Online Monitoring of Variable Polarity TIG Welding Penetration State Based on Fusion of Welding Characteristic Parameters and SVM

Abstract

In variable polarity TIG (VPTIG) welding of aluminum alloy, effective recognition of welding penetration states is a hot research topic. It is also one of the key factors for the quality of weld and the joint represent. We established an intelligent sensor system for VPTIG welding to obtain the welding current, misalignment and interval, the clear weld pool images and wire feed speed online. With an effective image processing algorithm, weld pool width is measured accurately online. To investigate the complicated relationships between the welding parameter and different welding condition, an improved Support Vector Machines (SVM) classification model based on artificial fish swarm algorithm is built. The work shows that the proposed Support Vector Machine model classifies aluminum alloy welding states effectively.