Simulation of pile driving by 3-dimensional Finite-Element

analysis

S. Henke & J. Grabe

Institute of Geotechnical and Constructional Engineering, University of Technology Hamburg, Germany

ABSTRACT: In the present paper a three-dimensional finite-element model to simulate the pile
installation process is proposed. With this model different installation methods (driving, vibra-
tion) are simulated to examine the effects of these methods on the soil parameters around the
driven pile like void ratio or effective stresses. Furthermore the vibration of a pile in the center
of an existing pile group is examined to show the resulting effect of the installation process on
the afore driven piles. The results are compared with a pile jacking analysis. Concluding the
paper planned in-situ measurements are presented which shall provide a basis to verify the nu-
merical models.

1 INTRODUCTION

The use of piles for deep foundations is state-of-the-art for many different kinds of buildings.
Nowadays a multiplicity of quay constructions is founded on complex pile grillages with a lot
of interbreeding piles. But also in structural engineering and bridge construction piles are often
used as foundation elements.
Nevertheless the variation of soil parameters like the void ratio around the driven pile due to
the different driving methods is not fully explored. Particularly in the near field around the
driven pile extreme dilation or compaction of the ground can be expected depending on the
driving method as shown by GRABE et Al. (2005, 2006) and MAHUTKA et Al. (2006). In
these articles an axisymmetric finite-element model is used to simulate the pile driving process.
The approach of using an axisymmetric model has also been used in the simulations of MAB-
SOUT (1995, 1999, 2003) and TO & SMITH (1985). With these finite-element analyses the au-
thors are capable to predict the effect of e.g. one hammer blow, see MABSOUT (1999) or the
effects of different driving methods on the surrounding soil, see MAHUTKA et Al. (2006). But
these models are not well suited to simulate the interaction between existing structures and the
newly driven pile. To investigate these phenomena a three-dimensional analysis is needed.
Especially for quay constructions with many interbreeding piles the technique of three-
dimensional modelling of pile driving via finite-element analysis delivers a basis to investigate
the interaction of interbreeding piles.
The only experiences on this field of research are received by in-situ measurements, see
GATTERMANN et Al. (2005). GATTERMANN et Al. (2005) describe the measurement of the
deformations of inclined piles due to the driving of interbreeding driven concrete piles at dif-
ferent quay constructions in Hamburg by using an inclinometer. It is shown that the extra de-
formation due to driving interbreeding piles can reach values higher than 20 cm which can en-
danger the stability of the whole construction, see figure 1.
Because of this the present paper introduces a three-dimensional finite-element approach
which makes it possible to investigate the pile installation process and its effects on adjacent
structures.
Figure 1. Deformation of an inclined pile at Europakai in Hamburg during construction, see GATTER-
MANN (2005)

First the numerical models for the different installation methods are described. The main results
of pile jacking and vibratory pile driving on the surrounding soil are presented and compared.
Using a three-dimensional finite-element model the example of driving a pile in the centre of
an existing pile group is examined and the effects on the afore driven piles are presented con-
sidering the additional deformations and bending moments.
Concluding the paper planned in-situ measurements at the newly build container-terminal
CT IV in Bremerhaven are presented. The results of these measurements will provide the basis
to verify the finite-element models.

2 FINITE ELEMENT MODELLING

The finite-element method is used to simulate the pile installation process. The solution of the
boundary value problem is computed numerically with the commercial code ABAQUS. In the
following section the main aspects of modelling are described. Special purposes for the differ-
ent driving methods are explained in detail in the belonging sections.

2.1 Modelling of pile penetration

The penetration process into a continuous material like the soil in the present study can only
be performed using a special technique. In the axis of penetration a tube of 1.0 mm radius is
modelled. At the beginning of the analysis this tube is in frictionless contact with the surround-
ing soil. During pile penetration the pile slides over the tube and the soil separates from the
tube so that contact between the penetrating pile and the surrounding soil can be established,
see figure 2.
Figure 2. Modelling the pile penetration into the soil-continuum.

