Friction Stir Welding of Aluminum Using a Multi-Objective Optimization Approach Based on Both Taguchi Method and Grey Relational Analysis

This work deals with the use of a multi-objective optimization method using a hybrid statistic algorithm to improve the Friction Stir Welding of aluminum alloy AA2195-T8. The hybrid approach combines the Taguchi Method with the Relational Grey Analysis technic. In order to optimize the Friction Stir Welding process, the axial force, the rotational tool velocity, the welding velocity and the shoulder diameter were considered as input parameters while the heat input, the maximal temperature value and the Heat Affected Zone length were chosen as output parameters. In this method, the minimization of the heat input, the HAZ length and the temperature value in the stir zone is the main goal. In the process of improving the aluminum welding by FSW, the axial force is the most influential parameter with a contribution of 52.4%, followed by the rotational tool velocity with 37.4%, then the welding velocity with 6.3% and finally the tool diameter with a contribution of 3.6%. The obtained results from the application of three-dimensional numerical thermal model have confirmed the effectiveness and the robustness of the used optimization approach.