Soils and Rocks

www.soilsandrocks.com

ISSN 1980-9743
An International Journal of Geotechnical and Geoenvironmental Engineering ISSN-e 2675-5475

Validation of a 3D numerical model for piled raft systems

founded in soft soils undergoing regional subsidence
Andrea J. Alarcón Posse1# , Juan F. Rodríguez Rebolledo1 , Julián A. Buriticá García1 ,
Bernardo Caicedo Hormaza2 , Edgar Rodríguez-Rincón3  Article

Keywords Abstract
Numerical analysis In this paper a 3D numerical model using a software based on the Finite Element Method
Piled raft (FEM), was developed and validated using the results obtained in a geotechnical centrifuge
Centrifuge modeling model of a piled raft system founded in soft soils undergoing regional subsidence. The piled
Hardening soil model raft configuration had nine piles distributed in the center of the raft. The kaolin parameters
Regional subsidence were obtained, calibrated, and validated for the Hardening Soil Model (HSM), based on
Small scale laboratory triaxial and oedometer test results. Also, a single pile load test was carried out in
the centrifuge to get the resistance parameters used in the FEM. The developed numerical
model reproduced satisfactorily soil and foundation consolidation displacements due,
not only by the structural service load but also by the pore pressure drawdown. For load
distribution on piles and raft, the model reproduces with good agreement the foundation
behavior only for the structural service load, for pore pressure drawdown some adjustments
on the embedded piles elements shaft and base resistance had to be done. The developed
model allowed to identify the most sensitive parameters for this type of simulation, to define
the types and stages of analysis that had the best fit for the physical model, and to obtain
additional results to those measured in the physical model, e.g., the axial load distribution
developed along the piles and therefore the magnitude of the negative skin friction, that
is an important load that should be considered for the structural safety review of piled
foundations subjected to this complex conditions.

1. Introduction states and pore pressure, and the definition of the analysis
stages. The result of this process allows developing a better
On many occasions, commercial software based on
understanding of the sensitivity of the different parameters
numerical models are used indiscriminately for the analysis
and a more realistic analysis methodology.
of complex problems without a real understanding of the
In this paper, the case of a piled raft system used on
problem. Also, by ignoring the influence that different
soft soils undergoing regional subsidence was studied.
geotechnical parameters have on the simulation results.
According to Alnuaim et al. (2018), a piled raft is a composite
The cases of instrumented structures and/or physical models structure with three components: subsoil, raft, and piles.
in laboratory allow obtaining results closer to reality and The structural components interact with each other and
with a clearer understanding of the phenomenon. However, with the surrounding soil (pile-soil, raft-soil, and pile-raft)
these results may be limited by the number of case studies, to bear vertical, horizontal, and moment loads coming from
model dimensions, number of variables, time of assembly and the superstructure. Luo et al. (2018) refer to this system
execution, type and quantity of instruments and problems related as an effective foundation due to its efficiency in reducing
to the installation of the instruments and during monitoring. settlements and improving bearing capacity.
Regarding the numerical models, they can be calibrated and Many papers have been presented to understand the
validated through the results produced by the physical tests behavior of piled raft systems using different ways of approaching
and, at the same time, be used to obtain additional results. (field test, laboratory test, and numerical modeling). The use
This calibration and validation process is complex since it of numerical modeling has increased considerably, and it has
must consider the selection of the constitutive model, the been used as a tool that allows simulating the behavior of
adjustment of the parameters, the definition of the initial stress complex structures in real projects. Some models have been
#
Corresponding author. E-mail address: [email protected]
1
Universidade de Brasília, Brasilia, DF, Brasil.
2
Universidad de los Andes, Bogotá, Colombia.
3
Universidad Nacional de Colombia, Bogotá, Colombia.
Submitted on July 6, 2020; Final Acceptance on January 20, 2021; Discussion open until May 31, 2021.
DOI: https://doi.org/10.28927/SR.2021.053620
This is an Open Access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided
the original work is properly cited.

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 1

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

developed using different software and constitutive models magnitude of positive and negative skin fractions and point
mainly to evaluate how the pile spacing, load sharing, pile load. According to Auvinet & Rodríguez-Rebolledo (2017),
length, and diameter affect the settlement of the foundation the effect of the negative skin friction developed on piles
(Cui et al., 2010b; Lee et al., 2010; El-Mossallamy, 2008; Roy shafts should be considered for the structural safety review
& Chattopadhyay, 2011; Cho et al., 2012; van Tran et al., 2012b; and for the estimation of the long-term displacements of
Rodríguez-Rebolledo et al., 2015; Watcharasawe et al., 2015; piled foundations.
Banerjee et al., 2016; Sinha & Hanna, 2017; Zhang & Liu, Being one of the most complete constitutive models
2017; Alnuaim et al., 2017; Khanmohammadi & Fakharian, of Plaxis, the Hardening Soil Model (HSM) was chosen to
2018; Luo et al., 2018; Mali & Singh, 2018). Although some simulate the soil behavior. To complete the data needed for the
of those works consider consolidation analyses, few studies numerical simulation, new laboratory tests, and a load test in
have really focus on simulating the subsidence process in a a single pile in the centrifuge were performed. The parameters
more precise way using more accurate constitutive models obtained for the HSM were calibrated through numerical
that represent the soil behavior, which can lead to a closer modeling of the tests using the SoilTest module of Plaxis
analysis of the system’s behavior. software. Based on the evaluation and calibration of these
Geotechnical centrifuge modeling is an advanced physical parameters, a geotechnical model profile to represent the
modeling technique that provides data for investigating centrifuge experimental test is proposed. The calibration by
mechanisms of deformation and failure and for validating displacements and by loads distribution of the 3D numerical
analytical and numerical methods (Ng, 2014). Some authors model by comparison with the centrifuge test results is
have presented centrifuge tests that evaluate the influence presented and discussed. Finally, the axial loads developed
of regional subsidence in a different type of constructions along the center, border and corner piles, for the different
(Sun et al., 2008; Cui et al., 2010a; Cheng et al., 2011; stages of the problem, are presented and interpreted.
Tang et al., 2012; Wang et al., 2013; Zhang et al., 2017a, b). 2. Materials and methods
On the other hand, various researches have been done using
piled raft system on centrifuge like Thaher & Jessberger 2.1 Case study - centrifuge model
(1991), Horikoshi & Randolph (1996), Bajad & Sahu (2008), The case study is based on a centrifuge model developed
Goh & Zhang (2017), among others. van Tran et al. (2012a), by Rodríguez-Rincón (2016); Rodríguez-Rincón et al. (2020)
Rodríguez-Rincón (2016) and Rodríguez-Rincón et al. at the Geotechnical Models Laboratory of the Universidad
(2020), focused specifically on the behavior of the piled raft de Los Andes, Bogota, Colombia. The model is focused
under the effects of regional subsidence, assessing not only on the evaluation of the behavior of piled raft systems in
settlements but also load distribution. soft soils along the consolidation process generated both
The aim of this work is to develop and validate a three by the structural load and by the pore-pressures drawdown.
dimensional (3D) numerical model based on the Finite The decrease of the pore pressure value was associated with
Element Method (FEM, Plaxis 3D) capable of simulating the subsidence process induced by the extraction of water
the complex behavior of a piled raft system founded in soft from deep permeable layers (Figure 1).
soils undergoing regional subsidence. For this purpose, the The soil profile used was composed of three layers of
results obtained by Rodríguez-Rincón (2016) of a geotechnical a mixture of kaolin with water content at 1.5 times the liquid
centrifuge model were used. This model allows to identify limit, divided by two sand layers that work as a filter and a
the most sensitive parameters for this type of simulation, bottom layer as drainage. This profile is intended to represent
to define the types and stages of analysis that had the best a soft clay soil typical of the city of Bogotá. To physically
fit to the physical model, and to obtain additional results to model a piled raft foundation, a 70 g centrifugal acceleration
those measured in the physical model, e.g., the axial load was adopted due to the capacity of the modeling box (boundary
distribution developed along the piles and therefore the conditions), the size of elements sections after scaled and

Figure 1. Representation of pore pressure conditions at testing stages. Adapted from Rodríguez-Rincón (2016).

2 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

the size and capacity of the available instrumentation. a perfectly plastic linear elastic medium. The HSM is an
The configuration of the piled raft is a model with nine isotropic hardening double surface plasticity model that
piles distributed in the center of the raft, with a pile spacing gives more accurate displacements patterns for conditions
of two diameters. Table 1 summarizes the dimensions and at working load (Schanz et al., 1999). This model considers
parameters of the piled raft elements. both theories of the non-linear elasticity and the plasticity,
The test setup that was employed to evaluate the performance representing a significant advance in comparison with the basic
of the piled raft is shown in Figure 2. The instrumentation linear elastic models (LE) and the elastic-perfectly plastic
used were composed of three linear variable differential model of Mohr-Coulomb (MC). This model is available in
transducers (LVDT) on the soil and three on the raft, four the Plaxis software and was implemented by the program
piezometers and a load cell. Four piles were also instrumented initially as an extension of the MC model (Nordal, 1999).
with miniature load cells to measure the load transmitted Although the results obtained with this model are closer to
to the top piles. Important results were obtained regarding “reality”, it requires a greater number of input parameters
the piled raft behavior and were used for the present paper. that demand more experimental tests. The HSM basic
2.2 Hardening Soil Model (HSM) characteristics are given by:
The research was conducted by using Plaxis 3D • Total strains are calculated using a stress-dependent
software, which is widely used for geotechnical analysis. stiffness according to a power law (input parameter m);
As mentioned by Rodríguez-Rebolledo et al. (2019), soil • Shear hardening: plastic straining is due to primary
constitutive models have advanced significantly from basic deviatoric loading (input parameter E50 ref
);
models that idealize the soil as a linear elastic medium or • Compression hardening: plastic straining is due to
primary compression (input parameter Eoed ref
);
• Failure according to MC criterion (input parameters
Table 1. Elements dimensions of the piled raft for models with a
scale factor of 70g. c’ and φ›);
• Stiffness defined by loading and unloading/reloading
Element Parameter Model
conditions (input parameters Eurref and νur);
Raft Material Aluminum
• Non-associated flow rule assumed for shear hardening
Thickness 13 mm
(input parameter ψ);
Young’s modulus 70000 MPa
• Associated flow rule assumed for compression
Width 200 mm
hardening.
Length 200 mm
Piles Material Aluminum 2.3 Parameters determination from laboratory tests
Diameter 9 mm With the aim of numerically reproduce the behavior
Young’s modulus 70000 MPa of a pile raft foundation system and to take into account the
Length 320 mm need to determine the mechanical parameters of the HSM,

Figure 2. Distribution of the instrumentation on the centrifuge model M3. Adapted after Rodríguez-Rincón (2016).

