Sustainability-based optimization of dissimilar friction stir welding parameters in terms of energy saving, product quality, and cost-effectiveness

The friction stir welding (FSW) process is a prominent approach to fabricate high-quality welds for materials having low melting temperatures. Welding quality criteria have been extensively considered in former publications. The purpose of the current work is to optimize process parameters, including the tool rotational speed (N), travel speed (V), plunging depth (P), and tilting angle (A) for decreasing the energy consumption in the welding time (EW) and improving the ultimate tensile strength (UTS) as well as and percent elongation (PEL) for the FSW operation of dissimilar aluminum alloys. The adaptive neuro-based fuzzy inference system (ANFIS) approach is utilized to develop the FSW responses in terms of optimizing inputs, while a novel model is developed to compute the production cost (PC). The grey relational analysis (GRA) is applied to calculate the weight of each response. The vibration and communication particle swarm optimization (VCPSO) algorithm and combined compromise solution (CCS) are applied to produce feasible solutions and determine the best optimal point. The obtained outcomes presented that the optimum outcomes of the N, V, P, and A are 1500 RPM, 56 mm/min, 0.98 mm, and 2 deg., respectively, while the EW, UTS, and PEL are enhanced by 13.3, 7.4, and 55.7 at the optimal solution. The optimal ANFIS models were trustworthy and ensure accurate predictions. The developed method using the ANFIS, VCPSO, and CCS could be effectively utilized to determine the optimal outcomes instead of the trial–error and/or human experience. The observed findings provided efficient information, which could help operators to select the optimal FSW parameters and enhance the welding responses. © 2022, The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature.