论文信息 - Comparison of ANN and Regression Modeling for Predicting the Responses of Friction Stir Welded Dissimilar AA5083-AA6063 Aluminum Alloys Joint

Comparison of ANN and Regression Modeling for Predicting the Responses of Friction Stir Welded Dissimilar AA5083-AA6063 Aluminum Alloys Joint

Joining of dissimilar aluminum alloys are widely used in automobile, aerospace and shipbuilding industries. Friction Stir Welding (FSW) has been established as one of the most promising processes to defects free joining of aluminum alloys. The aim of present work is to compare the predicted results of FS welded joint through Artificial Neural Network (ANN) modeling and regression modeling. Three responses tensile strength, average microhardness at weld nugget zone (WNZ) and average grain size at WNZ have been selected. The predicted values by ANN modeling and regression modeling of TS, MH and GS values have been found close to the experimental values. The overall average percentage prediction error of ANN model is small as compared to regression model.

Saurabh Gupta | Rajneesh Kumar | K. N. Pandey | Rajneesh Kumar | S. Gupta

[1] Mostafa Akbari,et al. Modelling and Pareto optimization of mechanical properties of friction stir welded AA7075/AA5083 butt joints using neural network and particle swarm algorithm , 2013 .

[2] Anthony P. Reynolds,et al. Visualisation of material flow in autogenous friction stir welds , 2000 .

[3] W. Thomas,et al. Friction stir welding for the transportation industries , 1997 .

[4] Erol Arcaklioğlu,et al. Artificial neural network application to the friction stir welding of aluminum plates , 2007 .

[5] Joseph R. Davis. Properties and selection : nonferrous alloys and special-purpose materials , 1990 .

[6] V. Balasubramanian,et al. Comparison of RSM with ANN in predicting tensile strength of friction stir welded AA7039 aluminium alloy joints , 2009 .