B 1 Trkeletal 2020 TheEffectofNaturalFrequency
B 1 Trkeletal 2020 TheEffectofNaturalFrequency
net/publication/339414993
CITATIONS READS
2 191
3 authors:
Erol Guler
George Mason University
93 PUBLICATIONS 957 CITATIONS
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
All content following this page was uploaded by Irem Zeynep Yildirim on 07 April 2022.
The Effect of Natural Frequency on the Seismic Behavior of an 8 m High MSE Wall
Berk Turkel, S.M.ASCE1; Irem Zeynep Yildirim, Ph.D., P.E., M.ASCE2;
and Erol Guler, Ph.D., M.ASCE3
1
Graduate Student, Dept. of Civil Engineering, Bogazici Univ., Bebek, Istanbul, Turkey. E-mail:
[email protected]
2
Assistant Professor, Dept. of Civil Engineering, Bogazici Univ., Bebek, Istanbul, Turkey. E-
mail: [email protected]
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
3
Professor, Dept. of Civil Engineering, Bogazici Univ., Bebek, Istanbul, Turkey. E-mail:
[email protected]
ABSTRACT
In recent years, the use of mechanical stabilized earth (MSE) wall systems have been
becoming dramatically popular worldwide. When MSE walls are compared with conventional
retaining walls, MSE walls have various economic and aesthetic advantages. As MSE walls
have a higher tolerance for deformations in comparison to rigid retaining walls, they are much
preferred specially in earthquake-prone areas. A better understanding of their behavior under
static and seismic loading conditions is required to better predict their field performance. This
study presents various methodologies developed for calculating the natural frequency of MSE
walls. Upon estimating its natural frequency, an 8 m high prototype MSE wall was modeled
using two-dimensional (2D) finite element method (FEM) computer program PLAXIS.
Numerical analysis of the MSE wall was performed by applying input ground motions with
varying frequency content (i.e., earthquake cases). Seismic response of MSE walls was discussed
in the context of the frequency of the harmonic loading and the natural frequency of MSE walls
by comparing deformations and tensile stresses in the reinforcement for each seismic loading
case. Based on the results, the maximum acceleration amplification was observed at the
estimated natural frequency of the MSE wall and the maximum horizontal displacements were
greatest at the uppermost reinforcement.
INTRODUCTION
Soils are coupled with materials (e.g., geosynthetics, steel bars) that possess higher tensile
strength in MSE Walls that became an alternative to traditional retaining structures. According to
FHWA, these reinforcements are generally layered with a reinforcement spacing of 0.3 m to 1.0
m in soils and MSE Wall is a generic term for reinforced soils (Berg et al. 2009).
MSE Wall case history studies have revealed that MSE Walls show satisfactory performance
during seismic activities such as the 1995 Great Kobe and the 2011 Great East Japan
Earthquakes. Additionally, numerous traditionally constructed retaining walls collapsed as a
result of these earthquakes and they were rebuilt as MSE Walls (Tatsuoka et al. 2012).
Numerous MSE Walls have been constructed in recent years as they have many superiorities,
such as tolerating deformations and displaying higher resistance to seismic loads. However, in
order to perform reliable MSE design, an understanding of their behavior under earthquake
loading conditions is necessary (Koseki et al. 2006). Recent developments in software utilizing
numerical methods have made it possible to analyze the seismic behavior of MSE Walls. In
essence, the effect of design parameters on MSE Walls either under static or seismic loading
conditions have been investigated in previous studies performed; however, their behavior under
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 407
harmonic loading conditions with various frequencies have not been investigated.
As the seismic response of short-period structures is dominated by their natural frequency,
we initially aimed to find the natural frequency of the MSE Wall in this study (Hatami and
Bathurst 2000). First, several methods from the literature were investigated for estimating the
natural frequency and, then an 8 m high MSE Wall was analyzed using numerical methods to
better understand its response under earthquake loading conditions. The resonance frequency
obtained from the empirical formula and the FE analyses were compared. Reinforcement tensile
stresses and displacements of the MSE Wall were compared for various harmonic earthquake
input ground motions.
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
NUMERICAL MODEL
Numerical modeling can be used for analyzing and understanding the seismic behavior of
MSE Walls. One of the most preferred methods is the Finite Element Method (FEM) which
provides approximate solutions to boundary value problems by iteration processes. It helps to
analyze complex models to get approximate solutions by estimating the response of physical
structures under external loads. MSE Wall models were prepared in the well-known two-
dimensional finite element program PLAXIS 2D version 8.6 using plane strain model for this
study. According to a sensitivity analysis performed considering both 6-noded and 15-noded
triangular elements to compare the results, the parametric study was proceeded with 6-noded
elements to minimize the computation time. The model dimensions of the 8 m high MSE Wall
with 5.6 m long reinforcements are demonstrated in Figure 1.
