Development of constitutive relationship model of Ti600 alloy using artificial neural network

• Published in: Computational materials science 2010-05, Vol.48 (3), p.686-691
• Main Authors: Sun, Y., Zeng, W.D., Zhao, Y.Q., Qi, Y.L., Ma, X., Han, Y.F.
• Format: Article
• Language: English
• Subjects: BP neural network; Condensed matter: structure, mechanical and thermal properties; Constitutive relationship; Deformation and plasticity (including yield, ductility, and superplasticity); Exact sciences and technology; Mechanical and acoustical properties of condensed matter; Mechanical properties of solids; Physics; Ti600 alloy
• ISSN: 0927-0256; 1879-0801
• DOI: 10.1016/j.commatsci.2010.03.007

Constitutive equation which reflects the highly non-linear relationship of flow stress as function of strain, strain rate and temperature is a necessary mathematical model that describes basic information of materials deformation and finite element simulation. In this paper, based on the compression experiment data obtained from Gleeble-1500 thermal simulator, the prediction model for the constitutive relationship existed between flow stress and true strain, strain rate and deformation temperature for Ti600 alloy has been developed using back-propagation (BP) neural network method. A comparative evaluation of the traditional regression method and the trained network model was carried out. It was found that the established network model can not only predict flow stress better than the traditional hyperbolic sine constitutive relationship equation but also describe the whole deforming process for Ti600 alloy. Moreover, the ANN model provides a convenient and effective way to establish the constitutive relationship for Ti600 alloy.