2.2 Contact Formulation

In order to cope with the large deformations which occur during the pile penetration process
a kinematic contact formulation is used. This formulation is based on a master-slave-principle
implemented in ABAQUS. In this formulation no contact elements are used so that it suits well
for problems with large deformations.
As constitutive contact law the COULOMB friction model is used throughout all analyses.

2.3 Constitutive Law

In this study the penetration into non-cohesive, granular materials like sand is investigated.
For this purpose the constitutive equation of hypoplasticity after GUDEHUS (1996) and VON
WOLFFERSDORFF (1996) is used for the soil in the analyses. Hypoplasticity suits well the
non-linear and anelastic behaviour of granular materials. This constitutive law is well suited to
describe mechanical soil properties like dilatancy, contractancy and different stiffnesses for
loading and unloading.
The rate dependant formulation depends on the current stress state T and the void ratio e.
In order to model the accumulation effects and the hysteretic material behaviour under cyclic
loading NIEMUNIS & HERLE (1997) proposed the intergranular strain δ. This enhancement is
used in the present study. Pore pressure effects have not been taken into account in this analy-
sis.

2.4 Finite Element Meshes

All finite element meshes are discretised with three-dimensional solid brick elements with
reduced integration and hourglass control. A linear approach for the deformations is used.
The far field around the soil in analyses considering dynamic pile installation (vibratory pile
driving) is modelled by using infinite elements so that ground waves can disappear nearly un-
hindered out of the model. In the case of pile jacking a static analysis is executed so that the use
of infinite elements is not necessary.
In all analyses the penetrating pile is modelled as a rigid surface which is laterally supported.
The rigid surface only contains one reference point on which the loads are applied. In figure 3
the different meshes for pile jacking and vibratory pile driving are displayed.
Figure 3. Finite-Element mesh for a) pile jacking and b) vibratory pile driving.

It is evident that only a quarter of the soil body is modelled because of symmetry to reduce
calculation time.

2.5 Integration Schemes

ABAQUS provides two different solution techniques to solve non-linear systems. On the one
hand an implicit integration scheme which is used for the calculation of the pile jacking proc-
ess. On the other hand an explicit time integration scheme is used to simulate the dynamic in-
stallation method. For this scheme the central difference rule is used to solve the non-linear sys-
tem of differential equations in the manner that the unknown solution for the next time step can
directly be calculated by knowing the solution of the previous time step. This leads to the ad-
vantage that no iteration is needed.
Both methods have already been approved in different applications, see GRABE & KÖNIG
(2004), KELM & GRABE (2004) and MAHUTKA & GRABE (2005).

3 NUMERICAL MODELLING OF PILE JACKING

3.1 Configuration
The jacked pile has a diameter D = 30 cm and is jacked displacement-controlled into the
ground. The penetration depth reaches up to 5 m into the ground. For the jacked pile contact
friction of tan δ = 0.0 is considered.
For the soil body the material parameters of Karlsruher Sand are used, see table 1. The inter-
granular strain is considered using the parameters listed in table 2.
Table 1. Soil parameters for Karlsruher Sand.
Material ϕc hs n ed0 ec0 ei0 α β
[°] [MPa]
Karlsruher Sand 30 5800 0.28 0.53 0.84 1.00 0.13 1.05

Table 2. Parameters for intergranular strain.

Material m2 m5 Rmax βr χ

Karlsruher Sand 2 5 0.0001 0.5 6.0

The void ratio of the sand is varied between a loose and a medium dense state to show the
main effects of pile jacking on the surrounding soil considering the relative density.

3.2 Results
In the following section the main results of the pile jacking analysis are described. In figure 4
the void ratio around the jacked pile after four meters of penetration in initially medium dense
Karlsruher Sand is shown. It is evident that the void ratio increases around the jacked pile up to
a distance of approximately 0.5-1.0 times the diameter of the pile. In greater distance from the
pile the soil body is compacted so that the void ratio decreases in this zone of compaction.

Figure 4: Calculated void ratio distribution after 4 m of pile jacking for initially medium dense Karlsruher
sand.
Figure 5: Void ratio distribution around the jacked pile in different depths after four meters of penetration
in initially medium dense Karlsruher Sand.