Figure 1. Preliminary design of 8 meters high MSE Wall and connectivity plot of the model
MSE model had a foundation depth (Fd) of 15 m, and the distance in front of the reinforced
fill (Fw) was set about 40 m in order to place boundaries away from MSE Wall. In the model, the
reinforcement had the same length from bottom to the top of the wall and they were placed with
a vertical spacing 0.4 m.
In this study, two types of soils were defined: (i) backfill soil and (ii) natural soil. They were
both modeled with Hardening Soil (HS) Material Model as this model allows for more advanced
and realistic definition of the stiffness matrix when compared to the Mohr-Coulomb Material
Model. In the model, backfill soil was used both in the reinforced zone and the backward area of
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 408
the reinforced zone of the wall. Federal Highway Administration (FHWA) recommends an upper
limit of 34° for the internal friction of the reinforced zone unless there is a triaxial or direct shear
test result. For this reason, 34° was selected for the friction angle of the granular backfill soil
(Berg et al. 2019. The soil parameters used in the model are presented in Table 1.
Table 1. Material properties of soils in PLAXIS model
Backfill Natural Soil
Title Material Properties Symbol Unit
Soil (Foundation)
dry
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
The reinforcement elements were modeled as elastic materials, and accordingly, their axial
normal stiffness parameter was defined. The nominal stiffness, denoted by EA, is calculated by
the product of modulus of elasticity and the reinforcement thickness; therefore, its unit is force
per unit width. In PLAXIS, soil reinforcements were modeled using the geogrid feature. The
geogrid properties were based on the Tensar’s Structural Geogrid element “UX1100MSE”
(Tensar 2013). Two parameters are used to define the elastic behavior of the material: tensile
strength and strain. Axial stiffness value was defined as 600 kN/m for all of the reinforcements
to investigate the general behavior along the wall height (International Organization for
Standardization 2015). Additionally, it is important to note that 0.5 m of the reinforcements were
placed between the modular blocks.
Facing unit was modeled with line elements and defined as concrete modular block type.
Unlike soils - that require more than E and to evaluate their loading response - concrete walls,
intact rock formations or stiff volumes can be defined using Linear Elastic Model (LE). This
model is based on Hooke’s law of isotropic elasticity (Brinkgreve 2019a). Modular blocks were
modeled as a soil element with 0.5 m width and 0.2 m height, the material type was selected as
non-porous. Material properties of the facing unit are given in Table 2.
Table 2. Facing properties
Material Properties Symbol Unit Modular Block
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 409
and the geogrid and soils in the models. The strength reduction factor ( Rinter ) defines the
interface strength. For backfill and natural soil, Rinter was assigned as 1.0 which means that the
interfaces defined around the material behave rigid, it shows the same properties on the interface
as soil properties except for Poisson’s ratio. For facing units Rinter was defined as 0.7
(Brinkgreve 2019b).
For dynamic calculations, “standard earthquake boundaries” were utilized for the absorption
of stress rises at boundaries by preventing the reflection inside the soil body. In order to perform
finite element analysis, mesh generation was accomplished automatically. Considering previous
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
studies, global coarseness was selected as “fine.” The reinforced soil and 1-meter back of the
reinforced soil were refined together by refining the cluster feature one time entirely. The
approximate coarseness of the models was shown in Figure 1. The model had 2044 elements,
5315 nodes, and 6132 stress points. In the model, the groundwater table was considered below
the foundation and the initial stress of the soil occurred by the self-weight.
The staged construction method was used to perform the analysis. To execute the elastic-
plastic deformation analysis, the construction stages of the models were prepared with the plastic
calculation option. The reinforced was filled with 0.2 meter soil layers in stages, and each layer
was compacted with 15 kN/m2 distributed load extended to 5.1 m from the back of the facing.
After the plastic analysis, a stage was defined to reset the displacements to zero just before the
dynamic analysis to study net seismic displacements that occur during the dynamic load
application. The natural frequency of the modeled MSE Wall was estimated to determine the
appropriate frequency content of the input harmonic motions ground motions for the dynamic
analysis.
NATURAL FREQUENCY OF MSE WALLS
Various methods in the literature can be used to estimate the natural frequency of MSE
Walls. In order to determine the natural frequency of the prepared model, several equations that
are proposed based on these methods, can be utilized. Wu (1994) proposed an equation as
follows:
2
1 G 2 H
f11 1 (1)
4H 1 B
where f11 is the estimated natural frequency, G is shear modulus, H is wall height, B is model
width, is density, and is Poisson ratio (Bathurst and Hatami 1998).
