system identification of steel jacket type offshore platforms … · and new systems. in this...
TRANSCRIPT
9th
European Workshop on Structural Health Monitoring
July 10-13, 2018, Manchester, United Kingdom
Creative Commons CC-BY-NC licence https://creativecommons.org/licenses/by-nc/4.0/
System Identification of Steel Jacket Type Offshore Platforms using
Vibration Test
Hamed Rahman Shokrgozar 1, Behrouz Asgarian
2
1 Department of Civil Engineering, University of Mohaghegh Ardabili, Ardabil, Iran,
2 Department of Civil Engineering, K.N.Toosi University of Technology, Tehran, Iran,
Abstract
Offshore platforms during their life-time due to severe environmental conditions need to
monitoring continuity than the onshore structures. Structural health monitoring is a new
method that used for most infrastructures in recent years. This method is a two-step
process including the system identification and damage detection. The purpose of this
paper is study the health monitoring of an experimental steel jacket type offshore
platform in order to propose a dynamic system identification method for this type of
infrastructures. Due to the time-varying loads and severe environmental condition of sea
state affected on offshore platforms, system identification of this type of structure has
principal role in conforming the quality of construction, validation or updating
analytical finite element structure models, and specially conducting damage detection.
In this paper, dynamic response measurement of a prototype jacket type offshore
platform, SP, is performed using forced vibration imposed from an eccentric mass
shaker. The prototype platform SP is installed on eight skirt piles embedded on
continuum monolayer sand. Dynamic characteristics of the platform are identified using
signal processing approach. Numerical simulation of responses for the studied structure
is also performed using capability of ABAQUS software. The 3D model of ABAQUS is
created using continuum elements for soil and piles, and beam elements for jacket, deck
and pile elements. It can be seen that dynamic characteristics of such a platform can be
extracted from experimental result of the system subjected to dynamic loading.
1. Introduction
Steel jacket-type offshore platforms are the most common kind of offshore structures
that been widely used in offshore oil and gas industry. These large and complex
structures during their service life are subjected to random sea environmental loads such
as wave, current, wind and earthquake; therefore, proper maintenance to ensure the
safety of their operation is an important issue. Using the common methods such as
visual inspection for monitoring platform integrity are often hard and sometimes
impossible due to sea condition or increasing the depth of water. These problems led to
the development of simpler techniques for evaluation of structural performance.
Structural health monitoring (SHM) is a new method for assessment of the integrity of
structures in the last two decades. The goal of this process is to maximize the structure's
performance and minimize risk. The major step of SHM is system identification. Most
important infrastructure components such as bridges, tall buildings, dams and offshore
Mor
e in
fo a
bout
this
art
icle
: ht
tp://
ww
w.n
dt.n
et/?
id=
2343
8
2
platforms can be conveniently instrumented to estimate their dynamic characteristics.
The major advantage of using system identification is improvement fidelity of finite
element models, minimizing inaccuracies and calibrating design codes. Inaccuracies in
finite element models may be due to several factors such as the approximations made in
the discretization, miss-modelling of structural elements, the difference in material
properties and inaccuracy of dimension. System identification is also help to check the
integrity of structures subjected to ever increasing loads and monitor the changes of
structural characteristics and responses, and subsequently detects damages in the
structure and predict the expected remaining life. The role of Structural Identification is
also presented in the state of the art report by American Society of Civil Engineers
(ASCE) and Structural Engineering Institute (SEI) committee (ASCE-SEI 2012).
Many studies have focused to develop techniques for system identification and show
reliable achievement (Arici and Mosalam 2005a and 2005b; Mohanty and Rixen 2006;
Carden and Brownjohn 2008; Gul and Catbas 2008; Elshafey et al. 2009; Mojtahedi et
al. 2011). In system identification process, a mathematical model is generally
determined by observing its input –output relationships. This model may be classified to
graphic or mathematical models. Sometimes a mathematical model can be constructed
based on the physical laws that govern the system behaviour, but often such direct
modelling may not be possible due to incomplete knowledge about system’s mechanism
or change of properties of the system in an unpredictable manner.
