Sie sind auf Seite 1von 8

JournalofofGeotechnical

Journal GeotechnicalandandTransportation
TransportationEngineering
Engineering Received 8/28/2015 Accepted
Received 8/28/2015 Accepted11/3/2015
11/3/2015
Volume1 1| Issue
Volume | Issue2 2 Published12/15/2015
Published 12/15/2015
Lateral
LateralSoil-Pile
Soil-Pile Stiffness Subjected
Stiffness to Vertical
Subjected and and
to Vertical CorrespondingAuthor:
Corresponding Author:
Lateral
Lteral Loading
Loading R.R.Ziaie
ZiaieMoayed;
Moayed;reza_ziaie_moayed@yahoo.com
reza_ziaie_moayed@yahoo.com
Taheri
Taherietetal.al. www.jgtte.com
www.jgtte.com

Lateral Soil-Pile Stiffness Subjected to Vertical and


Lateral Loading
Omid Taheri
M.S., Imam Khomeini International University, Civil Eng. Dept., Qazvin, Iran
Email: o.taheri@yahoo.com

Reza Ziaie Moayed


Associate Professor, Imam Khomeini International University, Civil Eng. Dept., Qazvin, Iran
Email: reza_ziaie_moayed@yahoo.com

Mohammadamin Nozari
Ph.D. Student, Imam Khomeini International University, Civil Eng. Dept., Qazvin, Iran
Email: mohammadaminnozari@gmail.com

Abstract shallow depths. Single piles or piles in groups are often subjected
to both axial and lateral loads. Such foundations are required to be
Soil-structure interaction is one of the most important factors in the analyzed for combined vertical and lateral loads. Piles which are
analysis; especially for complicated problems such piles are subjected to lateral loads in offshore structures, waterfront
subjected to different loads. Owing to the sensitivity of pile design structures, bridges, buildings, industrial plants, locks, dams, and
and analysis to geometry, loading type and soil behavior, it is retaining walls. Piles used to stabilize slopes are also subjected to
unavoidable to consider the soil-structure interaction. Using lateral loading. Lateral load–deflection response of long piles is
subgrade reaction approach is one of the most practical method in commonly predicted using subgrade reaction approach, although
predicting the deflection and bearing capacity of piles subjected to other approaches are available. In subgrade reaction approach, it is
lateral loads. Therefore, it is necessary to predict the exact value of assumed that soil acts as a series of independent linear elastic
modulus of subgrade reaction. Many full-scale pile test have been springs. This method has the advantage that it is relatively simple
performed in order to investigate the behavior of lateral resistance and factors like soil nonlinearity, variation of subgrade reaction
on cast-in-place concrete piles. Finite element (FE) methods also with depth, and soil layers can be incorporated relatively easily.
can be applied to predict the variation of horizontal subgrade Several empirical coefficients developed from pile test are used in
reaction (�ℎ ) in three-dimensional analysi. Accordingly, the FE the analysis of piles using subgrade reaction modulus approach.
Palmer and Thompson’s [1] formula for subgrade reaction kh was
results are compared with the results of some loading tests
documented in the literature. The obtained results from numerical used to interpret the instrumented results of the lateral load test.
analyses are found to be in good agreement with the experimental Based on analysis of 22 full-scale lateral load tests on piles of
measurements. The purposes of this paper are to estimate the different materials embedded in different types of soils, Mwindo [2]
modulus of horizontal subgrade reaction (kh) for single piles by developed five empirical relations to estimate the strain dependent
back-analysis method and the effect of vertical load, pile length and modulus of horizontal subgrade reaction kh. This approach was
amount of applied loads on the kh value. The results of the back- further analyzed by Kumar [3] and Prakash and Kumar [4] to
analysis method are compared to those obtained using existing predict the load–displacement relationship for piles under lateral
relationship. Moreover, the results show FE method can be helpful loads. Kumar et al. [5] proposed khmax approach, which is defined as
to estimating soil-pile interaction behavior. a low strain modulus of horizontal subgrade reaction. Chin et al. [6]
presented the results and interpretation of a lateral load test on a
Keywords: Modulus of horizontal subgrade reaction, Back-analysis fully instrumented spun pile in soft ground for the land viaduct
method, Lateral pile behavior, Soil-pile interaction section of a high speed train project. The test results were further
analyzed to interpret the subgrade reaction profile (�� ) along the
pile depth. Generally, interpreted subgrade reaction profile had
1. Introduction shown a peak at about five to six meters depth for all loading except
Pile foundations are often proposed as supporting structures, where for initial loading up to 20kN. This was due to the horizontal soil
adequate bearing capacity is not available for the designers at pressures not fully developed. In addition, the initial interpretation

