New *Abdelkader Noui , Toufik Karech and Tayeb Bouzid3) · 2018. 8. 19. · shallow foundation...

The 2018 World Congress on Advances in Civil, Environmental, & Materials Research (ACEM18) Songdo Convensia, Incheon, Korea, August 27 - 31, 2018

Numerical investigation of dynamic behavior of foundation reinforced

by stone columns using Finn model: Liquefaction mitigation

*Abdelkader Noui1), Toufik Karech2) and Tayeb Bouzid3)

1), 2), 3) Department of Civil Engineering, Faculty of Technology, BATNA 2 UNIVERSITY,

Batna 05078, Algeria 1)

[email protected]

ABSTRACT

This work investigated the behavior of shallow foundation model resting on liquefiable loose sand soil reinforced with stone columns. The foundation subjected to the variable-amplitude harmonic ground motion. Using these columns is an efficient method to decrease the excess pore water pressure, thus a liquefaction mitigation. In this study, a 3-D difference element analysis in combination with a Finn model is used to identify the liquefaction by the relation between the pore-pressure build-up and the volumetric response. Parametric study is included geometrically the effect of the thickness of the mattress and the diameter of the stone columns, it is included also the effect of the overload. The numerical simulation demonstrates the stiffening benefit of stone columns more than liquefaction mitigation under the foundation where the vertical stresses increase under the effect of overload and adjusting the acceleration in the zones where the great negative excess pore water pressure (EPWP) build-up in the soil. the effect of stone columns diameter and mattress thickness need more study and attention under another conditions to know more its role and determine the range for their effective values. In the other hand, when the depth close to the effect of surcharge the magnitude of EPWP decreases. Keywords: shallow foundation; loose sand; stone columns; liquefaction; Finn model;

accelerations. 1. INTRODUCTION The stone columns are flexible and vertical inclusions with a circular cross-section. It is one of the geotechnical techniques, which allows the improvement the mechanical properties in the ground and plays an important role in liquefiable soils (Dhouib and Blondeau 2005). As known, the liquefaction is a very dangerous event which may be the origin of large sinister in soils and structures. This has imposed to the researchers

1)

PhD student 2)

Professor 3)

PhD degree

mailto:[email protected]


to find ways to mitigate their damage, and one of these ways is the stone columns solution. There are no sufficiently developed and well-known methods for studying the behavior of stone columns in seismic zones. However, in spite of the limited development and research results in this area, some authors have developed approaches to study the earthquake behavior of soils treated with stone columns and as a liquefaction mitigation procedure. There are studies developed by Seed and Booker (1977) and others like Priebe (1978, 1998), Ishihara and Yamazaki (1980) …. Etc. There is also centrifuge and shaking table tests at the laboratory (Adalier et al. 2003; Adalier and Elgamal 2004; Sadrekarimi and Ghalandarzadah 2005; Qu et al. 2016). Numerical methods have experienced a remarkable development in this domain. In the dynamic field, most simulations are for the Centrifuge experiment of sand soil (Tang et al. 2015; Esmaeili and Hakimpour 2015; Meshkinghalam et al. 2017) and other ones show the mitigation of liquefaction in term of reducing in the lateral deformation in a sloping stratum (Elgamal and Forcellini 2009) or in the excess pore pressure generation taking into account a group of influencing parameters (Asgari et al. 2013). This paper proposes a numerical simulation by the program FLAC3D to design a shallow foundation resting on liquefiable soil of Boudouaou [Boumerdes-Algeria] (Bouafia 2014) with and without reinforced by stone columns under the effect of overload (shallow foundation + mattress + surcharge). This paper proposes also to evaluate the dissipation of pore water pressure using of the Finn model with the criterion of Byrne (1991) under the effect of several factors, including the thickness of the mattress, the diameter of the stone columns and the effect of surcharge. 2. THEORETICAL BACKGROUND 2.1 Behaviour of the stone columns in a seismic zone Stone columns were proposed as the principal ground improvement method to potentially liquefiable sand deposits (loose fine sand with a high percentage of fines but less than 35%) (Dhouib and Blondeau 2005) They are designed to improve the surrounding soil by increasing the shear stiffening and decrease of excess pore water pressure by the drainage process (Priebe 1998). The study in the behavior of the stone columns under seismic load did not get a significant development in order to have a precise and clear idea of this behavior (Dhouib and Blondeau 2005). Nevertheless, there are certain analytical approaches that analyze seismic behavior of soil treated by stone columns and evaluated the risk of liquefaction. As mentioned earlier in the introduction, Seed and Booker (1977) were the first who studied the use of stone columns as a liquefaction mitigation procedure. They propose a method for the assessment of the effects of radial drainage on the pore-water pressure caused by the earthquake (Fig. 1) (Bouckovalas et al. 2009) The dominant mechanism in the operation of a gravel drain system is often a pure horizontal drainage system; the equation can be written as following (Murali Krishna et al. 2006):


uu

g

rrNN

u

2c o s

2s i n

1

121

'0

(1)

