Process parameters settings of friction stir welding using multi-response optimization for aluminum alloys

Abstract

The defense and aerospace industries utilize the AA2050, an aluminum-lithium alloy of the third generation. Principal component analysis (PCA) and Taguchi-grey relational analysis (GRA) are utilized in this study to enhance the process constraints for friction stir welding (FSW). The quality of the weld can be influenced by various process factors such as the speed at which the weld is traversed, the speed at which the tool is rotated, the angle at which the tool is tilted, the diameter of the shoulder, and the shape of the tool pin. The combination of PCA and the well-known Taguchi-GRA approach enables the objective estimation of response weights. Here, a Taguchi L16 orthogonal array was designed, and sixteen tests were carried out using it. These experiments were conducted with various assessments including tensile strength, yield strength, elongation percentage, hardness of the weld zone, bending load, etc. Through the utilization of Taguchi-GRA-PCA analysis, the most suitable process parameters were identified as a traverse speed (TS) of 160 mm/min, a rotating speed (RS) of 900 rpm, a tilt angle (TA) of 2 degrees, a shoulder diameter (SD) of 16 mm, and with a straight square tool pin profile (TPP). ANOVA revealed the relevance of all five characteristics, with rotational speed being the most influential, accounting for 43.56% of the entire result. A confirmation experiment done under ideal conditions revealed a significant improvement in total weld quality of 19.06%.