30
30
Taheri et al.

using subgrade reaction did not show any visible trend with applied simplifications without creating an overly conservative or incorrect
lateral load. Iftekharuzzaman and Hawlade [7] examined the design. Representing the full spectrum of three-dimensional soils
behavior of steel pipe pile under pure lateral load in sand subjected foundation-structure interaction is a laudable goal, but rarely
to lateral load by (FE) analysis. achievable within the present design environment, hardware, and
The subgrade reaction approach treats a laterally loaded pile as a software. As a reasonable approximation, one may model the
beam on elastic foundation Fig. 1 [8]. It is assumed that the beam foundation system using sophisticated geotechnical software
is supported by a Winkler soil model, according to which the elastic (Plaxis, Midas, FLAC) and produce a series of foundation response
soil medium is replaced by a series of infinitely closely spaced, curves which can then be approximated in the structural design
independent and elastic springs. According to Winkler’s model with simpler elements and material behaviors [10].
assumption the stiffness of the springs, kh can be expressed as In this study the back-analysis method is suggested to predict
follows: modulus of horizontal subgrade reaction in single piles. The back-
� analysis method is a way in which pile and soil are modeled in
�� = (1) numerical method. Then the pile displacement-depth and soil

reaction-depth curves can be achieved from outputs. By dividing
Where p is the soil reaction per unit length of pile and y is the the soil reaction to displacement in any depth of pile, the horizontal
deflection. subgrade reaction curve is achieved. This approach could calculate
kh for using in structural software.

2. Analysis method
2. 1. Back-analysis method
The coefficient of horizontal subgrade reaction, �� , is influenced
by many factors among which are the pile deflection, the effective
overburden pressure, the relative density of soil, groundwater
condition, nature of applied load (static or dynamic), and the
properties and shape of the pile section. To calculate the modulus
of subgrade reaction there are three ways: Full scale pile test
program; Plate load test; empirical relationships. Among these
methods, the full scale pile test is accurate. But it’s time consuming
and is cost a lot. Therefore it cannot be used in all projects.
Empirical relationships are less accurate. Because in many of them
a series of simple assumption are used. In this study, the back-
Fig. 1. Subgrade reaction Model of (1) actual soil reaction on pile analysis method is proposed to predict the subgrade reaction
and (2) Elastic spring model of soil reaction after Prakash and modulus. In back-analysis method, soil and pile with considering
Sharma (1990) [8] interactions between them and all of the conditions are completely
The behavior of a laterally loaded pile can thus be analyzed by modeled in numerical software. After numerical analysis the soil
using the equation of an elastic beam supported on an elastic reaction-depth curve and pile deflection-depth curve will be
foundation and is given by the equation; obtained. By dividing soil reaction to pile deflection for each depth,
the coefficient of horizontal subgrade reaction curve will be get. In
�� � this paper the numerical modeling was performed by using
�� +� = 0 (2)
�� � ABAQUS software packages.
Where E is the modulus of elasticity of the pile material and I is the 2. 2. The numerical approach
moment of inertia of the pile.
Eq. (2) can be rewritten as: To investigate a reinforced concrete (R.C.) pile under vertical and
horizontal loading, the finite element software Abaqus [11] is used
�� � �� � in this study. In this software, several constitutive models are
+ =0 (3)
�� � �� involved to model the behavior of the pile and soil. However, the
choice of a constitutive model depend on the material parameters
An exact solution of the fourth order differential equation is not that are available for the case study. Considering that, the
always available as it depends on the variation of modulus of constitutive models used in the present study are relatively simple
subgrade reaction, k, with deflection. Matlock and Reese [9] and require few material parameters as input data. These parameters
performed dimensional analysis and suggested relations to can also be easily obtained from conventional geotechnical and
determine pile deflection, pile rotations, bending moments, shear structural tests.
forces, and soil reactions. The p–y curve approach incorporates the A linear elastic perfectly plastic model with Mohr–Coulomb failure
nonlinear behavior of soils in the analysis of laterally loaded piles criterion and flow rule of nonassociated type is considered for
by predicting nonlinear soil reaction–deflection relationship. modelling the soil behavior. The soil parameters required by this
Geotechnical engineers prefer to consider soils as nonlinear constitutive model are the modulus of elasticity E, Poisson’s ratio
materials with properties that vary throughout a site. These � � , shearing resistance angle � � , effective cohesion � � , and angle of
properties may also vary over time and loading conditions, yielding dilatancy �.
a highly complex set of behaviors. While this is an encouraging The pile is specified by different stress–strain relationships for
trend, structural engineers are often required to take a step back concrete and reinforcement. The constitutive model for concrete
from such a sophisticated and time-consuming approach. This is contain the plasticity theory for compressive stresses and the
often reflected in structural design software that use beam elements fracture mechanics for tensile stresses. This model is appropriate
and elasto-plastic spring elements to represent soil support for R.C. structures, which are subjected to monotonic loading. For
behavior. This paper examines the methods to achieve these
31

