|Thursday, February 20|
|PS1 Poster Session I & Opening Mixer||
Thu, Feb 20, 5:15 PM - 6:45 PM
A General Statistical Model for Corrosion Pit Depth Analysis for Threshold Data (302739)Todd Graves, Berry Consultants
*Elizabeth Josephine Kelly, Los Alamos National Laboratory
Douglas Kirk Veirs, Los Alamos National Laboratory
Keywords: Generalized Extreme Value (GEV); Pitting corrosion; Pit depth growth, Bayesian modeling,YADAS, 3013 Containers
In this paper we develop a general statistical predictive modeling framework for corrosion pit depth threshold data. This framework incorporates Bayesian analysis with the generalized extreme value distribution. This approach provides a context for comparing results from corrosion pit depth data collected under varying experimental and/or environmental conditions. This work is motivated by the need to determine if pitting can result in the loss of containment of stainless steel containers used to store nuclear material during the fifty-year lifetime of the containers. This framework is used to develop a statistical predictive model for threshold pit depth data from laboratory studies with nuclear material and surrogate material (no radiation) and from a container used to store nuclear material for over five years. Maximum observed pit depths collected from nuclear material containers at times between five and eight years are compared to predictions from the statistical model. An attractive property of the approach is that the presence or absence of pits shallower than a threshold value does not substantially impact predictions for deepest pits.