Fatigue analysis of offshore platforms subject to sea wave loadings

The fatigue damage calculation for random loading on offshore platforms takes the form of a rainflow analysis of the dynamic response of individual members to various sea states. This procedure is lengthu and consequently this paper attempts to provide a theoretical method for determining random load fatigue damage. This dynamic response for many joints leads to a broad band random loading but despite this, previous theoretical methods have simplified the loading to narrow band. This has not been done in the present case; instead, an analysis based on broad band random loading has been produced. This theoretical approach gives a fatigue life estimate which is slightly (6.6%) more conservative, for a typical example, than a rainflow analysis.     

A model of corrosion fatigue crack growth in ship and offshore steels

A model describing corrosion fatigue crack growth rate da/dN has been proposed. The crack growth rate is assumed to be proportional to current flowing through the electrolyte within the crack during a loading cycle. The Shoji formula for the crack tip strain rate has been assumed in the model. The obtained formula for the corrosion fatigue crack growth rate is formally similar to the author's empirical formulae established previously. The different effects of ΔK and the fatigue loading frequency f on da/dN, in region I as compared to region II of the corrosion fatigue crack growth rate characteristics can be described by a change of one parameter only: the crack tip repassivation rate exponent.     

Predicting the fatigue life of offshore structures by the single-moment spectral method

The dependence of fatigue damage accumulation on power spectral density (psd) is investigated for Gaussian random processes representing stresses in offshore structures. This involves extensive computer simulation of representative stress time histories and related analyses to predict the time of fatigue failure. The recently introduced single-moment (SM) spectral method is shown to be very attractive for predicting fatigue failure. The SM method predicts the damage accumulation rate based only on a single calculation from the psd curve (a moment integral). The accuracy of the SM method and of other spectral methods is studied by comparing results with ones calculated from the much more expensive alternative of using simulated stress time histories, rainflow analysis, and Palmgren-Miner calculations.The SM method is shown to be even simpler than most spectral methods, and all spectral methods are much simpler than any method involving simulation of stress time histories. The accuracy of the SM method is shown to be quite good for a wide variety of situations. Furthermore, there are situations in which the SM method is significantly more accurate than other simple spectral methods, including the very commonly used Rayleigh approximation.It appears that the SM method may provide a valuable tool for predicting the fatigue life of offshore structures subject to complicated time histories of loading, as is required both in the design of new structures and in the reassessment of aging structures. In order to allow a proper balance between fatigue life and initial cost, there is a need for a fatigue prediction method which is both accurate and easy to apply, and the SM method appears to be adequate on both bases.     

The fatigue crack growth of a ship steel in seawater under spectrum loading

Fatigue crack growth of ABS EH36 steel under spectrum loading intended to simulate sea loading of offshore structures in the North Sea was studied using fracture mechanics. A digital simulation technique was used to generate samples of load/time histories from a power spectrum characteristic of the North Sea environment. In constant load-amplitude tests, the effects of specimen orientation and stress ratio on fatigue crack growth rates were found to be negligible in the range 2 × 10^?5 to 10^?3 mm/cycle. Fatigue crack growth rates in a 3.5% NaCl solution were two to five times faster than those observed in air in the stress intensity range 25 to 60 MPa √m. The average fatigue crack growth rates under spectrum loading and constant-amplitude loading were in excellent agreement when the fatigue crack growth rate was plotted as a function of the appropriately defined equivalent stress intensity range. This procedure is equivalent to applying Miner's summation rule in fatigue life calculations.     

