Modelling and prediction of thermal deformation behaviors in a pearlitic steel

Abstract

To predict the thermal deformation behavior in pearlitic steel, the stress–strain data from hot compression tests on a Gleeble-1500D thermo-simulation machine, with the strain rate range of 0.001–1 s−1 and the deformation temperature range of 1223–1373 K, were employed to develop the Arrhenius-type constitutive model and BP neural network model. Then a comparative study was carried out to further evaluate the predictability of the Arrhenius-type constitutive model and BP neural network model. The results show that the correlation coefficient (R) and average absolute relative error (AARE) for the improved Arrhenius-type constitutive equations were found to be 0.98884, 6.41%, respectively, while the R and AARE for BP neural network model were 0.99986, 0.5519%, respectively. The BP neural network model can accurately track the experimental data across the whole hot working process and exhibits better performance than improved Arrhenius-type constitutive equation. Then, the experimental results at 1323 K demonstrate the predictive capability of the developed BP neural network model. Finally, the processing maps suggests that the flow instability tends to occur at low deformation temperature and high strain rate. The optimum processing parameters of the pearlitic steel are in the temperature range of 1223–1293 K and strain rate range of 0.001–0.004 s−1.