Where ru the excess pore water pressure ratio (ru = u/σ0’)

Fig. 1 Method of Seed and Booker principle (1977)

Priebe (1978, 1998) (Priebe 1998; Dhouib and Blondeau 2005) suggests in seismic areas an approach based on correlations that are similar to those of Seed et al (1983) that allows to express the ratio of cyclic shear stress generated by the

earthquake h to the effective vertical stress σ΄v0 in the ground, by the following expression, based on the theory of Suzuki et al. (1997):

d

v

v

v

h rg

aM

'0

0m a x

'0

11.0

(2) Where:

M: is the magnitude of the earthquake;

g: is the acceleration of gravity;

σv0: is the total vertical stress of the soil;

rd.: is a dependent reduction coefficient of the depth z. To take account of the influence of the stone column on the shear stress induced by the earthquake, Priebe (1998) suggests a correction to the ratio of the cyclic constraints through the factor of improvement n0 (Dhouib and Blondeau 2005):


d

v

v

C o r r e c t e dv

h rg

aM

n '0

0m a x

0'0

11.01

(3) Where the factor of improvement n0 is dependent on the incorporation rate a = (Ac/A), which for some strains at constant volume (ν = 0.5) (Dhouib and Blondeau 2005) is given by:

1

1

110

aKan

ac (4) With:

24t a n2 cacK

(5)

zrd 015.01 (6) Robertson and Campanella (1985) (Dhouib and Blondeau 2005):, who consisted to link the ratio of cyclic stress to the diameter corresponding to 50% of pass (D50) by the following equation:

35.0

lg146.0 50'0

D

v

h

(With D50 in mm) (7) Adalier and Elgamal (2004) reviewed the different analytical methods and testing studies of the stone column method as a liquefaction reduction by its drainage property with a mention of historical cases. 2.2 Finn model Finn Model was used for the liquefiable materials when the pressure of the fluid increases and the effective stress acting on the grain matrix decreases. Martin et al. (1975) approach note that the relation between irrecoverable volume-strain and cyclic shear-strain amplitude is independent of confining stress. The principle in this method is the capture skeleton behavior under cyclic loading that imposes a volumetric constraint to consider EPWP. This is a coupled equation between the incremental shear and the increment of volumetric strain under simple shear load (Martin et al. 1975; Byrne 1991; Itasca Consulting Group, Inc. 2005)

vd

vdvdvd

C

CCC

.

...

4

23

21

(8)


C1, C2, C3 and C4 are constants depending on the relative density of the sand and are related as follows C1. C2. C4 = C3 when Δεvd = 0 (Itasca Consulting Group, Inc. 2005) Byrne (1991) proposed a modified and simpler volume change model with two constants (C1 and C2).

vdvd CC 21 exp.

(9) There is also the index form which allows the computing of the incremental volumetric strain induced in freely draining soil by each half cycle in a shear strain time history by the Eq. (9) (Carter et al. 2013):

t h r e s hi

ivt h r e s hiivd

CC

21 exp..5.0

(10) Where (Δεvd)i is the incremental volumetric strain induced by the i

th half cycle in

the shear strain time history; is the amplitude of the ith half cycle in the shear strain

time history; γthresh is the threshold shear strain below which no volumetric strain

will occur ; ( ) is the cumulative volumetric strain before the ith half cycle in the shear

strain time history is applied. The parameter C1 controls the amount of volume change (volumetric strain increment), and C2 controls the shape of variation for the accumulated volume by the number of cycles (volumetric strain curve). Byrne (1991) recommended a correlation equation to obtain the model constant C1 in terms of sand density Dr as:

5.2

1 .7600

rDC (11) As the form is the same for all densities, the C2 parameter is a constant fraction of C1 for all relative densities and can be prescribed as follows:

12

4.0

CC

(12) The coefficient C1 is assigned from the SPT value (N60) as below (Byrne 1991; Itasca Consulting Group, Inc. 2005):

