Although Artificial Neural Network (ANN) models have been able to predict the flow strength of austenitic steels, its prediction accuracy is largely dependent on the training schemes and model structure because the flow strength varies with deformation conditions and chemical compositions in a very complex way. This is hard to simulate precisely with traditional artificial neural network models. In this work, ANN model strategy was developed to predict the hot strength of a series of austenitic steels with different carbon content deformed under a wide range of conditions. The work hardening coefficient and Zener-Hollomon parameter, developed from phenomenological and empirical models, were incorporated into the model to provide more information in the training data set. The scheme for selecting training data of every independent input was optimised, so that a generalised model could be achieved with less training data. With the technique introduced in this work, the effect of the carbon content and deformation conditions on flow stress, peak strain and peak stress was accurately presented in both the work hardening and dynamic recrystallisation regimes.

purchase the full-text of this paper (price £20)

go to the previous paper
go to the next paper
return to the table of contents
return to the book description
purchase this book (price £75 +P&P)