STRUCTURAL RESPONSE INCLUDING

SEISMIC SOIL-PILESTRUCTURE
INTERACTION
M. G. Durante, L. Di Sarno, A.L. Simonelli
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

INDEX
• SERIES Project PILESI Project (at Bristol) :
“Experimental investigation of soil-pile-structure seismic interaction”

• Experimental set-up and test programme

• Sample Experimental results

• Numerical analyses and comparison

2|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

SERIES PILESI Project (at Bristol)

Experimental investigation of soil-pile-structure seismic interaction

Project leader: Prof. A. L. Simonelli (University of Sannio, Italy)

Objective: To investigate the seismic response of pile groups in layered
soils in order to provide a benchmark for numerical-model
validation purposes.

A shear stack is used to apply appropriate boundary conditions

3|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Bristol SHEAR STACK

Input motion direction

1-g shaking table has two main advantages:

• to work on a larger and more reliable physical
model, which allows detailed measurements of pile
response and numerous combinations regarding
soil profile, pile-head boundary conditions,
superstructure features
• to plan and realize a wide set of tests by
reasonable operating costs
4|32
4|3
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

SOA: Unsolved issues on seismic soil-pile interaction

Less research has been carried out on short piles and/or interfaces located
close to the pile head.

Lack of simple formulations for assessing kinematic moments at the pile head
for non homogeneous soils.

Material nonlinearity and plasticity need to be explored in a more systematic

way.
The effect of the transient nature of the input motion on both kinematic and
inertial bending moments along the pile requires further investigation.

Proper “combination” rule between kinematic and inertial bending

moments should be defined.

Need of experimental data to validate the huge amount of numerical

findings.

The reliability of existing analytical formulations to predict the inertial

and kinematical bending moment along the pile, as a function of the
5|32
soil and pile properties and the input ground motion characteristics,
 STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

a) Model details and

a)
instrumentations
a)
b)
b) b)
140 140 70 70 mm
a) a) a) 140 140
140 70 140 70 70 mm
70 mm
Upper layer: Bottom layer:
340 mm thick 460 mm thick
Leighton Buzzard sand Leighton Buzzard sand

340 mm
340 mm
E E Fraction E + Fraction
340 mm

E LB-Fraction E

750 mm
750 mm
B
750 mm

Shear wave
Relative Dry unit weight γd

460 mm
460 mm velocity
Soil B+E
Thickness
B+E density Dr (%) (kN/m3)
Vs2/Vs1
Void Vs (m/s)
ratio e
460 mm

layers H (mm)
B+E Nov June Nov June Nov June Nov June
2010 2011 2010 2011 2010 2011 2010 2011
1190 mm 1190 mm
Top
340 0.9 28 10 13.63 13.13 51 54
LB(E)
1.59 1.57
b) 1190 mm
Bottom
460 0.48 41 50 17.46 17.92 81 85
LB(E+B)

c) c) d) d)

c) d)

y direction

Strain- gauges accelerometers z direction

LVDT
x direction 6|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Instrumentations details

SOIL DEPOSITION

7|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Instrumentations details

Piles
and
strain gauges

8|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Model configurations
1) 2) 3)

4) 5) 6)

7) 1) Free head pile

5) Short cap connecting piles 3,4,5
2) SDOF on pile 1 6) Short cap with SDOF on pile 5
3) SDOF on pile 4 7) Long cap from pile 1 to pile 3
4) SDOF on
Element pile 5 details
Geometrical Unit weight Length Young’s modulus E
[type] [mm] [kN/m3] [mm] [GPa]
De = 22.23
Pile 27 750 70
t = 0.71
100 (phase I)
Aluminum column Cross section 3x12 27 50 -100 70
(phase II)
100
Steel column Cross section 3x12 80 21
(phase II)
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Model
Free Head Pile Short
Short cap
cap
configurations

Free Head Pile

10|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Model
configurations
with
SDOF

Free Head Piles + SDOF

11|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Free Head Piles + SDOF

SETRA Accel.

12|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Model
configurations

Short cap

13|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Short cap

Strain gauge wires

LVDT

SETRA Accel.
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Model
configurations

Short cap + SDOF

15|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Model
configurations

Large cap

16|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Test objectives
test - input motion objective

Horizontal natural frequency of the

White noise signals applied on the
Horizontal white noise test vibrating table along y direction
system; Shear stack influence on the
system motion
Sinedwell signals applied on the Horizontal kinematic interaction
Horizontal sinedwell test vibrating table along y direction factors
White noise signals applied on the Vertical natural frequency of the
Vertical white noise test vibrating table along z direction system
Sinedwell signals applied on the
Vertical sinedwell test vibrating table along z direction
Vertical kinematic interaction factors
Apply small increments of lateral load
Static impedance test (Pullover Khh, Krr, Krh components of stiffness
and measure pile head deflection at
test) two points matrix
Horizontal motion applied to the
Kvv, Krr, Krv components of stiffness
Dynamic impedance test shaking table. The shear stack is
arranged in configuration LC matrix