The traditional fast Fourier transform (FFT) was proposed by Bendat and Piersol,
(1993) for recognition dynamic properties of civil structures. The peaks of the averaged
normalized power-spectral densities are determined the natural frequency of structures.
In this method, the system properties are obtained by converting the measured data to
the frequency domain by a Discrete Fourier Transform (DFT). Andersen et al. (1997)
proposed Auto-Regressive Moving Average Vector (ARMAV) models. In his proposed
model, it is assumed that the structure is linearly, time-invariant, and the unknown input
force can be modelled by a white noise filtered through a linear and time-invariant
shaping filter. The ARMAV is calibrated to the measured time signals by minimizing
the prediction error. Van Overschee and De Moor (1996) suggested a method in which
dynamic behaviour of a structure excited by white noise can be described by a
stochastic state space model. The state space matrices identify based on the
measurements and by using robust numerical techniques such as least squares. James et
al (1995) demonstrated that the correlation functions can be used in the identification
algorithms of traditional modal analysis. Classical techniques such as Least Squares
Complex Exponential (LSCE) and Eigen system Realization Algorithm (ERA) are
appropriate to extract the modal parameters from the measured response data of
structures undergoing unknown forces.
Identification the dynamic characteristic of offshore platforms is so important problem
both in the analytical analysis and implementation of structural health monitoring
methods. The response of these structures is a function of several parameters such as
loading properties and dynamic pile-soil structure interaction behaviour. Experimental
investigations of scale models provide means to validate numerical calculations and
evaluate the existing approaches and measuring instruments for measurement its
response. They also provide a controlled environment in which the effect of specific
parameters can be studied. The present paper reports the results of an experimental
program that was carried out at K.N.Toosi University of Technology to investigate the
dynamic response of a scaled model of a fixed offshore structure. For this purpose, a
3
three dimensional 1:15 scaled model is fabricated as welded-steel space frame. The
scale model is installed on eight skirt piles embedded in monolayer continuum soil.
Experimental Force vibration tests were performed by an eccentric mass shaker.
Accelerometers and LVDTs were used to estimate the response of structure.
Signal based identification, based on analysis of response signals of structures, has also
been developed. The classical method of frequency domain analysis is applied by means
of Fourier transform, and simultaneous its four properties. Finally, experimental results
were compared with the results obtained from a finite element model.
2. Identification of dynamic characteristics
Dynamic Response and behaviour of structural systems may be studied from response
observation during earthquakes, experimental results on reduced scaled or actual size
models of the structures and finally from analytical or numerical modelling of
structures. Performing experimental investigation is very useful in complex structure
such as offshore platforms to verify the numerical result. The use of experimental
techniques has been widely applied to detect dynamic response of structural systems
subjected to dynamic loading. These techniques are being investigated in several fields
such as mechanical engineering (Mohiuddin and Khulief, 2002; Sabnavis et al., 2004),
aerospace applications (Ghoshal et al, 2001) and the offshore industry (Idichandy and
Ganapathy, 1990; Mangal et al. 2001 and Ruotolo et al., 2000).
Experimental dynamic investigation offers the opportunity to obtain information about
the whole system dynamic characteristics based on small number of measurements.
Usually, dynamic tests lead to determination of system modal characteristics.
Experimental modal analysis is the term used for determination of modal properties
such as natural frequencies, mode shape and modal damping experimentally. Two
common methods of the structural modal testing are the force vibration and ambient
vibration (Salawu and Williams, 1995). In ambient vibration tests, the excitation is not
under the control and it is usually considered as a stationary random process. Ambient
excitations are from different sources such as wind, pedestrian or vehicular traffic,
earthquakes, waves or similar excitation. For very large and massive structures, ambient
excitation is often the only practical choice. Structural identification through ambient
vibrations has been successful in numerous cases (Ivanovic et al., 2000 and Ventura et
al., 2003). Usually, experimental test of full-scale structures under dynamic loads is not
practical. Experimental measurement of laboratory scale model of the structure is one of
the alternative methods for comparison and investigation of the performance of existing
and new systems. In this paper, scaled model of a jacket type offshore platform is
fabricated and experimental forced dynamic test is performed on it.
