Thermal deformation behavior of as-cast Mg-2Nd alloy: constitutive equation, microstructural analysis, and rheological stress prediction based on support vector regression

The Arrhenius model and support vector regression (SVR) method were utilized to investigate the thermal deformation behavior of Mg-2Nd alloy. Hot compression experiments were conducted on the Mg-2Nd alloy within the deformation range of 150–300 °C and 0.001–1 s⁻¹ using a Gleeble-3500 thermal simulator. The experimental results indicate that with increasing temperature, non-basal slip was activated, leading to a transition in the deformation mechanism of the alloy from low-temperature twinning dominance to slip dominance. Evaluation of the two models were carried out individually using the coefficient of determination (R²) and mean square error (MSE), with the SVR model (R² = 0.99975, MSE = 0.9375) showing better accuracy in predicting rheological stress compared to the Arrhenius model (R² = 0.8955, MSE = 60.75). To validate the generalization ability of the SVR model, rheological stress predictions were made for 100 randomly selected points within the 0–0.3 true strain range, achieving a high prediction accuracy with a correlation coefficient R² of 0.99985.