Multi-Objective Optimization of Friction Stir Welding of Aluminium Alloy 6082-T6 Using hybrid Taguchi-Grey Relation Analysis- ANN Method

Abstract This paper aimed to multi-response optimization of friction stir welding (FSW) process for an optimal parametric combination to yield favourable tensile strength and impact strength using the Taguchi based Grey Relational Analysis (GRA) and the Artificial Neural Network (ANN). The objective functions have been selected in relation with FSW parameters; tool rotation speed, welding speed and tilt angle for newly developed composite pin profile tool. The experiments were planned using Taguchi’s L27 orthogonal array for three different tools. The optimal tool and process parameters for friction stir welding were determined by simulating parameters using a well-trained ANN model with the help of grey relational grade obtained from the GRA. This study has also showed the application feasibility of the ANN-Grey relation analysis in combination with Taguchi technique for continuous improvement in welding quality.