25.1

601 .7.8

NC (13) 2.3 Numerical simulation background Sasaki and Taniguchi (1982) The finite element software under the name "SADAP 2D" has been modified to permit generation and dissipation of pore pressure during dynamic loading. SADAP used the method of direct integration taking into account the


nonlinearity of soil behavior law, for this the model of behavior used is Hardin-Drnevich (Hardin and Drnevich, 1972). The generation of pore pressure is calculated by "SADAP 2D" gives a reasonable result on the accumulation of pore pressure during cyclic loading. Millea (1990) Millea (1990) makes a numerical modelization of the centrifuge test of a foundation based on Leighton-Buzzard sand 120/200 by the finite element software 'DYNAMFLOW' with constitutive models developed by Prevost (1985) who can predict the generation of excess pore pressure in saturated sand deposits. Two models were considered, 1- sand not loaded by the foundation without and with ballasted column and 2- sand loaded by the foundation without and with four ballasted columns. After a parametric study contains the permeability, the length for the stone columns and the vertical and horizontal stresses, shear stresses.....etc. Millea concludes that the stone columns have successfully reduced excess pore pressure under the footing with load displacement to the columns during dynamic loading. Esmaeili et Hakimpour (2015): The work of Esmaeili and Hakimpour (2015) analyzes the use of ballasted columns as a liquefaction reduction method for sand, the numerical study is done by the finite difference code FLAC 3D for sand NEVADA with a density 40% with the Finn model. An analysis was performed for a single column to examine the effect of radius (diameter) on the reduction of excess pore pressure, the second analysis was performed for a group of stone columns installed in a square arrangement with different column diameter values and in addition different values of spacing between columns. The results show that the stone column with the diameters 0.9, 1.20 and 1.50 m acts well to control liquefaction. Lateral columns have helped the central column to reduce excess pore pressure and drainage so there is collective work in the group stone columns except for spacing between columns values 4 and 5 or each column is working in an individual way. In addition, the individual way in the group forms more efficient than a single column, therefore, the behavior of group stone columns was better than that of a single column, so that the maximum effectiveness of the risk reduction liquefaction is 92%. Tang et al. (2015): A three-dimensional finite element analysis by the OpenSees computer code was used to simulate an experimental test to make a comparison with the work of Adalier et al. (2003) for models 1 and 2, a parametric study to know the effect of the permeability of the stone columns and the surface charge on the efficiency of the liquefaction reduction.

According to Tang et al. (2015), the response of the silty sand with and without columns corresponds well to the experimental data, the same for the results of the verification of the effect of the stone columns on the reinforcement according to the acceleration where the response more rigid is found with columns and also for excess pore pressure in the upper half of silty sand in both cases, with and without columns where complete liquefaction in this area did not avoid. The study shows also that stone columns with permeability exceed the threshold value can significantly reduce the risk of liquefaction because the low permeability does not reduce the risk especially to the


near surface, on the other hand, the stiffening benefit due to the greatest load applied to the stone column area was low but did not prevent liquefaction of the soil. Meshkinghalam et al. (2017): This work is very similar to the work of Esmaeili and Hakimpour (2015), the difference is in the scale of model (1/50 compared to Meshkinghalam et al. (2017)) and in the arrangement group of ballasted columns, for Meshkinghalam et al. (2017) the arrangement is triangular. Meshkinghalam et al. (2017) concluded that the variation in pore pressure decreases and reaches stability at several cycles as the depth increases, in addition the single stone column makes a drainage at a distance of 2.5 m from its center, but this drainage effect disappears at more than 2.5 m. Meshkinghalam et al. (2017) find also that the increase in column’s diameter (0.6 to 1.2 m) causes the increase of drainage at a distance of about 1-1.5 m from the column center, for a distance greater than 1.5 m, the increase in the column’s diameter does not influence pore water drainage. For the stone columns group, the increasing in the distance between columns the effect of group is decreased, so each column of group behaves like a single one. 3. NUMERICAL SIMULATION 3.1 Model geometry, soil properties and constitutive soil model A finite difference software, (FLAC 3D - Fast Lagrangian Analysis of Continua), is used during this study for the analysis of the liquefaction of a footing on a saturated loose sand deposit without and with reinforced by stone columns, coupled with a dynamic-groundwater flow (effective stress), the model is represented by 2648 zones and 3697 grid-points. Calculations’ code (FLAC3D) for this study uses the Finn model to carry out pore-pressure build-up (Itasca Consulting Group, Inc. 2005). The shallow foundation is plan rectangular of 5.5 m in length, 4.0 m in width and 0.8 m in thickness resting on loose sand layer without and with six stone columns, the stone column’s diameter is 0.8 m, the matress with 0.55 m in thickness supporting a shallow foundation and a load of 100 KPa (Fig. 2). The frequency content of the input wave and the wavespeed characteristics of the system will affect the numerical accuracy of wave transmission, and they can result in a numerical distortion of the propagating wave. To solve this problem, the spatial element size, Δl, must be smaller than approximately one-tenth (1/10) to one-eighth (1/8) of the wave length associated with the highest frequency component of the input wave (Δl ≤

) (Lysmer and Kuhlemeyer 1969; Itasca Consulting Group, Inc. 2005).

