Sie sind auf Seite 1von 22

Accepted in Geotechnique

doi: 10.1680/geot./15-P-116
Construction of simplified design p-y curves for liquefied soils
D.Lombardi1, S.R.Dash2, S.Bhattacharya3, E.Ibraim4, D Muir Wood5, C.A.Taylor6
1

Lecturer, University of Manchester, UK


Assistant Professor, Indian Institute of Technology Bhubaneswar, India
3
Chair in Geomechanics, University of Surrey, UK
4
Reader, University of Bristol, UK
5
Emeritus Professor of Civil Engineering, University of Bristol, UK
6
Professor of Earthquake Engineering, University of Bristol, UK
2

Corresponding author:
Dr Domenico Lombardi
Lecturer of Geotechnical Engineering
School of Mechanical, Aerospace and Civil Engineering
University of Manchester
Manchester
M13 9PL
Email: domenico.lombardi@manchester.ac.uk

ABSTRACT
In practice, laterally loaded piles are most often modelled using a Beam-on-NonlinearWinkler-Foundation (BNWF) approach. While well calibrated p-y curves exist for nonliquefied soils (e.g. soft clay and sands), the profession still lacks reliable p-y curves for
liquefied soils. In fact, the latter should be consistent with the observed strain-hardening
behaviour exhibited by liquefied samples in both element and physical model tests. It is
recognised that this unusual strain-hardening behaviour is induced by the tendency of the
liquefied soil to dilate upon undrained shearing, which ultimately results in a gradual
decrease of excess pore pressure and consequent increase in stiffness and strength.
The aim of this paper is twofold. First it proposes an easy-to-use empirical model for
constructing stress-strain relationships for liquefied soils. This only requires three soil
parameters which can be conveniently determined by means of laboratory tests, such as a
cyclic triaxial and cyclic simple shear tests. Secondly, a method is illustrated for the
construction of p-y curves for liquefiable soils from the proposed stress-strain model. This
involves scaling of stress and strain into compatible soil reaction p and pile deflection y,
respectively. The scaling factors for stress and strain axis are computed following an energybased approach, analogous to the upper-bound method used in classical plasticity theory.
Finally, a series of results from centrifuge tests are presented, whereby p-y curves are backcalculated from available experimental data and qualitatively compared with that proposed by
the authors.
Key words: pile foundation, stress-strain response, p-y curves, liquefied soil, dynamic soilstructure interaction, geotechnical centrifuge.

INTRODUCTION
Poor performance of pile foundations, ranging from cracks to plastic hinge formations, is still
observed in liquefiable soils after most major earthquakes. As a result, the behaviour of piled
foundations during liquefaction phenomena is an area of active research (Haigh, 2002;
Bhattacharya, 2003; Bhattacharya et al., 2004; . Bhattacharya et al., 2005a; Bhattacharya et
al., 2005b; Cubrinovski et al., 2006; Knappett and Madabhushi, 2009; Dash et al. 2010;
Madabhushi et al., 2010; Stringer and Madabhushi, 2012; Lombardi, 2013; Lombardi and
Bhattacharya, 2014a).
To model laterally loaded piles, practicing engineers often use a simplified method
normally referred to as Beam-on-Nonlinear-Winkler-Foundation, BNWF (Winkler, 1867;
Hetnyi, 1946). This method stands on the hypothesis that the soil reaction p, exerted by the
soil at a certain elevation on the pile shaft, is proportional to the relative pile-soil deflection,
y. According to the BNWF method, the pile is modelled by means of consecutive beamcolumn elements, whereas the lateral pile-soil interaction (LPSI) is modelled through
nonlinear springs attached to nodal points between two consecutive elements. Each spring is
defined by means of a nonlinear relationship between soil reaction per unit length of the pile
p and corresponding relative soil-pile horizontal displacement y. The coefficient of
proportionality between p and y is the modulus of subgrade reaction k, with dimension of
pressure divided by length. This relationship is normally referred to as p-y curve or reaction
curve. Despite its limitations of discrete nature, the BNWF method is extensively used in
practice because of its mathematical convenience and ability to incorporate nonlinearity of
the soil and ground stratification. The validity of BNWF approach is based on the assumed
similarity between two mechanical system responses: (i) load-deformation response of the
pile, which takes into account the overall macro behavior of the soil-pile system; (ii) stressstrain response of the adjacent soil being sheared as the pile moves laterally. This is related to
the micro behaviour of the deforming material. In theory, the transformation from micro to
macro can be made by applying appropriate scaling factors, whereby stress is converted into
equivalent soil reaction, p; and strain is converted into equivalent relative pile-soil
displacement y. Bouzid et al. (2013) demonstrated that appropriate scaling factors can be
derived from the so-called Mobilisable Strength Design (MSD) method (Bolton & Powrie,
1988; Osman and Bolton, 2004; Vardanega and Bolton, 2011). In routine practice, however,
p-y curves are constructed by means of empirical relationships. These were originally
developed in the 70-80s from a relatively limited number of full-scale tests carried out on
flexible steel piles (Matlock, 1970; Reese et al., 1974; Reese et al., 1975; ONeil and
Murchison, 1983).
One of the first attempts to propose p-y curves for liquefiable soils is provided by
Dobry et al. (1995). The method, illustrated in Figure 1, consists of applying to the
conventional p-y curve for non-liquefied sand a reduction factor mp. The latter can be
conveniently determined from the equivalent clean sand blow count (N1)60 using empirical
charts, such as the one depicted in Figure 1b. An alternative method is to use conventional py curves for soft clays, such as the ones recommended by API (2000), but replacing the
undrained shear strength by the residual strength of the liquefied soil, Sr (Goh and ORourke,
1999). This approach is normally referred to as residual strength approach and is
schematically illustrated in Figure 2. The residual strength of the liquefied soil can be
estimated by means of empirical charts (see Figure 2b) that have been proposed by several
researchers (Seed and Harder, 1990; Pillai & Salgado, 1994; Olson and Stark, 2002;
Brandenberg, 2005).

