Investigation and optimization of friction stir welding process of aluminum 5010 to 6061

The feasibility of joining AA6061 and AA5010 by FSW is one of main purposes of the current research. First, 9 experiments were conducted with aim of optimal pin profile design. These experiments were designed by using various straight cylindrical, square and triangular pin profiles under different frictional conditions. Results of the first stage indicated that the tool with square pin profile leads to the highest tensile strength, hardness and elongation. All process steps were simulated in finite element software and the results were analyzed. Hence, this tool was applied for conducting the second stage of experiments. Experiments of second stage were designed and performed with the aim of investigating the effect of tool rotary speed, welding speed and plunge depth on tensile strength, hardness and elongation and selection of optimal factors combination which result in maximal mechanical properties. In the second stage, the central composite design of experiments was used to conduct experiments and to model mechanical properties accurately and find the optimal solution, the neural network was integrated with particle swarm optimization (PSO). Results demonstrated that the neural network with topography of 3-11-3 yields superior prediction. In addition, tool rotary speed of 800 RPM, welding speed of 60 mm/min and plunge depth of 0.2 mm is the optimal combination that leads to maximal tensile strength, hardness and elongation.