Application of Neural Network for the Prediction of Tensile Properties of Friction Stir Welded Composites

Shettigar, Arun Kumar and Prabhu, Subramanya R B and Malghan, Rashmi L and Rao, Srikantha and Herbert, Mervin A (2016) Application of Neural Network for the Prediction of Tensile Properties of Friction Stir Welded Composites. In: Asia Conference on Mechanical and Materials Engineering, 14/07/2016, Malaysia.

[img] PDF
633.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

In this paper, an attempt has been made to apply the neural network (NN) techniques to predict the mechanical properties of friction stir welded composite materials. Nowadays, friction striwelding of composites are predominatally used in aerospace, automobile and shipbuilding applications. The welding process parameters like rotational speed, welding speed, tool pin profile and type of material play a foremost role in determining the weld strength of the base material. An error back propagation algorithm based model is developed to map the input and output relation of friction stir welded composite material. The proposed model is able to predict the joint strength with minimum error.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: Neural Network, Friction Stir Welding, Composite, Error Back Propagation
Subjects: Engineering > MIT Manipal > Mechanical and Manufacturing
Engineering > MIT Manipal > Mechatronics
Depositing User: MIT Library
Date Deposited: 13 Jan 2017 13:34
Last Modified: 13 Jan 2017 13:34
URI: http://eprints.manipal.edu/id/eprint/148091

Actions (login required)

View Item View Item