It can be concluded that p-y curves for liquefied soils used in routine practice exhibit
strain-softening behaviour, characterised by a relatively high stiffness mobilised at small
displacements, which gradually reduces upon shearing. This response, however, is in contrast
to the strain-hardening behaviour of liquefied sands sheared in undrained conditions. A
number of studies (Wilson et al., 2000; Tokimatsu et al., 2001; Ashford and Rollins, 2002;
Boulanger et al., 2003) have shown that back-calculated p-y curves of liquefied soils have a
more complex shape, which is concave upward, hence, characterised by practically zero
initial stiffness that gradually increases upon shearing. It is worth noting that this strain
hardening response is consistent with the post-liquefaction behaviour of sands as observed in
element tests (Yasuda et al., 1994; Vaid & Thomas, 1995; Sivathayalan & Vaid, 2004;
Sitharam et al., 2009; Dash, 2010; Lombardi et al., 2014b).
p
pu tanh

kz
y
pu

liquefied soil

pu
p

mp

pu tanh

AIJ
(2001)

0.8

p-multiplier mp

mp

1.0

non-liquefied soil

pu

kz
y
pu

0.6

0.4

0.2

Brandenberg
(2005)

10

20

30

(N1)60

(a)
(b)
Figure 1: Construction of p-y curves for liquefiable soils according to p-multiplier approach:
(a) application of degradation factor mp to p-y curves recommended by API(2000) for nonliquefied sands; (b) mp vs. (N1)60 (after Brandenberg, 2005).
60

p-y curve for clay (API, 2000)

Residual shear strength, Sr

p
pu
Sr

0.72 pu

50
40
30
20
10
0

10

15

20

Equivalent clean sand SPT blow count, ( N1 ) 60

Figure 2: Construction of p-y curves for liquefiable soils according to residual strength
approach: (a) schematic p-y curves; (b) residual strength Sr vs. (N1)60 (after Cubrinovski &
Bradley, 2008).
Figure 3 schematically illustrates the effects of different shapes of p-y curves on the
seismic response of piled foundations. Starting from the concave-downward strain-softening
p-y curve illustrated in Figure 3a, it can be noted that when the relative soil-pile displacement
is small, the resistance experienced by the pile depends on the initial stiffness of the soil and
4

corresponding value of deflection. For large displacement, however, the resistance offered by
the adjacent soil is governed by the ultimate strength of the soil. Differently, if the shape of
the p-y curve is concave-upward, i.e. strain-hardening (see Figure 3b), the pile response is
much more complex and may be significantly different from that described above. In fact,
owing to the practically zero stiffness mobilised at small displacements, the soil may offer
minimal opposition to any lateral movement of the pile. This may result in buckling mode of
failure of the foundation. On the other hand, the higher stiffness and strength mobilised at
larger displacements may prevent a complete collapse of the structure.
Axial Load

Small amplitude
vibration

Large displacement

Lateral Load

Pile

Lateral soil
spring
(
spring)

End bearing
soil spring

(a)

(b)

Figure 3: Possible soil-pile response resulting from (a) p-y curve from p-multiplier approach
(strain-softening response); (b) proposed p-y curve (strain-hardening response)
The motivation behind this research is to propose a simplified p-y curve for the
analysis of soil-structure interaction problems in liquefied soils. The proposed curves can be
constructed from a simplified stress-strain model that requires only three parameters. These
are hereafter referred to as: take-off strain to, initial shear modulus, G1 and shear modulus at
large strains, G2. The advantage of this model is twofold. Firstly, the proposed stress-strain
model requires only four parameters, which can be conveniently determined by means of
conventional element tests, such as triaxial and simple shear tests. Secondly, the stress-strain
relationship is consistent with the strain-hardening behaviour observed in liquefied soils in
both element and physical modelling tests. The proposed stress-strain relationships are
subsequently used to construct a novel family of p-y curves for liquefiable soils. This
involves scaling of stress and strain into compatible soil reaction p and pile deflection y,
respectively. The proposed p-y curves are finally compared to those back-calculated from
centrifuge model tests.

POST-LIQUEFACTION STRESS-STRAIN RESPONSE


Only a limited number of studies have focused on the post-liquefaction behaviour of liquefied
soils. One pioneering work was published by Seed (1979) and focused on the post-earthquake
stability of dams in liquefiable deposits. It was concluded that liquefied soils gradually
mobilised increasing strength and stiffness upon shearing. This unusual strain-hardening
behaviour was later confirmed by subsequent studies by Yoshida et al. (1994) and Kiku &

Tsujino (1996). Studies by Thomas (1992) and Vaid & Thomas (1995) showed that the postliquefaction behaviour of sands subjected to undrained monotonic loading was dilative, even
if the sand was contractive and strain softening before the onset of liquefaction. This
observation was confirmed by subsequent tests carried out by Sivathayalan & Vaid (2004) on
in-situ frozen sands from alluvial environments. It was concluded that the post-liquefaction
behaviour exhibits a continually stiffening response, i.e. strain-hardening, owing to the
tendency of the liquefied soil to dilate upon shearing. Yasuda et al. (1994) observed that the
onset of dilative behaviour occurred only beyond a threshold strain, which was referred to as
reference strain at the transformation point. The latter was found to be strongly dependent
upon the initial degree of packing of the sand, whereby denser sands tended to mobilise
higher strength at smaller strain levels. Kokusho et al. (2004) investigated the effects of
different particle gradations on the post-liquefaction undrained behaviour of sands. The
experimental results suggested that the tendency to dilate was more pronounced in wellgraded soils. A recent study by Sitharam et al. (2009) showed that the post-liquefaction
monotonic behaviour of the soil was influenced by the amplitude of the cyclic axial strain
applied to cause liquefaction, however, it was found independent of the initial confining
stress. Dash (2010) compared the undrained monotonic behaviour of liquefied and nonliquefied samples. Figures 4 shows typical results for samples of Toyoura sand, prepared at
same initial relative density (Dr=50%) and consolidated at same effective stress ('c=50kPa).
The monotonic stress-strain response (see Figure 4b) showed a significant lower initial
stiffness mobilised by the liquefied sample (this is depicted by the dark line in Figure 4). At
large strains, however, the two samples exhibited a similar behaviour, with both samples
mobilising a comparable stiffness. Lombardi et al. (2014c) published results from a series of
multi-stages cyclic triaxial tests in which samples were initially liquefied by cyclic loading
and subsequently sheared monotonically, with no relief of excess pore pressure between the
two stages. It was found that, due to the tendency of the liquefied soil to dilate upon
undrained shearing, which resulted in a gradual decrease of excess pore pressure and increase
in effective stress, the post- liquefaction stress-strain response was always characterised by a
distinct strain-hardening behaviour.