This effect can also be seen in figure 5. Up to a distance of approximately 0.5-1.0 times the
diameter of the jacked pile the void ratio increases significantly. In greater distance up to 5-10D
around the jacked pile the soil body is densified. This can be explained by strong shear forces
around the driven pile during the pile jacking process which act on the surrounding soil. As a
result delation occurs in this area of strong shear forces. Regarding the main direction of the
vertical stresses around the driven pile it can be found out that these stresses are directed on a
kind of arch from the pile toe, see figure 6.

Figure 6: Distribution of vertical stresses around the jacked pile after four meters of penetration in initially
medium dense Karlsruher Sand.
 As a result of this effect the main vertical stresses cannot be found directly around the driven
pile but in a distance of about 2-3D around the jacked pile. In this distance the soil is com-
pacted the most which can be seen in figure 5.
In figure 7 the distribution of the change in void ratio in different distances around the pile
after four meters of pile jacking can be seen. On the left side initially loose soil and on the right
side initially medium dense soil is considered.

Figure 7: Change in void ratio for different distances to the jacked pile after four meters of jacking in ini-
tially loose (left) and medium dense (right) Karlsruher Sand.

It becomes clear that for initially medium dense soil in a distance of 0.5D the void ratio in-
creases because of the dilation explained before. Only in a depth of about 4-5 m the sand is
densified. This can be explained by the great stresses around the pile toe. As seen in figure 6
the maximum of the vertical stresses is located at the bottom of the pile. This maximum of
stresses must lead to a compaction of the surrounding soil directly at the pile toe.
In contrary for an initially loose soil no loosening of the soil is noticeable. In all distances
from 0.5D to 4D the change in void ratio is negative or nearly zero so that the soil body is com-
pacted. Furthermore the area which is affected by the penetrating pile is wider for a medium
dense soil than for a loose soil. In a loose soil there are more voids so that the grains are rear-
ranged due to the soil displacement caused by the jacked pile. The denser the soil there are less
voids but the displaced volume of sand because of the jacked pile is the same so that the area of
influence is wider for a medium dense soil than for a loose soil.
Figure 8: Contour plot of radial stress after 4 m of pile jacking into initially medium dense Karlsruher
Sand.

In figure 8 the distribution of radial stress around the jacked pile for medium dense soil is
displayed. The highest radial stresses are located at the pile toe. Besides, it can be stated that
the radial stress increases around the jacked pile during pile penetration. This effect is also
shown in figure 9.

Figure 9: Radial stresses in different depths around the jacked pile after 3.5 m of pile jacking for initially
medium dense soil.
 It can be seen that the radial stress increases significantly up to a distance of more than 10D.
The lateral earth pressure around the pile shaft of the jacked pile is about 2.25 times higher as
the initial value.

Figure 10: Lateral earth pressure at the jacked pile after 4 m of pile jacking.

In figure 10 the lateral earth pressure distribution at a vertical cross section around the jacked
pile is displayed. It is evident that the radial stresses increase around the jacked pile shaft com-
pared with the initial state. Furthermore at the pile toe the radial stress increases significantly
with a peak value of about 600 kN/m². Below the pile toe the radial stresses decrease under the
K0-value as already stated at MAHUTKA et Al. (2006).

4 NUMERICAL MODELLING OF VIBRATORY PILE DRIVING

4.1 Configuration
Comparing the pile used for the pile jacking analysis the diameter of the vibratory driven pile
is also chosen with D = 30 cm. The length of the pile is l = 10 m which corresponds to a con-
crete pile with a mass of m = 1.77 t. This mass is applied at the reference point of the rigid pile.
Furthermore a static force due to the weight of the vibrator is set to Fst = 20 kN. In this study
an excitation frequency of f = 40 Hz is considered. For all cases the centrifugal force is set to
Fd = mu ru ω ² sin(ωt ) . This leads to a centrifugal force Fd = 292 kN.
In all calculations the contact friction coefficient between soil and driven pile is set to tan δ =
0.176. The void ratio of the sand is varied between a loose and a medium dense state.
The material parameters are those of the Karlsruher Sand introduced in section 3.1.
4.2 Results