The loose sand representing a liquefiable site which is located in the region of Boudouaou-Algiers, the geotechnical characteristics of the sandy material are (Bouafia 2014):

Friction angle ϕ = 29o;

Shear velocity Vs = 185 m/s;

Poisson’s ratio ν = 0.4;

Saturated unit weight ϒsat = 17kN/m3.


Fig. 2 (a) Plan view of Model, (b) Cross sectional area of Model.


Fig. 3 (a) Typical difference element mesh for a shallow foundation resting on loose sand, (b) Zoomed view of mesh for foundation resting on a stone columns.

The ballasted column has the following properties (Dhouib and Blondeau 2005):

Unit weight of ballast ϒ = 20kN/m3;

Cohesion C = 0 kPa, ϕ = 380;

Module of deformation Ec/Es = 10, Es is the dynamic deformation module of the soil calculated in function of Vs.

A Rayleigh damping of 5% is considered in this model. The mattress has the same

properties of stone column.

The behavior of the soil is studied by the Finn model for modeling the liquefaction

phenomenon using a built-in pore pressure generation, the classic Mohr-Coulomb criterion for

stone column and Elastic behavior for mattress and raft foundation.

3.2 Effect of the free field The limits of the free field impose in which lateral limits of the main grid of the model are coupled into the grid of the free field by viscous dashpots to simulate a quiet boundary (Fig. 4). With this model, plane waves propagating upward suffer no distortion at the boundary

because the field-free grid supplies conditions that are identical to those in an infinite model

(Itasca Consulting Group, Inc. 2005). 3.3 The seismic load On May 21, 2003 at 19:44:19 local time, Zemmouri area in the Boumerdes region about 70 km east of the capital Algiers has suffered an earthquake of magnitude in the order of 6.8. The location of the epicenter is (36,90 North, 3.71 Est) determined by the U.S.G.S (united states geological survey). The focal depth of the earthquake was about 10 km. The recorded amplitude is given by the station of Dar El Beida (0.52 g) source (C.G.S. Center of earthquake engineering, Algeria) (Japanese Reconnaissance Team. Boumerdes earthquake, May 21, 2003). The earthquake lasted 27.675 seconds, and the strong accelerations were between 6 and 10 seconds. The maximum acceleration is of 556.79 cm/s2 (0.57 g) at 7.70 seconds


Fig. 4 Field-free limit.

after the Baseline Correction made by SeismoSignal software (Fig. 6). In the response spectrum,

the strong horizontal seismic acceleration corresponds nearly to 3 Hz and 6 to 8 Hz (Fig. 5).

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28

-600

-400

-200

0

200

400

Ac

ce

lera

tio

n (

m/s

2)

Time (sec)

Dar El Beida E-W

0,1 1 100

500

1000

1500

2000

Pseu

do

-Accele

rati

on

Sp

ectr

a

Frequency (Hz) Fig. 5 Accelerogram Signal Est-West at the Dar El Beida station and corresponding

response spectrum.

The earthquake is modelized by a sinusoidal acceleration applied at the base of the model

in the horizontal direction after static equilibrium was achieved. For the long calculation time

when the accelerogram is applied, the accelerogram is converted to the variable-amplitude

harmonic ground motion record illustrated in Fig. 6 and expressed by the following equation.

ftteeamplta ttft 2sin....)( / (14)

Where = 0.73 and tf = 6 (time of ground motion). This equation is a contribution from the expression of Bathurst and Hatami (1998):

ftteampl

ta t 2sin..2

)( (15)

FLAC3D 3.00

Itasca Consulting Group, Inc.Minneapolis, MN USA

Step 1554740 Model Perspective19:26:05 Sun May 13 2018

Center: X: -2.223e-004 Y: -2.620e-003 Z: 5.565e+000

Rotation: X: 30.000 Y: 0.000 Z: 30.000

Dist: 1.627e+002 Mag.: 0.8Ang.: 22.500

Block GroupNoneloose_sands-column1s-column2

s-column3s-column4s-column5

s-column6matressraft_foundation


The wave has a frequency (f) of 3.3 Hz with an amplitude (ampl) equal to the maximum acceleration of the accelerogram of Dar El Beida EW = 0.57g. Where: α = 5.5, β = 55, and ζ = 12 are constant coefficients; f = frequency; and t = time = 6 seconds.

