Abstract
– Friction stir welding is an innovative method for welding technology and it provides several benefits in terms of joint strength, especially over traditional welding. This work focuses on the optimization of the multi-response parameters of Friction Stir Welding (FSW). A Taguchi based Gray relational and Hybrid GRA-ANN process has been introduced in order to achieve the appropriate tension strength, elongation, hardness and temperature (top, mid and bottom surfaces) of the joint. In this study, tool rotational speed, travel speed, and shape of the tool have been the input parameters. The testing has been planned to utilize Taguchi's L18 orthogonal array. The findings demonstrate that the pin shape has the greatest influence on the performance and reliability of the tensile strength of the joint. For the welding application, the convex pin has been better than a simple one. At a tool rotating speed of 1,400 rpm and at a travel speed of 10 mm/min, the optimal value for the FSW joint has been obtained. The findings have revealed that the hybrid Taguchi-Grey Relationship Analysis – ANN approach provides the highest grey relation grade of 0.903787 with a convex pin, which is 4.15 percent more than the conventional Taguchi based grey relational analysis.
Original language | English |
---|---|
Pages (from-to) | 32-43 |
Number of pages | 12 |
Journal | International Review on Modelling and Simulations |
Volume | 14 |
Issue number | 1 |
DOIs | |
Publication status | Published - Feb 2021 |
Keywords
- Artificial Neural Networks
- Friction Stir Welding
- Grey Relational Analysis
- Hybrid Grey Relational Analysis
ASJC Scopus subject areas
- Modelling and Simulation
- General Chemical Engineering
- Mechanical Engineering
- Logic
- Discrete Mathematics and Combinatorics
- Electrical and Electronic Engineering
- Applied Mathematics