200

100

Toyoura sand
Dr =50%

Toyoura sand
Dr =50%

80

150

60

100

40

monotonic test

50

multi-stage test

20

0
0

multi-stage test
-50
0

20

40

60

80

100

120

Mean effective stress, p' [kPa]

(a)

-20
-5

5
Axial strain, a [%]

10

15

(b)

Figure 4: Undrained behaviour of liquefied and non-liquefied samples. (a) effective stress
path; (b) stress-strain response (after Dash, 2010).

PROPOSED MONOTONIC POST-LIQUEFACTION STRESS-STRAIN


CURVE
With the current understanding of the behaviour of liquefied samples and its influencing
factors, a simplified monotonic post-liquefaction stress-strain model is proposed. The aim of
this model is to provide the engineer/designer with a reasonable stress strain curve for
liquefied soil for use in design and decision making. The proposed model, depicted in Figure
5, requires only four parameters that can be derived from conventional element tests, these
are:
1. Take-off shear strain to: engineering shear strain upon which the shear strength of the
liquefied soil exceeds 1kPa when the soil is subjected to monotonic shearing without
drainage. The observed stress-strain response (see Figure 4), shows that the transition
between G1 and G2 is quite smooth and may occur over 1-2% of shear strain. In order
to make some allowance for this transition, the proposed value of to has been
increased by 25%. As a result, the second linear segment of the stress-strain model
starts from 1.25to with a constant slope (i.e. stiffness) given by the shear modulus G2.
2. Initial shear modulus G1: shear modulus in the initial stage of shearing, i.e. <to.
3. Shear modulus at large strains G2: tangent shear modulus. This is practically a
constant value while the soil is continuously sheared at relatively large strains, >to.
4. Maximum shear stress max: maximum shear stress mobilised at very large strains, i.e.
>>to.
It is worth noting that only the first three parameters, i.e. to, G1 and G2, can be obtained
from conventional element tests. The limiting value of shear stress in the tests, i.e. max, is
required due to the tests conditions are not representative of in-situ conditions. For example,
laboratory triaxial tests on sands are typically performed with the application of a back
pressure in order to improve sample saturation. Samples which wish to dilate as they are
sheared - usually at stress ratios higher than the critical state stress ratio - tend to generate
negative pore pressures in undrained tests. The back pressure provides some guard against
the pore pressure actually falling to zero. However, continued suppressed dilation may
eventually lead to a pore pressure below -100kPa (typical atmospheric pressure) at which
point the pore water will cavitate and boil. At this point, the presence of gas in the water will
turn the undrained test into a drained test. While the pore pressure is becoming increasingly
negative, the effective mean effective stress becomes increasingly positive and the strength
increases. The potential to mobilise such high negative pressures is not usually present in the
field and the fourth parameter represents a capped shear stress. It must be mentioned that
these newly proposed mechanics-based (rather than empirical-based) p-y curves have been
used by Lombardi and Bhattacharya (2016) to analyse shaking table tests on piled
foundations.

Shear stress
max

G2

G1
1kPa

to

1.25

to

Shear strain

Figure 5: Proposed post-cyclic stress-strain model for liquefied soil.

EVALUATION OF PARAMETERS to, G1 AND G2


The advantage of the proposed stress-strain model for liquefied soils lies in its simplicity and
ease of implementation. In fact, the empirical parameters to, G1 and G2 can be conveniently
determined by performing multi-stages tests, whereby the sample is firstly liquefied by means
of cyclic loading, and subsequently sheared in undrained monotonic condition. In the absence
of adequate facilities (e.g. cyclic triaxial, cyclic simple shear, etc.), and for preliminary
design considerations, the three parameters can be estimated from data available in the
literature. In this paper, the three parameters have been estimated from published test results
obtained on sands whose characteristics are listed in Table 1. Figure 6 shows the particle
distributions of the sands, conjointly with grain size distributions of liquefaction-prone sands.
A summary of test data is given in Table 2.
Table 1. Index properties
Sand
Redhill 110
Toyoura
Fraser river

Gs
2.65
2.64
2.72

emax
1.035
0.973
1.00

emin
0.608
0.635
0.68

D50 [mm]
0.14
0.18
0.30

Uc
1.63
1.66
1.68

Particle shape
angular
sub-angular
sub-angular to
sub-rounded
Syncrude
2.62
0.962
0.552
0.20
3.34
Angular to subangular
Narita
2.70
1.371
0.773
0.12
3.08
Angular to subangular
Gs: Specific gravity; emax, emin: maximum and minimum void ratio, D50: 50%
uniformity

Reference
Lombardi et al. (2014c)
Lombardi et al. (2014c)
Vaid & Thomas (1995)
Sivathayalan (1994)
Yasuda et al. (1994)
finer size; Uc: coefficient of

100

Legend

90

Redhill 110
Toyoura

80

Fraser River

70

Narita

60

Liquefiable
Highly-liquefiable

Percent passing [%]

Syncrude

50
40
30
20
10
0
0.001

0.01

0.1

1.0

10

Particle size [mm]

Figure 6: Particle size distributions of sands considered in this work and grain size
distributions of liquefaction-prone sands according to Japanese Seismic Code for Harbor
Structures.
Take-off shear strain, to and initial shear modulus G1
The take-off shear strain to, which has been previously defined as the shear strain required to
mobilise a shear strength of 1kPa, can be directly determined from the stress-strain response
exhibited by the liquefied sample during undrained monotonic loading (see Figure 4b). In the
interpretation of the collated data, the deviator stress q and axial strain a from triaxial tests
have been converted into equivalent shear stress =q/2 and engineering shear strain, =1.5a,
respectively. Figure 7(a) plots the available data of to versus initial relative density Dr.
Beyond the observed scatter, it appears that the take-off shear strain decreases with
increasing Dr. Following the definition of take-off strain, the initial shear modulus can be
defined by G1=1/to, expressed in kPa. Figure 7(b) displays the computed G1 versus initial
relative density Dr. It can be observed that the initial stiffness seems to increase exponentially
with increasing Dr.

