Multi-Objective Optimization of Parameters during Friction Stir Welding of AM50A Mg Alloy Using Genetic Algorithm and Desirability Function Approach for Maximizing Tensile Strength and Hardness
摘要
This research work aims to optimize the friction stir welding (FSW) parameters to enhance the mechanical performance of AM50A magnesium alloy joints. Central composite design (CCD) methodology was adopted to systematically design the experiments, focusing on a butt joint configuration. FSW parameters welding speed, axial force, depth of plunge, and tool RPM were varied to evaluate their influence on critical mechanical attributes, namely ultimate tensile strength (UTS) and microhardness (HV). Two advanced optimization methodologies, genetic algorithm (GA) and desirability function approach (DFA), were employed to determine the optimized parameter combinations maximizing UTS and HV. Regression models with perfect accuracy (R2 values of 0.9819 for UTS and 0.9497 for HV) were formulated for understanding the relationship between parameters and mechanical attributes. GA-derived optimal parameters 1100 rpm, 1.19 mm·s−1, 3 kN during the tool plunge depth of 0.187 mm yielded superior mechanical properties. Similarly, DFA optimization suggested slightly different parameters for maximum desirability, including a 1.11 mm/s, 1100 rpm, 3.51 kN, and 0.192 mm depth of plunge. Experimental validation under optimal conditions confirmed perfect agreement with the anticipated values, ensuring the robustness of employed optimization methodologies. Scanning electron microscopy (SEM) revealed grain refinement in the nugget zone, which had contributed for enhancement in the mechanical attributes. Optimized FSW parameters contributed for flaw-free joints exhibiting a 193.25 MPa tensile strength, microhardness of 58.35 HV, substantiating the efficiency of the formulated model.