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 3

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

it was necessary to carry out tests on a kaolin soil mixture 2.4 Calibration of parameters
whose profile represented the one proposed by Rodríguez- To calibrate the soil parameters listed in Tables 2 and 3,
Rincón (2016). In this way, it was possible to experimentally the CID triaxial and oedometer tests were modeled in Plaxis
determine the behavior of the soil in a different stress state, as using the SoilTest tool. This tool is a quick and convenient
well as the value of the axial pile resistance. The procedure procedure to simulate basic soil lab tests based on a single
described by Rodríguez-Rincón et al. (2020) was used for the point algorithm, i.e., without the need to create a complete
fabrication of the soil mixture in the experiments. The results finite element model (Brinkgreve et al., 2018). It works with
of the oedometer, triaxial tests, and the pile load test in the the inputted soil parameters obtained from a site investigation
centrifuge are presented next. to compare with the behavior as defined by the soil model
a) Oedometer tests data chosen (HSM in this case).
The oedometer tests were conducted on three samples In order to obtain suitable parameters to give the best
at different layers of the fabricated soil labeled M1, M2 and fit results, the input parameters were adjusted, as presented in
M3. Table 2 shows the calculated values of the reference Figures 5, 6 and 7, for layers M1, M2 and M3, respectively.
oedometer modulus ( Eoed ref
, Eurref, oed ) and the parameter that The results from the three layers reveal good agreements
defines the dependency level of the strains on the stress among all the stress-strain and stress path behavior for
state (m). The methodology to calculate the parameters was different confining pressure values (σ’3 = 100, 200, and
the one suggested by Surarak et al. (2012) and Rodríguez- 300 kPa). Although the M3 layer results (Figure 7) calculated
Rebolledo et al. (2019). The results are plotted in Figure 3. were not as successful as those of the M1 (Figure 5) and
b) Triaxial tests data
Three isotropically drained consolidated triaxial
Table 2. Parameters calculated from oedometer tests.
tests (CID) were conducted at the three distinct layers
ref ref
of the experiment M1 to M3. The confining pressures Layer Eur (kPa) m Eoed
(kPa) m
σ 3 used for the M1 and M2 samples were 100, 200, M1 4,976 1.13 830 0.99
300 kPa, and for the M3, σ’3 was equal to 200, 300, and M2 6,164 1.08 1,347 0.82
500 kPa. The friction angle (φ’) obtained were 25°, 22°, M3 7,707 0.91 2,214 0.5
and 18°; whereas the cohesion (c’) was 21, 40 and 1 kPa,
respectively. The reference modulus at 50% of strength Table 3. Parameters calculated from triaxial tests.
( E50
ref
) and power m determined from the CID tests using ref
Layer E50
(kPa) m ϕ’ (°) c’ (kPa)
double log scale plots are given in Figure 4. These values
are summarized in Table 3 and were also obtained following M1 1,413 0.8 25 21
the methodology described by Surarak et al. (2012) and M2 2,044 0.5 22 40
M3 843 1 18 1
Rodríguez-Rebolledo et al. (2019).

Figure 3. Oedometer Modulus versus consolidation pressure calculated from one-dimensional consolidation tests.

Figure 4. Variation in E50 with confining pressure.

4 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

Figure 5. CID triaxial and oedometer test results and their FEM simulations with HSM for layer M1.

Figure 6. CID triaxial and oedometer test results and their FEM simulations with HSM for layer M2.

Figure 7. CID triaxial and oedometer test results and their FEM simulations with HSM for layer M3.

M2 (Figure 6) layer, since they were underestimated for parameters adopted for this layer were the ones calibrated
the confining pressure of 500 kPa, nevertheless, it can be for a stiff clay by Surarak et al. (2012). The geotechnical
stated that the HSM predictions agree reasonably well with parameters for the different soil layers obtained for the
the triaxial test results. numerical simulations are presented in Table 4.
As the stress state has a significant variation throughout
2.5 Long term bearing capacity estimation
the depth, for the numerical simulation, the soil profile was
divided into several layers using the over consolidation ratio Having the load measured on the top of the piles from
(OCR) values as a criterion. Besides being an indicator of the centrifuge, it was considered fundamental to calibrate
the stress state, the OCR is one of the input parameters of the model not only with the displacements but also with
the HSM. Also, the ground-water table was considered at the load distribution, so to have a more accurate model.
3.5m of depth as originally proposed by Rodríguez-Rincón In consequence, a pile load test on an isolated pile was
(2016). Since in the process of fabrication and lowering of carried out in the centrifuge to better establish the long-term
the water level a stiff layer was formed on the surface, the parameters for the numerical model.

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 5

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

Table 4. Geotechnical parameters for the soil profile.

ref ref
z γ σ’0 σ σP c’ ϕ E50 ref
Eur Eoed k
Layer k0 k0nc m ν OCR
m kN/m 3
kPa kPa kPa kPa ° MPa MPa MPa m/h
M1 L-1 0 - 3.5 18.67 10 28 170 11.5 28 1.24 0.53 9.5 30 12 1 0.2 8x10-6 6.07
L-2 3.5 - 6 16.68 30 75 170 21 25 1.2 0.58 1.41 10 1 1 0.2 8x10-6 5.66
L-3 6 - 9 16.68 50 125 191 21 25 0.97 0.58 1.41 10 1 1 0.2 8x10-6 3.39
M2 L-4 9 - 19 17.03 81 200 223 20 22 0.92 0.62 2.4 15 1.55 1 0.2 8x10-6 2.77
M3 L-5 19-23 17.03 141 342 342 20 17 0.74 0.71 2.5 16 1.58 0.8 0.2 8x10-6 1.27
L-6 23-28 17.03 176 425 382 20 17 0.69 0.71 2.5 16 1.58 0.8 0.2 8x10-6 1.01
L-1: Crust, over consolidated high plasticity clay subject to wetting and drying cycles; L-2 to L-6: Soft clay formation, from over to normally consolidated high plasticity
saturated clay.

Table 5. Parameters of the pile. The installation of the model pile was carried out at
Parameter Value Unit lg with a compression rate of about 0.5 mm/s. This model
Axial skin resistance 11.38 kN/m was tested in two stages: first, without loading the pile till
Base resistance 205 kN stabilization of the readings so to guarantee the adherence of
the pile shaft with the soil; and second, with the subsequent
The test was performed in a cylindrical container with development of the load test. In general, each load increment
an inner diameter of 30 cm and 60 cm in height, and a model was held until the cells had reached their steady state condition
scale of 1/70 was used with a centrifugal acceleration of 70g. before another load increment was further applied. After
The instrumented pile was made of an aluminum bar with a stopping the centrifuge, vane tests were conducted at different
6 mm diameter and Young’s modulus of 70 GPa. The outer depths to check on the undrained shear strength.
diameter of this pile was 10 mm with 400 mm of length. The load and displacement data are shown in Figure 9.
The applied axial load was monitored by a central load cell, Test results are expressed in the prototype scale unless
and four extra lateral units were used to measure the axial stated otherwise. The maximum applied load was 539 kN.
load transfer along the pile shaft during the tests. Also, a Table 5 presents the input parameters that were needed for
linear variable differential transducer (LVDT) was employed the model, in terms of pile shaft and base resistance for long
to track the pile displacement during test (Figure 8). term behavior. It was observed that the pile-soil adherence

Figure 8. Centrifuge model assembly and instrumentation.

