Microstructural evolution and hot deformation behavior of W-3Re-5HfC alloy

The hot compression behavior of sintered W-3Re-5HfC alloy was investigated using the Gleeble1500D thermo-mechanical simulator at deformation temperatures and strain rates from 1200 °C to 1500 °C and 0.001 s−1 to 1 s−1, respectively. The microstructural evolution and flow behavior were studied. The microstructural observations reveal that the original grains are initially squashed, and then refined, which can be attributed to the occurrence of dynamic recovery and continuous dynamic recrystallization. Moreover, the electron backscattered diffraction analysis demonstrates a typical fibrous texture at low temperatures and high strain rates, which gradually transforms into the plate-like texture with the increase of temperature and decrease of strain rate. Based on the experimental results, arrhenius constitutive equations and an artificial neural network (ANN) model were developed for the characterization and prediction of the high-temperature deformation behavior in the W-3Re-5HfC alloy. The results indicate that the ANN model is more efficient in predicting the hot compressive behavior of W-3Re-5HfC alloy than the constitutive equations. The correlation coefficient and average absolute relative error in constitutive equations are found to be 0.99299 and 4.09%, respectively, whereas the corresponding values in the ANN model are found to be 0.99892 and 0.99%. In order to meet the performance requirements of different high-temperature structure parts, thermal deformation of W-3Re-5HfC alloy is needed to obtain the required microstructure and mechanical properties. ANN model provide theoretical basis for reducing deformation defects, saving process design time and obtaining machining parts with excellent microstructure.