Citation for published version:

Pelecanos, L & Soga, K 2017, 'Innovative Structural Health Monitoring Of Foundation Piles Using Distributed
Fibre-Optic Sensing', Paper presented at 8th International Conference on Structural Engineering and
Construction Management 2017, Kandy, Sri Lanka, 7/12/17 - 10/12/17 pp. 1-8.

Publication date:
2017

Document Version
Peer reviewed version

Link to publication

University of Bath

Alternative formats
If you require this document in an alternative format, please contact:
[email protected]
General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.

Download date: 07. Oct. 2022

 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

INNOVATIVE STRUCTURAL HEALTH MONITORING OF FOUNDATION PILES

USING DISTRIBUTED FIBRE-OPTIC SENSING
L. Pelecanos1*, Kenichi Soga2
1Lecturer, University of Bath, UK. Formerly: Research Associate, University of Cambridge, UK.
2Chancellor’s Professor, University of California, Berkeley, US. Formerly: Professor, University of Cambridge, UK.
*Correspondence E-mail: [email protected], TP: +44-1225-386250
Abstract: Tall buildings in congested urban areas need deep pile foundations to carry the huge
building and operational loads. The current state-of-the-art design of foundation piles is based
on previous knowledge gained from small-scale laboratory tests and some instrumented field
tests. In some cases the as-built pile might be different from the initial design because of various
construction constraints, which might results in a smaller pile capacity. Also, pile bearing
capacity may be compromised over time because of material strength degradation or additional
unexpected loads. It is therefore essential that we monitor the response of foundation piles in
the future so that we are able to assess their structural health and take necessary engineering
remedial actions on time.
Distributed fibre-optic strain sensing using BOTDR or BOTDA is a novel instrumentation
technique that offers a spatially-continuous data. An optical fibre cable is installed along the
depth of the pile which serves as a sensor itself, as it provides distributed strain data every 5-
10cm along the cable length. This is superior to more conventional discrete point-based sensors
which provided limited monitoring information at pre-specified location points. The
availability of a distributed strain regime offers a number of advantages when it comes to
studying soil-structure interaction problems such as foundation piles.
This paper describes the use of distributed fibre-optic sensing in monitoring foundation piles
and presents some recent case studies in London. The monitoring data is also complemented by
relevant finite-element pile models that are able to reproduce the response of the examined
foundation piles.
Keywords: structural health monitoring; fibre-optic strain sensing; foundation piles;
finite-element analysis
1. Introduction strain localisation, abnormal pile diameter
etc.
Instrumented pile load tests are used widely
to establish relevant geotechnical Recent developments in optical strain
parameters for foundation piles, such as the sensing, such as the Brillouin Optical Time-
ultimate bearing capacity, or shaft and base Domain Reflectometry (BOTDR) or Analysis
resistance. Traditionally, pile load tests were (BOTDA) offer the opportunity to obtain
monitored with a number of discrete point distributed values of monitored strain. Fibre
sensors, such as vibrating-wire-strain- optic strain sensing provides a wealth of
gauges (VWSGs). Although these kind of information that is very useful in
sensors have been proved to be accurate understanding the actual behaviour of piles.
and reliable they still lack in term of
This paper presents the method of
continuity of the data. That is because they
distributed fibre optic sensing and its
provide discrete values of strain, i.e. only at
application in monitoring pile foundations.
the point where the VWSG sensor in
The principle of the method is first
installed.
introduced and then a recent example of a
In many cases of complicated soil-structure relevant application is presented.
interaction, a spatially-continuous profile
2. Distributed fibre optic sensing
along the entire length of the pile is very
useful in illustrating some finer details of This section provides a brief description of
the performance of the pile, such as necking, the principles of BOTDR/A. The complete
description of the physical principle (i.e.
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