0 1 2 3 4 5 6-6

-5

-4

-3

-2

-1

0

1

2

3

4

5

6

Accele

rati

on

(m

/s2)

Time (sec)

Fig. 6 Variable-Amplitude harmonic ground motion

4. RESULTS AND DISCUSSION 4.1 Study of the liquefaction phenomenon After a numerical analysis of the variation of the EPWP at different depths from the surface (2.5 m, 7.5 m and 9.5m from the surface) without and with improved by stone columns, under the centerline and the edge of foundation. Fig. 7 displays the recorded EPWP and accelerations during the motion under the centerline of foundation (P1, P4 and P7) (Fig. 2). As seen in this figure, a negative EPWP build-up tendency without reinforced by stone columns at depths 7.5 m and 2.5 m (P4 and P7) from the surface (this negative build-up increases when the depth close from the foundation). However, at depth 9.5 m (P1) (the zones close to base where the motion applied) significant positive EPWP was attained and the liquefaction occurs. With reinforced by stone columns, the negative EPWP reducing strongly, at depth 7.5 m (P4), the EPWP equal the effective stress σ0

’, so the liquefaction begins but at depth 2.5 m (P7) there is no liquefaction occurs with stone columns, at depth 9.5 m from the surface (P1) the computed EPWP of the loose sand with stone columns is about 20% less than without column but the liquefaction still occurs. With stone columns the acceleration in the soil must be slightly stronger than without them (the soil reinforced is stiffer) (Adalier et al. 2003), but with loose sand it is


Fig. 7 EPWP and acceleration measured under the centerline of foundation

the inverse, the acceleration was much stronger (6 to 8 times) without stone columns than with them, this can be explained by the increase in the magnitude and the spatial extent of horizontal normal strains in the foundation with stronger base input excitation (negative EPWP build-up tendency was observed) (Adalier et al. 2003), in other words dilatation of soil under the centerline of foundation.

In general, the stiffening effect of stone columns was obvious compared to liquefaction mitigation. Fig. 8 displays the recorded EPWP and accelerations during the motion under the edge of foundation (P2, P5 and P8) (Fig. 2). The same thing almost happened like in centerline foundation, the difference that:

1. the values of negative EPWP build-up is lower compared to the values under the centerline of foundation without stone columns and the same thing for the positive

σ0’= 1.40. 10

5 Pa

σ0’= 1.25. 10

5 Pa

σ0’= 1.60. 10

5 Pa


Fig. 8 EPWP and acceleration measured under the edge of foundation

EPWP build-up; 2. the difference between the accelerations in the two cases (without and with

stone columns) was decrease (2 times). 4.2 Effect of stone column diameter Asgari et al. (2013) studied the effect of the diameter in reducing the liquefaction in function of lateral displacement in medium saturated Nevada sand soil reinforced by stone columns subjected to the Loma Prieta (1989) earthquakes with a scaled maximum acceleration of 0.35 g and a ground slope of 4°. The general result shows that when the diameter increases, the lateral displacement decreases. In this paper, to examine the influence of the diameter of stone columns in mitigation of liquefaction. A series of modeling, each one is modeled with a value of diameter. The diameter values considered vary between 0.6 –1.4 m. Fig. 9 presents change of the magnitude of EPWP at the end of motion for different diameters under the centerline and the edge of foundation.

σ0’= 1.40. 10

5 Pa

σ0’= 1.25. 10

5 Pa

σ0’= 1.00. 10

5 Pa


(a) (b)

(c)

Fig. 9 Effect of the stone column diameter on the mitigation of liquefaction. (a) 9.5 m depth, (b)

7.5 m depth, (c) 2.5 m depth

As Fig. 9 shows, the magnitude of EPWP depends on the column diameter, it is changing

in a variable manner depending on the depth:

1. at depth of 9.5 m (P1 and P2), the magnitude of EPWP under the centerline and edge of foundation, resulting in general a decrease in the EPWP as the diameter increases, in this

depth the decrease in the diameter did not prevent soil liquefaction (EPWP > σ0’ = 140

kPa);

2. at depth of 7.5 m (P4 and P5), the inverse happened, the magnitude of EPWP under the centerline and edge of foundation increases as diameter increases, the results were

similar at the range [0.6-0.8] m under the centerline and edge of foundation, in this depth

the liquefaction occurs whatever the column diameter values (EPWP > σ0’ = 125 kPa);

3. at depth of 2.5m (P7 and P8), the same thing happened at a depth of 2.5 m, in general, the magnitude of EPWP under the centerline and edge of foundation decreases as

diameter increases, but in this case the liquefaction does not occur (EPWP < 1- σ0’ = 160

0,6 0,8 1,0 1,2 1,4

180

190

200

210

220

Excess P

ore

Wate

r P

ressu

re (

kP

a)

Diameter (m)

