Prediction of primary water stress corrosion crack growth rates in Alloy 600 using artificial neural networks. (March 2015)

• Record Type: Journal Article
• Authors: Shi, Jiangbo; Wang, Jihui; Macdonald, Digby D.
• Is Part Of: Corrosion science. Volume 92(2015)
• Page Start: 217
• Page End: 227
• Publication Date: 2015-03
• Subjects: A. Alloy -- B. Modeling studies -- C. Stress corrosion; Corrosion and anti-corrosives -- Periodicals; 620.11223
• DOI: 10.1016/j.corsci.2014.12.007
• Languages: English
• ISSNs: 0010-938X

Abstract

Highlights: An IGSCC crack growth rate database for Alloy 600 has been developed. An ANN model has been trained to predict the crack growth rate. The characteristic of IGSCC has been determined by sensitivity analyses. Abstract: After reviewing of the data for primary water stress corrosion cracking (PWSCC) for Alloy 600 in the literature, a crack growth rate (CGR) database was assembled, and an ANN model was developed and trained upon the data, in order to model PWSCC in Alloy 600. The dependence of PWSCC CGR on each of the principal independent variables of the system has been predicted. Sensitivity analyses were conducted via "fuzzy logic" and the importance of each variable was analyzed and show that IGSCC in Alloy 600 is primarily mechanical in character with the electrochemistry being a significant contributor.