Figure 11: Contour plot of the void ratio after 3 m of vibratory pile driving (f = 40 Hz) for loose soil

Figure 11 shows the void ratio distribution for an initially loose soil after 3 m of vibratory
pile driving. It can be seen that the soil is compacted directly around the driven pile. The influ-
ence of compaction is limited to an area of about 2 times the diameter of the driven pile for a
loose soil. For an initially medium dense soil this area is limited to 2.5 times the diameter of the
driven pile.
Only at the ground surface the zone of compaction reaches much farther. This is caused due
to surface waves.
In figure 12 the void ratio distribution around the vibratory driven pile in different depths for
a loose and a medium dense soil can be seen.
It is evident that the soil around the vibratory driven pile is compacted up to a distance of
about 2D for initially loose soil. Farther away from the pile no effects on the void ratio distribu-
tion can be seen. Only in a depth of 1 m the influence reaches up to a distance of 4D. This is
caused by surface waves mentioned before.
The same effects can be seen regarding the void ratio distribution for initially medium dense
soil, see figure 12.
By comparing the medium dense and the loose soil it becomes clear that the change of void
ratio is higher for the loose soil because there are more voids in loose soil compared with me-
dium dense soil so that the grains can be rearranged more during the pile installation process.
Figure 12: Void ratio distribution around the vibratory driven pile in different depths after 3.5 meters of
vibratory pile driving in initially loose and 3 meters in medium dense Karlsruher Sand.

Figure 13: Contour plot of radial stress after 3.5 m of vibratory pile driving in initially loose Karlsruher
Sand.

Figure 13 shows the radial stresses around the vibratory driven pile. It becomes clear that the
radial stresses decrease below the K0-state around the vibratory driven pile. In figure 14 the dis-
tribution of radial stresses around the vibratory driven pile in different distances is presented.
Regarding figure 14 it can be seen that the radial stresses decrease below the K0-state up to a
distance of 4D. The calculated values decrease up to a fourth of the initial value. Thus, the area
of influence is about two times bigger for the change in radial stresses than for the change in
void ratio.
Figure 14: Distribution of radial stresses around the vibratory driven pile after 3.5 m of vibratory pile
driving in initially loose Karlsruher Sand.

Figure 15: Lateral earth pressure at the vibratory driven pile after 3.5 m of vibratory pile driving.

In figure 15 a vertical cross section of the lateral earth pressure around the vibratory driven
pile is displayed. It can be found out that the radial stress increases at the pile toe up to a value
of 130 kN/m². Below the pile toe there is nearly no influence on the radial stress distribution
noticeable.
Furthermore, the before mentioned decrease of the radial stresses at the pile shaft can be seen
in this figure.

5 COMPARISON OF THE DIFFERENT DRIVING METHODS

Comparing the results of the two investigated driving methods many differences in the influ-
ence on the surrounding soil can be found out.
First of all the change in void ratio has to be mentioned. Vibratory pile driving causes higher
densification of the surrounding soil. In contrast pile jacking causes less compaction and in the
case of medium dense soil even a loosening of the soil directly around the jacked pile is notice-
able.
On the other hand the area of influence is much wider for pile jacking than for vibratory pile
driving. For vibratory pile driving the void ratio is influenced up to 2-2.5D but for pile jacking
the area of influence can be wider than 5D.
Furthermore, the radial stresses at the pile shaft can be compared. It was possible to find out
that pile jacking leads to an increase of radial stresses and this increase reaches up to more than
30D. Comparing vibratory pile driving leads to decreasing radial stresses at the pile shaft and
the influence reaches up to 4D. This is two times the radius as for the influence on void ratio
but it is much shorter than the area of influence for pile jacking.
At last the radial stresses at the pile toe show significant differences depending on the driv-
ing method. It is evident that the highest radial stresses result at the pile toe for both driving
methods. But the value is about 4.5 times higher regarding a jacked pile compared to a vibra-
tory driven pile.