theoretical physics, photonics and optics) Changes in temperature and/or strain

and the relevant experimental approaches induce a density change in the cable and
followed for calibration are not included, as therefore change in the acoustic velocity, vα,
they can be found elsewhere (Horiguchi et of the light too. As the strain or temperature
al. 1995) (Mohamad 2007) (Iten 2011) (Soga at a given location change, the frequency of
2014) (Soga et al. 2015). A detailed the backscattered light is shifted by an
description of the theory of distributed FO amount which is approximately linearly
strain sensing and its applications in civil proportional to the applied strain, Δε, or
and geotechnical infrastructure is given by temperature, ΔT.
Kechavarzi et al. (2016), whereas examples
Therefore knowledge about of this
of application to foundation piles are given
frequency difference can provide
by Klar et al. (2007), Ouyang et al. (2015)
information about the applied strain and
and Pelecanos et al. (2016, 2017a, 2017b) and
temperature changes at the location where
energy piles by Ouyang et al. (2018a,b,c).
the back-scattered light was generated. As
More examples on monitoring soil-structure
the speed of light is constant, the location
interaction may be found by Acikgoz et al.
can be evaluated by measuring the time
(2016, 2017), Cheung et al. (2010) Di Murro
since the light was initially sent into the
(2016), Schwamb et al. (2014) and Soga et al.
fibre. Back-scattered light is generated at
(2017).
every point along the entire length of the
2.1 Principle of BOTDR/A fibre and therefore by resolving both time
A fibre optic (FO) cable allows light waves and frequency a continuous strain profile
from a FO analyser to travel along its entire along the fibre can be determined.
length through total internal reflection. This
allows a signal to be carried over very long
distances, similar to broadband Internet.
Backscattered signals are generated as the
light wave passes through the optical fibre
and presents itself as Rayleigh, Raman and
Brillouin spectrum. Within the Brillouin
backscatter, the peak frequency experiences
a shift that is considered to be linearly
proportional to applied strain. Using the
measured time required for the
backscattered signal to return to the
analyser, the specific location at which this
frequency shift is observed can be estimated Figure 1: Principle of BOTDR/A.
accurately. Therefore, the entire fibre optic
2.2 Fibre Optic cables
cable is practically serving as a distributed
strain sensor. Strain on an optical fibre can be generated
from two sources, mechanical or thermal.
The FO analyser sends a light with of 1550 Therefore, two types of optical fibre cables
nm wavelength into an optical fibre and the are installed and are shown in Figure 2:
generated Brillouin spectrum of the back- Fujikura 4-core single mode fibres
scattered light has 25-27 MHz bandwidth reinforced ribbon cable for strain sensing
and around 11 GHz central peak frequency (strain sensing cable) and Excel 8-core single
when no strain is applied on the fibre. The mode fibres loose tube for temperature
back-scattered Brillouin central frequency, compensation (temperature cable). While
vb, is related to the input light according to they are both attached to the reinforcement
Eq. 1 and this is provided directly from the cage, the fibre optic cores of the temperature
FO analyser. cable sit in a gel which isolates any transfer
of mechanical strains from the outer
coating. Thus it is only subjected to thermal
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

changes. These measurements are used to With the pile loaded axially, it is assumed
compensate the readings measured from the that the concrete pile will have negligible
strain cables to provide an accurate reading hoop strain across its cross section and
of interest, the actual mechanical strain. therefore a 10m loop cable for both strain
and temperature is prepared and secured at
the end of the bottom reinforcement cage to
serve as a zero-strain loop for referencing
and compensation purposes.
For the ease of data interpretation, a pre-
strain of about 1000-2000με is often
introduced to the strain cable. Anchorage is
provided on the bottom loop end by cable
wire clamps before stretching the strain
(a) cable to the predetermined pre-strain. Strain
cable is then secured with another set of
cable wire clamps at the top of the
reinforcement cage before supplementing
the anchorage by either spot gluing with
epoxy glue or using cable ties at
approximately every 0.5-1.0m interval.
Temperature cables are loosely secured next
to the strain cables with cable ties as they
are routed to the top of the cage.
Once the bottom cage has been
instrumented, it is lowered into the
borehole. The fibre optic cables are then
unwound from the reels on each side of the
(b)
borehole as the cage is lowered. Pre-
Figure 2: Fibre Optic cables: (a) strain cable, (b) straining is carried out for the strain cables
temperature cable. for subsequent reinforcement cages as well
2.3. Sensor installation without epoxy glue due to time constraints.
Concrete is subsequently poured in the
Installation of FO cables is usually done on borehole and as the concrete cures the FO
site. Long pile foundations typically consist cables become securely embedded within
of a number of steel reinforcement cage the pile. Further details of FO cable
segments and therefore the bottom steel installation in piles established at the
cage is instrumented on the ground. The FO University of Cambridge can be found in
cables are running along the entire length of (Klar et al. 2006), (Soga 2014) and (Soga et
the bottom segment on two opposite sides al. 2015).
of the pile and a loop of some FO cable is
made close to the bottom of the segment. 2.4. Fibre Optic data analysis
The longitudinal cables are pre-strained (i.e. Measuring the Brillouin frequency difference,
a tensile strain is applied) using cable one can obtain the applied strain on the cable.
clamps at the two ends of the steel cage. FO cables are able to detect strains due to both
Once the borehole is dug, the bottom cage is mechanical and thermal loads, the two
inserted and while the other cages are components can be analysed separately. The
spliced onto the bottom cage and the whole measured frequency difference from the
pile lowered down in the borehole, the “temperature cable”, ΔvbT, is influenced only by
remaining FO cable is attached to them. changes in temperature, whereas that from the
Finally, the two ends of the FO cable run “strain cable”, ΔvbS, is influenced by changes in
from the top of the pile to the FO analyser. both mechanical load and temperature.
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

