Marine Science 2017, 7(1): 10-16

DOI: 10.5923/j.ms.20170701.02

The Process of Fatigue Analysis on Fixed Metal Offshore

Platforms
Sajed Nemati Havigh, Mehdi Behdarvandi Askar*

Department of Offshore Structures, Faculty of Marine Engineering, Khorramshahr University of Marine Science and Technology, Iran

Abstract One of the important reasons of fracture in jacket pelatforms that makes the sadenly failure and structual
resistanc reduce at long time, is Fatigu phenomena. Because of existence of peridice loads derived by waves, sea current,
wind, etc. hence, fatigue analysis is important item in offshore structual design. The known methodes of fatigue analys is
published at design codes suche as, API and DNV and ABS and other relavant codes, that are in acceptable relationship whit
toghether. This paper will reviwe the fatigue analysis methodes and will tries to present a simple and accurate methode that be
according to Persian Gulf candition. Whereas, the most of jacket pelatformes in Persian Gulf, has natural ferequence about
less than 3s, that’s wisely to using Deterministic methhode for fatigue analysis. However, the mentioned offshore design
codes, offer the Spectural methode for all of environmental condition, although the Spectural methode has more accuracy in
comparition whit other methodes, but it’s take larg time to analys and needs to high knowledge of mathmaticul
transformation functions, frecuncy-domain analysis and dynamic of structures. And in the end, the paper offers that Persian
Gulf’s pelatformes, consider whit Erthquake analys, because, most of pelatformes are in the end part of those useful life time.
Keywords Tubular joints, Natural ferequnce, SCF, Determinstic analys, Fatigue, Jacket

existing platforms. Due to the recurring nature of most

1. Introduction environmental loads in the seas and oceans, these structures
are constantly under multiple and recurring loadings. Loads
Nowadays, the oil and gas industry, or in fact in more such as winds, waves, and earthquakes that have a dynamic
common terms, economy and energy, is one of the most nature, are the dominant loads in the design of these
important and most strategic issues that all countries are structures [1]. In this paper, some of the existing methods for
trying to access this important factor in order to be able to fatigue analysis will be reviewed and subsequently a useful
play a decisive role in the region and the world. Therefore, method in terms of cost and precision will be presented for
the oil and gas industry is of great importance, and the the Persian Gulf region.
capability of the development and exploitation of oil and gas The phenomenon of fatigue in materials caught the
fields could be considered as one of the powerful arms of any attention of many researchers from the middle of 19th century,
country. Due to the existence of energy resources in seas, the which in particular Wohler's research led to S-N curves [2].
exploitation of energy from marine environments requires These curves are widely used in determining the fatigue life
high scientific and technological capabilities. Offshore [3].
platforms have a very important and strategic role in
providing the national development. Therefore, firm steps
can be taken in the direction of national and scientific 2. The Importance of Detection and
interests by the preservation and quantitative and qualitative
improvement of these structures. The excessive cost of
Explanation of Fatigue
building new platforms causes the subject of the renovation In engineering sciences, fatigue is defined as the failure of
and retrofitting of the existing platforms to be considered the structure caused by repeatedly applied loads. The random
more than ever. Template fixed platforms are the most nature of this phenomenon has caused sensitivity in detecting
common and most widely used existing platforms in the seas, this phenomenon. In offshore structures, including oil
which entered the oil and gas industry from the late 19th platforms, there have been disastrous catastrophes during the
century and today account for a large percentage of the history of oil and gas industry, which have been mainly
blamed on the fatigue phenomenon. Of these, we can
* Corresponding author: mention the following examples. The disaster of the
sazehenteghal@yahoo.com (Mehdi Behdarvandi Askar)
Published online at http://journal.sapub.org/ms
Alexander L. Kielland semi-submersible Norwegian oil
Copyright © 2017 Scientific & Academic Publishing. All Rights Reserved platform in the North Sea, killing 123 crew members. One of
 Marine Science 2017, 7(1): 10-16 11

the main braces connected to one of the pontoon bases the piles of a sample template platform under the
completely failed and separated from the platform and led to environmental conditions of the Persian Gulf. In this study,
the complete capsize of the platform (Almar-Naess, 1985). SACS software was used for static and dynamic analysis,
This platform and its capsizing status can be seen in figure and the GRLWEAP software was used to analyze the
(1). In 1979, a jack-up rig in the Gulf of Mexico suffered an installation and driving of the piles. The fatigue analysis of
accident, which the cause of that accident was also stated as the pile was performed by the classical method on the South
the onset and expansion of cracks due to fatigue in welded Pars Phase 12 platform with two profile types of clay and
joints. Such that, about 500 millimeters of a fatigue crack, sandy soils. This study showed that the peripheral and
initially with an angle of 270 degrees and then with an angle driving fatigues of the pile in the sand are more than in the
of about 90 degrees had stretched and had led to the failure of clay. Also, to reduce the damage of the driving fatigue in the
the structure (Publication Document 6493, 1980). pile, two types of the hammer with the same performance
can be used at different depths.