Traction and quick release applied on Natural frequency and damping of

Snapback test pile 5 head the embedded pile

Frequency scaled seismic motions Observe the behavior under seismic

Earthquake tests applied on the shaking table motion

17|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Phase I: test details

Short Cap + Free Head Pile Free Head Pile
Single Degree + Single + Single
Free Head Pile of Freedom on Degree of Degree of
Short Cap (SC) pile 5 Freedom on Freedom on
(FHP) pile 5 pile 4
(SC+SDOF pile (FHP+SDOF (FHP+SDOF
5) pile 5) pile 4)

White Noise 4 3 15 12 11
Sinedwell 21 15 75 90 90
Phase I + Phase II ≈ 600 tests
Phase II: test details
Short Cap + Free Head Pile + Free Head Pile + Free Head Pile +
Single Degree Single Degree of Single Degree of Single Degree of
Free Head
Short Cap of Freedom on Freedom on Freedom on Freedom on Long Cap
Pile
(SC) pile 5 pile 5 pile 4 pile 1 (LC)
(FHP)
(SC+SDOF pile (FHP+SDOF pile (FHP+SDOF pile (FHP+SDOF pile
5) 5) 4) 1)
White Noise 8 1 9 2 3 3 2
Sinedwell 45 1 58 6 26 6 24
Vertical White
2 - - - - - -
Noise
Vertical
26 - - - - - -
Sinedwell
Earthquake 5 - 9 5 - - -
Snapback 1 - - 5 - - -
Pullover 1 - - - - - -

18|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

SAMPLE EXPERIMENTAL RESULTS

FHP+SDOF 5
CONFIGURATION

19|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION
Sinedwell f=20Hz a=0.1g
HARMONIC SDOF masses from 75 to 975 grams
FHP+SDOF 5
about Resonance between input motion and Pile+SDOF system !!!
INPUT
m=75g m=175g m=475g Soil surface
m=125g m=275g m=975g Soil interface

800
Free-field Pile 4 Pile 5

600
z [mm]

400

200

a b c
0
0 0.1 0.2 0.3 0.4 0.5 0 100 200 300 400 500 0 1000 2000 3000
amax [g] M [Nmm] M [Nmm]
20|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

NUMERICAL SIMULATION: FLAC 2D CODE

I) SOIL MODEL

II) PILES MODEL

III) OSCILLATOR
MODEL
IV) PILE CONNECTION
FHP+SDOF 5
CONFIGURATION
21|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

SOIL MODEL  Ramberg-Osgood constitutive model

 Rayleigh damping (5%) at small strain

1.2
1
0.8

G/Gmax
0.6
0.4
0.2
0
30
25
20

D (%)
15

( )
𝑏1 10
2 ( 2 − 𝑏1 )2 ( 5 − 2 𝑏1 ) Vs 1 ,h 12 𝑉 𝑠 1 , 𝐵 =𝑉 𝑠 1 , h 1
𝐻
𝜔1 = h1
5
2 ( 𝑏1 − 4 𝑏1 +4 )
2 2
H 0
-6 -5 -4 -3 -2
10 10 10 10 10
2
2 A 1 + α α 1 A2 V s 1,B g
Soil Dynamics and ω = 2 2 Ramberg-Osgood
Earthquake A 3 + α α 1 A4 H
Engineering Vucetic&Dobry (1991), for Ip=0-7.5%
22|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

PILES MODEL
Normal coupling springs Shear coupling springs

P 𝑘𝑛 =𝑛h 𝑧
𝑝 𝑢𝑙𝑡 = 𝑓 ( 𝛾 ,𝜑 , 𝑘0 ,𝑘𝑎 , 𝐷 ) Failure in the soil
y
0 0

0.2 0.2

z [m]
z [m]

0.4 0.4

0.6 0.6

8 9 9
0.0 7 7 7
2.0x10 4.0x10 6.0x10 0.0 3 4
7.5x10 1.5x10
4
2.3x10 0.0 8.0x10 1.6x10 2.4x10
kn/D [N/m3] pult [Pa] ks [N/m2]

23|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

OSCILLATOR MODEL: FIXED BASE RESPONSE

Validation using the finite element code
SAP2000
50
Experimental data
Frequency [Hz] 40 SAP2000

30

20

10
Rectangular
section: 3x12 mm 0 200 400 600 800 1000
Height: 100mm Massa [g]
E = 70 GPa
24|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

