Indian Geotechnical Conference IGC2016

15-17 December 2016, IIT Madras, Chennai, India

APPLICATION OF PARTICLE IMAGE VELOCIMETRY TO STUDY

PILE SOIL INTERACTION

Sreelakshmi G1
Asha M N2
Suraj S3
1
Assistant Professor, Department of Civil Engineering, CMR Institute of Technology, Bangalore – 560037
Email:[email protected]
2
Associate Professor, Department of Civil Engineering, CMR Institute of Technology, Bangalore – 560037
Email:[email protected]
3
B.E Student, Department of Civil Engineering, CMR Institute of Technology, Bangalore – 560037
Email: [email protected]

ABSTRACT: The behaviour of pile foundations under different loading conditions has been a challenge for engineers as
the design philosophy of these systems are based on the principles of elastic theory and do not consider soil deformations
at soil-pile interface. In this paper attempts are made to investigate soil-pile interaction through PIV (Particle image
velocimetry) technique. The experimental studies are carried out in a steel tank of dimensions 450 mm × 200 mm in plan
and 450 mm deep. The front side of the steel tank is made of perspex to facilitate image capture. Two half section
aluminum piles of hollow and solid sections with same flexural rigidity are used in experiments. Wood’s scaling law is
employed to model pile dimensions. The infill material has an average size of 3.8 mm and experimental studies are carried
out at a uniform density. Under axi-symmetric conditions, the piles are driven at a uniform rate and the deformations of
the infill, surrounding the pile are seized using a high resolution digital camera. The images are used as inputs in Geo-PIV
software, and analysis is performed using MAT LAB code to capture soil-pile interaction.
Keywords: Axi-symmetric, Particle Image Velocimetry, Geo-PIV, Soil-Pile interactions.

1 INTRODUCTION Chore et. al. (2009), Khodair and Abdel-Mohti (2014)

carried out finite element analysis on framed structures
Pile foundations are used to transfer loads from with piles embedded in sand and clay. Lemnitzer (2008)
a superstructure, through weak, compressible strata or carried out full-scale cyclic field test on a 3x3-pile group
water onto stronger, more compact, less compressible, and and a separate single fixed-head pile to evaluate group
stiffer soil or rock at depth. There are many examples interaction effects. Experimental and conventional method
where, the structures with pile foundations have failed. of interaction studies done by Ravi and Gunneswara
For example, Mexico city pile foundation collapse (1985), (2011) on model building frames with pile foundation
13 storeyed building collapse in Shangai (2009), Delhi under vertical loads revealed that conventional method
Metro pillar collapse (2009), Wilmington Bridge pile failed to capture non linear soil structure interaction
foundation settlement (2014) are some of the cases of pile behaviour. Boominathan et. al. (2010) carried out the
foundation failures. Failure of a pile can be due to lateral load tests on 1X2, 2X2, 1X4 and 3X3 model pile
different conditions such as collapse of pile during driving groups embedded in soft clay. It had been found that the
itself, failure due to buckling or due to excessive lateral capacity of piles in 3X3 group at three diameter
settlements that surpass the serviceability criteria, shear spacing was 40% less than that of the single pile. Reddy
failure of piles, tilting of piles or due to rotational failure et. al. (2016) conducted 1-g model experiments in the
of piles. The different loading conditions that initiate such laboratory to investigate the behavior of a single pile in
failures are sudden loading, gradual loading and lateral sand under combined uplift and lateral load. Dimensions
loading due to wind and earthquake forces. In any loading of the model pile were ascertained using physical scaling
condition, capacity of pile is dependent upon the laws, based on adopted material properties of model and
behaviour of surrounding soil and therefore, it is necessary prototype pile foundations. The suitability of Particle
to study the soil-pile interaction behaviour. Image Velocimetry to understand the deformation
2 LITERATURE REVIEW mechanism has been a matter of research in the recent
years. Plane strain analysis was used to study the
Several studies carried out by Meyerhof (1976), Poulos interaction between pile and soil using PIV technique by a
and Davis (1980), Kraft (1991) concentrated on the few researchers. White et. al. (2003, 2004) has studied the
performance of axial capacity of piles. Limited studies are resistance offered by sand to penetration of a pile by
reported that demonstrates structure-pile-soil interaction. carrying out model studies in a tank of 1000 mm length
Several investigators like Buragohain and Shah (1977), and 745 mm depth using PIV technique. Tran (2005)