2.1. Fatigue
In general, the fatigue phenomenon is considered and
studied in three main stages:
 Initial cracking
 Spread of cracking
 Final fracture
In summary, the initial stage of the cracking in the
member and the joints is within the range of the microscopic
behavior of the materials. These cracks are usually formed
Figure (1). Crack caused by fatigue in a joint due to the special geometry of the joint and the secondary
effects and imperfections of the weld site, which occur either
Based on the studies of Ali Yaghteen and Hassan Sayadi in the form of very tiny internal cracks or at the welding
(1398), it was determined that the life of the Iran Khazar surface. Compared to the initial cracking, the stage of the
jack-up platform is about 18 years. This study was conducted spread of the crack is more recognizable, which fits in the
under conditions prevailing in the Caspian Sea. So that the framework of the theory of fracture mechanics. In addition,
significant wave height and the speed of sea current have the depth of the initial cracks and the geometry of welds also
been assumed 12 meters and one meter per second, have a significant effect on the spread of the cracks. This
respectively. And the interaction between soil and structure stage has the most important impact on the calculation of the
has been modeled with elastic springs. This research has fatigue life. The stage of final fracture takes place when the
been carried out using ANSYS’s finite element software. spread of cracks reaches its critical level. The final failure
Prior to that, the life of the studied platform had been depends on some factors such as stress level, crack size, and
evaluated using Sacs software, which the life of the platform the hardness of the material. Like the initial cracking stage,
had been determined to be about 20 years. This discrepancy the final fracture stage has also little importance in the
represents an error of about 7.5% between the results of the fatigue life.
two software that is acceptable. In the abovementioned In general, fatigue analysis is performed for structures that
research, the recommendations of the API guidelines have are sensitive to alternative loads such as wave loads (in
been used, which after the nonlinear dynamic analysis and jackets, floating structures, etc.), wind (in the burner boom,
the determination of the critical joint, the fatigue analysis staircase towers, inter-platform bridges, etc.), rotary
was performed on that joint. Also, the researchers concluded equipment (the structures on which such equipment is
that in designing the present platform, considering the connected to), repetitive stresses (driven piles, etc.). The
direction of the dominant wave alone could satisfy the design most important parameters for the determination of the
conditions. Behrouz Asgarian and Parviz Kuzehgar (2011), fatigue life are, stress range, the number of cycles, stress
in their studies on the subject of fatigue in the joints of a concentration coefficient, dynamic behavior, S-N curve or
sample platform (SPD 10) present in the Persian Gulf, they repetition-stress curve, figure (2).
showed that considering the local joint flexibility causes a
decrease in the predicted breakdown due to fatigue in the 2.2. Fatigue in the Offshore Platforms
joints, and the results of the research have advantageous
The wave load is the major force applied to the offshore
effects on the optimization of underwater inspection
platforms. The wave load, due to its repetition, causes fatigue
schedule. The above-mentioned research has been carried
in the members and joints of the structure. Due to the
out using finite element software OPENSEES and by
concentration of stress and welding defects in the platform
spectral analysis method. Mohammad Hossein Mohassel
joints areas, they have a high sensitivity to fatigue failure.
et al. (2010) performed a study on the fatigue phenomenon in
12 Sajed Nemati Havigh et al.: The Process of Fatigue Analysis on Fixed Metal Offshore Platforms

3. Fatigue Analysis Methods these wave loading states (which is the ratio of the
number of loading cycles to the number of joint failure
Fatigue analysis of platforms is mainly performed with the cycles) during the useful life of the structure.
two deterministic and spectral methods. According to the
Typically, due to the ease of use, the deterministic method
API-RP 2A regulations and the paragraph 1-5 of this
is more practical than the spectral method, nevertheless, both
regulation, in the design of tubular joints, the necessary
DNV (Det Norske Veritas) and API (American Petroleum
assessments for fatigue caused by alternative stresses in the
Institute) regulations consider deterministic method and
joints must be performed, and for this purpose, the detailed
static equivalent analysis under the conditions that the depth
fatigue analysis for the structure should be performed. Of
of water is less than 122 meters, the materials are soft and
course, the regulation suggests that spectral analysis should
formable, and the natural period of the structure is less than
be used. Also, in platforms with a period less than three
three seconds. To determine the amount of damage in a
seconds and the height of less than 122 meters, and in case of
platform the cumulative damage models for failures caused
the absence of access to an accurate analysis, the regulation
by fatigue that is known as Miner’s rule should be used:
recommends a simplified method in accordance with section
𝑛𝑖
C5.2 [3]. 𝐷= ≤1 (1)
𝑁𝑖

