Prediction of nickel-base superalloys’ rheological behaviour under hot forging conditions using artificial neural networks

In this paper neural networks are utilised to represent the rheological behaviour of the Nickel-base superalloy Nimonic 80A under deformation conditions approximating thermo-mechanical cycles of industrial hot forging operations. A feed-forward back-propagation neural network has been trained and tested on rheological data, obtained from hot compression experiments, performed under single- and multi-step deformation conditions, both at constant and varying strain rate. The good agreement between experimental and calculated flow curves shows that a properly trained neural network can be successfully employed in representing material response to hot forging cycles.