Besides Wu (1994), other methods are also proposed by researchers, such as those by Matsuo
and Ohara (1960) and Scott (1973) in which natural frequency of the soil-wall system is
calculated by the multiplication of the frequency of infinitely-long uniform soil ( f1 ) with a
geometric factor ( GF ). GF is uniquely defined in each method and the equation of f1 is
calculated as follows (Hatami and Bathurst 2000):
1 G
f1 (2)
4H
Matsuo and Ohara (1960) proposed that the natural frequency of the wall is located between
two extreme cases of the soil-rigid wall system. They defined the first case’s geometric factor ,
GFv0 , corresponding to 0 , as given below:
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 410
8 1 v H 2
GFv0 1 (3)
1 2v B
They provided the second case’s geometric factor, GF v 0 , corresponding to v 0 , as
given below :
2
2v H
GF v 0 1 (4)
1 v B
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
Scott (1973) proposed formulation for two-dimensional soil wall systems with uniform depth
by assuming the backfill as a one-dimensional shear beam which is attached to the wall with
elastic springs. Geometric factor formulation ( GFs ) can be determined by:
2
64 1 v H
GFs 1 2 (5)
1 2v B
There are also other methods that estimate the natural frequency of the retaining walls
systems defining empirical relationships. As an example, Richardson (1978) proposed an
empirical formula for natural frequency ( f11R ) in which the formulation is directly dependent on
the wall height as given below:
38.1
f11R (6)
H
Chopra (2001) assumed a plane strain cantilever shear deformation mode and also assumed
that the center of gravity of the walls is about their mid-height. He recommended the following
relationship for the first natural period (T) of the walls (Walthall et al. 2012):
T 2 H (7)
2G
The natural frequency values estimated for 8 m MSE Wall model using the equations listed
above are summarized in Table 3. The estimated average natural frequency of the MSE Wall
from these methods is about 4.6 Hz.
Table 3. Natural frequency of the 8 m high geosynthetic reinforced MSE Wall
Method Frequency (in Hz)
Wu (1994); Bathurst and Hatami (1998) 4.41
Case 1 5.39
Matsuo and Ohara (1960)
Case 2 4.37
Scott (1973) 5.16
Richardson (1978) 4.76
Chopra (2001) 3.66
Average 4.62
DYNAMIC ANALYSIS
Since the average natural frequency of the MSE Wall was estimated as 4.62 Hz, the range of
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 411
frequencies used in the FEM analysis were selected around this value in the 3 to 6 Hz range (i.e.,
3, 4, 5, and 6 Hz). Corresponding to these frequencies, four (4) different input ground motion
record accelerograms were generated using the equation published by Bathurst and Hatami
(1998) as below:
ü t et t sin 2 ft (8)
where , , and are coefficients, and t is the time in seconds.
The harmonic loading was applied at the prescribed displacements at the bottom boundary of
the model. The peak ground acceleration (PGA) was selected as 0.2g (g= gravitational
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
acceleration) which provides PGA of 2 m/s2 ( i.e., 200 Gal) and the input duration was set to 5
seconds. Each generated accelerogram was applied to the model. The accelerogram generated for
the Seismic Loading with 4 Hz frequency (i.e., SL=4 Hz) is shown in Figure 2.
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 412
for the reinforcements are presented with absolute values to ease the comparison.
Several points in the model were selected for detailed time-history results prior to the
analysis. One of those points (J), is located 1.2 m below the top of the wall and 0.8 m away from
the modular blocks (see Figure 1). For each seismic loading case (i.e., frequency contents
analyzed), frequency versus acceleration amplification factor is plotted for Point J in Figure 3.
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 413
reinforcement that is 0.4 m below the top of the wall. In Figure 5a, frequency versus extreme
horizontal displacement (absolute) at uppermost reinforcement values are shown. The maximum
horizontal displacements at the uppermost reinforcement was observed for SL=4 Hz and were in
the 2.01 cm to 4.52 cm range. In Figure 5b, frequency versus extreme axial force (envelope)
values at the uppermost reinforcement is illustrated. The maximum values were in the 5.21 kN/m
to 7.53 kN/m range for all seismic conditions and the maximum axial force was attained for
SL=4 Hz. As the maximum acceleration at top of the wall is for SL=4 Hz, extreme horizontal
displacements and axial forces also occurred for this frequency.