The effect of pitting corrosion on fatigue life

Fatigue testing of pre‐pitted 2024‐T3 aluminium alloy specimens is performed in laboratory air at 22 °C and 40% RH to characterize the effect of pitting corrosion on fatigue life. Specimens, pre‐corroded in a 0.5 M NaCl solution from 48 to 384 h, have fatigue lives that are reduced by more than one order of magnitude after 384 h pre‐corrosion as compared to those of uncorroded specimens. The reduction in fatigue life is interpreted in terms of the influence of the time of exposure to the corrosive environment or pit size. The crack‐nucleating pit sizes, ranging from 20 to 70 μm, are determined from post‐fracture examinations by scanning electron microscopy. Fatigue lives are estimated using a fracture mechanics approach and are shown to be in good agreement with the actual data. A probabilistic analysis shows that the distribution of fatigue life is strongly correlated to the distribution in nucleating pit size.     

The effects of saline aqueous corrosion on fatigue crack growth rates in 316 grade stainless steels

A study of fatigue crack propagation rates of 316 grade stainless steels in air and in an aqueous saline environment was carried out in an attempt to assess the fatigue properties encountered when such materials are used as surgical implants. The effects of variables such as temperature, pH, oxygenation level, bulk electrode potential, mean stress, frequency and stress waveform on the Paris crack growth law parameters were determined. Corrosion fatigue effects were observed in the aqueous saline environment, and a mechanism to describe this effect is proposed.     

A numerical analysis of constraint effects in fatigue crack growth by use of an irreversible cohesive zone model

The analysis of constraint effects in fatigue crack growth in multi-layer structures is discussed. The process of material separation under cyclic loading is described by a cohesive zone model (CZM) with an irreversible constitutive relationship. The traction–separation behavior does not follow a predefined path, but is dependent on the evolution of the damage dependent cohesive zone properties. A modified boundary layer model is used in simulations of fatigue crack growth along the centerline crack of the metal layer sandwiched between two elastic substrates. Fatigue crack growth is computed for a series of values of metal layer thickness under constant and variable amplitude loading conditions. The results of the computations demonstrate that certain combinations of load magnitude, layer thickness and material properties results in significant constrain effects in fatigue crack growth. The influence of these constraint effects on fatigue crack growth rates and on crack closure processes is determined. The evolutions of the traction–separation law, the accumulated and current plastic zones, as well as the stress fields during the crack propagation are discussed.     

Bifurcation analysis of the Hobson short fatigue crack growth law

A bifurcation analysis of the Hobson short fatigue crack growth law is presented. The analysis reveals that, although the growth law is non‐linear, it contains no bifurcation points.     

Corrosive fatigue of structures with dimple holes under spectrum loading

ABSTRACT Experimental investigation and life prediction are made for structures with straight holes and dimple holes in both laboratory ambient conditions and 3.5% NaCl solution under spectrum loading. With the aid of acoustic emission, label loading and fractography technique, the crack‐growth data are obtained in both environments. Stress‐intensity factors for the corner crack emanated from the dimple hole are evaluated by three‐dimensional (3D) finite element (FE) method. Life prediction of the tested structures is made by using the modified FASTRAN‐II code on the basis of corresponding fatigue crack‐growth rate curves. It is shown that once the initial defect size a₀ is determined by one set of test data, it can be used together with the 3D crack‐growth method in order to provide accurate life prediction in alternative structure and loading condition. By comparison, life prediction by the traditional local strain method is also made which is shown to be less reliable than the 3D crack‐growth method.     

Influence of foreign object damage on fatigue crack growth of gas turbine aerofoils under complex loading conditions

Foreign object damage (FOD) has been identified as one of the main life limiting factors for aeroengine blades, with the leading edge of aerofoils particularly susceptible. In this work, a generic edge ‘aerofoil’ geometry was utilized in a study of early fatigue crack growth behaviour due to FOD under low cycle fatigue (LCF), high cycle fatigue (HCF) and combined LCF and HCF loading conditions. Residual stresses due to FOD were analyzed using the finite element method. The longitudinal residual stress component along the crack path was introduced as a nodal temperature distribution, and used in the correction of the stress intensity factor range. The crack growth was monitored using a nanodirect current potential drop (DCPD) system and crack growth rates were correlated with the corrected stress intensity factor considering the residual stresses. The results were discussed with regard to the role of residual stresses in the characterization of fatigue crack growth. Small crack growth behaviour in FODed specimens was revealed only after the residual stresses were taken into account in the calculation of the stress intensity factor, a feature common to LCF, HCF and combined LCF + HCF loading conditions.     