31
Journal of Geotechnical and Transportation Engineering - 2015 vol. 1 (2)
(2)

compressive stresses, it is supposed that concrete behaves as an than 0.3�� . For higher values of stress, plastic deformations occur
elastic plastic material with isotropic hardening and associated flow owing to the hardening behavior of the material. Failure is
rule [11], [12]. The equation of the yield surface is encountered at a strain of 1.81 �� /�� , after which the material
response is strain-softening. For tensile stresses, the behavior is
� − √3�� �� − √3�� = 0 (4) linear elastic until stress reaches the tensile strength. After this
condition, the material is characterized by a strain-softening
In which �� and � are defined, respectively, as behavior with strength that progressively reduces with increasing
1 strain to account for approximately the tension stiffening effect
�� = − �� (5) owing to the concrete-reinforcement interaction. A more detailed
3
description of the constitutive model considered in this study for
concrete can be found in the Abaqus manuals [11]. The stress-strain
� = �3�� (6)
relationship considered for the steel bars is plotted in Fig. 3.
Where �� is the first invariant of the stress tensor and �� is the
second invariant of the deviatoric stress tensor. It should be noted
that the sign convention used in the present study is that usually
accepted in mechanics of solids, according to which tensile stresses
and strains are considered positive. In addition, because of the fact
that Eq. (4) does not depend on the third invariant of stresses, this
equation should be used when concrete is subjected to low
confining stress (less than four to five times the largest compressive
stress that can be carried by the concrete in uniaxial compression
[11]). In Eq. (4), �� is a constant that could be calculated as:

1 − ���
�� = √3 � (7)
1 − 2���

Where ��� is the ratio of the compressive strength of concrete under

biaxial conditions to that under uniaxial conditions (typically ��� ≅
1.16 [11]). In addition, �� accounts for the dependency of the yield
surface on the plastic deformations owning to the hardening
parameter �� the expression of which is: Fig. 2. Stress–strain relationship considered in the present study
� for concrete under uniaxial conditions
��
�� = �� �� (8) As shown, the behavior of steel is elastic perfectly plastic and it is
(1 + )( − 1)
� √� identical for tension and compression. The material parameters
Where required by this latter constitutive model are Young’s modulus �� ,
Poisson’ ratio �� , and yield strength of steel �� . The reinforcement
√�

��� �√3 − �� � + (�� − ) is assumed to be smeared on a pipe which is embedded in concrete
�� = 9 � (9) and is perfectly bonded to it.
� �
��� ��� − √3� + ��� (2√3 − 4�� ) Finally, a shear-strength criterion of Mohr–Coulomb type is

