Title: Effects of squeeze casting parameters on solidification time based on neural network

Authors: Rong Ji Wang; Wen Fang Tan; Dian-Wu Zhou

Addresses: College of Mechanical and Electrical Engineering, Central South University of Forestry and Technology, Changsha, 410004, China ' College of Mechanical and Electrical Engineering, Central South University of Forestry and Technology, Changsha, 410004, China ' State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body, Hunan University, Changsha 410082, China

Abstract: Based on artificial neural network (ANN) and ProCast software, the effects of different process parameter on the solidification time of squeeze casting hot die steel were investigated, such as interfacial heat transfer coefficient of metal/cavity die (h1), applied pressure (Pa), interfacial heat transfer coefficient of metal/male die (h2), die pre-heat temperature (Td) and pouring temperature (Tp). An ANN model on the relationship between process parameters and solidification time was constructed. The test results show that the ANN model is reasonable and can accurately predict the solidification time and the influence of process parameters on solidification time. The most important parameter is Td, and the secondary is Tp. While Td and Tp increasing within a certain range, the solidification time is found to increase, in contrast, Pa causes the solidification time to decrease. However, h1 and h2 increasing within a certain range, the solidification time is found to decrease. Moreover, the solidification time increases rapidly when h1 and h2 are above their respective critical point. The critical value increases with an increase in mould thickness.

Keywords: squeeze casting; process parameters; artificial neural networks; ANNs; solidification time; hot die steel; applied pressure; interfacial heat transfer coefficient; die pre-heat temperature; pouring temperature.

DOI: 10.1504/IJMPT.2013.056302

International Journal of Materials and Product Technology, 2013 Vol.46 No.2/3, pp.124 - 140

Received: 17 Apr 2012
Accepted: 15 May 2013
Published online: 21 Jun 2014 *

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