Prediction of Responses in FSW Processed Hybrid Composites Using Soft Computing Technique

Abstract

In this research a proper step taken to predict friction stir welding process parameters for the aluminium hybrid composites. In FSW Weld quality is influenced by welding input parameters. The welding parameters taken for this work are axial load, tool profile, spindle speed and material composition. In the present work different combination of aluminium hybrid composite of 6 mm thick is attempted to weld using vertical milling machine. By taking the input parameters such as axial load, tool profile, spindle speed and material composition whereas the response of the model tensile strength and hardness is calculated. The ANN was trained with experimental data. Testing of the ANN is carried out using experimental data not used during training. The result shows the comparison of experimental values of tensile strength and hardness and predicted value using ANN. This indicates that the developed neural network can be used as alternative way for calculating mechanical properties for given process.