3.1. Deterministic Method n: number of cycles of each stress

D: damage rate
The foundation of definitive analysis is based on this N: number of permissible cycles of a specific stress
principle that a limited number of waves with a specified
height and period from all peripheral states of the fatigue or, 3.2. Spectral Method
in other words, the sea state existing in the wave dispersion Spectral fatigue analysis is used to evaluate the dynamic
diagram, are selected as deterministic waves. The selection response of the structure to the wave height range and the
of these waves is based on the contribution of each sea state frequencies corresponding to the dispersion diagram. This
in the wavelength dispersion diagram. These waves are structural response considerably includes all states of the
assessed discretely and are combined such that they have the waves influential in the damage caused by fatigue. The
highest impact in fatigue damage. Each of these specific spectral dynamic analysis is performed based on the spectral
waves is considered at various wave crest states and in (wave spectrum) density function in each sea state. Such that,
sufficient numbers. Then, these crest wave states are the sea waves are modeled according to the wave dispersion
combined in such a way that the maximum and minimum diagram model as a set of sea states. And instead of applying
values of stress and consequently the range of stress in the each sea state with a specific height and period to the
members and joints of the structure are achieved. The platform, its spectrum (in the frequency domain) is applied to
structural responses, which include the range of nominal the platform and transformed to the hot spot stress spectrum
stresses in the joints and members of the structure, are in the peripheral points of the joint by the transfer function.
converted separately into the stresses of hot spots at the The spectral dynamic analysis method is a method in the
peripheral spots of the joint for each of these waves by frequency domain and it is based on the assumption that for a
applying the proper stress concentration factor (SCF) for specific period, the force is considered proportional to the
each joint. After identifying the stresses of the hot spots, wave height. Therefore, in order to create a transfer function
according to the number of occurrences of each wave in a for spectral analysis, the expression for the nonlinear drag
given time period (in accordance with the waves dispersion force in the Morrison equation should be linearized with one
diagram), and using the Palmgren -Miner law (relationship 1) of the proposed methods of the regulation such as the use of
of the linear cumulative fatigue damage theory, the fatigue the steepness of the continuous wave. In a the dynamic
damage rate for each wave is obtained. Finally, the overall analysis in deep waters under the influence of large storm
fatigue damage of a spot in the joint periphery, for all of the waves, the dominant wave force consists of the drag force
waves during the useful life of the structure is obtained from that is nonlinear, therefore, for a proper response, these
the sum of fatigue damage for each of the waves. The structures should be dynamically analyzed in the time
description of this approach is in the following general form: domain [4]. Among the important parameters in fatigue
 It is performed in the time domain. analysis, we can mention the number of wave cycles,
 In this method, the wave energy is simulated using the nominal stresses, hot point stress range, stress concentration
sum of a number of independent waves with specific coefficients and repetition-stress curves. These definitions
frequency, height and event numbers, and are applied have briefly been summarized below. Due to the geometry of
separately to the structure, and the structural response the constituent members of the marine structures, and
and critical stresses are obtained separately for each of especially the template platforms, which are often
these waves in at least eight peripheral joint points accompanied with shape and cross-sectional variations in the
(with a distance of 45 degrees). joints, and also, the use of welding in the joints, most of these
 Then, the damage due to the overall fatigue of the joints get lots of stress concentrations. Therefore, in the
structure is calculated from the sum of the damages of majority of structural joints, tensions at the location of the
 Marine Science 2017, 7(1): 10-16 13

