Elsevier

Structures

Volume 20, August 2019, Pages 779-793
Structures

Stress concentration factors in FRP-strengthened offshore steel tubular T-joints under various brace loadings

https://doi.org/10.1016/j.istruc.2019.07.004Get rights and content

Abstract

The stress concentration factors (SCF) in FRP strengthened tubular T-joints subjected to brace axial loading, in-plane and out-of-plane bending moments were investigated. The numerical analyses were performed using ABAQUS Finite Element software package. The benchmark joints were validated against the Lloyd's Register and API equations together with the experimental results. Six different types of FRP materials such as Glass/Vinyl ester, Glass/Epoxy (Scotch ply 1002), S-Glass/Epoxy, Aramid/Epoxy (Kevlar 49/Epoxy), Carbon/Epoxy (T300-5208) and Carbon/Epoxy (AS/3501) were used as strengthening material in order to investigate the SCF values on the chord member of the tubular T-joints. Results derived from analyses are promising and show that the FRP strengthening method could be considered as an effective method to reduce the SCFs and consequently extend the fatigue life cycle of tubular T-joints. Results of the analyses for a 6 mm CFRP layup show that under axial loading (AX), the FRP strengthening decreases SCFs up to 30% and 50% at Crown and Saddle points on the chord, respectively. Moreover, under the action of in-plane bending (IPB) and out-of-plane bending (OPB) moments, SCF reductions of about 45% and 50% were observed, respectively.

Introduction

Bottom founded offshore steel structures, such as jacket type platforms are constructed using circular hollow sections (CHS) due to their adequate structural performance, high strength-to-weight ratio, high buoyancy and lower drag coefficient [1]. An offshore jacket platform consists of several joints which are made by connecting CHSs. T-joint is the most common and basic joint configuration in these structures which is made by welding the cross-section of one tube (brace) perpendicular to the undisturbed exterior surface of the other tube (chord) by butt-welding. Typically, residual stresses attributed to welding are eliminated by post heating the joint-can (joint chord) before connecting to the offshore jacket structure.

Generally, in a jacket type offshore structure, joints are the most susceptible parts to the fatigue phenomenon due to the cyclic nature of ocean waves loading. Hence, accurate fatigue life estimation and innovations for enhancing the fatigue life is of crucial importance. Practically, fatigue life of an offshore tubular joint can be estimated from stresses at the weld toe. In the design practice, in order to quantify the stress concentration at the weld toe, a parameter called the stress concentration factor (SCF) is used. According to API-Part 8.3.1 [2], “For each tubular joint configuration and each type of brace loading, SCF is defined as the hot spot stress range (HSSR) divided by the nominal brace stress range”. In this research focus is made on the estimation of SCFs at Crown and Saddle points of the chord member in FRP strengthened tubular T-joints. Thus, the HSSR is the hot spot stress range on the chord which must be divided by the nominal direct stress in the brace member to reach the SCF.

In this study, the relative SCF values (SCFs in the strengthened joint that are divided by the SCFs in the benchmark joint) were extracted from the numerical models and results were discussed in details. The numerical models consist of 152 steel tubular T-joints strengthened with a 6 mm FRP layup under the action of brace axial, IPB and OPB loads using Finite Element Method performed by general purpose ABAQUS software package. Six different FRP materials such as Glass/Vinyl ester, Glass/Epoxy (Scotch ply 1002), S-Glass/Epoxy, Aramid/Epoxy (Kevlar 49/Epoxy), Carbon/Epoxy (T300-5208) and Carbon/Epoxy (AS/3501) were used to investigate the effect of FRP material properties on the SCF values at Crown and Saddle points of the chord member. Moreover, the effect of combined loads and fibers orientation of the FRP layup was investigated, and the results were presented.

According to an earlier study [3], the most effective fiber orientations in the FRP layup, under axial loading proved to be 0° and 90°. Also, owing to the nature of the loading (AX, IPB and OPB), the FRP materials were modeled such that the fibers placed in the two main cross directions would lead to the most appropriate FRP layup of 0° and 90° combination under IPB and OPB moments as well as axial loading. It is worth to note that the finite element models of the benchmark joints were verified against the experimental results extracted from HSE OTH 354 report [4] followed by comparing the results with the predictions of Lloyd's Register [4] and API [2] design equations. FRPs were applied on the verified benchmark joints. The results of analysis of strengthened finite element models addressed how FRP properties could affect the SCF values. In addition, more studies are being undertaken to investigate the combined effect of joint geometry and FRP layup properties.

Section snippets

Literature review

Several studies have been conducted on the SCF calculation in tubular joints since the 1970s. The main objective of these researches was to derive parametric equations for SCF calculation. Most of the researches were dedicated to derive SCF equations for the un-strengthened tubular joints; but owing to the new stricter code provisions, some joints of the existing tubular structures may not meet the code checks. Thus, reinforcing methods were presented to remove possible deficiencies.

A variety

Benchmark joint FE model verification

Finite element models of the benchmark joints were developed and analyses were carried out. In order to verify the accuracy of the finite element modeling and analysis, T-joints JISSP joint 1.3 (for axial loading) and JISSP joint 1.13 (for IPB and OPB moments) were chosen among the test results published in HSE OTH 354 report [4]. The finite element geometries of the benchmark joints were exactly the same as the experimental models. Table 1 presents the benchmark joints properties and

Verification of the strengthening method

In this part of the study, FRP strengthening method in the finite element analysis is verified against an experimental as well as numerical model performed by Lesani et al. [25].

FE modeling of the strengthened joint

In this section, some key aspects of the FRP strengthened numerical model are presented.

FRPs are composed of two distinct parts namely fibers and matrix. Thus, various compositions could be made. In this study, six types of common FRP materials namely, Glass/Vinyl ester, Glass/Epoxy (Scotch ply 1002), S-Glass/Epoxy, Aramid/Epoxy (Kevlar 49/Epoxy), Carbon/Epoxy (T300-5208) and Carbon/Epoxy (AS/3501), were used as strengthening material on the T-joints to find out how different FRP materials

Analyses and results

In order to evaluate contribution from each component to the final result, analyses were carried out in three phases, strengthening the chord member only, strengthening the brace member only and finally strengthening both members. The details and results of the analyses on SCF values in strengthened T-joints for each type of loadings and strengthening schemes are presented in the following subsections. The abbreviations used in the subsequent Figures are explained in Table 6.

Summary and conclusions

In this paper, changes in SCF values at Crown and Saddle points on the chord member of the tubular T-joint due to change in FRP material used for wrapping the joint were investigated. The numerical models of the benchmark joints were verified based on existing experimental data. In addition, another analysis performed to verify the strengthening method against an experimental as well as numerical analysis. Finally, the benchmark model was strengthened using different FRP materials and a

References (38)

Cited by (0)

View full text