OSCILLATOR MODEL: FIXED BASE RESPONSE

-2
1x10
SAP2000
8x10
-3 m=175g FLAC2D
Harmonic input
Displacement [m]

-3
4x10

0
-3
-4x10
-3
-8x10
-2
-1x10
0.0 0.2 0.4 0.6 0.8 1.0 4x10
-4

Time [s] SAP2000

FLAC2D
-4
2x10

Displacement [m]
Satisfactory match 0

SAP2000 – FLAC2D -4
-2x10
m=975g
-4
-4x10
0.0 0.2 0.4 0.6 0.8 1.0
Time [s]

Experimental and numerical assessment of dynamic soil-pile-structure-interaction - Maria Giovanna Durante

STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

COMPARISON BETWEEN
EXPERIMENTAL AND NUMERICAL RESULTS

FHP+SDOF 5
CONFIGURATION

26|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

HARMONIC INPUT Input: sinedwell f=20Hz a=0.1g FHP+SDOF 5

about Resonance between input motion and Pile+SDOF system !!!
Experimental m=175g Experimental m=975g Soil surface
Numerical m=175g Numerical m=975g Soil interface

800
Free-field Pile 4 Pile 5

600
z [mm]

400

200

a b c
0
0 0.1 0.2 0.3 0.4 0.5 0 200 400 600 0 1000 2000 3000 4000
amax [g] M [Nmm] M [Nmm]

27|32
STRUCTURAL RESPONSE INCLUDING
Experimental data
SEISMIC SOIL-PILESTRUCTURENumerical
INTERACTION
simulations

HARMONIC INPUT Input: sinedwell f=20Hz a=0.1g

4 Pile 5 : close to the head

Displacement [mm]

2 m=175g
Mass

-2
Surface
-4 Experimental data
3000 Numerical simulations
2000
M5(100mm)
M [Nmm]

1000
0
-1000 4
cement [mm]

-2000
-3000 Interface 2 Mass
2 0 0.4 0.8 1.2 1.6 2
t [s] 0
28|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

PERIOD ELONGATION FHP

COMPARISON AMONG 3 FHP+SDOF
CONFIGURATIONS SC+SDOF
Input Tolmezzo accelerogram SDOF H=100mm
SF 12, amax= 0.1g Mass=150 gr, ffix=27.0 Hz

Free-field Bendingthe
The cap (SC) reparts moment
action pile 5 the 3 piles !
among
a max [g] a max [g]
0.05 0.10 0.15 0.20 0.25 0.30 0.35 0 100 200 300 400 500 600 700 800
900 0

800 100

700
200
600
300
z 500 z
[mm] [mm] 400
400
500
300 FHP+SDOF FHP+SDO
5 F5
600
200 FHP
FHP
SC+SDOF
100 700 5
SC+SDOF5

0 800
29|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

PERIOD ELONGATION
FHP+SDOF
EFFECT ON TSSI FOR THE 2 SC+SDOF
CONFIGURATIONS f
SSI (T)SSI /Tfix
SDOF [Hz] [%]
ffix=27.0 Hz
FHP+SDOF 23.1 14.51
SC+SDOF 26.9 0.44
100

90

80
FHP+SDOF
70 SC+SDOF
Fixed base
60
Amplification

50

40

30

20

10

0
0 5 10 15 20 25 30 35 40 45 50
f [Hz] 30|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Main Conclusions
Experimental results

• FF response Free field is not affected by the mass on the oscillator (as expected)

• Pile 5+SDOF The oscillator at the top of the central pile significantly affected the
response of the close lateral piles, notwithstanding piles
are not connected

• Bending moments The SDOF on Pile 5 gave two primary effects on the lateral Pile 4:
1. the maximum bending location is not at the soil interface (as
expected for a purely kinematic interaction), but is clearly at a
shallower depth
2. the maximum bending moment varies with the mass of the
oscillator on the central pile; it achieves its maximum value for
the oscillator mass which induces the resonance condition
between the oscillator-pile system and the input motion

• Period elongation The SPS interaction gives a measured period elongation of 42%
!
31|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Main Conclusions
Numerical simulation

• FF response The numerical model is able to adequately reproduce the free-field

response

• Pile 5+SDOF The resonance effects and the bending moment can be reproduced
fairly accurately with the numerical model for the pile connected to the
oscillator, both in terms of envelope and time history response

• Bending moments The numerical model is able to effectively catch the bending
moment values due to kinematic interaction ;
the maximum bending location is slightly different probably because
the influence of the SDOF is not well accounted for

• Period elongation The measured elongation of 42% , is well reproduced by the

numerical analysis !!!

32|32
STRUCTURAL RESPONSE INCLUDING
SEISMIC SOIL-PILESTRUCTURE INTERACTION

Thank you