Constitutive modeling and workability characterization of pre-deformed AZ31 magnesium alloy during hot shear-compression deformation
Pre-deformation is a cost-efficient method to control the microstructure and mechanical properties of wrought magnesium (Mg) alloys. It introduces twins that promote dynamic recrystallization during hot deformation, thereby improving hot workability and weakening texture. However, uniaxial hot compr...
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-07-01
|
| Series: | Journal of Materials Research and Technology |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2238785425017855 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Pre-deformation is a cost-efficient method to control the microstructure and mechanical properties of wrought magnesium (Mg) alloys. It introduces twins that promote dynamic recrystallization during hot deformation, thereby improving hot workability and weakening texture. However, uniaxial hot compression tests fail to reflect the deformation behavior of Mg alloys under hot shear-compression process. In this study, shear-compression samples (SCS) were employed, and finite element simulations were combined with physical experiments to determine the Mises coefficients in the equivalent stress-strain conversion equation. The deformation characteristics, flow behavior and microstructure/texture evolution mechanisms of pre-deformed AZ31 alloy were systematically investigated under varying process parameters. A genetic algorithm-optimized artificial neural network (GA-ANN) constitutive model was developed using machine learning methods, and hot processing maps were established based on this model. The results show that the proposed method can accurately determine the Mises coefficients, describing the mechanical response of SCS at high temperatures. The shear-compression state causes the material's flow behavior to exhibit a distinct single-peak characteristic, with peak stress and peak strain increasing and then decreasing with the amount of pre-deformation.The GA-ANN model demonstrates good prediction accuracy and generalization ability for the hot shear-compression behavior of pre-deformed AZ31 alloy. The developed hot processing map can precisely predict microstructure evolution. The optimal pre-deformation amount is determined to be 2 %, with a recommended hot processing window defined as 300–400 °C and 0.07–7 s−1. |
|---|---|
| ISSN: | 2238-7854 |