Therefore, changes in temperature, ΔT, can be inserted in the ground after the drilling process.
obtained from Eq. 1 (where, CTT is a property The pile tested is 0.305m diameter (0.343m at
of the cable, obtained by calibrating the the top 6m because of a steel casing around the
“temperature cable”, which determines how pile) and 25m long, as shown in Figure 3 (a). On
temperature affects the Brillouin frequency the same figure, the soil stratigraphy is also
reading of the cable and it is usually around included with some known material properties
1.1∙10-3 GHz/oC). obtained from relevant triaxial and simple
shear laboratory tests. The pile test was carried
(1) out once the concrete material achieved a
specified value of minimum strength. The pile
The thermal strain, εtemp, (the strain that test consists of three consecutive cycles of
corresponds to free thermal expansion strain applied load (at the top of the pile) of up to
due to temperature change) is then given by Eq. 720kN, 1080kN and 1985kN for each of the
2 (where, αc is the thermal expansion three cycles, achieved after several loading and
coefficient of concrete and it is usually around unloading steps (Figure 3 (b)). The pile was
9.65 με/oC). instrumented with distributed FO cables on
(2) two opposite sides of the pile and a number of
discrete VWSGs along the pile depth.
The real (observed) strain, εreal, (the actual strain
that the pile experiences in the field) is then
given by Eq. 3 (where, CE is a property of the
fibre, obtained by calibrating the “strain cable”,
which determines how strain affects the
Brillouin frequency and it is usually around
5∙10-4 GHz/με; and CT is a property of the fibre
that determines how the Brillouin frequency is
affected by temperature difference, and it is
usually around 1.0∙10-3 GHz/oC).

( ) (3)

The mechanical (constrained) strain, εmech, (the

reaction strain that is the result of both the
applied mechanical load and temperature) is
then given by Eq. 4 (a)
(4)
[ ]

3. Application
This section provides a recent application of the
BOTDR/A method to monitor a pile load test in
London.
3.1. Description of pile load test
This project in London was designed to house a
fourteen-storey office building with two
basement levels. A number of mini piles of (b)
0.305m diameter were constructed in close Figure 3: Pile load test: (a) geometry of the pile
proximity to support the superstructure. A and soil stratigraphy, (b) history of applied
high-strength steel reinforcing case was load.
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

3.2. Fibre optic monitoring data

Figure 4(a) shows the axial strain in the pile for
the three peak values of the three cycles as it
was captured by the FO cables and the VWSGs,
whereas Figure 4(b) shows the corresponding
axial force profiles (calculated from strains
multiplied by the pile axial rigidity, EA, as
described by Eq. 7 and using E=30000MPa. No
VWSG data were obtained for the largest cycle
(i.e. for loading of 1985kN), as there was a
malfunction of the VWSG instruments, and
therefore only FO data is available for this load
case. It is also shown that there is some scatter
in the FO data values which is currently a
known issue with distributed FO strain sensing
systems. This is because the standard resolution (b)
of FO is constant and about 30-50με and
therefore this becomes relatively less significant
for larger applied loads (which imply larger
induced strains). The waviness of FO strains
may offer a challenge when differentiating
strain data profiles to obtain shaft friction
values, but their spatial continuity allows for a
distributed sensing of localised strains, e.g.
necking, fracture etc., whereas, such localised
features would not be identified by discrete
monitoring systems (such as VWSGs). Figure
4(c) shows the vertical displacements, u, of the
pile from the FO cables. The values from the
FOs were obtained by integrating the strain
profiles and adding those to absolute
displacement measurements from displacement
transducers at the top of the pile. (c)
Figure 4: Monitoring data: profiles of (a) axial
strain, (b) axial force and (c) vertical
displacement.
The results of a simplified numerical finite
element (FE) beam-spring model are included
for comparison in all profiles in Figure 4. The
simplified FE analysis considered a single
vertical pile loaded axially from the top
modelled with linear beam elements and
represented the surrounding soil with non-
linear springs which is a practical approach as
opposed to the more common way of
modelling the soil with solid elements. The
nonlinear springs follow a 4-parameter load-
transfer curve (t-z), as described by Eq. 5,
(a) (where tm and km are dependent on
geotechnical properties and are related to the
maximum soil stiffness and shaft capacity
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

