Study of Thermal Compression Constitutive Relation for 5182-Sc-Zr Alloy Based on Arrhenius-Type and ANN Model

Hot compression experiments were performed on alloy 5182 with small additions of Sc and Zr. The containing Zr is critical for expanding the alloy’s range applications, a thorough understanding its thermal processing behavior great importance to avoid defects. Alloy microstructure, including grain structures Al3(ScxZr1−x) dispersoids analyzed by EBSD TEM. Stable flow stresses observed below strain rate 1 s−1 Sc-Zr alloy. results constitutive models, without strain−compensation, artificial neural network (ANN) used compare experimental results. dispersoid data was introduced into ANN model as nonlinear influence factor. Addition information input improved accuracy practicality in predicting deformation squared correlation coefficients prediction reached 0.99.