6 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

embedded beam element, it is defined as a structural object

with special interface elements providing the interaction
between the beam and the surrounding soil. The interaction
involves a skin friction as well as a base resistance, which
is determined by the relative displacement between soil and
pile. This element type was chosen instead of the volume
elements since with them it is possible to generate a mesh
with fewer finite elements, thus decreasing the analysis
time (Oliveira, 2018).
The geometry of the piled raft and the boundary
conditions of the soil body are presented in Figure 10. Their
Figure 9. Displacement and time versus load curves.
properties are listed in Table 6. The horizontal movements
in the four boundaries were fixed as well as the vertical
displacement at the lower frontier. Regarding the water
boundary conditions, it shall be noticed that the water flow
exit was restricted in the lower edge in all phases before
pore pressure drawdown.
2.7 Stages of analysis
A graphic representation of the stages of the centrifuge
test performed and the conditions of each of them are
presented in Figure 11, where time intervals are also
specified. The numerical model was analyzed in terms of
effective stresses, with drained parameters and initial drained
conditions. According to Rodríguez-Rebolledo (2011), this
type of analysis is applied to obtain stresses, strains, and
displacements before, during, and after the consolidation
Figure 10. Geometry and mesh of the proposed 3D FEM model. process, which is the purpose in the present work.
To represent the centrifuge test, the considered calculation
Table 6. Parameters of the structural elements phases are described below:
Element Parameter Value • Initial Phase: at this stage, the initial stress of the soil
Unit weight 25 kN/m3 is generated. This stress state is usually characterized
Thickness 1.147 m by an initial vertical effective stress. In Plaxis, initial
Plate (Raft) Young’s modulus 35 GPa stresses may be generated by using the K0 procedure
Width 14 m that is a special calculation method to define these
Length 14 m stresses, considering the loading history of the soil
Unit weight 25 kN/m3 (Brinkgreve et al., 2018) (Plaxis, 2018);
Diameter 0.63 m • Phase 1, construction and loading: in this phase
Embedded beams Young’s modulus 30 GPa it was simulated the construction of the piled raft
(Piles) Length 22.4 m and the application of the load along the foundation
Axial skin resistance 11.38 kN/m surface, in accordance with the experimental test.
Base resistance 205 kN A consolidation calculation was used to analyze
the development of pore pressure as a function of
time. As it is possible to apply load in this analysis,
is low, which has consequently generated a significant
a value of 38.25 kPa was applied in 5000 hours
displacement of the pile.
corresponding to the interval time from tC-tE of the
2.6 Proposed model centrifuge test, Figure 11;
To model the structural components, such as concrete • Phase 2, consolidation: in this phase, the same
piles and raft, a linear elastic constitutive model was analysis was used as in the previous one to represent
assumed. Regarding the element type used for the design the interval time from tE-tF (Figure 11) in which the
of the piled raft foundation, a plate element for the raft load has reached its maximum value. The load was
and embedded beams for the piles, were assumed. Plates maintained for more 8,914 hours;
are structural objects used to model structures in the • Phases 3 to 6, pore water pressure drawdown: these
ground with a significant flexural rigidity that does not phases correspond to the Stage 3 previously explained
allow plastification, only linear elastic behavior. As for an in Figure 11 in which a drawdown of pore pressure

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 7

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

Figure 11. Description of the test carried out in the centrifuge and the pore pressure condition in the three stages. Adapted from
Rodríguez-Rincón (2016).

Table 7. Description of the phases that simulated the drawdown

pore pressure.
Phase Consolidation degree (%) Time (hr)
3 20 606,8
4 40 1736
5 60 2868
6 88 7787

is generated, Figure 12. Table 7 summarizes each of

these phases, that correspond to a particular degree
of consolidation to be reached in a certain period.
In the last stage of the centrifuge test, drawdown phase,
the soil was brought to an 88% degree of consolidation. This
stage was divided in four parts in the numerical model, where
pore pressures were sequentially imposed to reach 20, 40,
60 and 88% of degree of consolidation. The piezometer data
obtained in the centrifuge test was measured very close to
the filter layers, so it could not be used as an initial input
in the numerical simulation to model the exact decrease of
pore pressure. Consequently, the isochrones presented in
Figure 12 were established by using the finite difference
Figure 12. Pore water pressure conditions. method.

8 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

3 Results and discussions

3.1 Calibration by displacements
The displacement-time curve for the piled raft foundation,
under vertical loading and pore pressure drawdown, obtained
from the centrifuge test is presented along with the results
obtained from FEM. Figure 13 shows the displacements
measured at a point over the soil near the raft (Es1) with
respect to time. In the first stage of the test, the results in the
prototype are reasonably close to those from the centrifuge.
Regarding the drawdown pore pressure phase, the results
move away slightly, although the tendency is similar.
The displacements on the foundation system in the model
were measured in three corners on top of the raft, labeled
as Er1, Er2 and Er3. The results are plotted in Figure 14.
The experimental centrifuged results measured in the three Figure 14. Displacements versus time curves obtained at raft corner
corners of the raft were slightly different, and this can be points Er1, Er2 and Er3 by centrifuge and FEM models.
due to a possible uneven load application, that probably
caused an unequal load distribution among the system’s
components. When comparing the model and FEM results,
it is observed that the two paths are reasonably close. Hence,
with close results presented for soil and foundation, and a
quite accurate representation of the phenomenon, therefore
is possible to consider that the numerical model is calibrated
by displacements for the centrifuge tests carried out in the
laboratory.
FEM displacement results obtained from phase
2 (consolidation with load) and the final phase 6 (consolidation
with load and decrease of pore pressure) are presented in
Figures 15 and 16, respectively. The subsidence due to
drawdown can be seen in Figure 16, where the settlements
were considerably larger than in the previous phase (Figure 15).
In phase 2 the maximum settlements reached up to 6 cm Figure 15. Vertical section at the center of the FEM model showing
in the raft region, and plastification points are observed the total vertical displacements obtained at the end of Phase 2.
at the tip of the piles. At the end of phase 6 the maximum
settlement was approximately 50 cm, which is 8 times the one
obtained in phase 2. These comparative results do evidence

Figure 13. Displacements versus time curves obtained at point Es1 Figure 16. Vertical section at the center of the FEM model showing
by centrifuge and FEM models. the total vertical displacements obtained at the end of Phase 6.