Fatigue strength of small-notched specimens under variable amplitude loading within the fatigue limit diagram

Although the fatigue limit diagram is defined in principle for constant stress amplitude, it is often considered that fatigue failure would not occur, even in varying loading, if applied stresses were kept within the fatigue limit diagram. However, it was shown in the case of small‐notched specimens that fatigue failure occurred in some special cases of variable amplitude loading, even when all stress amplitudes were kept within the fatigue limit diagram. The cause of this phenomenon was examined using two‐step stress and repeated two‐step stress patterns in which the first step stress was chosen to be equal to the fatigue limit with zero mean stress and a mean stress was superposed on the second step stress. A non‐propagating crack was formed by the first step stress. This crack functioned as a pre‐crack for the second step stress with high mean stress. Consequently, fatigue failure occurred even when all stress amplitudes were kept within the fatigue limit diagram. It was an unexpected fracture caused by the interference effect of a non‐propagating crack and a mean stress change.     

Can pure mode III fatigue loading contribute to crack propagation in metallic materials?

The possibility of pure mode III crack growth is analysed on the background of theoretical and experimental results obtained in the last 20 years. Unlike for modes I and II, there is no plausible micromechanistic model explaining a pure mode III crack growth in ductile metals. In order to realize 'plain' mode III fracture surface, we propose the propagation of a series of pure mode II cracks along the crack front. Fractographical observations on crack initiation and propagation in a low alloy steel under cyclic torsion support such a model. The authors have not seen any clear indication of a pure mode III crack growth micromechanism in metals until now.     

Role of elasto-plastic analysis under cyclic loading in fatigue crack growth studies

Linear Elastic Fracture Mechanics (LEFM) has been widely used in the past for fatigue crack growth studies, but this is acceptable only in situations which are within small scale yielding (SSY). In many practical structural components, conditions of SSY could be violated and one has to look for fracture criteria based on elasto-plastic analysis. Crack closure phenomenon, one of the most striking discoveries based on inelastic deformations during crack growth, has significant effect on fatigue crack growth rate. Numerical simulation of this phenomenon is computationally intensive and involved but has been successfully implemented. Stress intensity factors and strain energy release rates lose their meaning,J-integral (or its incremental) values are applicable only in specific situations, whereas alternate path independent integrals have been proposed in the literature for use with elasto-plastic fracture mechanics (EPFM) based criteria. This paper presents certain salient features of two independent finite element (numerical) studies of relevance to fatigue crack growth, where elasto-plastic analysis becomes significant. These problems can only be handled in the current day computational environment, and would have been only a dream just a few years ago. The work presented in this paper is supported by sponsored research projects of the Aeronautics R & D Board, Government of India and their support is acknowledged.     

On the corrosion fatigue crack initiation model and expression of metallic notched elements

In the present study, attempts are made to extend the application of the mechanical model for the fatigue crack initiation (FCI) and the FCI life formula of metallic notched elements in laboratory air to those in the corrosive environment. The test results and analysis of the corrosion FCI (CFCI) life of aluminum alloys and Ti---6A1---4V show that the expression of the CFCI life obtained by modifying the FCI life formula in laboratory air can give a good fit to the test results of the CFCI life. The salt water (3.5% NaCl) environment has no effects on the CFCI resistant coefficient compared with the FCI resistant coefficient in laboratory air. However, 3.5% NaCl environment greatly decreases the CFCI threshold of aluminum alloy, but has little effect on the CFCI threshold of Ti---6A1---4V. The loading frequency ranging from 1 Hz to 10 Hz has no appreciable effect on the CFCI life, and thus, the CFCI threshold of aluminum alloys investigated. Hence, the expression for the CFCI life of metallic notched elements proposed in this study is a better one, which reveals a correlation between the CFCI life and the governing parameters, such as, the geometry of the notched elements, the nominal stress range, the stress ratio, the tensile properties and the CFCI threshold. However, this new expression of the CFCI life needs to be verified by more test results.     

