Friction Stir Processing of Al 2124 Reinforced Graphene Metal Matrix Composites and Multi Characteristic Optimization Through Desirability Approach Integrated with ANN

Multi-response optimization of composites was performed using the Taguchi-based desirability approach. Parameters, namely tool rotation speed (900 rpm, 1120 rpm, and 1400 rpm), feed rate (20 mm/min, 40 mm/min, and 60 mm/min), graphene nanoplates (GNPs) content (6.5 vol.%GNPs, 11 vol.%GNPs, and 17.4 vol.%GNPs) each at three levels were considered to optimize and enhance tensile strength and hardness using Taguchi’s L9 orthogonal array. Besides Taguchi’s analysis and the developed regression model, the artificial neural network was used to validate the results. A confirmation test was compared with the predicted value of composite desirability ( \(d_G\) $d_G$ ), and a comparison was drawn between experimental and predicted values. From ANOVA, results inferred that vol.%GNPs significantly impacts \(d_G\) $d_G$ . The obtained tensile strength and hardness values effectively reflect the macrostructural, microstructural studies and XRD results. Fracture analysis of the confirmation sample showed no dimples but continuous deep ridges with sharp, torn edges indicating brittle failure.