30
Legend
Redhill 110

25

Toyoura

Take off strain,

to

[%]

Fraser River
Syncrude

20

Narita
Fitting line

to = 74.34 - 17.71 ln(Dr)

15

R =0.876
10

10

20

30

40

50

60

70

80

60

70

80

Relative density, Dr [%]

(a)
1000

Legend
Redhill 110

Initial shear modulus, G1 [kPa]

Toyoura
Fraser River
Syncrude

100

Narita
Fitting line

G1 = exp(0.0657*Dr)
2

R =0.686
10

1
10

20

30

40

50

Relative density, Dr [%]

(b)

Figure 7: Take-off strain to vs. initial relative density Dr.


Shear modulus at large strains G2
As shown in Figure 4(b), the shear modulus of the liquefied sample is practically constant for
shear strains larger than to. The shear modulus at large strains G2 may be derived from
theoretical considerations that should take into account the effects of the rate of dilation and
level of confinement. However, such an approach would involve a rather extensive
mathematical derivation which seems excessive for the present initial simplified analysis.
Looking for alternative correlations that may exist between G2 and any other soil parameters,
it is proposed to estimate the G2 based on the ratio G2/Gmax, where Gmax denotes the tangent
shear modulus at small strains, i.e. <10-4 % (Kramer, 1996), which can be estimated from the
initial void ratio of the sample e and confining stress 'c according to expression in equation
(1) by Kokusho (1980). Figure 8 plots G2/Gmax versus Dr for the collated data.

Gmax 8400

2.17 e2 '0.5
c
1 e

10

('c and Gmax in kPa)

(1)

Normalised shear modulus G2 / Gmax

1.0

Legend
Redhill 110

Toyoura

Fraser River
Narita

Syncrude

0.1

0.01

0.001
10

20

30

40

50

60

70

80

Relative density, Dr [%]

Figure 8: G2/Gmax versus initial relative density Dr


Table 2: Test data

Toyoura

Dr
'c
[%]
[kPa]
23
50

Toyoura

30

Toyoura

Cyclic torsional shear

Monotonic
strain rate
10%/min

Yasuda et al. (1994)

50

Cyclic torsional shear

10%/min

Yoshida et al. 1994

33

50

Cyclic torsional shear

10%/min

Yasuda et al. (1994)

Toyoura

38

50

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Toyoura

44

100

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Toyoura

50

98

Cyclic torsional shear

10%/min

Yasuda et al. (1998)

Toyoura

70

50

Cyclic triaxial

10%/min

Yoshida et al. (1994)

Redhill 110

33

100

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Redhill 110

37

100

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Redhill 110

40

50

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Redhill 110

48

100

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Redhill 110

59

100

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Redhill 110

67

110

Cyclic triaxial

1%/min

Lombardi et al. (2014c)

Fraser River

19

100

Cyclic triaxial

1%/min

Vaid & Thomas (1995)

Fraser River

28

100

Cyclic simple shear

1%/min

Sivathayalan (1994)

Fraser River

39

100

Cyclic simple shear

1%/min

Sivathayalan (1994)

Fraser River

60

100

Cyclic simple shear

1%/min

Sivathayalan (1994)

Syncrude

51

100

Cyclic simple shear

1%/min

Sivathayalan (1994)

Narita

38

50

Cyclic torsional shear

10%/min

Yasuda et al. (1994)

Narita

53

50

Cyclic torsional shear

10%/min

Yasuda et al. (1994)

Material

Apparatus

Reference

Figure 8 shows that the data is randomly scattered, presumably due to variations in excess
pore pressure between different sands and relative densities. The figure identifies three
different ranges of G2/Gmax, i.e. 0.1, 0.01 and 0.001, depending on relative density. It is worth
noting that G2 is inversely proportional to effective mean principal stress p' and directly

11

proportional to Gmax. To compute the actual value of G2 the empirical expression suggested
by Dash (2010), given by (2), can be used.
1 G2
(2)
Gmax
5 p'
Maximum shear stress max
The proposed stress-strain model for liquefied soils (see Figure 5) suggests that the increase
in shear stress cannot continue indefinitely since the undrained shearing accompanied by
dilation cannot continue to large strains. Theoretically, the maximum shear stress would
reach its maximum physical value when the pore pressure reaches a negative value of about 100 kPa and cavitation of the pore water occurs. In this situation, max can be computed as
follows:
q
M p' 100kPa
(3)
max max c ini
2
2
in which Mc denotes the stress ratio at critical state under conditions of triaxial
compression and p'ini is the mean effective stress in geostatic condition.
An alternative approach consists of equating max to the residual strength Sr of the
liquefied. The latter can be conveniently estimated based on empirical correlations, such as
the one given in Figure 2b.

CONSTRUCTION OF p-y CURVES FROM STRESS-STRAIN CURVE OF


LIQUEFIED SOILS
The proposed method for the construction of p-y curves from stress-strain curves relies on the
similarity between load-deflection characteristics of the pile and mechanical behaviour of the
deforming soil. This involves scaling of stress and strain into compatible soil reaction p and
pile deflection y, respectively. It is assumed that plane strain conditions are established
around the pile at any depth. As a result, soil is expected to flow around the pile from front to
back. Although such an assumption is acceptable at full liquefaction condition, the same may
not be valid prior to the onset of liquefaction, when wedge type failure is likely to occur
particularly at shallow depths. In accordance with the postulated collapse mechanism, the SSI
problem reduces to a series of decoupled plane strain problems as schematically illustrated in
Figure 9. The model considered here is therefore a disc having an outer radius of R,
representing the soil, and a rigid disc with outer radius r0 that moves laterally in the
deforming soil. The soil is further supposed to adhere perfectly to the pile. It should be
emphasized the conceptualised problem is analogous to the plane strain problem used in
plasticity theory for the evaluation of the undrained lateral capacity of a cylinder moving
through an infinite medium (Randolph & Houlsby, 1984).