Uncertainty modeling in the fatigue reliability calculation of offshore structures

This paper presents a procedure of modeling uncertainties in the spectral fatigue analysis of offshore structures with reference to the reliability assessment. Uncertainties of the fatigue damage are generally embedded in response characteristics of the stress process and the damage-model used. Besides commonly accepted uncertainties in offshore structural analysis, which are associated with the modeling of structures and the random wave environment, there are also uncertainties arising from joint flexibilities that occur during the response, the wave–current and water–structure interactions. Uncertainties in joint flexibilities are associated with degradation of member connectivities during a response process. Uncertainties introduced by the wave–current interaction are related to the modeling of a random sea state, applied wave loads and water–structure interaction effects in general. The water–structure interaction, which is an important phenomenon to be considered in the analysis of dynamic-sensitive structures, introduces some added hydrodynamic damping. The associated uncertainties are reflected in the response analysis via the damping term. Therefore, in a quasi-static response analysis, these uncertainties disappear. In the spectral fatigue damage, in addition to the uncertainties of stress statistical characteristics there are some other uncertainties associating with the damage-model used. These uncertainties are related to experimentally determined fatigue data and configurations of selected joints at which damages are likely to occur due to high stress concentrations. This paper presents these uncertainty issues with emphasis on the application of a reliability assessment. However, some other uncertainties arise from approximations inherent in the model. They are assumed to be either comparatively negligible or can be considered within the current uncertainty models so that they are not treated further in this paper. In the calculation of the fatigue damage, a non-narrow banded stress process is used.     

Probabilistic fatigue crack growth analysis of spot‐welded joints

This paper presents a probabilistic fatigue crack growth life prediction methodology for spot‐welded joints under variable amplitude loading history. The loading is multi‐axial and is obtained from transient response analysis of a vehicle model using finite‐element analysis. A three‐dimensional (3D) finite element model of a simplified joint with four spot welds is developed, and the static stress analysis of this joint is performed. Then the fatigue crack inside the base material sheet is modelled as a surface crack. Probabilistic crack growth model is combined with the stress analysis result to develop a probabilistic fatigue crack growth life prediction methodology for spot welds. This new method is implemented with MSC/NASTRAN and MSC/FATIGUE and is useful for the reliability assessment of spot‐welded joints against fatigue crack growth.     

Assessing the accuracy of fatigue life in surface‐cracked straight pipes

The demonstration of leak before brake (LBB) based on fracture mechanics requires information on the initial size of a defect, initiation of crack growth from the inherent defect and subsequent crack growth rates. In the present paper the prediction methodologies have been tested for three different full scale pipes geometry experimentally tested data. The prediction accuracy of two SIF solutions available in the literature has also been judged. The effect of fatigue crack closure and corrections needed in the numerical prediction methodology using FEM have also been included. The results showed that the FEM could fairly predict the fatigue crack initiation and crack growth life of full‐scale piping components having a constant depth crack profile.     

Dual boundary element analysis for fatigue behavior of missile structures

Abstract

The fatigue behavior of a crack in a missile structure is studied using the dual boundary integral equations developed by Hong and Chen (1988). This method, which incorporates two independent boundary integral equations, uses the displacement equation to model one of the crack boundaries and the traction equation to the other. A single domain approach can be performed efficiently. The stress intensity factors are calculated and the paths of crack growth are predicted. In order to evaluate the results of dual BEM, four examples with FEM results are provided. Two practical examples, cracks in a V‐band and a solid propellant motor are studied and are compared with experimental data. Good agreement is found.