Mathematical modelling to predict angular distortion in butt-welded stainless steel 304 thin plates

Abstract

Arc welding processes are characterized by the input of huge amount of heat into the workpieces. The heating and cooling cycles associated with this process are quite rapid and non-uniform in nature. This leaves different parts of the weldment subjected to non-uniform thermally generated stresses resulting in weld distortions. These distortions account for multiple rejections and expensive remedial procedures in MIG welding of thin plates, particularly in the butt joint configuration plus cause difficulty in assembling and spoil the aesthetic appeal of the product. Thus, it is crucial for the designer to be able to foresee the extent of these distortions to a certain level of accuracy so that pre-remedial steps can be taken while selecting the input parameters to minimize the final angular distortion. Thus, there is a need for a mathematical model capable of predicting the same. Hence in the present work, a mathematical model was developed using the statistical technique of Design of Experiments to predict these effects to minimize their consequences. Accordingly, an investigative work was carried out that involved carrying out a number ofexperimental runs in a structured manner. The material selected for the study is SS304 plates (2mm thick), as Stainless Steel recently is finding ever increasing demand in engineering and structural applications because of its many advantagesover low alloy steels. A mathematical equation was developed whose adequacy was checked using ANOVA technique. Response surface methodology was used to developgraphical representation ofdirect and interactive effects of input parameters on angular distortion.