respectively, whereas parameters d and h are develop relevant load-transfer (t-z) curves
related to degradation of soil stiffness and for further finite element analyses of the
hardening). foundation piles.
(5) It is therefore suggested that similar
( ) monitoring schemes may be used in the
√[ ( ) ] future for monitoring axially loaded piles as
a standard sensing system. Finally, the
All the beam elements and non-linear springs wealth of monitoring data may also be used
contribute to the global stiffness matrix (which to derive relevant computational models for
is a sum of the pile stiffness matric, Kp, and the axially loaded foundation piles.
soil stiffness matrix, Ks) and therefore the global
FE equilibrium equations, as described in Eq. 6.
Acknowledgements
Due to the nonlinear nature of the soil-spring
the external load, P, is applied incrementally This work was carried out at the EPSRC and
and the solution provides the nodal Innovate UK sponsored Centre for Smart
displacements, u. More details about the Infrastructure and Construction (CSIC) at
numerical model may be found from Pelecanos the University of Cambridge, when both
et al. (2017b) and Pelecanos & Soga (2017a,b, Authors were working there. Their
2018). contribution is gratefully acknowledged.
([ ] , -) * + * + (6) References

It is shown from Figure 4 that the FE model is Acikgoz, MS, Pelecanos, L, Giardina, G, Aitken,
able to reproduce the observed profiles of axial J & Soga, K 2017, 'Distributed sensing of
strain and force and vertical displacement very a masonry vault during nearby piling',
well. This suggests that the FO strain data were Structural Control and Health
able to provide a very good insight into the real Monitoring, vol 24, no. 3, p. e1872.
performance of the monitored pile and that Acikgoz, MS, Pelecanos, L, Giardina, G & Soga,
they can be reliably used in the future to K 2016, 'Field monitoring of piling effects
establish the response of foundation piles. on a nearby masonry vault using
4. Conclusion distributed sensing.', International
Conference of Smart Infrastructure and
This paper presents the principle of Construction, ICE Publishing,
distributed fibre optic sensing using the Cambridge.
Brillouin Optical Time Domain
Reflectometry (BOTDR) or Analysis Cheung, L, Soga, K, Bennett, PJ, Kobayashi, Y,
(BOTDA) method and its application on Amatya, B & Wright, P 2010, 'Optical
monitoring foundation piles. A recent fibre strain measurement for tunnel
example from monitoring a pile load test in lining monitoring', Proceedings of the
London is presented and the monitoring ICE - Geotechnical Engineering, vol 163,
results are discussed. A complementary no. 3, pp. 119-130.
finite element analysis with an in-house Di Murro, V, Pelecanos, L, Soga, K, Kechavarzi,
code is also included which confirms that C, Morton, RF & Scibile, L 2016,
the monitored data compare quite well with 'Distributed fibre optic long-term
the expected performance of the pile. monitoring of concrete-lined tunnel
It is shown that distributed fibre optic section TT10 at CERN.', International
sensing may be used reliably to monitor Conference of Smart Infrastructure and
axially loaded foundation piles. Spatially Construction, ICE Publishing,
continuous data about the axial strain from Cambridge.
the FO cables can provide profiles of axial Horiguchi, T, Shimizu, K, Kurashima, T,
force, axial displacement and shaft friction. Tateda, M & Koyamada, Y 1995,
The latter information can also be used to 'Development of a distributed sensing
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