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 9

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

Table 8. Comparison between the centrifuge and FEM load values measured at the top of the piles for each stage of the test.
Stage 2 Stage 3
Element Load Load
Dif. % Dif. %
Centrifuge FEM Centrifuge FEM
Pile 2 196 kN 226 kN +15 347 kN 156 kN -122
Pile 5 173 kN 228 kN +32 393 kN 222 kN -77
Pile 6 164 kN 227 kN +38 463 kN 172 kN -169
Pile 7 215 kN 241 kN +12 352 kN 158 kN -122
All piles 1,753 kN 2,098 kN +20 3,421 kN 707 kN -384
13% 16% +23 26% 5% -420
Raft 87% 84% -3 74% 95% +28

87 to 74%). As explained by Rodríguez-Rincón et al. (2018)

this is because, in stage 3, when pore pressure drawdown
occurs, the soil continues to settle, a movement that is not
accompanied by the raft, generating an apparent emersion
process, and hence, a reduction of contact between the
soil and the raft. This phenomenon is not developing in
the FEM model because the considered skin resistance
of the piles is not enough to allow the generation of the
negative skin friction necessary for this.
The pore pressure drawdown generated in stage
3 produces an increment in effective stresses throughout the
compressible soil, leading to an increase in shear resistance.
The FEM results show that the embedded pile element does
not consider the increase of the shear resistance parameters
related to skin friction that happen when the soil is subjected
to a consolidation process. To overcome this problem, the
input parameters for this element, in stage 3, were further
Figure 17. Load versus time curves obtained at the top of the adjusted to properly “match” the centrifuge data in a sort of
instrumented piles 2, 5, 6 and 7 by centrifuge and FEM models. back-analysis. This analysis was made running the model
increasing the base and axial skin resistance gradually until
satisfactory match the data measured.
the distinct phenomena and resulting engineering behavior Figure 18 shows the load measure at the top of
that take place on a typical system founded in this type of piles 2, 5, 6 and 7 with respect to time after adjustment
environment, where loading and subsequent drawdown of of base and skin resistance of embedded piles elements.
pore water pressure can happen. It is possible to observe a behavior more like that of the
3.2 Calibration by load distribution experimental model, mainly in the magnitude of the load
The load-time curves for each instrumented pile, obtained at the end of the stage. The differences between
under vertical loading and pore pressure drawdown, models during pore pressure drawdown are mainly due to
obtained from the centrifuge test are presented along the type of drainage considered in each one. In the model
with the results obtained from FEM. Figure 17 shows the of the centrifuge this is developed in a more “efficient”
load measure at the top of piles 2, 5, 6 and 7 with respect way because it occurs through three draining layers,
to time. In the first stage of the test, the results in the obtaining a stabilization of most of the loads for a time
prototype are reasonably close to those from the centrifuge. of approximately 20,000 hours. For the numerical model,
Regarding the drawdown pore pressure phase, the values the drainage was simulated in a more “realistic” way
obtained with the FEM are considerably lower than those considering only a draining layer down to the compressible
obtained experimentally. Table 8 shows a comparison of stratum, observing its stabilization only until the end of
the results obtained with both models at the end of each the consolidation process.
stage. For stage 2 differences between FEM and centrifuge Table 9 shows a comparison of the results obtained with
models from 12 to 38% were obtained, while for stage both models at the end of each stage, after being adjusted.
3 from -77 to -169%. Centrifuge model shows that from It is possible to observe that now, for stage 3, the numerical
stage 2 to 3 load transmitted by piles increases (from 13% model is simulating the same behavior as the experimental
to 26%) while load transmitted by raft decreases (from one, presenting variations in the piles loads from -4 to 21%.

10 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

Table 9. Comparison between the centrifuge and FEM load values measured at the top of the piles for stage 3, after adjustment.
Stage 2 Stage 3
Element Load Load
Dif. % Dif. %
Centrifuge FEM Centrifuge FEM
Pile 2 196 kN 226 kN +15 347 kN 382 kN +10
Pile 5 173 kN 228 kN +32 393 kN 477 kN +21
Pile 6 164 kN 227 kN +38 463 kN 430 kN -7
Pile 7 215 kN 241 kN +12 352 kN 298 kN -15
All piles 1,753 kN 2,098 kN +20 3,421 kN 3,293 kN -4
13% 16% +23 26% 25% -4
Raft 87% 84% -3 74% 76% +3

This was only possible by increasing the base and axial skin
resistance of the embedded piles elements, during stage 3,
in about 3.5 times, this research demonstrates the limitation
in the use of this type of elements for the simulation of this
kind of problems. To avoid this problem, two solutions are
proposed:
1) determine the long-term base and skin resistance of
the embedded piles using previously a simple 2D
FEM model (axisymmetric), through the simulation
of a load test of a single pile in a medium previously
subjected to pore pressures drawdown.
2) use of volume elements for problems with a relatively
small number of piles, the use of this type of elements
can lead to exceptionally large finite element meshes
and therefore high or even irrational computational
costs (time and memory).
3.3 Obtaining the axial load along the pile Figure 18. Load versus time curves obtained at the top of the
instrumented piles 2, 5, 6 and 7 by centrifuge and FEM models,
Having the model calibrated, it was possible to obtain
after adjustment.
the variation of axial loads with depth for center, border and
corner piles (Figure 19), which allows to properly access
the negative skin friction that can be eventually generated.
For phase 2 (Figure 19a), due to the high rigidity
of the raft and to the proximity between piles, the load
transmitted to the three elements is practically the same,
slightly higher (+13 kN) for the corner one. For phase
6 (pore pressure drawdown) the model evidences the
development of negative skin friction in the three piles
(Figure 19b), being higher in the corner and lower in the
center one. As excess pore pressure dissipates, the neutral
level of the piles stabilizes at a depth between 15 to 16 m.
These results agree with those reported by Auvinet &
Rodríguez-Rebolledo (2017), they also demonstrate that Figure 19. Axial forces developed along the piles with different
positions in the piled raft (border, corner and center), for: (a) Phase
the depth of such level depends significantly on the initial
2; and (b) Phase 6.
pile load conditions. The differences between piles in the
magnitude of the axial loads is related to the corresponding The obtained results show the importance of considering
influence area, e.g., the influence area of the corner pile the negative skin friction on the pile design, the maximum
is considerably larger than other piles leading to higher axial load transmitted by the piles due to the structural service
values of negative skin friction. Also, Lee (1993) stated load is approximately 240 kN, whereas, when porewater
that negative friction at an individual pile in the group is pressure drawdown develops, the axial load increases to
smaller than in an isolated pile due to the interaction effects. 560 kN, 2.3 times higher.