1
 Indian Geotechnical Conference IGC2016
15-17 December 2016, IIT Madras, Chennai, India

investigated the 1g study of suction caissons in layered facilitate image capture. The infill material is filled within
sand-silt soil, where caissons were installed by slow and the test tank by pluviation, and a constant density of 1450
rapid pumping. The mechanism of heave formation in kg/m3 is maintained in all the experiments. A clearance of
dense sand and deformation of the silt layer was studied 50 mm is also maintained on top of the fill. Photograph of
using a half-caisson model and particle image velocimetry the test set up is shown in Fig. 1. Experiments are carried
(PIV) technique. out for two different pile length to diameter (L/D) ratios of
10 and 15. Model piles are driven through the infill at a
The present study aims at investigating soil-pile uniform rate of loading (4kN/min) using Universal
interaction through image analysis. Half section solid and Testing machine. Deformations around the pile are seized
hollow piles are driven into the infill material and the zone with the help of high resolution digital camera. When the
of influence of each pile is estimated using Geo-PIV pile is driven into the infill, a video is recorded and using
software. an image grabber tool, still images at different micro
3 MATERIALS USED seconds are captured.

3.1 Infill Material

The infill material used for the model studies has an
average size of 3.8 mm. The material has an angle of
internal friction of 320.
3.2 Pile
A solid reinforced concrete pile of 350mm diameter is
modelled in the experimental studies. Model pile
dimensions are determined by using scaling law proposed
by Wood et al (2002). To investigate soil-pile interaction, Fig 1.Experimental test set up
scale factors are considered for flexural rigidity.
According to Wood’s scaling law as shown in Table 1, a 5 Geo -PIV – A TOOL FOR PARTICLE
factor of 1/10 is used for scaling length and 1/105 is used IMAGE VELOCIMETRY
for scaling flexural rigidity.
The Geo-PIV software implements the principles of
Table 1. Scaling law used in the present study Particle Image Velocimetry (PIV) in a style suited to the
analysis of geotechnical test results. This tool uses the
(Wood et al. 2002)
principles of PIV to gather displacement data from
sequences of digital images captured during elemental
Variable Scaling Factors tests. GeoPIV is a MatLab module, which runs at the
Length 1/10 MatLab command line. The development and
performance of the software are described in detail by
Density 1
White and Take (2002). The different screen images are
Stiffness 1/10 used as inputs for Geo-PIV to calculate displacement
Stress 1/10 vectors and strain contours.
Strain 1
3 6 EXPERIMENTAL RESULTS
Force 1/10
5
For analysing the images there are two methods in the
Flexural rigidity 1/10 Geo-PIV Software. They are control point and Mesh point
analysis. In the present study since images has low spatial
Half section aluminium model piles, closed at the bottom variation in brightness, calibration in the control point
are used for the experiments. Hollow aluminium pile has analysis method is not possible. Hence, attempts are made
an outer diameter of 15.8 mm and inner diameter of 13.4 to investigate soil-pile interaction using mesh analysis.
mm whereas solid aluminium pile has 14.3 mm diameter. The experimental results are analysed for two different
Model piles are provided with a pile cap of 13 mm aspects namely, displacement vectors and shear strain
thickness. The length of the pile includes the embedment contours.
length and an additional free-standing length of 50 mm for
avoiding contact of the pile cap with the soil. 6.1 Displacement vectors of Hollow and Solid
Pile for different Length / Diameter ratios
4 EXPERIMENTAL SET UP
The displacement vectors obtained in case of hollow and
The experiments are carried out in a steel tank of solid pile corresponding to an L/D ratio of 10 are shown in
dimensions 600 mm × 200 mm in plan and 450 mm Fig. 2 and Fig. 3. The displacement vectors could capture
depth (Parkin and Lunne, 1982). A perspex sheet is used the large displacements induced at the surface of the infill.
on one long side or face side of the test tank in order to From the displacement vectors, it is observed that around

2
 Indian Geotechnical Conference IGC2016
15-17 December 2016, IIT Madras, Chennai, India

the pile, displacements are more in hollow pile when pile. Fig. 4 and Fig. 5 compare shear strain contours for
compared to solid pile. These observations are visible in pile with L/D ratio of 15 for solid and hollow pile
the experimental studies as well. When the pile L/D is 15, respectively. From the figures it is observed that for same
the observations are quite similar. The displacement L/D ratio, hollow piles are having a larger zone of
vectors in solid pile are comparatively less as compared to influence, in comparison to solid pile. The results are
the hollow pile for both length/diameter ratios. similar for L /D ratio of 10.