12

Winkler approach

Plan strain problem

r
r0

(a)
(b)
Figure 9: Schematic representation of the problem: (a) Winkler approach; (b) Plane strain
model.
Following Osman and Bolton (2005) and Klar (2008), the soil resistance p developed at a
mobilised stress mob is given by
(4)
p N s mob D
in which Ns is a scaling factor for stress. The mobilised shear stress mob can be related to an
average mobilised engineering shear strain s,mob. This mathematically defined as the spatial
average of the shear strain s in the entire volume of the deforming medium:
s dV
y
V
(5)
s , mob
Mc

dV

It is noted that the engineering shear strain s is defined as the difference between the major 1
and minor principal 3 strains:
(6)
s 1 3
If the stress-strain curve to be converted into p-y curve is obtained from triaxial compression
tests, the major and minor principal strains correspond to the axial a and radial r strain,
respectively. Considering the undrained condition established during liquefaction, the
engineering shear strain s can be expressed in terms of axial strain a, such that
s 1 3 a r 1.5 a
(7)
Because the loading is axisymmetric and neglecting end effects and possible problem
associated with initial anisotropy of the sample, the major 1 and minor 3 principal stresses
correspond to the axial a and radial r, respectively. In this condition the mobilised stress
mob is
3 a r
(8)
mob 1

2
2
Derivation of scaling parameters
Drawing on the concept of the mobilisable strength design (MSD) method (Bolton & Powrie,
1988; Osman & Bolton, 2005), and its extension (EMSD) by Klar & Osman (2008), which
differs from the former in having a deformation mechanism that changes throughout the
loading sequence, an energy-based approach is used to obtain the scaling factors Nc and Mc.

13

Assuming that the soil-pile system is initially in equilibrium, the rate of input work W done
by the moving disc is
(9)
W py
in which y denotes the increment in lateral displacement and p the resulting soil reaction
(per unit length of the pile).
The rate of dissipation of energy E within the deforming soil mass with volume V is
(10)
E 11 33 dV
V

The hypothesis of incompressibility implies that major and minor principal strain rates are
equal and opposite 1 3 . Equation (9) can be therefore rearranged and the rate of
dissipation of energy E can be expressed as a function of mobilised parameters such that:
E dV 2 dV dV
(11)

mob 1

mob s , mob

The scaling factor for stress Ns is estimated by equating the rate of dissipation of energy
within the deforming soil to the work done by the externa load, i.e. W E . However, in
seeking an upper bound solution to the problem, Ns is sought, which minimise the internal
work, such that:

min mobs , mobdV


V

(12)
Ns
mob Dy
The upper bound calculation required an incremental numerical procedure. In this study this
procedure was implemented in the finite element code COMSOL Multiphysics (Comsol,
2009). The soil was modelled using a triangular mesh with Lagrange-quadratic elements (see
Figure 10a). A non-linear slip boundary was considered for the soil-pile interface, which
corresponds to a rough surface. The incremental displacement y was applied to the inner
boundary of the soil whereas the exterior boundary was fixed (see Figure 10). The deforming
medium was modelled using the simplified strain-hardening model proposed by the authors
(see Figure 5). Specifically, six stress-strain curves were chosen for this parametric analysis,
in which to, G1 and max were fixed, but the ratio G2/G1 was varied from 10 to 500. The
optimal upper bound solution required the evaluation of the lowest possible value of E , at
each incremental displacement and for different radial distances R.

(a)
(b)
Figure 10: Numerical model. (a) mesh in FE model; (b) deformation mechanism.
The result of the numerical analysis is plotted in Figure 11(a). It can be seen that computed
values of scaling factors varied for different G2/G1. Specifically, Ns ranged between 20.23

14

and 21.3, before attaining a practically constant value for large G2/G1. On the other hand, the
scaling parameter for strain Ms decreased from 2.41 to 1.71 with increasing G2/G1. For design
considerations, the lowest values of Ns and Ms can be considered conservative. The bearing
capacity factor derived from classical theory of plasticity can be then used for scaling stress
into compatible soil reaction. According to Martin & Randolph (2006), the bearing capacity
factor Ns is 9.2 for smooth pile-soil interface and 11.94 for rough pile-soil interface.
Figure 11(a) shows that the scaling factor Ms is nonlinear. However, for practical
applications, scaling of strain into compatible deflection can be considered as linear, hence a
single value of Ms can be used. This can be taken as the lowest value in Figure 11(a), i.e.
1.87. It should be noted that Ms obtained from the FE analysis for liquefied soils was lower
than that proposed by Klar (2008), i.e. Ms=2.6, derived for a perfectly plastic material. The
scaling procedure of stress-strain into compatible soil-reaction-deflection is depicted in
Figure 11(b).
22

Ns

20

2.5

Ms

2.0

18

1.5

16

1.0

14

0.5

12

50

100

150

200

250

300

350

400

450

Scaling factor for stress axis, Ns

Scaling factor for strain axis, M s

3.0

10
500

Stiffness ratio of liquefied soil, G2 G1

(a)
p

NS D

Shear stress

soil reaction, p

proposed stress-strain curve

Shear strain

proposed p-y curve

D / MS

pile deflection y

(a)
Figure 11: (a) Proposed scaling approach for derivation of p-y curves from stress-strain
response; (b) Scaling factors Ms and Ns for different stiffness ratio G2/G1.

