基于应变补偿和PSO-BP神经网络的Ti-2.7Cu合金本构关系
摘要
The isothermal compression tests of Ti-2.7Cu alloy were tested to study the hot deformation behavior in temperature range of 740-890℃ and strain rate range of 0.001-10s<sup>-1</sup> on a Gleeble-3500 thermomechanical simulator. Constitutive model based on strain compensation was established by the Arrhenius hyperbolic sine function equation, and set up a constitutive equation for PSO-BP neural network. The results show that the flow stress is more sensitive to deformation temperature and strain rate, the flow stress is decreased with the increase of deformation temperature and decrease of strain rate; the flow stress curves present stable states in high temperature and low strain rate. For a constitutive equation based on strain compensation, the data points with the predicted error less than 15% account for 85.28% of all test data by error calculation; and for the constitutive equation based on PSO-BP neural network, the data points with the predicted error less than 15% account for 96.67% of all test data. PSO-BP neural network model has higher accuracy, it can better predict the flow stress of Ti-2.7Cu at elevated temperature.