Modelling of Cr2N Age-precipitation in High Nitrogen Stainless Steels by Neural Networks

Abstract

It is very important to study the incubation time of alloys as it has direct relation to precipitation kinetics and affects morphology of structure. In spite of many efforts of scientists to develop methods to find incubation time, it is still difficult to measure experimentally. In addition, there is need to develop an approach to analyze existing data for many steels. On the other hand, high nitrogen steels have received a lot of attention for their unique properties, however, there is no clear model developed to predict its precipitation kinetics. In present work, Cr2N age-precipitation in high nitrogen austenitic steels is simulated using neural network (NN) analysis. The feed forward neural network with a back propagation algorithm was built to obtain the constitutive relation of Cr2N age precipitation with alloying elements and aging temperature. The simulated results show that the NN model can correctly reproduce the precipitation behavior of the steel. An analysis of model predictions and experimental data is presented.