Validation of proposed p-y curves for liquefiable soils


To validate the proposed p-y curve construction procedure, a series of centrifuge tests were
re-analysed in order to back-calculate p-y curves from available experimental data. The tests

15

considered in this paper were carried out in the centrifuge facility of Shimizu Corporation
(Japan). Detailed information of the centrifuge facility used in these tests can be found in
Sato (1994).
Centrifuge tests
The tests were carried out at a centrifugal acceleration of 30-g. The model container was a
laminar box with inner dimensions of 805mm in length, 475mm in width and 324mm in
depth. The soil deposit consisted of four layers of soil, prepared at different densities and
using different sands, namely silica sand and Toyoura sand. The thickness and relative
density of each layer are specified in Figure 12. It is noted that layer 1 was unsaturated (i.e.
degree of saturation of 10%) whereas layers 3 and 4 were prepared at higher densities (i.e.
Dr=90%) in order to restrict the occurrence of liquefaction to layer 2 only. A summary of the
scaling parameters used for the construction of the models and interpretation of experimental
results are given in Table 3. Derivations of these parameters can be found in Schofield
(1981). To satisfy the scaling requirement for the diffusion process, a high viscosity silicone
oil was used as the pore fluid.
Table 3: Scaling laws
Quantity
Dimension Model/prototype
Stress
ML-1T-2
1
Strain
1
Length
L
1/n
Time (dynamic)
T
1/n
-2
Acceleration
LT
n
Frequency
[T-1]
n
Diffusion time
T
1/n2
The model consisted of a 22 pile-group made of steel pipes. Each pile had an outer
diameter of 10mm, wall thickness of 0.2mm and total length of 270mm. Pile ends were fully
fixed to a rigid plate fitted onto the base of the model container. One pile was instrumented
with pairs of strain gauges at the locations shown in Figure 12. The physical model also
included a rigid quay-wall located in front of the pile-group, however, this was not
considered in the present study. Other transducers included pore pressure transducers,
accelerometers and displacement transducers. A schematic of the instrumentation layout is
given in Figure 12. In the present study, three tests were considered for the back-calculation
of p-y curves. Each test was carried out under nearly identical conditions in terms of input
motion and relative density of the liquefiable layer (i.e. layer 2). However, different distances
between quay-wall and piled structure were used in each test as specified in Figure 12. The
input motion consisted of a sine dwell with a constant frequency of 60Hz, which
corresponded to a forcing frequency of 2Hz at the prototype scale. The amplitude of the sine
dwell was gradually incremented, attaining a value of 8g (i.e. 0.27g at prototype scale) after
approximately 0.25s. In the three tests, the onset of liquefaction occurred in 5-6 cycles of
loading.
Derivation of p-y curves from test data
The derivation of p-y curve involved three major steps:
(i) Double integration of soil acceleration to compute soil displacement ys
(ii) Double integration of bending moment along the pile to obtain pile deflection yp. The
relative pile soil displacement y was subsequently computed as the difference yp-ys
(iii) Double differentiation of bending moment along the pile to obtain soil reaction p

16

As the numerical operations of integration and differentiation are sensitive to low frequency
and high frequency components, respectively, a band-pass Butterworth filter was used in the
signal processing of the recorded data. Furthermore, because the computed bending moments
were known only at discrete locations, a cubic spline interpolation function was used to
obtain a continuous bending function along the instrumented pile. This interpolation function
was used by Wilson (1998) and Jeanjean (2009) for the same purpose. The back-calculated py curves are illustrated in Figure 13. In the figure, the soil resistance p has been normalised
by the effective stress 'v and pile diameter D. On the other hand, displacement y is normalized
by the pile diameter D. Few points may be noted:
(1) It is noted that three depths considered in Figure 13 (i.e. 2.1m, 4.1m and 5.7m) are at
prototype scale and located within the liquefiable layer (i.e. layer 2 in Figure 12). It can be
seen that the back-calculated p-y curves exhibited practically zero stiffness at small
deflection. This is in contrast to p-multiplier based p-y curves as shown in Figure 1 and
currently used in practice where there is considerable stiffness at small deflection. The
downside of using such approach is shown in Figure 3(a) and it is clear that using such model
would overestimate the stiffness of the foundation and ultimately underestimate the seismic
displacement demand as shown in Lombardi and Bhattacharya (2016).
(2) There is increasing stiffness and lateral resistance with increasing pile-soil relative
displacement. This strain-hardening behaviour is consistent with the element test results of
liquefied soil (see Figure 4b), proposed stress-strain model (see Figure 5) and the proposed py curves.
dimensions in mm
a

soil 1

130

soil 2

Test ID

Dr=50%

Dr=50%

80

soil 3

Dr=90%

30

soil 4

Dr=90%

Test 1

250 200

Test 2

250

50

Test 3

250

50

Quay wall

60

accelerometer
pore pressure sensor
strain gauge
850

Figure 12: Test setup and instrumental layout used in centrifuge tests

17

Test 1

Test 2
depth=2.1m

4 depth=2.1m

Test 3

'D

2 depth=2.1m
0

p/

-2
-0.4
4

depth=4.1m

-5
0.4 -0.4
4

depth=4.1m

-4
0.4 -0.4
4

0.4

0.4

0.4

'D

depth=4.1m

p/

-4
0.4 -0.4
4

depth=5.7m

-4
0.4 -0.4
4

depth=5.7m

0
-4

p/

'D

-4
-0.4
4 depth=5.7m

-4
-0.4

y/D

-4
0.4 -0.4

0.4 -0.4

y/D

y/D

Figure 13: Back-calculated p-y curves from centrifuge tests


Conclusions
The paper proposes a simplified p-y curve construction procedure for the analysis of soil-pile
interaction problems in liquefiable soils. The procedure requires the stress-strain of the
liquefied soil, and appropriate scaling factors for the conversion of stress-strain into
compatible soil resistance p and soil-pile relative deflection y. In contrast to the existing p-y
curves for liquefiable soils, the proposed curves retain the essential features of liquefied soil
as observed in both element and physical model tests. This include practically zero strength
and stiffness at low strain level (i.e. <to); and strain-hardening behaviour at large strain with
increasing stiffness upon shearing. It can be noted that the proposed p-y curves are consistent
with the behaviour exhibited by p-y curves back-calculated from model tests carried out in
centrifuge. Despite the simplicity of the method, the proposed p-y curve can be conveniently
implemented in routine practice for use in design and decision making.

Acknowledgements
The authors would like to thank the anonymous reviewers for their helpful and constructive
comments that greatly contributed to improving the final version of the manuscript. The work
is partially funded by the EPSRC (Engineering and Physical Sciences Research Council,
United Kingdom) through the grant EP/H015345/2. The first author would like to
acknowledge the funding received through University of Bristol Centenary PhD Scholarship
for 3 years. The third author would also like to acknowledge University of Bristol for the
University Senior Research Fellowship for the year 2011 to 2012.

