Prediction of the friction stir welding temperature and ultimate tensile stress for 6061 T6 aluminum

Friction stir welding (FSW) is a solid-state welding process that eliminates the need for filler metal, offering advantages in joining materials that are difficult to weld using traditional methods. This paper focuses on the thermomechanical behavior of FSW for 6061-T6 aluminum alloy, employing numerical simulations, response surface methodology (RSM), and artificial neural networks (ANNs) to predict key welding outcomes. The model successfully predicts the welding temperature with a peak of 509.78°C and the ultimate tensile strength (UTS) reaching up to 278 MPa. The influence of parameters such as tool profile, welding speed, and rotational speed is quantified, showing that rotational speed contributes 66.3% to UTS variability while welding speed accounts for 63% of the temperature variation. The ANN predictions align closely with experimental and simulation results, with less than 5% deviations. This integrated approach demonstrates a robust framework for optimizing FSW processes, providing actionable insights for improving weld quality in high-strength aluminum alloys.