Optimization of Process Parameters to Minimize Angular Distortion in Gas Tungsten Arc Welded Stainless Steel 202 Grade Plates Using Genetic Algorithms

This paper presents a study on optimization of process parameters using genetic algorithm to minimize angular distortion in 202 grade stainless steel gas tungsten arc welded plates. Angular distortion is a major problem and most pronounced among different types of distortion in butt welded plates. The extent of distortion depends on the welding process control parameters. The important process control parameters chosen for study are gun angle (θ), welding speed (V), plate length (L), welding current (I) and gas flow rate (Q). The experiments are conducted based on five factor five level central composite rotatable designs with full replication technique. A mathematical model was developed correlating the process parameters and the angular distortion. The developed model is checked for the adequacy based on ANOVA analysis and accuracy of prediction by confirmatory test. The optimization of process parameters was done using genetic algorithms (GA). A source code was developed using C language to do the optimization. The optimal process parameters gave a value of 0.000379° for angular distortion which demonstrates the accuracy and effectiveness of the model presented and program developed. The obtained results indicate that the optimized parameters are capable of producing weld with minimum distortion.