SCC ANALYSIS OF AUSTENITIC STAINLESS-STEELS IN CHLORIDE-BEARING WATER BY NEURAL NETWORK TECHNIQUES

Although case histories, reflecting engineers' experiences with chloride-induced stress corrosion cracking of austenitic stainless steels in water, are multifold, this source of information is virtually untouched since suitable technologies for the processing of these data have only recently evolved. In this paper, it will be shown how neural networks, which are capable of extracting regularities from these case histories, can predict the SCC risk of austenitic stainless steels in chloride-bearing water in an automated way. The parameters considered in this study are the temperature, the chloride concentration, and the oxygen concentration.