Center

Edge

0,6 0,8 1,0 1,2 1,4120

130

140

150

160

170

Excess P

ore

Wa

ter

Pre

ssu

re (

kP

a)

Diameter (m)

Centre

Edge

0,6 0,8 1,0 1,2 1,4-20

-10

0

10

20

30

40

Excess P

ore

Wa

ter

Pre

ssu

re (

Pa)

Diameter (m)

Center

Edge


kPa under the centerline, 2- σ0’ = 100 kPa under the edge), but when the diameter greater

than 1.3 m the magnitude of EPWP becomes negative under the whole foundation, as a

result the increase in the diameter after 1.3 m becomes a factor in increasing the

magnitude and the spatial extent of horizontal normal strains (Adalier et al. 2003).

4.3 Effect of distribution mattress thickness The thickness of the distribution mattress depends mainly on different characteristics (materials, geometry) whether for the stone column or the surrounding soil or also the mattress itself, this thickness must be at least 0.5 m under the loads distributed with spacings between columns of at most 3 m knowing that the mattress is not usable for isolated or strip foundations (Dhouib and Blondeau 2005). In order to find out the effect of the thickness of the mattress, a series of numerical simulations were performed. The thickness values considered were 0.35 –0.95 m. Fig. 10 displays the magnitude of EPWP in the end of the motion for each value of thickness:

1. under the centerline of foundation, at depth of 9.5 m the magnitude of EPWP decreases as diameter increases, the same thing continues to occur at the depth 7.5 m but with a difference that the magnitude increases after the thickness 0.55 m, in the two depths the increase in the value of the thickness of mattress did not stop soil liquefaction (EPWP > 1- σ0

’ = 140 kPa at depth of 9.5 m, 2- σ0’ =

125 kPa at depth of 9.5 m). At depth of 2.5 m, the magnitude of EPWP decreases as diameter increases and the liquefaction does not occur for all values of thickness.

2. Under the edge of foundation, at all depth the magnitude of EPWP decreases as thickness of mattress increases except for what happens at depth 9.5 m when the thickness value exceeds 0.70 m where the relationship between EPWP and Thickness becomes reverse.

It is also noted that the difference between the values of EPWP under the centerline and edge (Centerline’ EPWP > Edge’ EPWP) decreases as the depth approaches from the overload effect until it becomes under the edge is greater than under the centerline. 4.3 Effect of surcharge Previous studies of the effect of overload (surcharge) on soil liquefaction (Adalier et al. 2003, Asgari et al. 2013, Tang et al. 2015) have proved a decrease in EPWP as surcharge increases with a negative EPWP build-up trend for the silty soil. Fig. 11 shows the effect of the surcharge at the loose sand reinforced with stone columns in the range from 0 to 150 kPa on the magnitude of EPWP at the end of ground motion:

1. Under the centerline and in the greater depth (P1) where the liquefaction occurs and the positive EPWP build-up, the increase in the surcharge leads to increase in the EPWP, the same thing happened under the edge but with lower values;

2. For the other zones where the negative EPWP build-up tendency, the EPWP decreased with increasing surcharge under the centerline of foundation, but under


(a) (b)

(c)

Fig. 10 Effect of the distribution mattress thickness on the mitigation of liquefaction. (a) 9.5 m

depth, (b) 7.5 m depth, (c) 2.5 m depth

the edge of foundation the value of the EPWP is somewhat stable.

It is also noted that the difference between the values of EPWP under the centerline and

edge (Centerline’ EPWP > Edge’ EPWP) decreases as the depth approaches from the overload

effect until it becomes under the edge is greater than under the centerline.

5. ADDITIONAL EXPLANATIONS The study results are somewhat different from previous studies especially in terms of negative EPWP build-up without stone columns, as a result the values of acceleration were much stronger than with stone columns, this can be explained by the differences in properties between loose sand and silty soil used in those studies, the difference in the arrangement of the columns and their number. In addition, the great amplitude given by the station of Dar El Beida (0.57 g) source (Boumerdes earthquake) Compared to those used in other studies.

