Staff Members - Benha University
Comparison of RSM and RA with ANN in Predicting Mechanical Properties of Friction Stir Welded Aluminum Alloy Pipes,
Engineering and Technology in India • 2017
Publication Information
Authors
Ibrahim Sabry, Ahmed. M. El-Kassas and A.M. Khourshid
Keywords
Friction stir welding, Aluminium pipe, Regression analysis, Response surface methodology, Artificial
neural network
Journal
Engineering and Technology in India
Publisher
Not Available
Volume
1
Issue
1
Pages
1-14
publication.type
International
Paper Link
Open Link
Supplementary Materials
Not Available
Abstract
Aluminum can’t successfully be arc welded in an air environment, due to the affinity for oxygen. If fusion welded in normal
atmosphere oxidization readily happens and this outcome in both slag inclusion and porosity in the weld, greatly reducing its
mechanical properties. This work presents a systematic approach to develop the suggestion model by three (ANN), response
surface methodology (RSM) and regression analysis (RA) for predicting the ultimate tensile strength, percentage of elongation
and hardness of 6061 aluminum alloy which is widely used in automotive, aircraft and defense industries by incorporating
(FSW) friction stir welding process parameter such as tool rotation speed, welding speed and material thickness. The results
obtained through regression analysis and response surface methodology were compared with those through artificial neural
networks.
atmosphere oxidization readily happens and this outcome in both slag inclusion and porosity in the weld, greatly reducing its
mechanical properties. This work presents a systematic approach to develop the suggestion model by three (ANN), response
surface methodology (RSM) and regression analysis (RA) for predicting the ultimate tensile strength, percentage of elongation
and hardness of 6061 aluminum alloy which is widely used in automotive, aircraft and defense industries by incorporating
(FSW) friction stir welding process parameter such as tool rotation speed, welding speed and material thickness. The results
obtained through regression analysis and response surface methodology were compared with those through artificial neural
networks.