A probability distribution model for stress concentration factors in multi-planar tubular DKT-joints of steel offshore structures

Abstract

The stress concentration factor (SCF) is one of the most important parameters in the fatigue reliability analysis of steel offshore structures. This parameter exhibits considerable scatter which calls for greater emphasis in accurately computing the SCFs for predicting the fatigue life of these structures. As far as the authors are aware, no comprehensive research has been carried out on the probability distribution of SCFs in tubular joints, especially multi-planar ones, which cover the majority of practical applications. What has been used so far as the probability distribution of SCFs in the reliability analysis of offshore structures is mainly based on assumptions and limited observations, especially in terms of distribution parameters. In the present paper, results of performing FE analysis on 81 steel multi-planar tubular DKT-joints have been used to propose a probability distribution model for SCFs along the weld toe of the central brace under axial loads. Based on the multi-planar DKT-joint FE models which are verified against experimental results and the predictions of Lloyd's Register (LR) equations, a set of SCF sample databases was constructed. 15 different probability density functions (PDFs) were fitted to the relative frequency histograms of the SCF samples. The maximum likelihood method was used to determine the parameters of each distribution. Standard goodness-of-fit tests led to the conclusion that the Inverse Gaussian distribution is the best probability model for the peripheral distribution of SCFs along the weld toe and the Birnbaum–Saunders distribution is the best one for the maximum value of weld toe SCF.