Analysis of S-N data for new and corroded mooring chains at varying mean load levels using a hierarchical model

Abstract

Results from full scale fatigue tests of offshore mooring chains are considered. The data set includes new and used chains, tested at a range of mean load levels. The used chains have been retrieved after operation offshore and include samples with varying surface conditions, ranging from as-new to heavily corroded. Based on a parameterized S-N curve intercept parameter, the effects of mean load and chain condition are estimated empirically by regression analysis. A hierarchical model is used, to properly account for and quantify correlations within subsets of the data. Results show that the mean load and corrosion effects are both  significant. Differences in the fatigue performance of new versus used chains are quantified and discussed. The paper also  includes an assessment of the impact from correlations between chain links on the fatigue capacity and reliability of a chain  segment.