Title: Mathematical modelling for prediction of tube hydroforming process using RSM and ANN

Authors: P. Venkateshwar Reddy; B. Veerabhadra Reddy; P. Janaki Ramulu

Addresses: Department of Mechanical Engineering, JNTU Ananthapuramu, A.P., 515002, India ' Department of Mechanical Engineering, G. Pulla Reddy Engineering College (Autonomous), Kurnool, A.P., 518007, India ' School of Mechanical, Chemical and Materials Engineering, Adama Science and Technology University, Adama, Ethiopia

Abstract: Tube hydroforming (THF) is a special manufacturing process used to produce tubular components having applications in aerospace and automotive industries. The present study investigates the effect of process parameters such as coefficient of friction (CF), corner radius (CR) of the die and the axial feeding (AF) of the punch. The bulge ratio and thinning ratio has been evaluated to minimise the defects like bursting, wrinkling and buckling in the tubes. Apart from many parameters, these parameters are chosen to know the effect of each individual parameter on the outcomes namely bulge ratio and thinning ratio. Each factor has varied with three levels and a total of 27 simulations were carried out based on full factorial design. RSM and ANN were applied on the obtained results in order to predict the process parameters effect on the tube hydroforming process. The R-square value of ANN (0.9524 and 0.9517) is much closure to 1 when compared to R-square value of RSM (0.9539 and 0.9509).

Keywords: tube hydroforming; THF; FEM; RSM; artificial neural network; ANN; optimisation.

DOI: 10.1504/IJISE.2020.106848

International Journal of Industrial and Systems Engineering, 2020 Vol.35 No.1, pp.13 - 27

Received: 20 Apr 2018
Accepted: 22 Aug 2018

Published online: 23 Apr 2020 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article