2.1. Description of Scaled Model
Scaled model of a newly installed jacket type offshore platform in Persian Gulf SP1 is
considered for this study. The laboratory model is a welded-steel space frame with six
legs, horizontal and vertical braces, two-story decks and eight skirt piles. The details of
the model are presented in Figure 1. Due to available pipes and laboratory facilities, the
geometric scale of 1:15 is used. The modulus of elasticity and yield stress of steel piles
are determined as 2x108 KN/m2 and between 265-285 MN/m2 (respectively).
4
For considering real boundary condition in the experimental model, a square cubes hole
with approximately 7.00 m dimension is excavated in the yard of civil engineering
faculty of K.N.Toosi University of Technology. In order to simulate the stiff bedrock in
the bottom of the hole, a thick concrete layer with compression strength of 28MPa is
constructed. The hole is filled with homogenous sand with a volume of about 320 m3.
The process of filling the hole with sand is done in 10 months before performing driving
skirt piles and installing jacket. The soil profile consisted of a monolayer uniformly
graded sand with an angle of internal friction φ of 38° and the undrained shear strength,
cu of 9.8 KN/m2. Young’s modulus for the sand is varied between 12,000-25,000
KN/m2 along the depth of the pile. The installation process of the scaled platform is
conducted by driving eight piles through their sleeves using pile driving tool. Figure 2
shows picture of the scaled model of the SP1 platform installed on 8 skirt piles.
Figure 1. General view of scaled model SP1
5
In the forced vibration tests, one vibration generating system having an eccentric
mass shaker is used to excite the structure. This system can apply harmonic excitation
across a wide frequency range in one or two horizontal directions and can induce weak
to strong forced vibration to structures. The eccentric mass shaker is capable to impose
and hold a frequency in the range 0.0 to 10 Hz within 0.1 Hz intervals. The shaker is
installed in the second story of the platform deck and it is fixed to rigid plates of deck.
The response of the scaled model of SP1 platform to the frequency sweep kind of
excitation is measured using six two-dimensional accelerometers, four linear variable
differential transformers (LVDT's) and TMR-200 portable digital central recording
systems. The response of the scaled model of SP1 platform in two directions (Rows A&
C and Rows1& 2) is measured in several stages. The first stage of the test is conducted
by increasing the input frequency from 1 Hz to 4 Hz by 0.2 Hz intervals. In this stage,
the approximate modal frequency values are evaluated. In the next stage of force
vibration testing program, the exact value of natural frequencies and corresponding
mode shapes is evaluated. In these stages, the frequency of excitation is increased
around the preliminary resonant frequencies by a 0.1Hz intervals and the response of the
platform is measured using accelerometers.
Figure 2. Picture of scaled model, SP1 installed on soil-pile-supports.
2.2. Description of Test Cases
In order to change vertical bracing members of SP jacket, these members are fabricated
using flange type connection plates. Four different vertical bracing configurations are
selected for performing forced vibration tests. First case is the base case in which
members configuration is same as main design configuration of platform. In case No. 2
four vertical braces are added to top bay of jacket as shown in figure 3. In cases No.
3&4 vertical bracing of third and second bay of the jacket was removed as shown in
figure 3 in order to simulate damages may occur during platform life.
6
Figure 3. Four bracing configuration cases used in force vibration test
2.3. Numerical Modelling Description
Dynamic characteristics of the scaled platform SP1 is also computed numerically using
three- Dimensional modelling with ABAQUS software. The superstructure elements,
including legs, vertical and horizontal bracing members and deck beams are modelled
using two nodes beam elements. Three dimensional solid elements are applied to model
piles and soil media. Mohr-Coulomb geotechnical constructive model is assumed for
modelling soil material behaviour. The soil-pile interfaces are assumed as a frictional
interface where soil-pile slipping and gapping may occur. Generally, Coulomb’s law of
friction is used to model slipping and gapping in the soil and pile. If interface surface is
in contact, full transfer of shear stress is ensured and separation occurs when there is
tension between the soil and pile interface. Thus, in model using ABAQUS, the contact
7
constraint is applied between two surfaces of pile and soil. The surfaces are separated
when the contact pressure between them becomes zero or negative, and the constraint is
removed. As the system is subjected to a small force which does not induce slip, for the
tangential component, rough interaction is assumed between surfaces. Figure 4 is shown
3D model of the prototype in ABAQUS.