Fig.2 Displacement Vector Diagram for Hollow Pile

Fig. 4 Strain contour for Solid Pile L/D ratio =15

Fig. 3 Displacement Vector Diagram for Solid Pile

6.2 Shear strain contours of Hollow and Solid
Pile for different Length / Diameter ratios
Shear strain contours for different L/D ratios are Fig. 5 Strain contour for Hollow Pile L/D ratio =15
compared to investigate interference/smear zone around a

3
 Indian Geotechnical Conference IGC2016
15-17 December 2016, IIT Madras, Chennai, India

7 DISCUSSION OF RESULTS Geotech. Geoenviron. Eng, 10.1061/(ASCE)GT.1943-

5606.0001314, 04015030.
From the analysis of experimental results it is observed
that zone of inference of hollow pile is more than solid Meyerhof, G. G. (1976) ‘Bearing capacity and settlement
pile. This is mainly due to lesser mass density of hollow of pile foundations’. J.Geotech. Engineering,Div.,
pile which has resulted in increased disturbance of ASCE, 102(3), pp.197-228.
surrounding soil mass. Lesser the mass density of the pile, Parkin, A. K. and Lunne, T. (1982) ‘Boundary effects in
lesser will be the resistance to lateral soil pressures. the laboratory calibration of a cone penetrometer for
sand’. Proc.2ndEuropean Symposium on Penetration
8 CONCLUSIONS Testing, Netherlands.
From the present experimental study, the following Poulos, H. G and Davis, E. H (1980) ‘Pile Foundation
conclusions can be made: Analysis and Design’. Rainbow‐Bridge Book Co,
554786.
 PIV technique is successful in capturing soil-pile
Ravi, K.R. and Gunneswara, R (2011) ‘Experimental
interaction.
study of a modeled building frame supported by pile
 Though the flexural rigidity is same, zone of
groups embedded in cohesionless soil’. Interaction and
influence is more for hollow pile rather than solid pile
multiscale mechanics, 4(4), pp.321-336.
which could be due to lesser mass density of hollow
piles. Tran .M (2005) ‘Installation of Suction Caissons in Dense
Sand and the Influence of Silt and Cemented Layers’,
9 SCOPE FOR FUTURE WORK Phd Thesis.
This paper, presents a pilot study carried out to Wood, D.M., Crewe, A. and Taylor, C (2002) ‘Shaking
demonstrate the effectiveness of Geo-PIV as a tool to table testing of geotechnical models’. International
study pile-soil interaction. Even the size of infill material Journal of Physical Modelling in Geotechnics, 2(1),
used for the experimental studies is slightly large. Hence, pp.01-13.
detailed experimental studies are required to understand White, D. J., Take, W. A. & Bolton, M. D (2003) ‘Soil
soil-pile interaction for different in-situ soil conditions. deformation measurement using particle image
High resolution images facilitate use of control point velocimetry (PIV) and photogrammetry’.
analysis in Geo-PIV which will help us in correctly Ge´otechnique 53(7), pp.619–631.
estimating the zone of soil-pile interaction with respect to
diameter of pile. Yasser Khodair, and Ahmed Abdel-Mohti (2014)
‘Numerical Analysis of Pile–Soil Interaction under
References Axial and Lateral Loads’. International Journal of
Concrete Structures and Materials, 8(3), pp.239–249.
Buragohain, D. N. and Shah. V. L (1977) ‘ Curved
interface elements for interaction problems.’ Proc. Int.
Symposium on Soil‐Structure Interaction, Roorkee,
India, pp.197‐202.
Chandrasekaran, S., Boominathan, A., and Dodagoudar,
G (2010) ‘Group Interaction Effects on Laterally
Loaded Piles in Clay.’ J. Geotech. Geoenviron. Eng.,
10.1061/(ASCE)GT.1943-5606.0000245, pp.573-582.
Chore H.S., Ingle R.K., Sawant V.A (2009) ‘Building
frame- pile foundation – soil interaction analysis: a
parametric study’. Interact Multiscale Mech 3(1),
pp.55–79
Kraft, L. M., Jr (1991) ‘Computing axial pile capacity in
sands for offshore conditions’. Marine Geotechnology,
9(1), pp.61-92.
Lemnitzer, A (2008) ‘Experimental Testing of a Full-
Scale Pile Group Under Lateral Loading’, The 14
World Conference on Earthquake Engineering,
October 12-17, 2008, Beijing, China.
Madhusudan Reddy, K and Ayothiraman, R (2015)
‘Experimental Studies on Behavior of Single Pile
under Combined Uplift and Lateral Loading.’ J.