Abstract

Despite the numerous researches in Stress Corrosion Cracking (SCC) risk of austenitic stainless steels in aqueous chloride solution, no formulation or reliable method for prediction of time to failure as a result of SCC has yet been defined. In this paper, the capability of artificial neural network for estimation of the time to failure for SCC of 304 stainless steel in aqueous chloride solution together with sensitivity analysis has been expressed. The output results showed that artificial neural network can predict the time to failure for about 74% of the variance of SCC experimental data. Furthermore, the sensitivity analysis also demonstrated the effects of input parameters (Temperature, Applied stress and Cr concentration) on SCC of 304 stainless steel in aqueous chloride solutions.