet4 wavelet transform can better solve the interference problem of high frequency noise in the peak arc voltage data samples.(4)The experimental scheme for arc length-arc pressure collection of ring joints was designed.Arc pressure signals under different welding arc lengths were collected through welding experiments,and the sampled data were fitted by least-square linear method to obtain a mathematical model of arc length-arc pressure,and the model was applied to analyze abnormal welding ring joints.