Intelligent Control of Arc Welding Dynamics during Robotic Welding Process

Abstract

This paper presents a sequential research works on visual information acquirement and intelligent control of arc weld pool dynamics and seam formation during pulsed GTAW (Gas Tungsten Arc Welding) in robotic welding process. The visual information acquirement methods are focused in computer vision sensing, image processing and characteristic extraction of the weld pool surface from the single-item pool images by particular algorithms for robotic welding process. Based on acquired visual characteristics of weld pool and established neural network and knowledge models for predicting dynamical characteristics of weld pool during robotic welding, corresponding control methods, such adaptive control, self-learning and other composite intelligent control strategies are developed to control welding pool dynamics during pulsed GTAW by welding robot. Some experiments and applications of intelligent control methods in welding robot systems are shown in the paper.