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 11

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

4. Conclusions pressure drawdown produces an increment in effective stresses

In this work a 3D numerical model based on the throughout the compressible soil, leading to an increase in
Finite Element Method (FEM) capable of simulating the shear resistance. The FEM results show that the embedded pile
complex behavior of a piled raft system founded in soft element does not consider the increase of the shear resistance
soils undergoing regional subsidence was developed and parameters related to skin friction that happen when the soil
is subjected to a consolidation process. To approximately
validated by the results obtained by a geotechnical centrifuge
match the models results (variations in the piles loads from
model. This model allowed to identify the most sensitive
-4 to 21%), it was necessary to adjust the base and axial skin
parameters for this type of simulation, to define the types
resistance of the embedded piles elements in about 3.5 times,
and stages of analysis that had the best fit to the physical
demonstrating the limitation in the use of this type of elements
model, and to obtain additional results to those measured
for the simulation of this kind of problems. To avoid this
in the physical model as the magnitude of the developed
problem, two solutions were proposed:
negative skin fraction.
1) determine the long-term base and skin resistance of
For the simulation of the soft soil behavior (kaolin)
the embedded piles using previously a simple 2D
an advanced isotropic hardening double surface plasticity
FEM model (axisymmetric), through the simulation
model (Hardening Soil Model, HSM) implemented in Plaxis
of a load test of a single pile in a medium previously
software were used. The parameters for the HSM were
subjected to pore pressures drawdown;
obtained from oedometer and drained consolidated triaxial
2) use of volume elements for problems with a relatively
tests using the methodology proposed by Surarak et al.
small number of piles, the use of this type of elements
(2012) and Rodríguez-Rebolledo et al. (2019). The obtained
can lead to exceptionally large finite element meshes
parameters were satisfactorily calibrated and adjusted using
and therefore high or even irrational computational
the Soil Test tool from the Plaxis software.
costs (time and memory).
Pile shaft and base resistance for long term behavior
The model evidences the development of negative skin
were obtained from a pile load test carried out in a centrifuge
friction in the center, border and corner piles, being higher
model, as they were also needed as input parameters for the
in the corner and lower in the center one. As excess pore
chosen numerical model.
pressure dissipates, the neutral level of the piles stabilizes at
The developed numerical model reproduced satisfactorily
a depth between 15 to 16 m. These results agree with those
soil and foundation consolidation displacements due, not
reported by Auvinet & Rodríguez-Rebolledo (2017), they also
only by the structural service load but also by the pore
demonstrate that the depth of such level depends significantly
pressure drawdown (regional subsidence). For service load
on the initial pile load conditions. The differences between
the maximum settlements reached up to 6 cm in the raft piles in the magnitude of the axial loads is related to the
region. At the end of pore pressure drawdown, the maximum corresponding influence area, e.g., the influence area of the
settlement was approximately 50 cm (8 times bigger). These corner pile is considerably larger than other piles leading to
comparative results do evidence the distinct phenomena and higher values of negative skin friction.
resulting engineering behavior that take place on a typical It is finally concluded that a simulation model for
system founded in this type of environment. piled raft foundation systems founded on consolidation
For load distribution on piles and raft, the model soft strata, via loading or porewater pressure drawdown, is
reproduces with good agreement the foundation behavior only feasible with a quite reasonable approximation of the field
for the structural service load, for pore pressure drawdown behavior/site conditions. This will be extremely useful for
some adjustments on the shaft and base resistance of the future design scenarios via parametric analysis of this same
embedded piles elements had to be done. system, thus aiming to optimize the performance of this
For service loads differences between FEM and type of foundation structure when undergoing a regional
centrifuge models from 12 to 38% were obtained, while for subsidence phenomenon.
pore pressure drawdown from -77 to -169%. Centrifuge model
shows that from one stage to the other the load transmitted by Acknowledgements
piles increases (from 13% to 26%) while load transmitted by The research of the first author is possible through a
raft decreases (from 87 to 74%). As explained by Rodríguez- scholarship provided by the Brazilian sponsorship organization
Rincón et al. (2018) this is because, when pore pressure named Conselho Nacional de Desenvolvimento Científico
drawdown occurs, the soil continues to settle, a movement e Tecnológico - Brazil (CNPq), which is most appreciated.
that is not accompanied by the raft, generating an apparent The authors also want to acknowledge the Universidad de
emersion process, and hence, a reduction of contact between Los Andes - Colombia for their support for the development
the soil and the raft. This phenomenon was not developing of the centrifuge modeling, and Prof. Renato P. Cunha of
in the FEM model because the considered skin resistance the University of Brasília for the critical comments, and
of the piles was not enough to allow the generation of the diligent support provided throughout the development of
negative skin friction necessary for this. The generated pore the numerical stage of this research.