18

References
AIJ (2001). Recommendations for Design of Building Foundations. Architectural Institute
of Japan, Japan.
API (2000). Recommended Practice for Planning, Designing, and Constructing Fixed
Onshore Platforms - Working Stress Design. (RP 2A-WSD), 21st Edition. American
Petroleum Institute, USA.
Ashford, S. A., and Rollins, K. M. (2002). TILT: The Treasure Island liquefaction test.
Final Rep. No. SSRP 2001/17, Dept. of Structural Engineering, Univ. of California, San
Diego.
Bhattacharya, S (2003). Pile instability during earthquake liquefaction, PhD thesis,
University of Cambridge.
Bhattacharya, S., Madabhushi, S.P.G and Bolton, M.D. (2004). An alternative mechanism
of pile failure in liquefiable deposits during earthquakes, Gotechnique 54(3): 203-213.
Bhattacharya, S., Bolton, M.D. and Madabhushi, S.P.G (2005a). A reconsideration of the
safety of the existing piled bridge foundations in liquefiable soils, Soils and Foundations, 45
(4):13-26.
Bhattacharya, S., Madabhushi, S.P.G and Bolton, M.D. (2005b). Reply to the two
discussions on An alternative mechanism of pile failure in liquefiable deposits during
earthquakes, Gotechnique 55, 3: 259-263
Bolton, M. D., & Powrie, W. (1988). Behaviour of diaphragm walls in clay prior to
collapse. Gotechnique, 38(2), 167-189.
Boulanger, RW., Kutter, B. L., Brandenberg, S. J., Singh, P., and Chang, D. (2003). Pile
foundations in liquefied and laterally spreading ground during earthquakes: centrifuge
experiments and analyses (No. UCD/CGM-03/01). Center for Geotechnical Modeling,
Department of Civil and Environmental Engineering, University of California, Davis,
California.
Bouzid, D.J., Bhattacharya, S. and Dash, SR. (2013). Winkler Springs (p-y curves) for pile
design from stress-strain of soils: FE assessment of scaling coefficients using the Mobilized
Strength Design concept. Geomechanics and Engineering. 5(5): 379-399.
Brandeberg, S.J. (2005). Behaviour of pile foundations in liquefied and laterally spreading
ground. Ph.D. Thesis, University of California at Davis, California, USA.
Comsol (2009). COMSOL multyphysics user guide. Version 3-5a, COMSOL Stockholm,
Sweden.
Cubrinovski, M., & Bradley, B. (2008). Assessment of seismic performance of soilstructure systems. Proc. 18th New Zealand Geotechnical Society 2008 Symposium, 4-5 Sep
2008, Auckland, New Zealand, 111-127.
Cubrinovski, M., Kokusho, T. and Ishihara, K (2006). Interpretation from large scale
shake table tests on piles undergoing lateral spreading in liquefied soils, Soil Dynamics and
Earthquake Engineering, 26(2): 275-286.
Dash, S. (2010). Lateral Pile-Soil Interaction in liquefiable Soils. PhD thesis, University
of Oxford.
Dash, S.R., Bhattacharya, S. and Blakeborough, A (2010). Bending-buckling interaction
as a failure mechanism of piles in liquefiable soils, Soil Dynamics and Earthquake
Engineering, 30:32-39
Dobry, R., Taboada, V., and Liu, L. (1995). Centrifuge modeling of liquefaction effects
during earthquakes. Proc. 1st Intl. Conf. On Earthquake Geotechnical Engineering, ISTokyo. 14-16.
Goh, S and ORourke, T.D. (1999). Limit state model for soilpile interaction during
lateral spread. In Proc. 7th USJapan Workshop on Earthquake Resistant Design of Lifeline
Facilities and Countermeasures Against Soil Liquefaction: 237260.

19

Haigh, S.K (2002) Effects of Liquefaction on Pile Foundations in Sloping Ground. PhD
Thesis, University of Cambridge.
Hetnyi, M. (1946). Beams on Elastic Foundation. Theory with applications in the fields
of civil and mechanical engineering. Ann Arbor: The University of Michigan Press.
Jeanjean, P., (2009). Re-assessment of py curves for soft clays from centrifuge testing and
finite element modeling. Offshore Technology Conference.
Kiku, H. and Tsujino, S (1996). Post Liquefaction characteristics of sand, in Proceedings
11th World Conference on Earthquake Engineering, Acapulco, Paper no 1088.
Klar, A., 2008. Upper bound for cylinder movement using Elastic fields and its possible
application to pile deformation analysis. International Journal of Geomechanics, 8(2),
pp.162-167.
Klar, A. and Osman, A.S., 2008. Load-displacement solutions for piles and shallow
foundations based on deformation fields and energy conservation. Gotechnique, 58(7),
pp.581-590.
Knappett, J.A and Madabhushi, S.P.G (2009). Inuence of axial load on lateral pile
response in liqueable soils. Part II: numerical modelling. Gotechnique, 59(7):583-592.
Kokusho, T. (1980). Cyclic triaxial test of dynamic soil properties for wide strain range.
Soils and Foundations, 20, 45-60.
Kokusho, T., Hara, T., and Hiraoka, R. (2004). Undrained shear strength of granular soils
with different particle gradations. Journal of Geotechnical and Geoenvironmental
Engineering, 130(6), 621-629.
Kramer, S. L. (1996). Geotechnical Earthquake Engineering (Vol. 80). Upper Saddle
River, NJ: Prentice Hall.
Lombardi, D (2013). Dynamics of pile-supported structures in seismically liquefiable
soils, PhD Thesis, University of Bristol, UK.
Lombardi, D. and Bhattacharya, S. (2016) Evaluation of seismic performance of pilesupported models in liquefiable soils, Earthquake Engineering and Structural Dynamics
DOI: 10.1002/eqe.2716
Lombardi, D. and Bhattacharya, S. (2014a). Liquefaction of soil in the Emilia-Romagna
region after the 2012 Northern Italy earthquake sequence. Natural Hazards, 73: 1749-1770.
Lombardi, D. and Bhattacharya, S. (2014b). Modal Analysis of pile -supported structures
during seismic liquefaction, Earthquake Engineering and Structural Dynamics, 43 (1): 119138.
Lombardi, D., Bhattacharya, S., Hyodo, M., & Kaneko, T. (2014c). Undrained behaviour
of two silica sands and practical implications for modelling SSI in liquefiable soils. Soil
Dynamics and Earthquake Engineering, 66, 293-304.
Madabhushi, S.P.G, Knappett, J.A. and Haigh, S.K (2010). Design of pile foundations in
liquefiable soils. Imperial College Press.
Martin, C.M. and Randolph, M.F., 2006. Upper-bound analysis of lateral pile capacity in
cohesive soil. Gotechnique, 56(2), pp.141-145.
Matlock, H (1970). Correlation for design of laterally loaded piles in soft clay. In Offshore
Technology Conference: 77-94.
Olson, S.M. and Stark, T.D. (2002): Liquefied strength ratio from liquefaction flow failure
case histories, Canadian Geotechnical Journal, 39: 629-747.
ONeill and Murchison, J.M. (1983). An evaluation of p-y relationships in sands. Report
to American Petroleum Institute, University of Texas at Austin, 1983.
Osman, A. S., and Bolton, M. D. (2004). A new design method for retaining walls in clay.
Canadian Geotechnical Journal, 41(3): 451-466.
Osman, A.S. and Bolton, M.D., 2005. Simple plasticity-based prediction of the undrained
settlement of shallow circular foundations on clay. Gotechnique., 55(6), pp.435-447.