6 NUMERICAL INVESTIGATION OF A PILE GROUP

Three-dimensional modelling of pile driving enables the engineer to investigate problems of a

higher complexity e.g. studying the effects of pile driving on adjacent structures. In this study
the case of driving a pile in the centre of an existing pile group using vibratory pile driving is
examined.

6.1 General description

The pile group consists of five piles, four piles have already been installed and the last pile is
vibratory driven in the centre of the existing piles, see figure 16.

Figure 16. Geometry of the investigated pile group.

Each pile has a diameter of D = 30 cm and the direct distance L between the already driven
piles and the new pile is set to L = 1 m. The driven pile is treated as a rigid surface so that no
deformations can occur. The already installed piles are modelled with a linear elastic constitu-
tive law. The parameters of concrete are chosen to model the behaviour of these piles.
Figure 17. Finite-Element mesh for the analysis of a pile group with detail of the afore driven pile.

Furthermore no soil compaction around the already driven piles is considered. The sand is
idealised using the material parameters of Karlsruher Sand as presented in section 3.1. The
relative density is ID = 0.40 which means a medium dense state.
The friction coefficient between the afore installed pile and the surrounding soil is set to tan
δ = 0.176. The newly driven pile is considered to be frictionless.
The driven pile is installed with an excitation frequency of f = 40 Hz, the static load of the
vibrator is set to FSt = 20 kN and the centrifugal force results to Fd = 292 kN.
The results of the dynamic analysis will be compared with the results received by an analysis
regarding a jacked pile next to the afore driven piles. The dimensions are the same as in the
case of vibratory pile driving but for this configuration a static analysis is accomplished so that
no infinite elements are necessary.

6.2 Finite-Element mesh

In figure 17 the finite-element mesh with dimensions for the dynamic analysis is displayed. It
can be seen that only a quarter of the soil body with one afore installed pile is modelled because
of symmetry to reduce computational time. The whole continuum is discretised with three-
dimensional solid brick-elements with reduced integration and hourglass control.
To minimize wave reflection at the boundary of the soil continuum the far field is modelled
using infinite elements.
6.3 Results
Regarding the results of the analyses before it can be expected that the afore driven pile will
not be affected by the vibratory driven pile very much. The distance of L = 1 m is about 3.5
times the diameter of the driven pile. In the numerical analyses of vibratory pile driving it could
be found out that the area of compaction reaches up to 2-2.5D.

Figure 18: Horizontal deformation an bending moment at the afore driven pile after 4 m of vibratory pile
driving.

Looking at figure 18 the deformation of the afore driven pile after a penetration depth of 4 m
for the newly driven pile is displayed. It can be seen that the deformation at the pile head is less
than 1 mm towards the newly installed pile. Furthermore there is nearly no bending deforma-
tion recognisable.
It can be seen that the pile is rotated and no translative motion of the pile occurs. This can be
explained because the area of compaction leads up to 2.5D so that the afore driven piles are
placed out of this area. Besides the ground surface settles in the region around the newly in-
stalled pile because of the pile driving process, see MAHUTKA et Al. (2006). This causes the
rotation of the afore driven pile towards the newly driven pile because there is less soil resis-
tance in this direction caused by the surface settlement.
Moreover in figure 18 the bending moment at the afore driven pile is displayed. It can be
seen that there are nearly no bending moments visible. Only at the pile head a bending moment
of about 5 kNm occurs. This bending moment is caused by the surface waves described in sec-
tion 4. These waves hinder the rotation towards the newly driven pile at the pile head of the
afore driven pile so that the bending moment is established.
In figure 19 the horizontal deformation of the afore driven pile after 4 m of pile jacking is
displayed. There are significant differences recognisable comparing this result with the afore
discussed deformations of the pile due to vibratory pile driving. On the one hand the rotational
deformation of the afore driven pile caused by pile jacking is about four times higher than the
deformation of the pile due to vibratory pile driving. Furthermore it can be found out that the
pile does not only perform a rotational but also a translative motion. It can be seen that the
afore driven pile moves 12 mm away from the jacked pile. At last a bending deformation can be
seen at the toe of the afore driven pile.
Figure 19: Horizontal deformation of the afore driven pile after 4 m of pile jacking

