To read this content please select one of the options below:

Artificial neural network modeling of creep behavior in a rotating composite disc

V.K. Gupta (Department of Mechanical Engineering, University College of Engineering, Punjabi University, Patiala, India)
N. Kwatra (Department of Civil Engineering, TIET, Patiala, India)
S. Ray (Department of Metallurgical and Materials Engineering, IIT Roorkee, Roorkee, India)

Engineering Computations

ISSN: 0264-4401

Article publication date: 13 March 2007

340

Abstract

Purpose

This paper aims to explore the capabilities of artificial neural network (ANN) for predicting the creep response of a rotating Al‐SiCP composite disc operating at elevated temperature.

Design/methodology/approach

Mathematical modeling of the steady state creep behavior, as described by Sherby's law, of a rotating disc made of isotropic aluminium‐silicon carbide particulate composite has been carried out. The creep response has been calculated for various combinations of particle size, particle content and temperature by extracting creep parameters from the limited experimental creep data available on similar material. The results thus obtained are used to train the ANN based on back propagation learning algorithm with particle size, particle content and temperature as input and stress and strain rates as output parameters. The trained network is used to predict the stresses and strain rates in the disc for the data set not covered in the training of network. The predictions obtained from the ANN model have been compared with the corresponding analytical values.

Findings

A nice agreement between the ANN predicted and analytical values of the creep stresses and strain rates has been observed.

Originality/value

ANN can be used as a reliable tool for investigating the effect of operating temperature and, reinforcement‐size and ‐content, on the creep behavior of a rotating composite disc to reach at optimum design code.

Keywords

Citation

Gupta, V.K., Kwatra, N. and Ray, S. (2007), "Artificial neural network modeling of creep behavior in a rotating composite disc", Engineering Computations, Vol. 24 No. 2, pp. 151-164. https://doi.org/10.1108/02644400710729545

Publisher

:

Emerald Group Publishing Limited

Copyright © 2007, Emerald Group Publishing Limited

Related articles