joint on both sides of the main member and the minor this minimum of the axial and flexural stress concentration
member are significantly larger than the nominal stresses. It coefficient for the main and minor members, figure two has
should be noted that nominal stresses are directly derived been recommended. For internally reinforced joints figure
from the structural analysis. The type of welding profile and 2.5 and for overlapping tubular joints and externally
connection geometry are influential in determining the reinforced joints with pendentive or circular hardener, figure
amount of stress concentration coefficients. This coefficient six have been recommended for this coefficient. In offshore
is obtained using experimental results or finite element structures, most joints are multi-planar, which for the
analysis. Some of the relationships presented in the current calculation of stress concentration coefficients, they are
regulations are given below. Stresses do not remain uniform examined separately at each plan and the highest coefficient
in mechanical joints, including the main and minor members, among them is selected as the stress concentration
figure (2). coefficient at the joint. The geometric parameters for tubular
The stress at the toe of a joint weld, which is, in fact, the connections are given in the DNV regulation. The presented
same as the hot spot stress range. The stress concentration parameters include (figure 3):
factor is, in fact, the ratio of the hot spot stress range (HSSR) D: main member diameter
to the range of nominal brace stress at the junction. t: thickness of minor members
𝐻𝑆𝑆𝑅 𝑎𝑡 𝑡ℎ𝑒 𝐿𝑜𝑐𝑎𝑡𝑖𝑜𝑛 T: thickness of main member or base
𝑆𝐶𝐹 = Brace force equation
𝑅𝑎𝑛𝑔𝑒 𝑜𝑓 𝑁𝑜𝑚𝑖𝑛𝑎𝑙 𝑏𝑟𝑎𝑐𝑒 𝑠𝑡𝑟𝑒𝑠𝑠
L: length of the member
(2)
Θ: the angel between the two members
The hot spot stress range (HSSR) is computed at least at g: Horizontal distance between two braces at the toe of the
eight points of the periphery of the joining point of the main weld
member to the minor member. The API RP 2A API The forces applied to the members of a joint at the site of
regulation prefers Efthymiou's relationships for overlapping the weld, as well as the eight peripheral spots of the joint for
and non-overlapping non-reinforced tubular joints. It also calculating the hot stress spot, are shown in figure (4). The
recommends Lioyd equations for reinforced tubular joints. In axial load (N), in-plane bending moment (IPB), and out of
this regulation, some minimal values have been presented for plane bending moment (OPB) can be seen in the figure 4.
stress concentration, which generally for tubular joints, for

Figure (2). A view of a T joint and the illustration of stress in the inflamed area

Figure (3). Geometric Parameters of Tubular Joint (DNV)

14 Sajed Nemati Havigh et al.: The Process of Fatigue Analysis on Fixed Metal Offshore Platforms

Figure (4). Forces Applied on a Joint

Hot spot stress (HSS) in the periphery of a joint is ∆𝜎: existing stress
calculated from the recommended relationship in the DNV ∆𝜎𝑟𝑒𝑓 : experimental standard stress
regulation as follows and it is used in the repetition-stress m: curve slope
diagram [5]. where the slope of these diagrams is given in the following
𝐹𝑎𝑥 𝐹𝑖𝑝𝑏 table:
𝐻𝑆𝑆𝑅 = 𝐷𝐴𝐹 ( 𝑆𝐶𝐹𝑎𝑥𝑖 + 𝑆𝐶𝐹𝑖𝑝𝑏 𝑆𝑖𝑛𝜃𝑖
𝐴 𝑍
𝐹𝑜𝑝𝑏 Table (1). Characteristic values of the equation of the S-N diagrams
+ 𝑆𝐶𝐹𝑂𝑃𝐵 𝐶𝑜𝑠𝜃𝑖 ) (3)
𝑍
∆𝜎𝑟𝑒𝑓
DAF: dynamic amplification factor Curve m Application
Z: basis of the cross-section in two million
cycles mpa
𝑆𝐶𝐹𝑖𝑝𝑏 : in-plane bending moment stress concentration
factor Two sides with flat weld
X 4/38 100
profile and control with x-ray
𝑆𝐶𝐹𝑂𝑃𝐵 : Out of plane bending moment stress
concentration factor Susceptible weld with rugged
X' 3/74 79
𝜃: angle between peripheral spots sides

3.3. Repetition-Stress Diagrams The curve X is considered for a state where the weld
profile is controlled and the thickness of the minor member is
The recommended diagrams in API regulation, which are less than 25 millimeters. If the thickness of the minor
in the two x and x' states [3], figure (5). member is less than 25 mm, the correction of the effect of the
HSSR thickness is done according to equation (5) in the following.
(mpa) 𝑡 𝑟𝑒𝑓 0.25
𝑆 = 𝑆𝑜 ( ) (5)
𝑡

