Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Prediction of IGSCC in Type 304 SS using an artificial neural network

Book ·
OSTI ID:70096
;  [1]
  1. Pennsylvania State Univ., University Park, PA (United States). Dept. of Engineering Science and Mechanics

Intergranular stress corrosion cracking (IGSCC) of recirculating piping in boiling water reactors (BWRs) has been a major operating problem world-wide. An artificial neural network (ANN) has been developed to describe intergranular stress corrosion cracking in sensitized Type 304SS in high temperature aqueous solutions. The ANN predictions of crack growth rate (CGR) versus oxygen concentration, flow velocity, stress intensity, hydrogen concentration, and ECP are compared with the predictions of the deterministic Coupled Environment Fracture Model (CEFM). The predictions of these two approaches, which represent the extremes in the spectrum of predictive technologies, are generally in good accord, except that the CGRs obtained from the ANN are up to a factor of three higher than those predicted by the CEFM under some conditions. However, this difference is within the uncertainty in the experiment data used to train the ANN. 39 refs.

OSTI ID:
70096
Report Number(s):
CONF-940222--
Country of Publication:
United States
Language:
English