Figure 4. Three-Dimensional Model of SP1 in ABAQUS
3. Results
In this section, results of experimental and numerical analyses are presented in term of
dynamic characteristics. Response of scale model to excitation is derived from
experimental data and ABAQUS model analysis result. To identify modal
characteristics of the scaled platform SP1, signal processing method was used based on
the power spectral density function. Modal frequencies and mode shapes are also
obtained numerically using ABAQUS outputs.
3.1. Natural Frequencies of the Platform in different cases
Natural frequency of the scaled platform, SP1 is obtained using power spectral density
(PSD), cross correlation spectrum (CPS), phase relationship, and the Coherence
Spectrum (CS), (Equation 1 through 4). The CPS, phase relationship and CS are
estimated between all response measurement points and one reference point. The
segment averaging method (also known as Welch’s method) was used for better
correlation and minimizing errors. The method consists of dividing time-series data into
8
(possibly overlapping) segments, computing a modified spectrum of each segment, and
then averaging the PSD and CPS estimates. Therefore, PSD and CPS are estimated by
dividing each acceleration data to eight segments and using a Hanning window with
50% overlap. To distinguish the spectral peaks representing the platform vibration
modes from those corresponding to peaks in the input spectrum, the advantage
concurrency of the CPS peak and the orthogonality condition of mode shapes are used.
That is, all points of the platform in a lightly damped mode of vibration are in-phase or
180 out- phase with each other, depending only on the shape of the normal mode.
Moreover, the large amount (approximately 1) must be occurred in the value of the
coherence spectrum at the candidate frequency. A MATLAB subroutine is coded for
this aim.
∫+∞
∞−
τπ− ττ= deRfSPSD fj
xxxx
2)()(: (1)
∫+∞
∞−
τπ− ττ= deRfSCPS fj
xyxy
2)()(: (2)
)()()( fjQfCfS xyxyxy −= (3)
⎥⎥⎦
⎤
⎢⎢⎣
⎡=θ −
)(
)()(
21
fC
fQtgf
xy
xy
xy (4)
)()(
)(:
2
2
fSfS
fSCS
yyxx
xyxy =γ (5)
In the above equation Sxx and Rxx are power spectral density and the auto-correlation of
a measurement point of the structure respectively. Sxy and Rxy are cross correlation
spectrum and the cross-correlation between two response measurement points of the
structure respectively. Cross correlation spectrum, Sxy is a complex value that is shown
by the real value Cxy and imaginary value Qxy. The phase spectrum of the CPS is
calculated from equation 2 and detect that two-point of vibrations is in-phase or out-
phase. The degree of linear association between two signals is compared by ordinary
coherence function. Two signals are completely correlated, and its function is shown
unity value, if there is not any noise during the vibration recording and there is not any
computational error in the spectral calculation. The coherence spectrum has a peak in
the resonant frequency of the platform, and its value is larger than 0.5. Figure 5 shows
power spectral density, cross spectral density, coherence spectrum a phase spectrum that
obtained from above equations for Row A&C. Tables 1 present first and second natural
frequencies of scaled model platform for four vertical bracing configurations. The result
of Finite element modelling using ABAQUS software is also presented in this table. By
comparing the result of natural frequencies of second case with first case, it is observed
that adding vertical bracing in top bay of the jacket, increased significantly natural
frequencies. This confirms efficiency of adding such an offshore installed bracing
system in order to avoid soft story mechanism in top bay of such a jacket. It can be seen
that removing vertical bracing in second and third bay of jacket has significant change
in dynamic characteristics of platform and it can be concluded that results of dynamic
measurements can be used for damage detection of such structures.
