An Intelligent Algorithm to Control Welding Parameters for Lab Joint

Recently, the automated welding system is employed in high volume production industries where the cost of equipment is justified by the large number of pieces to be made. The system also provides the same time saving and precision welding, yet it can only be applied to small-lot production and to production of a single part. However, an intelligent model that predicts bead geometry and accomplishes the desired mechanical properties of the weldment in the automated GMA (Gas Metal Arc) welding is required. The algorithm should also cover a wide range of material thicknesses and be applicable for all welding position. Furthermore, the proposed algorithm for the automatic welding system must be available in the form of mathematical equations. In this study, an intelligent model, which employed the neural network, one of AI (Artificial Intelligence) technologies has been developed to study the effects of welding parameters on bead penetration area and predict bead penetration area for lab joint in the automated GMA welding process. BP (Back Propagation) and LM (Levenberg-Marquardt) neural network algorithm have been used to develop the intelligent model. Not only the fitting of these models have been checked and compared using variance test, but also the prediction on bead penetration area using the developed models have been verified.