technique using Brillouin scattering', Development-Future Challenges, Rome,

Journal of Light-wave Technology, vol Italy. (Accepted)
13, no. 7, pp. 1296-1302.
Pelecanos, L, Soga, K, Hardy, S, Blair, A &
Iten, M 2011, 'Novel applications of distributed Carter, K 2016, 'Distributed fibre optic
fiber-optic sensing in geotechnical monitoring of tension piles under a
engineering', PhD Thesis, ETH, Zurich. basement excavation at the V&A
museum in London.', International
Kechavarzi, C, Soga, K, de Battista, N,
Conference of Smart Infrastructure and
Pelecanos, L, Elshafie, MZEB & Mair, RJ
Construction, ICE Publishing,
2016, Distributed Fibre Optic Strain
Cambridge.
Sensing for Monitoring Civil
Infrastructure, Thomas Telford, London. Pelecanos, L, Soga, K, Chunge, MPM, Ouyang,
Y, Kwan, V, Kechavarzi, C & Nicholson,
Klar, A, Bennett, PJ, Soga, K, Mair, RJ, Tester, P,
D 2017a, 'Distributed fibre-optic
Fernie, R, St John, HD & Thorp-Peterson,
monitoring of an Osterberg-cell pile test
G 2006, 'Distributed strain measurement
in London.', Geotechnique Letters, vol 7,
for pile foundations', Proceedings of the
no. 2, pp. 1-9.
ICE - Geotechnical Engineering, vol 159,
no. 3, pp. 135-144. Pelecanos, L., Soga, K., Elshafie, M., Nicholas,
d. B., Kechavarzi, C., Gue, C. Y., Ouyang,
Mohamad, H 2007, 'Distributed optical fibre
Y. and Seo, H., 2017b. Distributed Fibre
strain sensing of geotechnical structures.
Optic Sensing of Axially Loaded Bored
PhD Thesis', University of Cambridge,
Piles. Journal of Geotechnical and
Cambridge, UK.
Geoenvironmental Engineering. (In
Ouyang, Y, Broadbent, K, Bell, A, Pelecanos, L Press).
& Soga, K 2015, 'The use of fibre optic
Pelecanos, L & Soga, K 2017a, 'Using
instrumentation to monitor the O-Cell
distributed strain data to evaluate load-
load test on a single working pile in
transfer curves for axially loaded piles',
London', Proceedings of the XVI
Journal of Geotechnical &
European Conference on Soil Mechanics
Geoenvironmental Engineering, ASCE
and Geotechnical Engineering,
(Under Review).
Edinburgh.
Pelecanos, L., Soga, K. 2017b, „The use of
Ouyang, Y., Pelecanos, L., Soga, K. 2018a,
distributed fibre-optic strain data to
'Thermo-mechanical response test of a
develop finite element models for
cement grouted column embedded in
foundation piles‟. 6th International Forum
London Clay', Geotechnique (Under
on Opto-electronic Sensor-based
Review).
Monitoring in Geoengineering. Nanjing,
Ouyang, Y., Pelecanos, L., Soga, K. 2018b, China, November, 2017.
„Finite element modelling of thermo-
Pelecanos, L., Soga, K. 2018, „Development of
mechanical behaviour and thermal soil-
load-transfer curves for axially-loaded
structure interaction of a thermo-active
piles based on inverse analysis of fibre-
cement column buried in London Clay‟.
optic strain data using finite element
9th European Conference on Numerical
analysis and optimisation‟. 9th European
Methods in Geotechnical Engineering,
Conference on Numerical Methods in
Porto, Portugal. (Accepted)
Geotechnical Engineering Porto,
Ouyang, Y., Pelecanos, L., Soga, K. 2018c, „The Portugal. (Accepted)
field investigation of tension cracks on a
Schwamb, T, Soga, K, Mair, RJ, Elshafie, MZEB,
slim cement grouted column‟. DFI-EFFC
R., S, Boquet, C & Greenwood, J 2014,
International Conference on Deep
'Fibre optic monitoring of a deep circular
Foundations and Ground Improvement:
excavation', Proceedings of the ICE -
Urbanization and Infrastructure
 th
8 International Conference on Structural Engineering and Construction Management, 2017

ICSECM2017-Paper120

Geotechnical Engineering, vol 167, no. 2,

pp. 144-154.
Soga, K 2014, 'XII Croce Lecture:
Understanding the real performance of
geotechnical structures using an
innovative fibre optic distributed strain
measurement technology', Rivista
Italiana di Geotechnica, vol 4, pp. 7-48.
Soga, K, Kechavarzi, C, Pelecanos, L, de
Battista, N, Williamson, M, Gue, CY, Di
Murro, V & Elshafie, M 2017, 'Distributed
fibre optic strain sensing for monitoring
underground structures - Tunnels Case
Studies ', in S Pamukcu, L Cheng (eds.),
Underground Sensing, 1st edn, Elsevier.
Soga, K, Kwan, V, Pelecanos, L, Rui, Y,
Schwamb, T, Seo, H & Wilcock, M 2015,
'The role of distributed sensing in
understanding the engineering
performance of geotechnical structures',
Proceedings of the XVI European
Conference on Soil Mechanics and
Geotechnical Engineering, Edinburgh.