GEOMETRY AND THERMAL REGULATION OF GMA WELDING VIA CONVENTIONAL AND NEURAL ADAPTIVE-CONTROL

This paper investigates the application of conventional and neural ada ptive control schemes to Gas Metal Are (GMA) welding. The goal is to p roduce welds of high quality and strength. This can be achieved throug h proper on-line control of the geometrical and thermal characteristic s of the process. The welding process is variant in time and strongly nonlinear, and is subject to many defects due to improper regulation o f parameters like arc voltage and current, or travel speed of the torc h. Adaptive control is thus naturally a very good candidate for the re gulation of the geometrical and thermal characteristics of the welding process. Here four adaptive control techniques are reviewed and teste d, namely: model reference adaptive control (MRAC), pseudogradient ada ptive control (PAC), multivariable self-tuning adaptive control (STC), and neural adaptive control (NAC). Extensive numerical results are pr ovided, together with a discussion of the relative merits and limitati ons of the above techniques.