Authors: A.H. Gandhi, P.P. Gohil, H.K. Raval
Addresses: Mechanical Engineering Department, C.K. Pithawalla College of Engineering and Technology, via. Magdalla Port, Surat-Dumas Road, Surat 395007, Gujarat, India. ' Mechanical Engineering Department, Charotar Institute of Technology, Changa, Taluka: Petlad, Anand 388421, Gujarat, India. ' Mechanical Engineering Department, Sardar Vallabhbhai National Institute of Technology, Ichchhanath, Surat 395007, Gujarat, India
Abstract: Three-roller bending aimed at plastic bending of plates into cylinder. Accurate prediction of springback is important for controlling product dimensions. Due to inconsistency in springback, online tool for controlling the final product dimensions are more reliable. Presented work discusses the development of artificial neural network for prediction of springback and top roller position for three-roller bending. Effect of network parameters and data normalisation on network performance was studied. Statistical analysis was performed to check the reliability of the model. Artificial neural network based predictions can be reliably used in the production environment for improving the accuracy and efficiency. [Received 12 January 2008; Revised 26 July 2008; Accepted 4 November 2008]
Keywords: three-roller bending; springback prediction; real material behaviour; ANNs; artificial neural networks; plastic bending; cylinders.
International Journal of Manufacturing Research, 2009 Vol.4 No.3, pp.265 - 280
Available online: 19 Jun 2009 *Full-text access for editors Access for subscribers Purchase this article Comment on this article