MODELING OF WELD BEAD GEOMETRY AND PENETRATION IN TWIN WIRE SUBMERGED ARC WELDING

Abstract

Submerged arc welding (SAW) is one of the chief metal joining processes, employed in heavy industries for the various applications because of its ability to produce high quality and reliable welds. Twin wire submerged arc welding process which is one of its variants is also widely being used for joining and surfacing applications. The present work comprises of modeling of weld bead geometry and penetration for welds produced by Twin wire submerged arc welding process by using Artificial Neural Networks (ANN). The models were developed to relate the important process variables like welding voltage, welding current, welding speed and contact tip to work distance to important bead features namely penetration, reinforcement, bead width, and width of HAZ. Results obtained from the models are used to determine the interaction effects between process variables and the bead features, these are presented graphically as well. The models developed, have achieved good agreement with the training and cross validation data. Predicted and actual experimental values for each model matched to a greater extent and this highlights the success of applying ANN for modeling weld bead features. The developed models are very useful in selecting the process parameters to achieve the desired weld bead quality