�� is the plastic strain measured in a uniaxial test, and ����
is the imposed at the soil–pile interface. For cohesionless soils, this
criterion is expressed by the following equation:
ratio of the plastic strain at failure under biaxial condition to that

under uniaxial conditions (typically ��� ≅ 1.28 [11]). The �� − �� ���� = �� ��� � (11)
relationship can be determined using Eq. (8) along with Eq. (4) that
under uniaxial conditions leads to Where ���� is limiting shear force, �� is the normal force, � is the
friction angle at soil-pile interface. In addition, separation between
1 �� pile and soil occurs when tensile force develops at the soil–pile
�� = ( − )�� (10)
√3 3 interface.

In which �� is the stress associated with �� in an uniaxial
compression test. Specifically, �� is calculated as a function of ��
� 3. Analysis of case study
using Eq. (8), and the associated value of �� is calculated as a The behavior of a long circular pile during some horizontal loading

function of �� (which is experimentally related to �� ) using Eq. tests and a short (rigid) circular pile subjected lateral and vertical
(10). After peak strength is reached, a strain-softening behavior is loading documented in the literature is analyzed. For each case
considered to simulate the strength reduction caused by concrete study, the available experimental results from these tests are
crushing. compared with those predicted by the numerical approach.
The main advantage of the described constitutive model is that it Considering that the subsoil at both test sites consists of sandy
only requires, as input data, a stress–strain relationship of the soils, the analyses are performed under drained conditions.
concrete under uniaxial conditions. The relationship adopted in the
present study is shown in Fig. 2. It consists of a piecewise linear
function which is completely defined by the following constitute
parameters: Young’s modulus �� (which together with Poisson’s
ratio, �� describe the elastic behavior of concrete), compressive
strength �� , and tensile strength ��� of concrete. As shown in Fig. 2,
the behavior of concrete is elastic when compressive stress is less
32

32
Taheri et al.

Loading is applied close to the pile head which is off the ground
surface by 1 m. A comparison in terms of the horizontal
displacement-load curve at the pile head is presented in Fig. 5.
Conte et al. [14] analyzed the same pile in his researches. He plotted
the validation results in his paper. As can be seen, there is a fairly
good agreement between simulation and observation.
Table 2: Soil properties considered for simulating the load test
[13].
G �� �� ��
Layer No ��
(���) (���) (°) (°)
1 30.8 0.3 0 33 0
2 57.7 0.3 0 34 0
3 57.8 0.3 0 28 0
4 87.7 0.3 0 33 0
5 87.7 0.3 0 28 0
Fig. 3. Stress–strain relationship considered in the present study 6 87.7 0.3 0 30 0
for the steel bars.

3. 1. Load test on a long circular pile


The finite element results are verified using the full-scale test
results reported by Huang et al. [13]. To optimize the design of the
pile foundations planned for the construction of a high speed rail
system in Taiwan, full-scale load tests were performed on two pile
groups and some single piles. The piles were either bored or driven.
In this study, the results from the horizontal loading test performed
on a single pile denoted as B7 are considered. Pile B7 was a bored
R.C. pile with a diameter of 1.5 m and length of 34.9 m. The
reinforcement consisted of 52Ф32 steel bars arranged in two rings.
The structural properties of the concrete and reinforcement are
indicated in Table 1. A hydraulic jack was used to apply the lateral
load at the pile head during the test. In addition, an inclinometer
casing was attached to the longitudinal bars of the reinforcement to
measure the lateral deflection profile of the pile.
Table 1: structural properties of the pile subjected to the load test Fig. 4. Soil properties considered for simulating the load test
[13]. documented by Huang et al. [13].
Concrete Steel
�� �� ��� �� �� 3500
�� ��
(���) (���) (���) (���) (���)
32,173 0.2 -27.5 2.7 210,000 0.3 471 3000

