Regression Modeling for Prediction of the Effect of Process Parameters on Bead Geometry and Bead Quality of Submerged Arc Weldment

Submerged arc welding (SAW) is one of the chief metal fabrication processes in industry. It works with high current density and can effect high metal deposition rate. The present work emphasizes on the study of influence of process parameters on quality and performance of submerged arc weldment, by incorporating multiple linear regression method. Based on 3³ factorial design without replication, experiments were conducted with three different levels of process parameters like voltage, welding current and electrode stick out to obtain butt joints from mild steel plates. Experimental data have been utilized to develop a mathematical model which reveals the linear relationship among various process control parameters and response variables in relation to submerged arc weldment. Graphical representations of the experimental data as well as the predicted data, obtained from the developed model, are supposed to contribute valuable information for quality control of submerged are welding process. This would help to obtain superior quality weld and also to achieve higher productivity.