12 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

Declaration of interest El-Mossallamy, Y. (2008). Modeling the behaviour of piled

The Authors declares that there is no conflict of interest raft applying Plaxis 3D Foundation Version 2. Plaxis
that could inappropriately bias the work presented. Bulletin, (23), 1-4.
Goh, S.H., & Zhang, L. (2017). Estimation of peak acceleration
Author’s contributions and bending moment for pile-raft systems embedded in
Andrea J. Alarcón Posse: conceptualization, methodology, soft clay subjected to far-field seismic excitation. Journal
investigation, validation, writing - original draft. Juan of Geotechnical and Geoenvironmental Engineering,
F. Rodríguez Rebolledo: supervision, conceptualization, 143(11), 04017082. http://dx.doi.org/10.1061/(ASCE)
methodology. Julián A. Buriticá García: investigation. GT.1943-5606.0001779.
Bernardo Caicedo Hormaza: resources. Edgar Rodríguez- Horikoshi, K., & Randolph, M.F. (1996). Centrifuge modelling
Rincón: resources. of piled raft foundations on clay. Geotechnique, 46(4),
References 741-752. http://dx.doi.org/10.1680/geot.1996.46.4.741.
Khanmohammadi, M., & Fakharian, K. (2018). Evaluation
Alnuaim, A.M., El Naggar, H., & El Naggar, M.H. (2017).
of performance of piled-raft foundations on soft clay: a
Evaluation of piled raft performance using a verified 3D
nonlinear numerical model. Geotechnical and Geological case study. Geomechanics and Engineering, 14(1), 43-50.
Engineering, 35(4), 1831-1845. http://dx.doi.org/10.1007/ http://dx.doi.org/10.12989/gae.2018.14.1.043.
s10706-017-0212-1. Lee, C.Y. (1993). Pile groups under negative skin friction.
Alnuaim, A.M., El Naggar, M.H., & El Naggar, H. (2018). Journal of Geotechnical Engineering, 119(10), 1587-1600.
Performance of micropiled rafts in clay: numerical Lee, J., Kim, Y., & Jeong, S. (2010). Three-dimensional
investigation. Computers and Geotechnics, 99, 42-54. analysis of bearing behavior of piled raft on soft clay.
http://dx.doi.org/10.1016/j.compgeo.2018.02.020. Computers and Geotechnics, 37(1-2), 103-114. http://
Auvinet, G., & Rodriguez-Rebolledo, J. (2017). Criteria for dx.doi.org/10.1016/j.compgeo.2009.07.009.
the design of friction piles subjected to negative skin Luo, R., Yang, M., & Li, W. (2018). Normalized settlement
friction and transient loads. Ingeniería, Investigación of piled raft in homogeneous clay. Computers and
y Tecnología, 18, 279-292. http://dx.doi.org/10.22201/ Geotechnics, 103, 165-178. http://dx.doi.org/10.1016/j.
fi.25940732e.2017.18n3.025. compgeo.2018.07.023.
Bajad, S.P., & Sahu, R.B. (2008). An experimental study on Mali, S., & Singh, B. (2018). Behavior of large piled-raft
the behaviour of vertically loaded piled raft on soft clay. foundation on clay soil. Ocean Engineering, 149, 205-
In Proceedings of the 12th International Conference of 216. http://dx.doi.org/10.1016/j.oceaneng.2017.12.029.
International Association for Computer Methods and Ng, C.W.W. (2014). The state-of-the-art centrifuge modelling
Advances in Geomechanics, Goa, India. of geotechnical problems at HKUST. Journal of Zhejiang
Banerjee, S., Joy, M., & Sarkar, D. (2016). Parametric study University. Science A, 15(1), 1-21. http://dx.doi.org/10.1631/
and centrifuge-test verification for amplification and jzus.A1300217.
bending moment of clay-pile system subject to earthquakes. Nordal, S. (1999). Present of PLAXIS. In R.B.J. Brinkgreve
Geotechnical and Geological Engineering, 34(6), 1899- (Ed.), Beyond 2000 in Computational Geotechnics: ten
1908. http://dx.doi.org/10.1007/s10706-016-9999-4. years of PLAXIS International (pp. 45-54). London:
Brinkgreve, R. B. J., Kumarswamy, S. & Swolfs, W. M. Routledge.
(2018). Reference manual and material models manual
Oliveira, B. (2018). Avaliação do atrito negativo em grupos
PLAXIS. The Netherlands.
de estacas assentes em solos moles [Master’s dissertation].
Cheng, S., Zhang, G., Zheng, R.H., & Sun, Z.Y. (2011).
Universidade de Brasília.
Centrifuge modeling of response of bridge due to exploiting
Rodríguez Rebolledo, J.F., Auvinet-Guichard, G.Y., &
groundwater. Yantu Lixue, 32(6), 1781-1786.
Cho, J., Lee, J.H., Jeong, S., & Lee, J. (2012). The settlement Martínez-Carvajal, H.E. (2015). Settlement analysis of
behavior of piled raft in clay soils. Ocean Engineering, 53, friction piles in consolidating soft soils. Dyna, 82(192),
153-163. http://dx.doi.org/10.1016/j.oceaneng.2012.06.003. 211-220. http://dx.doi.org/10.15446/dyna.v82n192.47752.
Cui, C., Luan, M., & Li, M. (2010a). A study on time-effects Rodríguez-Rebolledo, J.F. (2011). Modeling behavior of
of piled raft system by using computational methods. In piles and inclusions under regional consolidation in
Proceedings of the GeoShanghai International Conference the lake area of Mexico City (244 p.). Mexico: National
2010 (pp. 42-51), Shanghai, China. Autonomous University of Mexico. (In Spanish).
Cui, Z.-D., Tang, Y.-Q., & Yan, X.-X. (2010b). Centrifuge Rodríguez-Rebolledo, J.F., León, R.F.P., & Camapum de
modeling of land subsidence caused by the high-rise Carvalho, J. (2019). Obtaining the mechanical parameters
building group in the soft soil area. Environmental Earth for the hardening soil model of tropical soils in the city
Sciences, 59(8), 1819-1826. http://dx.doi.org/10.1007/ of Brasília. Soils and Rocks, 42(1), 61-74. http://dx.doi.
s12665-009-0163-9. org/10.28927/SR.421061.