A site investigation consisting of boreholes, penetration tests (SPT


2500
and CPT) and flat dilatometer tests (DMT) were performed before
pile installation. Two penetration tests with shear wave velocity
Depth (m)

measurements (SCPT) were also carried out. In addition, laboratory 2000


tests were conducted on the soil samples taken from the boreholes.
The soil within the 80 m depth can be classified in general as a silty 1500
sand with layers of sandy silt. These latter layers are characterized
by values of the tip resistance, �� in the order of 2–3 MPa. The 1000
average value of �� in sand is about 10 MPa. The dilatometer Present Study
modulus �� presents a similar profile to that of �� , with values of Huang et al. [13]
about 10 MPa in the silty layers, and 40–60 MPa in the sand. 500
Groundwater was found at approximately 1 m below the original Conte et al. [14]
ground surface. Three additional CPTs and DMTs were performed 0
after the pile construction, but before performing the loading test. 0 50 100 150
Comparing the results of these tests with those of the tests Lateral Displacement (mm)
performed before pile installation shows that the construction of the
Fig. 5. Comparison of measured and predicted load-displacement
bored pile B7 had a loosening effect on the surrounding soil at
curve at the pile head.
depths of less than 10 m from the ground surface [13]. On the basis
of the available data, in the present study the subsoil is subdivided 3. 2. Load test on a rigid circular pile
into six layers as shown in Fig. 4. The soil parameters used in the
analysis are shown in Table 2. It is assumed that the soil is The investigation were carried out on an experimental pile test
result at one of the construction sites in Kiev, composed to a depth
completely submerged with buoyant unit weight � � = 9��/�� .
33

33
Journal of Geotechnical and Transportation Engineering - 2015 vol. 1 (2)
(2)

of 6 m of sandy loam of very stiff consistency with a unit density 4. Results and discussion
� = 1.65 �/��� , shear parameters � = 18° and c = 18 kPa, under
lain by loam with a thickness of more than 7 m with � = A full three-dimensional FE analysis performed done for R.C. pile
1.71 �/��� , � = 14° and c = 24 kPa [15]. subjected to vertical and lateral loads in silty sand. The pile length
The Young’s modulus and the Poisson’s ratio of the top soil were is 34.9 m and 7 m respectively for long and short pile, and the pile
taken as 25,000 kPa and 0.35, respectively based on empirical diameter is 1.5 m for both of them. The bottom and the lateral sides
correlations [15]. The soil in the bottom layer was assumed to have of the domain are located far enough from the pile to avoid any
Young’s modulus of 20,000 kPa and Poisson’s ratio of 0.40. The significant boundary effect. In this connection, the suggestion by
dilation angle of the soil in both layers was assumed to be 0°. The Karthigeyan et al. [17] of locating the lateral sides and the soil’s
length and diameter of the concrete test pile considered by Karasev bottom respectively at a distance of 20B from the pile axis and
et al. were 3 m and 600 mm, respectively. The piles were reinforced L+20B (where B is the pile diameter and L is pile length) are found
over the entire length by circular three-dimensional cages with to be successful. The base of the domain is fully fixed, and the
eight 16-mm-diameter working rods of steel of class A-II. The field lateral sides are constrained by vertical rollers. To evaluate the
tests were conducted by first loading the pile in the vertical initial stress field in the subsoil before the loading test, the
direction and then the horizontal loads were applied while the coefficient of earth pressure at rest is estimated on the basis of the
vertical load was kept constant [15]. The same sequence of load value of �� = 0.72 can be assumed for any soil layer [14]. The
application was followed in the current finite element analysis. The dilatancy angle is assumed to be zero. It is also assumed that
behavior of the soil was modelled using the Mohr-Coulomb � = � � at the soil–pile interface.
constitutive model. The comparison between the finite element The finite element mesh adopted in the analyses for discretizing the
predicted and the experimented data is shown in Fig. 6. Karthigeyan soil and pile consists of 8 node linear brick reduced integration solid
et al [16] employed this pile for validation of the numerical elements with hourglass control [11]. The mesh adopted for soil,
modeling and his result is also plotted. pile and reinforcement is shown in Fig.7 a–c, respectively.
Structural properties of the pile are presented in Table 1. The soil
120 properties used in simulation are the first layer of the field that
Houang et al [13] performed load test. The geotechnical parameters
used in numerical analyses are shown in Table 3.
100
Table 3: Geotechnical properties of soil used in finite element
analysis [13].
Lateral Load (kN)