It can be said that the afore driven piles are more influenced by jacking a pile in the centre of
the pile group than by vibratory pile driving. This can be explained by the area of influence
which is much wider for pile jacking than for vibratory pile driving as shown in the sections 3
and 4. Regarding the distance L = 1 m ≅ 3.5D between the piles it can be seen that the afore
driven piles are located in the zone of compaction for a jacked pile. For a vibratory driven pile
they are beyond the area of main influence. In this case the afore driven piles are mainly influ-
enced by surface waves.

7 PLANNED IN-SITU MEASUREMENTS

Figure 20. Cross section of the container-terminal in Bremerhaven (bremenports GmbH).

 The numerical models have to be validated to prove the correctness of the made assumptions.
For this purpose in-situ measurements at the container-terminal CT IV in Bremerhaven are
planned, see figure 20.

7.1 Description of the planned measurements

Figure 21. Arrangement of accelerometers and strain gauges on an inclined pile.

Three-dimensional modelling of pile driving leads to the possibility of predicting the addi-
tional deformations of inclined piles in quay constructions due to driving interbreeding piles. At
CT IV in Bremerhaven one inclined pile is equipped with three tri-axial accelerometers along
the first 20 m of the pile. The first accelerometer is installed at the cross-over point with pile P2
the second at the cross-over point with pile P3 and the third between the cross-over points with
piles P3 and P4.
Additionally three strain gauges have been installed beside the accelerometers. The ar-
rangement of the sensors can be seen in figure 21.
During installation of the interbreeding piles the accelerometers measure the resulting accel-
erations at the inclined pile. Furthermore the normal forces in the inclined pile due to the driv-
ing process of the steel-piles will be measured with the help of the strain gauges. The received
data will be used to validate a model similar to CT IV in Bremerhaven.
Another part of the measuring program is the installation of one steel-pile with three tri-axial
accelerometers and three strain gauges. The accelerometers on this pile deliver the occurring
accelerations and the strain gauges record the dynamic forces at the driven pile during the in-
stallation process.
The results of the measurements can be taken to validate the Finite-Element results. On the
one hand the accelerations and dynamic forces during driving the steel-pile can be compared
with the numerically received accelerations and forces in the model presented in section 4.
Regarding the inclined pile the results of the accelerations and forces in this pile will be the
basis to calibrate a more complex FE model considering the effect of driving a pile next to an
inclined pile like it is used in pile grillages.
Figure 9. Installation of the inclined pile.

8 CONCLUSION

In the present paper a three-dimensional finite-element model to simulate the pile driving
process and its effects on the surrounding soil are presented. It is possible to show significant
differences on the soil parameters regarding the different installation methods. For example, the
change in void ratio is much higher for the soil around a vibratory driven pile than for the soil
around a jacked pile. The densification is much higher for the dynamic installation method be-
cause of cyclic and dynamic shearing effects.
In contrast, pile jacking has a much wider area of influence regarding the considered soil pa-
rameters like void ratio and radial stresses.
At the pile toe both installation methods show high lateral stresses due to stress interlocking.
The radial stresses at the pile toe are 4.5 times higher regarding a jacked pile.
The received results are comparable to those received with an axisymmetric finite-element
model as presented by MAHUTKA et Al. (2006). The main effects described in this publication
are reproducible using the presented three-dimensional finite-element model. It can be stated
that the investigation of the effects of different installation methods on the surrounding soil can
be analysed sufficiently correct with an axisymmetric finite-element model. It is preferable to
use an axisymmetric model because the calculation time increases disproportionately high using
a three-dimensional finite-element model.
The main advantage of three-dimensional modelling of pile installation processes is that the
effects on adjacent structures due to the driving process normally cannot be investigated using
an axisymmetric approach.
In the present paper driving a pile in the centre of an existing pile group is analysed. It can be
shown that jacking a pile in the centre of the pile group influences the existing piles much more
than vibrating the pile. Vibratory pile driving causes mainly rotational deformation of the exist-
ing piles but jacking causes a combination of rotational, translative and bending deformation.
The results can be explained regarding the different areas of influence comparing the different
installation methods.
Concluding the paper planned in-situ measurements are described which will provide the ba-
sis to verify the afore described numerical models.