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 13

 Validation of a 3D numerical model for piled raft systems founded in soft soils undergoing regional subsidence

Rodríguez-Rincón, E. (2016). Experimental analysis of piled Environmental Earth Sciences, 66(7), 1809-1816. http://
raft systems in consolidating soft soils (261 p.). Brasília: dx.doi.org/10.1007/s12665-011-1406-0.
Universidade de Brasília. (In Portuguese). Thaher, M., & Jessberger, H. (1991). Investigation of the
Rodríguez-Rincón, E., Cunha, R.P., & Caicedo Hormaza, behaviour of pile-raft foundation by centrifuge modelling.
B. (2020). Analysis of settlements in piled raft systems In Proceedings of the 10th European Conference on Soil
founded in soft soil under consolidation process. Canadian Mechanics and Foundation Engineering (Vol. 2, pp.
Geotechnical Journal, 57(4), 537-548. http://dx.doi. 597-603), Firenze, Itália.
org/10.1139/cgj-2018-0702. van Tran, T., Kimura, M., & Boonyatee, T. (2012b). 3D FE
Rodríguez-Rincón, E., Cunha, R.P., & Caicedo, B. (2018). analysis of effect of ground subsidence and piled spacing
Behaviour of piled raft foundation systems in soft soil with on ultimate bearing capacity of piled raft and axial force
consolidation process. In Proceedings 9th Int. Conf. on of piles in piled raft. Open Journal of Civil Engineering,
Physical Modelling in Geotechnics, London. 2(4), 206-213. http://dx.doi.org/10.4236/ojce.2012.24027.
Roy, S., & Chattopadhyay, B. C. (2011). Piled-Raft foundation van Tran, T., Teramoto, S., Kimura, M., Boonyatee, T., &
behaviour on consolidating soft soil. In Proceedings of Vinh, L.B. (2012a). Effect of ground subsidence on load
the International Conference on Structural Engineering sharing and settlement of raft and piled raft foundations.
Construction and Management, Kandy.
Stress, 1(3), 120-127.
Schanz, T., Vermeer, P. A., & Bonnier, P. G. (1999). The
Wang, J., Huang, T., & Sui, D. (2013). A case study on stratified
hardening soil model: formulation and verification. In
settlement and rebound characteristics due to Dewatering
R.B.J. Brinkgreve (Ed.), Beyond 2000 in Computational
in Shanghai Subway Station. The Scientific World Journal,
Geotechnics: ten years of PLAXIS International (pp.
2013, 213070. http://dx.doi.org/10.1155/2013/213070.
281-296). London: Routledge.
Watcharasawe, K., Kitiyodom, P., & Jongpradist, P. (2015).
Sinha, A., & Hanna, A.M. (2017). 3D numerical model for piled
raft foundation. International Journal of Geomechanics, Numerical analyses of piled raft foundation in soft soil
17(2), 04016055. http://dx.doi.org/10.1061/(ASCE) using 3D-FEM. Geotechnical Engineering, 46(1), 109-116.
GM.1943-5622.0000674. Zhang, L., & Liu, H. (2017). Seismic response of clay-pile-
Sun, Z., Zhang, G., Zhang, J., Li, G., & Zheng, R. (2008). raft-superstructure systems subjected to far-field ground
Centrifuge modeling of ground settlement due to motions. Soil Dynamics and Earthquake Engineering, 101,
groundwater pumping. China Civil Engineering Journal, 209-224. http://dx.doi.org/10.1016/j.soildyn.2017.08.004.
41(4), 67-72. Zhang, L., Goh, S.H., & Liu, H. (2017b). Seismic response
Surarak, C., Likitlersuang, S., Wanatowski, D., Balasubramaniam, of pile-raft-clay system subjected to a long-duration
A., Oh, E., & Guan, H. (2012). Stiffness and strength earthquake: centrifuge test and finite element analysis.
parameters for hardening soil model of soft and stiff Soil Dynamics and Earthquake Engineering, 92, 488-
Bangkok clays. Soil and Foundation, 52(4), 682-697. 502. http://dx.doi.org/10.1016/j.soildyn.2016.10.018.
http://dx.doi.org/10.1016/j.sandf.2012.07.009. Zhang, L., Goh, S.H., & Yi, J. (2017a). A centrifuge study
Tang, Y.Q., Ren, X., Chen, B., Song, S., Wang, J.X., & Yang, of the seismic response of pile-raft systems embedded
P. (2012). Study on land subsidence under different plot in soft clay. Geotechnique, 67(6), 479-490. http://dx.doi.
ratios through centrifuge model test in soft-soil territory. org/10.1680/jgeot.15.P.099.

14 Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021)

 Alarcón Posse et al.

List of symbols
α is an auxiliary parameter of the model
β auxiliary parameter of the model related to the reference
tangent stiffness modulus for oedometric loading
γp plastic shear strain
φ’ Internal friction angle
σ’3 confining stress in the triaxial test
ψ dilatancy angle
c’ Cohesion
ε axial strain
ε 1 vertical strain
ε 1p plastic axial strain
ε vpc volumetric plastic strains in isotropic compression
ε vp plastic volumetric strain
Ei initial stiffness
Eoed axial stress-dependent stiffness modulus for primary
oedometric loading
E50 is the confining stress-dependent stiffness modulus for
the primary load
Eur stress-dependent stiffness modulus for unloading and
reloading stress
ref reference secant stiffness modulus for the drained
E50
triaxial test
ref
Eoed reference tangent stiffness modulus for oedometric loading
Eurref
reference stiffness modulus for unloading and reloading
conditions
f c cap compression hardening yield function
fs shear hardening yield function
K onc coefficient of earth pressure at rest (NC state)
Ko coefficient of earth pressure at rest
m Exponential power
OCR Over Consolidation Ratio
p is the isotropic stress
pp is the pre-consolidation isotropic stress
pref Stress of reference
q deviatoric stress
qa asymptote of the shear strength
qf ultimate deviatoric stress at failure
q is the special stress measurement for deviatoric stresses
Rf failure ratio
νur Unloading/reloading Poisson’s Ratio

Alarcón Posse et al., Soils and Rocks 44(1):e2021053620 (2021) 15