Intelligent methodology for sensing, modeling and control of pulsed GTAW part 2 - Butt joint welding

This paper addresses intelligent techniques for the quality control of the pulsed gas tungsten are welding process for butt joints, and it is a develo pment to Ref. 1. Because there exist some important differences in butt joi nt welding and bead-on-plate welding, the modeling and control scheme in Re f. 1 does not completely fit for butt joint welding. In this paper, the dif ferences between the two were investigated. The shape and size parameters f or the weld pool were used to describe the weld pool geometry. A new real-r ime algorithm was developed for the size and shape parameters. A size and s hape neural network model (SSNNM) was established to predict the maximum ba ckside width. The model accuracy was verified. Furthermore, a self-learning fuzzy neural network controller (FNNC) was designed for control of the max imum backside width and the fuzzy rules were modified online. Based on the FNNC, and combined with an expert system, a double-input and double-output (DIDO) intelligent controller was developed for controlling the maximum bac kside width and the shape of the weld pool. Experiment results showed the D IDO intelligent controller could form a better butt joint weld.