Abstract

The integrity of the underground radioactive waste tanks at the Hanford site is managed through a variety of engineering assessments, laboratory testing, and field monitoring. Much of this assessment is deterministic in nature. However, the many uncertainties surrounding the historical tank conditions, waste chemistries, and the relationship between waste chemistry and corrosion demand a probabilistic approach to tank integrity. In this paper, several approaches are identified for such probabilistic assessment. The paper is conceptual in nature, with some illustrative examples. It does not attempt to predict the probability of failure of any given tank. Recommendations are made for improving the probabilistic assessments and the experimental data needed.

Keywords:
corrosion inhibition, machine learning, flowline corrosion, materials and corrosion, oilfield chemistry, riser corrosion, classification, classifier, prediction, scc
Subjects:
Production Chemistry, Metallurgy and Biology, Pipelines, Flowlines and Risers, Information Management and Systems, Materials and corrosion, Well Integrity, Subsurface corrosion (tubing, casing, completion equipment, conductor), Corrosion inhibition and management (including H2S and CO2)
This content is only available via PDF.
2015. NACE International
You can access this article if you purchase or spend a download.