Dynamic modeling and control of welding processes using neural network techniques

Summary form only given, as follows. A scheme for an adaptive network of nonlinear elements and delay lines is proposed, which can learn to model the time responses of a nonlinear, multivariable system. The structure has been applied to build a model of a highly coupled multivariable process: gas tungsten arc welding. The authors present the architecture, the learning algorithm, and the summary of experiments which show the feasibility of the approach, and propose a controller architecture that can regulate the welding process using neural network techniques.<>