Thermal forming properties of a Cr‐Mn‐Si‐Ni alloyed naval steel under different forming conditions by different constitutive models

Abstract A series of thermal compression tests on a Cr‐Mn‐Si‐Ni alloyed naval steel were carried out at different strain rates (0.0005–0.0100 s −1 ) at different temperatures (1023–1173 K). Based on the friction‐corrected data obtained from the compression tests, strain‐compensated Arrhenius‐type constitutive (SCAC) and backpropagation artificial neural network (BP‐ANN) models with the optimized structure of the Cr‐Mn‐Si‐Ni alloyed naval steel were established. The optimized BP‐ANN model, where the operation time and overfitting of BP‐ANN were shortened and avoided, respectively, exhibited improved predictive performance. The two models were assessed further in terms of the correlation coefficient ( R ), average absolute relative error, and root mean square error. The results validated that the optimized BP‐ANN model predicted the flow behavior of the Cr‐Mn‐Si‐Ni alloyed naval steel better than the SCAC model. The effect of the forming temperature and strain rate on the microstructural evolution behavior of the naval steel during thermoplastic deformation was investigated through the electron backscatter diffraction analysis of the compressed samples. It was observed that the dynamic recrystallization of the naval steel was promoted by an increase in the forming temperature and a decrease in the strain rate during thermoplastic deformation.