Modeling Constitutive Relationship of Ti-555211 Alloy by Artificial Neural Network during High-Temperature Deformation

Abstract:

Using experimental data gained from hot compression tests in the temperature range of 750~950 °C and strain rate range of 0.001~1 s-1, the constitutive relationship of Ti-555211 titanium alloy was investigated based on the back propagation artificial neural network constitutive model (ANN model). The capability of the model was measured by the average absolute relative error (AARE), and correlation coefficient (R). The simulated values were compared with experimental values. The results show that the R and AARE for the ANN model are 0.99938 and 1.60%, respectively, indicating that the hot deformation behavior of Ti-555211 titanium alloy can be predicted by the ANN model efficiently and accurately. Furthermore, the back propagation artificial neural network model is a more efficient quantitative way to predict the deformation behavior of the Ti-555211 titanium alloy compared to the mathematical equation. The results show that the peak stress of the alloy decreases with increasing of temperature and decreasing of strain rate, and the phenomenon of discontinuous yielding is more obvious with the increase of deformation temperature and strain rate. The flow curve characteristics under different deformation parameters show obvious differences.