80 E � �� �� �� ��

(MPa) (���) (°) (°) (��/�� )
60 80.08 0.3 0 33 0 9

4. 1. Lateral displacement
40 Present Study
Fig. 8 shows the computed lateral displacement of the long pile with
Karthigeyan et al. [16] depth for both pure lateral loads and combined vertical and lateral
loads. As can be seen, the deformation of the pile at initial depth is
20 Karaserv et al. [15] high and after about 8 or 9 meters depth, the deformation is very
close to zero. There is no rotation around a point for long pile due
0 to high overburden pressure. The lateral deflections have reduced
0 2 4 6 8 10 12 14 16 18 20 22 considerably due to the presence of vertical loads. The data on
lateral deflections clearly shows the influence of vertical loads on
Lateral Displacement (mm) the lateral behavior of piles.
Fig. 6. Comparison of measured and predicted load-displacement Fig. 9 shows the computed lateral displacement of the rigid pile
curve at the pile head with depth for both pure lateral loads and combined vertical and
lateral loads. As can be seen there is rotation around a point for rigid
pile. The lateral deflections have decreased due to the presence of
vertical loads. The point of rotation has also moved up as the
vertical load levels increase.

Fig.7. Finite element mesh for the (a) soil domain, (b) pile and (c) reinforcement used in FE analysis
34

34
Taheri et al.

Deflection (mm) Soil reaction (kN/m2)


-10 10 30 50 70 -200 0 200 400 600 800
0 0
Ground

2
5
4

10 6

Depth (m)
8
Depth (m)

15
10

20 H=1000kN-V=8800kN
12
H=1000kN-V=0
H=1000kN-V=8800kN
H=1500kN-V=8800kN 14
25 H=1000kN-V=0
H=1500kN-V=0 H=1500kN-V=8800kN
16
H=2100kN-V=8800kN H=1500kN-V=0
30
H=2100kN-V=0 18 H=2100kN-V=8800kN
H=2100kN-V=0
35 20
Fig. 8. Lateral displacement of long pile Fig. 10. Soil reaction on long pile

Deflection (mm) Soil reaction (kN/m2)


-4 0 4 8 12 16 -400 -300 -200 -100 0 100 200 300 400
0 0
Ground

1 1

2 2
H=370kN-V=880kN
3 H=370kN-V=0
Depth (m)

3
Depth (m)

H=480kN-V=880kN
H=480kN-V=0
4 4
H=560kN-V=880kN

H=370kN-V=880kN H=560kN-V=0
5 H=370kN-V=0 5
H=480kN-V=880kN
H=480kN-V=0 6
6
H=560kN-V=880kN
H=560kN-V=0
7 7
Fig. 9. Lateral displacement of short pile Fig. 11. Soil reaction on Short pile

4.2 Soil reaction 4. 3. Influence of applied loads changes on kh values


Lateral soil reaction (force per meter length of the pile) is plotted in By changing lateral load values, soil reaction and pile deflection
Fig. 10 and Fig. 11 for long and short pile respectively. In FE will be change. Therefore, according to Eq. (4), the subgrade
analysis, the x-component (lateral) of nodal force is calculated first reaction modulus definitely will change. Fig. 12 shows subgrade
for all the nodes at a given depth. Dividing the sum of the nodal reaction-depth curve for some values of lateral loads for long pile.
force in the x-direction by the vertical and horizontal distance By comparison the curves, it can be seen that with increasing lateral
between two sets of nodes in the pile, the lateral soil reaction is loads, the horizontal subgrade reaction modulus is reduced.
obtained. After reaching to the maximum value of soil reaction, it
decreases quickly with depth. The maximum soil pressure is Because by increasing the lateral load, the pile deflection increased.
developed at greater depth for larger value of lateral load. So based on Eq. (1), subgrade reaction modulus will be decreased.
Fig. 13 shows the effect of vertical and lateral loads on subgrade