9
PSD Using Welch Method
0
20
40
60
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
Accel-T2-1
0
1
2
3
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
Accel-D2-1
0
1
2
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
Accel-D1-1
0
2
4
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
Accel-T1-1
0.0
0.1
0.2
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0Frequency (Hz)
Accel-S3
Coherence Spectra
0.0
0.4
0.8
1.2
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
(Accel-T2-1)&(Accel-T1-1)
0.0
0.4
0.8
1.2
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
(Accel-T2-1)&(Accel-D2-1)
0.0
0.4
0.8
1.2
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
(Accel-T2-1)&(Accel-D1-1)
0.0
0.4
0.8
1.2
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
Frequency (Hz)
(Accel-T2-1)&(Accel-S3)
CPS Angle
-4
-2
0
2
4
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
(Accel-T2-1)&(Accel-T1-1)
-4
-2
0
2
4
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
(Accel-T2-1)&(Accel-D2-1)
-4
-2
0
2
4
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
(Accel-T2-1)&(Accel-D1-1)
-4-2024
0.0 1.5 3.0 4.5 6.0 7.5 9.0 10.5 12.0 13.5 15.0 16.5 18.0 19.5 21.0 22.5 24.0
Frequency (Hz)
(Accel-T2-1)&(Accel-S3)
Figure 5. Power spectral density, cross spectral density, coherence spectrum and phase spectrum
for Row A&C.
10
Table 1. Measured and computed Natural Frequency of the platform at each case
Case
No. Direction
Test ABAQUS
Mode 1 Mode 2 Mode 1 Mode 2
1
Row A&C 2.75 11.02 2.95 12.03
Row 1&2 2.89 11.78 3.10 12.12
Torsion 5.49 13.93 5.64 14.25
2
Row A&C 4.08 14.12 4.55 14.90
Row 1&2 4.67 16.88 4.87 17.30
Torsion 9.34 23.97 9.82 24.52
3
Row A&C 1.35 9.04 1.60 9.44
Row 1&2 1.70 9.20 1.85 9.53
Torsion 3.07 10.64 3.10 11.04
4
Row A&C 2.07 8.34 2.27 8.72
Row 1&2 2.19 8.69 2.39 9.63
Torsion 3.83 11.02 4.01 11.25
3.2. Platform Mode shapes
For estimation of the mode shapes of scaled model platform SP1, the amplitude of PSD
is obtained for each point, and its root is calculated. The phase difference of each point
relative to the reference point is determined by phase spectrum. If the phase angle is in
the first or fourth quarter of the unit circle, that point is in-phase with reference point,
but if the phase angle is in the second or third quarter of the unit circle, that point is out-
phase with the reference point. The mode shapes are determined according to the
obtained amplitude root of PSD and the phase difference (in or out- phase). Figures 6 to
9 show mode shape of scaled model platform SP1. In these figures, results of finite
element (ABAQUS) model are also shown. By comparison of the result of mode shapes
for cases 1 and 2, simpler dynamic behaviour can be observed, in the other words,
adding vertical bracing in top bay of such a jacket and strengthening platform leads to
similar mode shapes with shear regular structures. Removing vertical bracing in second
and third bay of jacket changes mode shapes configuration according to figures 8 and 9.
Figure 6. Mode shapes of platform - case 1
Row A&C Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
Ele
va
tio
n (
m)
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row A&C Mode No. 2
0
1
2
3
4
5
6
-0.3 0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 2
0
1
2
3
4
5
6
-0.3 0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
11
Figure 7. Mode shapes of platform - case 2
Figure 8. Mode shapes of platform - case 3
Figure 9. Mode shapes of platform - case 4
3.3. Estimation of Modal Damping
Half-power bandwidth method is used for calculating the modal damping. The damping
ratio is calculated using equation 6 in which frequencies fa and fb are illustrated in
figure 10. Table 2 shows damping ratio for different modes of both boundary condition
cases. It can be seen that generally ratio of damping increases by decreasing the lateral
stiffness of structure and the difference of damping ratios in the first mode is so
significantly than the higher modes.