20

Pillai, V.S. and Salgado, F.M (1994). Post-Liquefaction stability and deformation analysis
of Duncan Dam, Canadian Geotechnical Journal, 31:967-978.
Randolph, MF., and Houlsby, GT. (1984). Limiting pressure on a circular pile loaded
laterally in cohesive soil. Gotechnique, 34(4):613623.
Reese, L., Cox, W. and Koop, F (1974). Analysis of laterally loaded piles in sand. In
Offshore Technology Conference: 95-105.
Reese, L., Cox, W. and Koop, F. (1975). Field Testing and Analysis of Laterally Loaded
Piles in Stiff Clay. In Offshore Technology Conference.
Sato, M. (1994). A new dynamic geotechnical centrifuge and performance of shaking
table tests. In centrifuge, Vol. 94, 157-162.
Schofield, A.N., (1981). Dynamic and earthquake geotechnical centrifuge modelling.
Proceedings International Conference Recent Advances in Geotechnical Earthquake
Engineering and Soil Dynamics, Vol. 3, 1081-110.
Seed, H.B. (1979). Considerations in the earthquake-resistant design of earth and rockfill
dams. Gotechnique, Vol. 29 (3). 215 263.
Seed, R.B. and Harder, L.F., Jr. (1990): SPT-based analysis of cyclic pore pressure
generation and undrained residual strength, Proceedings of the H.B.Seed Memorial
Symposium, Bi-Tech Publishing Ltd., 2: 351-391.
Sitharam, T.G., Vinod, J.S. and Ravishankar, B.R. (2009). Post-liquefaction undrained
monotonic behaviour of sands: experiments and DEM simulations. Gotechnique, 59(9):739
749.
Sivathayalan (1994). Static cyclic and post liquefaction simple shear response of sands,
MSc Thesis, University of British Columbia.
Sivathayalan, S. and Vaid, Y.P. (2004). Cyclic resistance and post liquefaction response of
undisturbed in-situ sands. Proc. 13th World Conference on Earthquake Engineering,
Vancouver, B.C., Canada August 1-6, 2004. Paper No. 2940.
Stringer M.E and Madabhushi S.P.G. (2012). Axial load transfer in liquefiable soils for
free-standing piles, Gotechnique, 63(5): 400 4009.
Thomas, J. (1992). Static, cyclic and post liquefaction undrained behaviour of fraser river
sand. MSc Thesis, University of British Columbia, October 1992.
Tokimatsu, K., Suzuki, H., and Suzuki, Y. (2001). Back-calculated py relation of liquefied
soils from large shaking table tests. Proc. 4th International Conf. on Recent Advances in
Geotechnical Earthquake Engineering and Soil Dynamics. 1-6
Vaid, Y.P and Thomas, J. (1995). Liquefaction and post liquefaction behavior of sand.
Journal of Geotechnical Engineering, 121(2):163173.
Vardanega, P. J., and Bolton, M. D. (2011). Strength mobilization in clays and silts.
Canadian Geotechnical Journal, 48(10): 1485-1503.
Wilson, D.W. (1998). Soil-pile-superstructure interaction in liquefying sand and soft clay.
PhD thesis. University of California, Davis, USA.
Wilson, DW, Boulanger, R. W., and Kutter, B. L. (2000). Observed seismic lateral
resistance of liquefying sand. Journal of Geotechnical and Geoenvironmental Engineering,
126(10), 898-906.
Winkler, E. (1867). Die Lehre von der Elasticitaet und Festigkeit besonderer Rucksicht
auf ihre Anwendung in der Technik fur politechniche Schulen. Bauakademien, Ingenieure,
Maschinenbauer, architecten, etc, H. Dominicus.
Yasuda, S., Masuda, T., Yoshida, N., Nagase, H., Kiku, H., Itafuji, S., Mine K. and Sato,
K. (1994). Torsional shear and triaxial compression tests on deformation characters of sands
before and after liquefaction. Proc., 5th US-Japan workshop on earthquake resistant design
of lifelines and countermeasures against soil liquefaction. 249-265

21

Yasuda, S., Terauchi, T., Morimoto, M., Erken, A. and Yoshida, N. (1998). Post
liquefaction behavior of several sands. In Proc. 11th European Conference on Earthquake
Engineering, Balkema, Rotterdam.
Yoshida, N., Yasuda, S., Kiku, M., Masuda, T. and Finn, W.D.L. (1994). Behaviour of
sand after liquefaction. U.S.-Japan workshop on earthquake resistant design of lifeline
facilities and countermeasures against soil liquefaction, 5, Buffalo, N.Y., 29 Sep.-29 Oct.
1994.

22

Das könnte Ihnen auch gefallen