Modeling of thermal deformation behavior near γ′ solvus in a Ni-based powder metallurgy superalloy

We used two different methods to model the thermal deformation behavior of a Ni-based powder metallurgy superalloy at large strains (> 0.6) near the gamma' solvus. We concluded that the Artificial Neural Network model, rather than the strain-compensated Arrhenius model, describes thermal deformation behavior accurately under large strains during hot extrusion. Artificial neural network, a data-driven machine learning approach, is more suitable for predicting unknown deformation behavior in extreme conditions by data learning based on a known experimental dataset.