So: allowed stresse obtained from S-N Curves

S: allowed stress
t: thickness of the minor member
tref: 16 mm for X ' curve and 25 mm for X curve
X' curve, is for the case where the welding profile is not
controlled but conforms to the smooth profile of the AWS
standard. By the way, this curve is used when the thickness
of the minor tube is less than 16 millimeters. The correction
of thickness will be done similar to equation (5).
N (sycle)
3.4. Modeling a Sample Platform in Abacus Software
Figure (5). These curves are plotted according to the following equation:
The depths of the platforms in the Persian Gulf are
6 ∆σ −m generally less than 100 meters, and also due to the relative
N s = 2 × 10 × ( ) (4)
∆σ ref
rigidity of marine structures such as jackets because of the
N(s): Number of permitted cycles type of their welded joints, the natural period of these
 Marine Science 2017, 7(1): 10-16 15

structures often does not exceed 3 seconds [6]. profiles, and three-dimensional beam elements, which have
Due to the geometry of fixed metal platforms in the the capability of incorporating rotations, bending, and
Persian Gulf and the installation depth of these platforms, displacement at the ends of each element. This platform has
which are usually between 60 and 80 meters, an example been modeled using B32 element in the partitioning and in
fixed metal platform similar to the specifications of the the form of fixed at the end of the legs at the bottom of the
implementation plan of the SPD15 structure (Prepared at sea, figure (6). Due to the high hardness in this model, the
National Iranian Oil Company) has been modeled using natural period of this model is about 1.3 second. This
Abacus 2016 software under the environmental conditions of modeling has been done without taking into account the
the Persian Gulf (Fig. 6). The present platform (substructure), interaction between the soil and the structure. Also, the
with a total height of 73.85 meters from the sea bed and at a natural period of the Resalat template platform with four legs
depth of 66.6 meters, has been modeled. It has been drawn in and at an approximate depth of 68.2 meters, which is about
three-dimensional form, and of the coordinate type with wire 3.2, has been modeled using ANSYS software [7]. Also, the
members, which are joined together in the form of a fixed SPD19A platform, with a depth of 54.5 m and a height of
connection. These members have been defined with pipe 70.55 m, has a natural period of 2.32 using SACS. [8].

Figure 6. The Platform Modeled using Abacus Software

Figure 7. The First Six Modes of the Simulated Platform

16 Sajed Nemati Havigh et al.: The Process of Fatigue Analysis on Fixed Metal Offshore Platforms

4. Results [2] Almar-Naess, A., 1985. – Fatigue Handbook – Offshore Steel

Structure. Tapir, Trondheim, Norge, 520p.
In this research, first the importance and the process of
[3] American Petroleum Institute, 2002, “API RP 2A - WSD.
fatigue analysis in the fixed offshore platforms has been Recommended Practice for Planning Design and
discussed, and then the stages of analysis by deterministic Constructing Fixed Offshore Platforms. Working Stress
and spectral methods have been described. With regard to the Design,” 21st Edition.
similar geometry and hardness of metal platforms in the
[4] Anastogtopoulos, S. A., 1982. Dynamic Response of
Persian Gulf region and the natural period of these structures, Offshore Structures to Extrem Waves Including
which is generally less than three seconds, the deterministic Fluid-Structur Interaction. Eng structures, Vol.4, pp 179-185.
method can be used to analyze the fatigue of these platforms
with a good accuracy. Considering the applied method in the [5] Det Norske Veritas, October 2001, “DNV. Recommended
Practice RP- C203, Fatigue Strength Analysis Offshore Steel
analysis, the definite analysis has a lower time and cost Structures.”
compared to the spectral analysis. Therefore, with
confidence in the results of the deterministic analysis, we can [6] Zaynoddini, M., 2005. The Design and Implementation of
use this method for the platforms in the Persian Gulf. Fixed Oil Platforms. First print, National Center of
Oceanographic Sciences, Tehran, 340.
Considering the useful life of the existing platforms in the
Persian Gulf, which is about 25 to 50 years, most of these [7] Mohajer Nasab, S., Dastan, M. A., Tabeshpour, M. R., Seyf
platforms are in their final years of their life, which this fact M. S., Mehdigholi, H., 2013, The Application of New-Wave
makes the seismic analysis of these platform important as Theory in the Endurance Wave Analysis (EWA) for the
Evaluation of Fixed Base Platforms Under the Loading of the
well. Persian Gulf Waves. Journal of Marine Engineering, Fall, and
Winter of 2013, Ninth year, No. 18, pp. (71-82).
[8] Documents of the National Iranian Oil Company. The
Calculations of the In-Place Analysis of SPD19A Platform,
REFERENCES Document Number: SP19 -P0-ST-A-8021.

[1] Tabeshpour, M. R., Komachi, Y., Golafshani, A. A.,

Assesment and Rehabilitation of Jacket Pelatforms, 2012.
Prof, Abbas Mustafa (Ed), ISBN: 978-953-53-0123-9.