9 ACKNOWLEGEMENT

The authors thank the “Norddeutscher Verbund für Hoch- und Höchstleistungsrechnen
(HLRN)“ for providing the possibility of performing the presented calculations on their parallel
computers.
 References

Gattermann, J., Fritsch, M., Stahlmann, J. 2005, Auswahl- und Einbaukriterien geotechnischer Messgeber
zur Bestimmung des Normalkraftverlaufs der Wand/Verankerung bei Kaimauerkonstruktionen und de-
ren Ergebnisse. Pfahl-Symposium 2005, Mitteilungen des Instituts für Grundbau und Bodenmechanik,
Technische Universität Braunschweig, Heft 80, pp. 21-44.
Grabe, J. & König, F. 2004, Zur aushubbedingten Reduktion des Drucksondierwiderstandes, Bautechnik.
82(9), pp. 569-577.
Grabe, J. & Mahutka, K.-P. 2005, Finite-Elemente-Analyse zur Vibrationsrammung von Pfählen. Bau-
technik, 82(9):632-640
Grabe, J., Mardfeldt B. und Mahutka K.-P. 2006, Zur Lastabtragung von Pfahlkonstruktionen im Hafen-
bau. HANSA, 143(2) pp. 42-51.
Gudehus, G. 1996, A comprehensive constitutive equation for granular materials. Soils and Foundation.
Volume 36, Issue 1, pp. 1-12.
Kelm, M. & Grabe, J. 2004. Numerical simulation of the compaction of granular soils with vibratory roll-
ers. In Triantafyllidis (ed), Cyclic behaviour of soils and liquefaction phenomena, pp 661-664, Rotter-
dam:Balkema.
Mabsout, E. M. et Al. 1995. Study of pile driving by finite-element method. Journal of Geotechnical En-
gineering, Volume 121, Issue 7, pp. 535-543.
Mabsout, E.M. 1999. Pile Driving by Numerical Cavity Expansion. International journal for numerical
and analytical methods, Volume 23, Issue 11, pp. 1121-1140.
Mabsout, E.M. 2003. A study on the effect of driving on pre-bored piles. International journal for nu-
merical and analytical methods, Volume 27, Issue 2, pp. 133-146.
Mahutka, K.-P. & Grabe, J. 2005. Numerical investigations of soil compaction and vibration propagations
due to strong dynamic excitation. Electronic Proceedings of Joint ASCE/ASME/SES Conference on
Mechanics and Materials –McMat 2005. Baton Rouge (USA).
Mahutka, K.-P., König, F., Grabe, J. 2006. Numerical modelling of pile jacking, driving and vibratory
driving. Proceedings of International Conference on Numerical Simulation of Construction Processes
in Geotechnical Engineering for Urban Environment (NSC06), Bochum, ed. T. Triantafyllidis, Bal-
kema, Rotterdam, pp.~235—246.
Niemunis, A. & Herle, I. 1997. Hypoplastic model for cohesionless soils with elastic strain range. Me-
chanics of cohesive-frictional materials. Volume 2, pp. 279-299.
Sieffert, J.-G. 2003, Eigenschaften und Grenzen einer einfachen Modellierung der Vibrationsrammung.
Bauingenieur, Band 78 (März 2003), pp. 117-124.
To, P., Smith, I. M. 1988. A note on finite element simulations of pile driving. International journal for
numerical and analytical methods, Volume 12, Issue 2, pp. 213-219.
Von Wolffersdorff, P.-A. 1996. A hypoplastic relation for granular materials with a predefined limit state
surface. Mechanics of cohesive-frictional materials. Volume 1, pp. 251-271.

View publication stats