A Study on Intelligent Algorithm to Control Welding Parameters for Lap-joint

Abstract

With the advent of electrically driven mechanical man ipulators, multidisciplinary mechatronic engineering and the requirement of information acquisition has increased. The systems removed elements of human mechanical effort from the welding operation, improve consistency and got rid of the operator from the welding hazards. However, the system can only be applied to small- lot assembly and to the production of a single part until now. Therefore, an intelligent model that predicts bead geometry and accomplishes the desired mechanical properties of the weldment in the automated GMA(Gas Metal Arc) welding should be required. In this study, an intelligent model which employed the neural network, one of AI (Artificial Intelligence) technolo gies has been developed to study the effects of welding parameters on the bead penetration area and predict bead penetration area for Lap-joint in the automated GMA welding process. BP (Back Propagation) and LM (Levenberg-Marquardt) neural network algorithm have been employed for developing the intelligent model. Not only the fitting of these models have been checked and compared by using the variance test, but also the prediction on a bead penetration area using the developed models have been verified.