The Use of Artificial Neural Networks in Predicting Fatigue Life of Friction Stir Welded Lap Joints of AA 5754

Abstract

Friction stir welding (FSW) is currently being widely investigated in the aerospace industry for joining high strength Al-alloys that are difficult to weld using conventional fusion techniques. The quality of the process mainly depends on the pin diameter, pin height, tool rotation and traverse speed. In the present work, an artificial neural network (ANN) method was used for modeling and predicting the fatigue life of friction stir welded lap joints of AA5754 aluminum alloy. The ANN model of fatigue life is developed with the welding parameters such as pin diameter, pin height, tool rotation, traverse speed and with the weld property of fatigue strength. The experimental results were trained in an ANN program and the results were compared with experimental values. It is observed that the experimental results coincided with ANN results.