ab
ab
ff
ff
+
−=ζ (6)
Row A&C Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
Ele
va
tio
n (
m)
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row A&C Mode No. 2
0
1
2
3
4
5
6
-0.3 0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 2
0
1
2
3
4
5
6
-0.4 0.0 0.4 0.8 1.2
F.V. Test
N.A. -ABAQUS
Row A&C Mode No. 2
0
1
2
3
4
5
6
-0.3 0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 2
0
1
2
3
4
5
6
-0.3 0.0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row A&C Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
Ele
va
tio
n (
m)
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row A&C Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
Ele
va
tio
n (
m)
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 1
0
1
2
3
4
5
6
0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row A&C Mode No. 2
0
1
2
3
4
5
6
-0.3 0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
Row 1&2 Mode No. 2
0
1
2
3
4
5
6
-0.3 0.0 0.3 0.6 0.9 1.2
F.V. Test
N.A. -ABAQUS
12
Figure 10. Typical frequency response curve.
Table 2. Result of Modal Damping from forced Vibration Tests.
Mode 1 Mode 2
Case 1
Row A&C 0.01324 0.00409
Row 1&2 0.01689 0.00416
Torsion 0.00819 0.00287
Case 2
Row A&C 0.01203 0.00382
Row 1&2 0.01509 0.00407
Torsion 0.00536 0.00204
Case 3
Row A&C 0.02033 0.00497
Row 1&2 0.02196 0.00681
Torsion 0.01518 0.00414
Case 4
Row A&C 0.02051 0.00570
Row 1&2 0.02059 0.00546
Torsion 0.00654 0.00355
4. Conclusion
In this paper, dynamic system identification of a scaled model of steel jacket type
offshore platform newly installed in Persian Gulf are studied using experimental and
numerical simulation. The model is tested on a pile supported condition in order to
simulate real boundary conditions. Strengthening of jacket type offshore platforms by
adding vertical bracing members in top bay of jacket and weakness of such a platform
by removing vertical bracing members also studied from experimental and numerical
results. The force vibration dynamic test is conducted to identify modal properties of the
structure. Signal processing is used based on power spectral density function. The
numerical modelling of sample platform SP1 is performed using ABAQUS software.
The result of experimental analysis shows that soil-pile-structure interaction decrease
the natural frequency of structure. The same result is also observed in the other modal
properties and considering SPSI is increased the relative lateral displacement at mode
shapes and modal damping. The effects of SPSI are significantly illustrated at higher
modes compared to first mode for all the dynamic characteristics. Efficiency of offshore
installed vertical bracing members in top bay of jacket is also observed from
improvement of dynamic characteristics of structure experimentally and numerically.
The numerical results obtained from ABAQUS model matches more with experimental
observation at first mode compared to higher modes.
13
Acknowledgements
The research employed herein was sponsored under POGC (Pars Oil and Gas Company)
project No. 132 “Investigation of Structural Health Monitoring of Steel Jacket Offshore
Platforms”. The financial support of POGC is gratefully acknowledged.
References
1. Catbas, F. N., Kijewski-correa, T., Aktan, A. E., (2012). “Structural
identification of constructed facilities: Approaches, methods and technologies for
effective practice of St-Id.” ASCE SEI committee on structural identification of
constructed systems.
2. Abaqus Inc., Pawtucket, RI, Abaqus Analysis User’s manual, 2010, Version 6.8-2
3. American Petroleum Institute, Recommended practice for planning, designing and
constructing fixed offshore platforms (2000). API Recommended Practice 2A (RP-
2A), 21st ed. American Petroleum Institute, Washington, D.C.
4. Asgarian ,B. and Lesani, M., (2009) “pile–soil-structure interaction in pushover
analysis of jacket offshore platforms using fiber elements”. Journal of
Constructional Steel Research 65(1), 209–218.
5. Asgarian, B., rahman Shokrgozar, A., Shakeri Talarposhti, A., (2008) “Seismic
performance Evaluation of Jacket Type Offshore Platforms Through Incremental
Dynamic Analysis Considering Soil-Pile-Structure Interaction”, MERCEA’08,
Seismic Engineering Conference, Italy, 1787-1794
6. Bentley KJ, El Naggar MH., (2000) “Numerical analysis of kinematics response
of single piles’’. Canadian Geotechnical Journal, 37(6), 1368-1382.
