METAHEURISTIC APPROACH

The aim of this study is to optimize the impact energy of Tungsten Inert Gas (TIG) mild steel welds by identifying the most effective combination of welding parameters current, voltage, and gas flow rate to achieve the best mechanical performance. The specific objectives include developing a mathematical model to describe the relationship between these parameters and impact energy, applying a metaheuristic algorithm to determine the optimal settings, and validating the optimized results against existing experimental data. This research seeks to address the limitations of traditional trial-and-error and local statistical optimization techniques, which often fail to locate the true global optimum. The study employed a hybrid computational optimization approach that combines Response Surface Methodology (RSM) and Particle Swarm Optimization (PSO). RSM was first used to develop a second-order regression model of impact energy based on existing experimental data from TIG welding of mild steel. This model served as the objective function for the PSO algorithm, which was implemented in MATLAB. The PSO algorithm iteratively adjusted welding parameters to maximize the predicted impact energy, thereby exploring the solution space beyond the limits of conventional statistical methods. The results showed that the optimal welding parameters were 192.73 A (current), 19.12 V (voltage), and 20.23 L/min (gas flow rate), corresponding to a maximum predicted impact energy of 118.52 J. This value slightly exceeded the best experimental result of 116.48 J reported in literature, confirming the effectiveness and accuracy of the hybrid RSM–PSO framework. The optimized results not only align closely with existing research trends but also demonstrate that integrating metaheuristic algorithms into welding parameter selection can enhance weld toughness, minimize experimental effort, and improve process reliability