Artificial Neural Network Modeling for Predicting the Tensile Strength, Microhardness and Grain Size of Friction Stir Welded Dissimilar AA5083- AA6063 Aluminum Alloys Joints
Keywords:
Friction stir welding, aluminum alloys, artificial neural network, tensile strength, grain size.Abstract
Friction Stir Welding (FSW) has been established as one of the most promising processes for defects free joining of aluminum alloys. In present study artificial neural network (ANN) modeling for predicting the tensile strength, microhardness and average grain size at weld nugget zone of FS welded dissimilar AA5083-O/ AA6063-T6 aluminum alloys joint. Experiments are performed according to L27 OA which is decided based on process parameters and their levels. The developed ANN based model for tensile strength, microhardness and grain size has been found satisfactory with average percentage prediction errors of 1.094%, 1.078 and 1.583%, respectively. Analysis of Variance (ANOVA) is also used to find out the percentage contribution and significance of process parameters for quality characteristics. Based on ANOVA results, tool rotational speed is the significant parameter for tensile strength whereas welding speed is significant parameter for grain size.
References
2. Thomas WM, Nicholas ED. Friction stir welding for the transportation industries. Materials & Design 1997; 18: 269–73.
3. Mishra RS, Ma ZY. Friction stir welding and processing. Materials Science and Engineering R 2005; 50: 1-78.
4. Lakshminarayanan AK, Balasubramanian V. Comparison of RSM with ANN in predicting tensile strength of friction stir welded AA7039 aluminium alloy joints. Transactions of Nonferrous Metals Society of China 2009; 19: 9-18.
5. Jayaraman M, Sivasubramanian R, Balasubramanian V et al. Application of RSM and ANN to predict the tensile strength of
friction stir welded A319 cast aluminium alloy. International Journal of Manufacturing Research 2009; 4: 306-23. 6. Okuyucu H, Kurt A, Arcaklioglu E. Artificial neural network application to the friction stir welding of aluminum plates. Materials & Design 2007; 28: 78–84.
7. Shojaeefard MH, Behnagh RA, Akbari M et al. Modelling and Pareto optimization of mechanical properties of friction stir welded AA7075/AA5083 butt joints using neural network and particle swarm algorithm. Materials & Design 2013; 44: 190–98.
8. ASTM E8 M-04. Standard test method for tension testing of metallic materials. ASTM International, 2006.
9. Montgomery DC. Design and Analysis of Experiments. New York: John Wiley & Sons, 2004.
