Prediction of crack growth rate in Type 304 stainless steel using artificial neural networks and the coupled environment fracture model

A crack growth rate (CGR) database has been developed to train an artificial neural network (ANN). The trained ANN and the extended coupled environment fracture model (CEFM) were used to predict the CGR in Type 304SS as a function of each of the principal variables of the system. The ANN revealed the underlying relationships that map the dependencies of the CGR on the various input independent variables. A sensitivity analysis revealed that IGSCC in sensitized Type 304SS in high temperature aqueous environments is primarily electrochemical in character. Comparison between the ANN-predicted CGR and CEFM-predicted CGR reveal good agreement.