7. Cai YX, Gould PL, Desai CS., (2000) ‘‘Nonlinear analysis of 3D seismic
interaction of soil–pile-structure system and application’’, Engineering Structures,
22(2), 191–9.
8. De Sortis, A., et al. (2005) ‘‘Dynamic identification of a masonry building using
forced vibration tests’’, Engineering Structures, 27(2), 155-165.
9. El Naggar MH, Novak M. (1996) ‘‘Nonlinear analysis for dynamic lateral pile
response’’, Soil Dynamic Earthquake Engineering,15, 233–44.
10. El Naggar MH, Novak M., (1995) ‘‘Nonlinear lateral interaction in pile
dynamics’’, Soil Dynamic Earthquake Engineering, 14, 141–57.
11. Ghoshal, A., Harrison, J., Sundaresan, M. J., Hughes, D., and Schulz, M. J.,
(2001) “Damage Detection Testing on a Helicopter Flexbeam”. Journal of
Intelligent Material Systems and Structures, 12(5), 315-330.
12. Idichandy, V.G., Ganapathy, C., (1990) “Modal Parameters for Structural Integrity
Monitoring of Fixed Offshore Platforms”. Experimental Mechanics, 30(4), 382-
391
13. Ivanovic, S. S., et al. (2000). ‘‘Ambient vibration tests of a seven-story reinforced
concrete building in Van Nuys, California, damaged by the 1994 Northridge
earthquake’’, Soil Dynamics and Earthquake Engineering 19(6), 391-411.
14. Kunert, J., Nigbor R. (2012) “Force vibration testing to investigate structure soil
interaction”. seventh Nevada Undergraduate Research Symposium (NURS’12),
University of Nevada, Reno Joe Crowley Student Union Reno, Nevada USA.
14
15. Maheshwaria, B.K., Trumana, K.Z., El Naggarb, M.H. and Gould, P.L., (2004)
“Three-dimensional nonlinear analysis for seismic soil–pile-structure interaction’’
Soil Dynamics and Earthquake Engineering, 24, 343–356.
16. Mangal, L., Idichandy, V.G., Ganapathy, C., (2001), “Structural monitoring of
offshore platforms using impulse and relaxation response”. Ocean Engineering,
28, 689–705.
17. Memarpour, M.M., Kimiaei, M., Shayanfar, M., (2011), “Simplified numerical
model for analysis of offshore piles under cyclic lateral loading”. Frontiers in
Offshore Geotechnics II, The Netherlands, 531-536.
18. Mizuno H. (1987) “Pile damage during earthquake in Japan (1923-1983).”
Dynamic response of Pile Foundations, Geotech. Special Publ. No: 11, ASCE, 53-
78.
19. Mohiuddin, M. A. and Khulief, Y. A. (2002) “Dynamic Response Analysis of
Rotor-Bearing Systems with Cracked Shaft”. Journal of Mechanical Design,
Transactions of the ASME, 124(4), 690–696.
20. Nogami T, Konagai K. (1986) ‘‘Time domain axial response of dynamically
loaded single piles’’, J Engng Mech ASCE, 112(11), 1241–52.
21. Nogami T, Konagai K. (1988) ‘‘Time domain flexural response of dynamically
loaded single piles’’, J Engng Mech ASCE, 114(9), 1512–25.
22. Nogami T, Otani J, Konagai K, Chen HL. (1992) ‘‘Nonlinear soil–pile interaction
model for dynamic lateral motion’’, J Geotech Engng ASCE, 118(1), 89–106.
23. Rovithis, E., Kirtas, E. and Pitilakis, K., (2009), “Experimental p-y loops for
estimating seismic soil-pile interaction” Bull Earthquake Eng, 7, 719–736
24. Ruotolo, R., Surace, C. and Worden, K. (2000) ‘‘Application of two damage
detection techniques to an offshore platform’’, Shock and Vibration Digest, 32(1),
30-31.
25. Sabnavis, G., Kirk, R.G., Kasarda, M. and Quinn, D. (2004) “Cracked Shaft
Detection and Diagnostics: A Literature Review”. The Shock and Vibration
Digest, 36(4), 287–296.