35

35
Journal of Geotechnical and Transportation Engineering - 2015 vol. 1 (2)
(2)

reaction modulus for short pile. These curves show that presence of Subgrade reaction modulus (MN/m3)
vertical load reduces the subgrade reaction modulus, and with 0 20 40 60 80 100 120
increasing the vertical load value, there will be more reductions. 0
Therefore, vertical load plays an important role in the estimate of
subgrade reaction modulus and it cannot be ignored in practice. H=370kN - V=880kN
4. 4. Comparison with empirical relationships Ground surface
H=370kN - V=0
0.5
Most engineers use subgrade reaction approach in practice. For H=480kN - V=880kN
calculating the kh values, they use empirical relationships. But these H=480kN - V=0
relations have limitations and do not take into account all the
H=560kN - V=880kN
factors. Here the comparison between some empirical relationships 1
which are most useable in practice with back-analysis method is H=560kN - V=0
presented. The coefficient of horizontal subgrade reaction can be
obtained by means of various empirical and semiempirical

Z/D
relationships. Assuming that the coefficients of horizontal subgrade 1.5
is a function of depth Z and unit weight �, Terzaghi [18] showed
that

�� = �� ( ) (12) 2

Where �� is the constant of horizontal subgrade reaction and is
given by
2.5
�� = 2��/1.35 (13)
Where A is a constant. Various values of �� for sands are suggested
by Terzaghi [18], Reece [19], and Bowles [20]. Fig. 14 shows the
comparison between Eq. (12) values and the back-analysis 3
approach. In empirical relationships the subgrade reaction increases Fig. 13. Influence of applied loads on subgrade reaction modulus
with depth linear while the real curve is nonlinear, and by changing for short pile
the loads �� values certainly will be change which empirical
relationships does not consider it.

Subgrade reaction modulus (MN/m3)


Subgrade reaction modulus (MN/m3)
0 20 40 60 80 100 0 20 40 60 80 100
0 0
H=1000kN - V=8800kN H=1000kN - V=8800kN
Ground surface
0.5 H=1000kN - V=0 H=1000kN - V=0
0.5 Ground surface
H=1500kN - V=8800kN
H=1500kN - V=8800kN H=1500kN - V=0
1 H=1500kN - V=0 1 H=2100kN - V=8800kN
H=2100kN - V=8800kN H=2100kN - V=0
Reece et al. [19]
1.5 H=2100kN - V=0 1.5 Terzaghi [18]
Bowles [20]
2 2
Z/D

Z/D

2.5 2.5

3 3

3.5 3.5

4 4

4.5 4.5
Fig. 12. Influence of applied loads on subgrade reaction modulus Fig. 14. Back-analysis method vs empirical relationships
for long pile.

