Intelligent Control Methodology of Dynamic Process of the Pulsed GTAW Welding

摘要

This paper summaries the real-tie intelligent control emthodologies of weld pool dynamic process of the pulsed Gas Tungsten Arc Welding (GTAW). Based on the computer vision systems for sensoring feature parmeters of the weld pool. such as the weld face/root weldth, half-length, adn penetra-tion. some intelligent contro-llors, e.g. fuzzy control, neural control, self-learning control, are designed for control of the weld pool dynamic changes. Analysis and experiemnt results show that the self-learning fuzzy neural neural control scheme is more sutiable to the welding dynamic process.