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
Figure 5. (a) Frequency versus extreme horizontal displacement (absolute) and (b)
Frequency versus extreme axial force (envelope) at the uppermost reinforcement of MSE
Wall
Figure 6. (a) Frequency versus extreme horizontal displacement (absolute) and (b)
Frequency versus extreme axial force (envelope) at the reinforcement at about the mid-
height of MSE Wall
Figure 6a and 6b show frequency versus extreme horizontal displacement and frequency
versus axial force (absolute) for the reinforcement located at the height of 3.6 m (corresponding
the reinforcement at about mid-height of the MSE Wall), respectively.
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 414
At about the mid-height of the MSE Wall, extreme horizontal displacements varied in the
1.19 cm to 3.58 cm range and extreme axial force values varied between 10.59 kN/m to 11.65
kN/m.
CONCLUSIONS
Numerical analyses were conducted using FEM program PLAXIS 2D (dynamic module) to
better understand the effect of harmonic loading conditions with various frequencies on the
seismic response of an 8 m high MSE Wall. Hardening-Soil model was utilized for soils. The
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
methodologies used to estimate the natural frequency of wall systems were summarized and the
natural frequency of the 8 m high model MSE Wall was estimated to aid the selection of the
loading frequencies accordingly. The following main conclusions were drawn from this study:
Based on the methods presented in the literature, the natural frequency of the 8 m high
MSE Wall was estimated in the 3.7-5.2 Hz range.
The estimated average natural frequency was about 4.6 Hz. Accordingly, the harmonic
loading frequencies were selected as SL= 3 Hz, 4 Hz, 5 Hz, and 6 Hz for the seismic
loading.
Accelerations amplified for SL= 3 Hz, SL= 4 Hz, SL=5 Hz, and de-amplified for SL=6
Hz. The results show that maximum acceleration amplification occurred for SL= 4 Hz.
These results indicate that the maximum amplification occurred for input ground motion
frequencies near the empirically determined natural frequency of the MSE Wall.
The maximum horizontal displacements (absolute) were greatest at the uppermost
reinforcement.
REFERENCES
Bathurst, R. J., and Hatami, K. (1998). "Seismic Response Analysis of a Geosynthetic-
Reinforced Soil Retaining Wall." Geosynthetics International, 5(1-2), 127-166.
Berg, R. R., Christopher, B. R., and Samtani, N. C. (2009). Design and Construction of
Mechanically Stabilized Earth Walls and Reinforced Soil Slopes- Volume I, Report No.
FHWA-NHI-10-024, US Department of Transportation, Federal Highway Administration,
National Highway Institute.
Brinkgreve, R. B. J. (2019a). PLAXIS 2D Materials Model Manual 2019, Delft University of
Technology and PLAXIS bv The Netherlands.
Brinkgreve, R. B. J. (2019b). PLAXIS 2D Reference Manual 2019, Delft University of
Technology and PLAXIS bv The Netherlands.
Chopra, A. K. (2001). Dynamics of Structures: Theory and Applications to Earthquake
Engineering, Second Edition, Prentice Hall, Upper Saddle River, NJ.
Hatami, K., and Bathurst, R. J. (2000). "Effect of Structural Design on Fundamental Frequency
of Reinforced-Soil Retaining Walls." Soil Dynamics and Earthquake Engineering, 19(3),
137-157.
Hudson, M., Idriss, I. M., and Beirkae, M. (1994). QUAD4M User’s Manual- A Computer
Program to Evaluate the Seismic Response of Soil Structures Using Finite Element
Procedures and Incorporating a Compliant Base, Center for Geotechnical Modeling,
Department of Civil and Environmental Engineering, University of California, Davis, CA.
International Organization for Standardization (2015). ISO 10319 Geosynthetics- wide width
tensile test, Third Edition, Comité Européen de Normalisation, Brussels.
© ASCE
Geo-Congress 2020
Geo-Congress 2020 GSP 316 415
Koseki, J., Bathurst, R. J., Guler, E., Kuwano, J., and Maugeri, M. (2006). "Seismic stability of
reinforced soil walls." Proc., 8th International Conference on Geosynthetics, Yokohama,
Vol. 1, 51-77.
Kwok, A. O. L., Stewart, J. P., Hashash, Y. M. A., Matasovic, N., Pyke, R., Wang, Z., and Yang,
Z. (2007). "Use of Exact Solutions of Wave Propagation Problems to Guide Implementation
of Nonlinear Seismic Ground Response Analysis Procedures." Journal of Geotechnical and
Geoenvironmental Engineering, 133(11), 1385-1398.
Matsuo, H., and Ohara, S. (1960). "Lateral Earth Pressure and Stability of Quay Walls During
Earthquakes." Proc. of 2nd WCEE, Vol. 1, 165-183.
Downloaded from ascelibrary.org by Bogazici University on 03/06/20. Copyright ASCE. For personal use only; all rights reserved.
© ASCE