0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

190

195

200

205

210

215

220

225

Excess P

ore

Wa

ter

Pre

ssu

re (

kP

a)

Mattres Thickness (m)

Center

Edge

0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,0

125

130

135

140

145

Excess P

ore

Wa

ter

Pre

ssu

re (

kP

a)


Center

Edge

0,3 0,4 0,5 0,6 0,7 0,8 0,9 1,00

5000

10000

15000

20000

25000

30000

Excess P

ore

Wa

ter

Pre

ssu

re (

kP

a)


Center

Edge


(a) (b)

(c)

Fig. 11 Effect of the surcharge on the mitigation of liquefaction. (a) 9.5 m depth, (b) 7.5 m depth,

(c) 2.5 m depth

6. CONCLUSIONS The aim of this work is the analysis to the liquefaction mitigation for a foundation rests on loose sand reinforced by stone columns, using a numerical method of finite differences (FLAC 3D), a series of 3D numerical simulations were performed to know the effectiveness of stone columns to loose sand and a varying key parameters (column diameter, mattress thickness and surcharge). EPWP and acceleration results indicate a stiffer role of stone columns for the loose sand compared to unimproved case. However, full liquefaction occurs in the in the lower zones close to base where the ground motion applied in the remediated loose sand. The stone columns are supposed to increase acceleration in order to strengthen the soil but in this study the acceleration in the unimproved soil is greater

-20 0 20 40 60 80 100 120 140 160175

180

185

190

195

200

205

210

Excess P

ore

Wate

r P

ressu

re (

kP

a)

Surcharge (kPa)

Center

Edge

-20 0 20 40 60 80 100 120 140 160120

130

140

150

160

170

Excess P

ore

Wate

r P

ressu

re (

kP

a)

Surcharge (kPa)

Center

Edge

-20 0 20 40 60 80 100 120 140 160

5

10

15

20

25

30

35

Excess P

ore

Wate

r P

ressu

re (

kP

a)

Surcharge (kPa)

Center

Edge


than improved case due to the great negative EPWP build-up under the foundation (under the centerline > under the edge), this negative build-up caused by the great amplitude of the earthquake (0.57g) and the surcharge. From to the results obtained, the effect of stone columns diameter needs more study and attention under another conditions to know its role more and determine the range for its effective values in liquefaction mitigation, in this study the range of effective values is [0.60-0.80] m. The same thing for the effect of mattress thickness, in this study the range of effective values is [0.35-0.70] m. In the other hand, when the depth close to the effect of surcharge, whatever the value of the surcharge, the magnitude of EPWP decreases. Additional studies (numerical or experimental) and parametric studies are needed to explore liquefaction mitigation of loose sand with stone columns under another conditions specially for stone column diameter and other parameters like permeability of soil or stone column input motion parameters (amplitude, frequency, …). REFERENCES Adalier, K. and Elgamal, A. (2004), “Mitigation of liquefaction and associated ground

deformations by stone columns.”, Eng. Geo., Vol. 72, 275-291. Adalier, K., Elgamal, A., Meneses, J. and Baez, J. (2003) “Stone columns as

liquefaction countermeasure in non-plastic silty soils.” Soil. Dyn. Earthqu. Eng., Vol. 23, 571-584.

Asgari, A., Olaei, M. and Bagheri, M. (2013) “Numerical simulation of improvement of a liquefiable soil layer using stone column and pile-pinning techniques.” Soil. Dyn. Earthqu. Eng., Vol. 51, 77-96.

Bathurst, R.J. and Hatami, K. (1998), “seismic response analysis of a geosynthetic reinforced soil retaining wall.” Geosynth. Int., Vol. 5(1-2), 127-166.

Bouafia, A. (2014), Application de la Dynamique des Sols (Problèmes Résolus), Office des Publications Universitaires, ALG. (In French).

Bouckovalas, G.D., Papadimitriou, A.G. and Niarchos, D. (2009), “Gravel drains for the remediation of liquefiable sites: The Seed & Booker (1977) approach revisited”, In: Proceedings of the International Conference on Performance-Based Design in Earthquake Geotechnical Engineering, London, England.

Byrne, P.A. (1991), “A Cyclic shear-volume coupling and pore-pressure model for sand”, In: Second International Conference on Recent Advances in Geotechnical Earthquake Engineering and Soil Dynamics, St. Louis-Missouri, USA.

Carter, L., Green, R., Bradely, B. and Cubrinovski, M. (2014), The influence of near fault motions on liquefaction triggering during the Canterbury earthquake sequence, In: Orense, R.P., Towhata, I. and Chouw, N. Soil Liquefaction during Recent Large-Scale Earthquakes, Taylor & Francis Group, London, ENG.

Dhouib, A. and Blondeau, F. (2005), Colonnes Ballastées (Techniques de Mise en Oeuvre, Domaine d’Application, Comportement, Justification, Contrôle, Axes de Recherches et Développement), Presse de l’Ecole Nationale des Ponts et Chaussées, Paris. (In French).


Elgamal, A., Lu, J. and Forcellini, D. (2009), “Mitigation of Liquefaction-induced lateral deformation in a sloping stratum: three-dimensional numerical simulation.” J. Geotech. Geoenviron. Eng., Vol. 135, 1672-1682.

