Application of Artificial Neural Network to Predict the Hot Flow Behavior of Ti-Nb Microalloyed Steel During Hot Torsion Deformation

In the present study, the hot compressive deformation behavior of Ti-Nb bearing microalloyed steel was investigated in the temperature range of 850–1100 °C and strain rates of 0.01–1 s −1 . Flow curves obtained from hot compression tests were analyzed, and it was demonstrated that dynamic recrystallization is the main softening mechanism during deformation at 900, 1000, and 1100 °C at all strain rates. It was also demonstrated that dynamic recovery occurs during deformation at 800 °C. Based on the obtained results, an artificial neural network was constructed to predict the hot flow softening behavior of Ti-Nb microalloyed steel which is due to the occurrence of dynamic recrystallization and recovery. Results obtained from artificial neural network were compared with the experimental values of flow stress, and a very good correlation was observed between them. This indicates the excellent capability of the developed artificial neural network to predict the flow softening due to complex microstructural evolutions accompanying dynamic recrystallization and dynamic recovery.