36

36
Taheri et al.

5. Conclusions [8] Prakash, S. & Sharma, H.D., Pile Foundation in Engineering


Practice, Wiley-IEEE, 1990.
The behavior of a reinforced concrete pile subjected to vertical and
lateral loads is examined by FE analysis. The response of the piles [9] Matlock, H. and Reese, L. C. , "Generalized solutions for
in silty sand soil under lateral loads is influenced by the presence of laterally loaded piles," Journal of the Soil Mechanics and
vertical loads. The presence of vertical loads decreases the lateral Foundations Division, ASCE, vol. 86, pp. 63-91, 1960.
deflection and by increasing the lateral loads, the reduction [10] Strom R. W., Ebeling R. M, "State of the practice in the
magnitude will be increased. design of tall, stiff, and flexible tieback retaining walls,"
Back-analysis method has been presented for predicting the Technical Report ERDC/ITL TR-01-1, US Army Corps of
subgrade reaction modulus. Based on the analysis the modulus of Engineers 1-225., 2001.
horizontal subgrade reaction will be decreased by increasing lateral [11] K. B. S. P. Hibbit HD, "ABAQUS theory and user’s manual,"
loads and presence of vertical loads. It also will change with 2010.
changing the pile geometry. The available empirical relationships
[12] Hillerborg A, Modeer M, Petersson PE., "Analysis of crack
are reviewed. It is shown that a wide range of value of horizontal
formation and crack growth in concrete by means of fracture
subgrade modulus can be obtained depending on the table, chart, or
mechanics and finite elements.," Cem Concr Res, pp. 773-82,
relationship used. From the available field load test data in literature
1976.
and numerical results, it is observed that among the many factors
that influence the coefficient of horizontal subgrade reaction, the [13] Huang AB, Hsueh CK, O’Neill MW, Chern S, Chen C. ,
magnitude of pile deflection and the effective overburden pressure "Effects of construction on laterally loaded pile groups,"
play a significant role. Jouranal of Geotechnical and Geoenvironmental
engineering, ASCE, vol. 127, no. 5, pp. 385-397, 2001.
[14] E. Conte, A. Troncone and M. Vena, "Nonlinear three-
References dimensional analysis of reinforced concrete piles subjected
to horizontal loading," Computers and Geotechnics, vol. 49,
pp. 123-133, 2012.
[1] Palmer, L.A. and Thompson, J.B., "The earth pressure and
deflection along the embedded lengths of piles subjected to [15] Karasev OV, Talanov GP, Benda SF, "Investigation of the
lateral thrust.," in proceedings of the second international work of single situ-cast piles under different load
conference on soil mechanics and foundation engineering, combinations," Journal of Soil Mechanics and Foundation
Rotterdam, vol 5, pp156-161, 1948. Engineering, vol. 14, pp. 173-177, 1977.
[2] Mwindo, J.M., "Strain dependent soil modulus of horizontal [16] Karthigeyan S, Ramakrishna VVG, Rajagopal K.,
subgrade reaction," MS Thesis, University of Missouri, "Numerical investigation of the effect of vertical load on the
Rolla, MO. lateral response of piles.," J Geotech Geoenviron Eng ASCE,
pp. 512-521, 2007.
[3] Kumar, S., "Non-linear load deflection prediction of single
piles in sand using a subgrade reaction approach," MS [17] S. Karthigeyan; V. Ramakrishna; K. Rajagopal, "Influence of
Thesis, University of Missouri, Rolla, MO., 1993. vertical load on the lateral response of piles in sand,"
Computers and Geotechnics, vol. 33, pp. 121-131, 2006.
[4] Prakash, S. and Kumar, S., "Nonlinear lateral pile deflection
prediction in sands," Journal of Geotechnical Engineering, [18] K. Terzaghi, "Evaluation of Coefficients of Subgrade
ASCE, vol. 122, no. 2, pp. 130-138, 1996. Modulus," Geotechnique, pp. 297-326, 1955.
[5] Sanjeev Kumar, Latika Lalvani and Maher Omar, "Nonlinear [19] L. Reece, W. Cox and F. Koop, "Analysis of Laterally
response of single piles in sand subjected to lateral loads Loaded Piles in Sand," Houston, Texas, 1974.
using K_hMax approach," Geotechnical and Geological [20] J. Bowles, Foundation Analysis and Design, New York:
Engineering, vol. 24, pp. 163-181, 2006. McGraw Hill, 1982.
[6] Chin, Ir. Tan Yean, Ir. Sew, Gue See & Chung, Ir. , [21] N. Kimura, K. Kosa and Y. Morita, "Failure Tests on Lateral
"Interpretation of Subgrade Reaction from Lateral Load Loading Cast-in-place Concrete Piles.," 1994.
Tests on Spun Piles in Soft Ground G&P," Geotechnics Sdn
Bhd, 2010.
[7] Md. Iftekharuzzaman, Bipul C Hawlade, "Numerical
Modeling of Lateral Response of Long Flexible Piles in
Sand," Geotechnical Engineering Journal of the SEAGS &
AGSSEA, vol. 44, no. 3, pp. 25-31, 2013.

37

37

Das könnte Ihnen auch gefallen