Multi-Objective Optimization of Friction Stir Spot Welds of Aluminum Alloy Using Entropy Measurement

Abstract

Surface finish accrued extra-production cost, reduced effective sheet thickness, stir zone galling, undesirable flash-root stress concentration and fatigue cracks are consequences of bulk expulsion of flash during friction stir spot welding of aluminum alloys. This paper attempts to cutback the abovementioned challenges and improves the weld strength (shear failure load) of friction stir spot welded joints of an Al alloy by adopting an integrated Grey relational analysis-entropy measurement method as a multi-objective optimization tool. Shear failure load, and expelled flash properties (pushed out length and thickness) are the three examined quality characteristics of the joint while tool rotational speed (600-1400 rpm), dwell time (3-6 s) and plunge depth (1.5-1.7 mm) are the studied process parameters. The experiment was planned via the use of Taguchi method whereas the entropy measurement method facilitated the identification of the precise weighting values required for the estimation of the unified grey relational grade. The failure load of the joint was maximized while both flash height and pushed-out length were minimized. The optimized shear failure load and flash properties were attained at a parameter setting of 1400 rpm rotational speed, 6 s dwell time and 1.5 mm plunge depth. The tool rotational speed was found to have the most significant effect and percentage contribution on the combined responses with 67.75%, followed by plunge depth (12.88 %) and dwell time (11.94 %) respectively. The validation results confirm the robustness of the entropy measurement-based multi-objective optimization as a tool for improving the quality responses of friction stir spot welds.