26. Salawu, O. S. and C. Williams (1995) ‘‘Review of full-scale dynamic testing of
bridge structures”, Engineering Structures, 17(2), 113-121.
27. Shelley, S. J., Freudinger, L. C. and Allemang, R. J. (1993) ‘‘Development of an
on-line modal state monitor", Proceedings of the Eleventh International Modal
Analysis Conference, l581-1586.
28. Ventura, C. E., et al. (2003) ‘‘Dynamic characteristics of a base isolated building
from ambient vibration measurements and low-level earthquake shaking”, Soil
Dynamics and Earthquake Engineering 23(4), 313-322.
29. Wang, S., Kutter, B. L., Chacko, M. J., Wilson, D. W., Boulanger, R. W. and
Abghari, A. (1998). ‘‘Nonlinear seismic soil-pile structure interaction”.
Earthquake Spectra, 14(2), 377-396.
30. Wu, G. and Finn, W. D. L. (1997) ‘‘Dynamic nonlinear analysis of pile
foundations using finite element method in the time domain”, Canadian
Geotechnical Journal, 34, 44-52.
31. Yu, E., (2005). “Forced vibration testing and analytical modeling of a four-story
reinforced concrete frame building”. PhD thesis, Department of Civil and
Environmental Engineering, University of California, Los Angeles.
15
32. Andersen, P., (1997) Identification of Civil Engineering Structures using Vector
ARMA Models. Ph.D. thesis, Department of Building Technology and Structural
Engineering, Aalborg University, Denmark.
33. Bendat, JS., Piersol, AG., (1993), Engineering applications of correlation and
spectral analysis. 2nd Edition. New York, John Wiley & Sons.
34. [4] Brownjohn, JMW., (2003). Ambient vibration studies for system
identification of tall buildings. Earthquake Engineering and Structural Dynamics,
32, 71–95.
35. Van Overschee, P., De Moor, B., (1996), Subspace identification for linear
system: theory, implementation and applications. Netherlands: Kluwer Academic
Publishers.
36. Carden, E. P. and Brownjohn, J. M. (2008). “ARMA modelled time-series
classification for Structural Health Monitoring of civil infrastructure.” Mechanical
Systems and Signal Processing, 22 (2), 295-314.
37. Arici, Y. and Mosalam K. M. (2005a). “Modal identification of bridge systems
using state-space
38. methods.” Journal of Structural Control and Health Monitoring. 12, 381-404.
39. Arici, Y. and Mosalam K. M. (2005b). “Statistical significance of modal
parameters of bridge
40. systems identified from strong motion data.” Earthquake Engineering and
Structural
41. Dynamics, 34, 1323-1341.
42. Mohanty, P. and Rixen, D. J. (2006). “Modified ERA method for operational
modal analysis in
43. the presence of harmonic excitations.” Mechanical Systems and Signal Processing.
20, 114-130.
44. Mojtahedi, A., Lotfollahi Yaghin, M.A., Hassanzadeh, Y., Ettefagh, M.M.,
Aminfar, M.H., Aghdam, A.B., 2011, Developing a robust SHM method for
offshore jacket platform using model updating and fuzzy logic system, Applied
Ocean Research, 33, 398– 411.
45. Elshafey, A.A., Haddara, M.R., Marzouk, H., (2009) Dynamic response of
offshore jacket structures under random loads, Marine Structures, 22, 504–521
46. TH Tan, “Reciprocity relations for scattering of plane, elastic waves”, J. Acoust.
Soc. Am 61(4), pp 928-931, 1977.
47. BA Auld, “General electromechanical reciprocity relations applied to the
calculation of elastic wave scattering coefficients”, Wave Motion 1(1), pp 3-10,
1979.
48. B Köhler, W Müller, M Spies, V Schmitz, A Zimmer, KJ Langenberg and U
Metzko, “Ultrasonic Testing of Thick Walled Austenitic Welds: Modelling and
Experimental Verification”, AIP Conf. Proc. 820, pp 57-64, 2006.