Esmaeili, M. and Hakimpour, M.H. (2015), “Three dimensional numerical modelling of stone column to mitigate liquefaction potential of sands.” J. Seismlog. Earth. Eng., Vol. 17(2), 127-140.

Hardin, B.O. and Drnevich, V.P. (1972), “Shear moduls and damping in soils.” ASCE J. Soil. Mech. Found. Div., Vol. 98(SM7), 667-692.

Ishihara, K. and Yamazaki, F. (1980), “Cyclic simple shear tests on saturated sand in multidirectional loading.” Soils. Found., Vol. 20(1), 45-59.

Itasca Consulting Group, Inc. (2005), Fast Lagrangian Analysis of Continua in 3 Dimensions, Optional Features, Itasca Consulting Group, Inc., Minneapolis-Minnesota, USA.

Japanese Reconnaissance Team. Boumerdes earthquake, May 21, 2003 (2004), Japanese Report on the Boumerdes Earthquake May 21, 2003. Algiers, Algeria, Japan Association of Earthquake Engineering (JAEE), japan Society of Civil Engineering (JSCE), Architectural Institute of Japan (AIJ), Japan Geotechnical Engineering Society (JGES).

Lysmer, J. and Kuhlemeyer, R.L. (1969), “Finite dynamic model for infinite media.” J. Eng. Mech. Div., Vol. 95(4), 859-878.

Martin, G.R., Lam, I.P., McCaski, S.L. and Tsai, C.F. (1981), “A Parametric Study of an Effective Stress Liquefaction Model” In: Proceedings: First International Conference on Recent Advances in Geotechnical Earthquake Engineering & Soil Dynamics, St. Louis, Missouri, USA.

Meshkinghalam, H., Hajialilue-Bonab, M. and Azar, A.K. (2017), “Numerical investigation of stone columns system for liquefaction and settlement diminution potential.” Inter. J. Geo-Eng., Vol. 8, 11.

Millea, M.T. (1990), Liquefaction mitigation technology, National Civil Engineering Laboratory, Port Hueneme, CA.

Murali Krishna, A. and Madhav, M.R. (2009), “Engineering of Ground for Liquefaction Mitigation Using Granular Columnar Inclusions: Recent Developments.” Amer. J. Eng. App. Sci., Vol. 2(3), 526-536.

Murali Krishna, A., Madhav, M.R. and Madhavi Latha, G. (2006), “Liquefaction mitigation of ground treated with granular piles: densification effect.” ISET J. Earthqu. Tech. Vol. 43(4), 105-120.

Priebe, H.J. (1998), “Vibro-replacement to prevent earthquake induced liquefaction” In: Proceedings of the Geotechnique-Colloquium, Dramstadt, Germany.

Qu, M., Xie, Q., Cao, X., Zhao, W., He, J. and Jin, J. (2016), “Model test of stone columns as liquefaction countermeasure in sandy soils.” Fron. Struct. Civ. Eng., Vol. 10(4), 481-487.

Rathje, E.M., Abrahamson, N.A. and Bray, J.D. (1998), “Simplified frequency content estimates of earthquake ground motion.” J. Geotech. Geoenviron. Eng., Vol 124(2), 150-159.

Sadrekarimi, A. and Ghalandarzadah, A. (2005), “Evaluation of gravel drains and compacted sand piles in mitigating liquefaction.” Ground. Improv., Vol. 9(3), 91-104.


Sasaki, Y. and Taniguchi, E. (1980), “Shaking table tests on gravel drains to prevent liquefaction of sand deposits.” Soils. Found., Vol. 22(3), 1-14.

Seed, H.B. and Booker, J.R. (1977), “Stabilization of potentially liquefiable sand deposits using gravel drain systems.” J. Eng. Mech. Div., Vol. 103(7), 757-768.

Seed, H.B., Idriss, I.M. and Arango, I. (1983), “Evaluation of liquefaction potential using field performance Data.” J. Geotech. Eng., Vol. 109(3), 458-482.

Suzuki, Y. and Koyamada, K. (1997), “Prediction of Liquefiable resistance based on CPT tip resistance and sleeve friction” In: Proceedings of the 14th International Conference on Soil Mechanics and Foundation Engineering, Hamburg, Germany.

Tang, L., Zhang, X. and Ling, X. (2015), “Numerical simulation of centrifuge experiments on liquefaction mitigation of silty soils using stone columns.” KSCE J. Civ. Eng., Vol. 20(2), 631-638.

Date post:	23-Oct-2020
Category:	Documents
Upload:	others
View:	2 times
Download:	0 times

New *Abdelkader Noui , Toufik Karech and Tayeb Bouzid3) · 2018. 8. 19. · shallow foundation...

Documents