A study of microstructure and mechanical properties of friction stir welded joint of Al-Ce-Si-Mg aluminium alloy plates and optimization cum prediction techniques using Taguchi and ANN

Abstract

In the current research work, the effect of friction stir welding (FSW) on joint strength and evolution of microstructure at the weld zone of the friction stir welded of Al-Ce-Si-Mg aluminium alloy (Al10Mg8Ce3.5Si and Al5Mg8Ce3.5Si) is studied. The experimentations demonstrated that the speed of tool revolution, contour of tool pin and tool movement rate have influence on the quality of the FSW joint of the alloy. It was noticed that the size of grains at nugget zone (NZ) depends upon the speed of tool spin, speed, of tool feed, tool pin contour and composition of the aluminium alloy. The grain size at the bottom of the NZ was discovered to be reducing when compared to that at the top of the NZ. It was also noted that highest hardness was reached at NZ. Lowest hardness was found at heat affected zone (HAZ) and most of the tensile specimens fractured at HAZ. The Taguchi orthogonal array-based design has demonstrated that the tool pin contour had the greatest influence in enhancing the joint strength, followed by speed of tool feed, material composition and speed of tool spin. A speed of revolution of tool of 1000 rpm, tool movement rate of 20 mm/min, triangular contour pin (TCP) tool and Al10Mg8Ce3.5Si aluminium alloy were concluded as the optimum variables of the process. The percentage contribution of each one of these input process parameters on optimum output quality characteristics was also found out and noticed to be lying well within the confidence interval of 95% suggested by the Taguchi design. The current study has shown that the prediction results with artificial neural network (ANN) are better compared to those predicted using statistical methods like Taguchi Techniques.