Research on hot deformation, dynamic recrystallization and numerical simulation of Fe–28Mn–10Al–1.5C–0.6V steel

Peng Wan,Pengfei Gao,Meng Wang,Zhengzhi Zhao

DOI: https://doi.org/10.1007/s10853-022-07692-w

IF: 4.5

2022-09-14

Journal of Materials Science

Abstract:The hot deformation behaviors of Fe–28Mn–10Al–1.5C–0.6V steel at T = 950–1150 °C and = 0.01–10 s −1 were studied using the Gleeble-3500 thermo-mechanical simulator. The constitutive model of Fe–28Mn–10Al–1.5C–0.6V steel was established by an artificial neural network (ANN) based on the experimental flow stress data. The critical conditions of dynamic recrystallization (DRX) of the steel under experimental conditions were studied by using both the Najafizadeh–Jonas work hardening rate model and the Cingara–McQueen rheological stress model. The true stress–true strain curve of the steel was analyzed, the work hardening rate was calculated, and the critical strain model was established. At the same time, the relevant material parameters were calculated by linear regression, and the DRX kinetics model of JMAK was also established. This model then was combined with the finite element software DEFORM-3D to simulate the DRX behavior of the steel during hot deformation. These results indicated that the rheological stress of Fe–28Mn–10Al–1.5C–0.6V steel at low shows higher temperature sensitivity and that high T and low are conducive to the occurrence of DRX. The average relative error of the ANN constitutive model is 2.40%, which has high prediction accuracy. The dynamic recrystallization conditions determined by the above models are reliable and conform to the experimental rules. By verifying the microstructure of the steel after thermal compression, the prediction error of the finite element model to the DRX volume fraction is mostly within 10%. Thus, the model has a good predictive ability and provides an effective tool for predicting plastic deformation and microstructure in industrial production.

materials science, multidisciplinary