ArticlePDF Available

Abstract and Figures

This paper investigates the use of distributed optical fiber sensors (DOFS) based on Optical Frequency Domain Reflectometry of Rayleigh backscattering for Structural Health Monitoring purposes in civil engineering structures. More specifically, the results of a series of laboratory experiments aimed at assessing the suitability and accuracy of DOFS for crack monitoring in reinforced concrete members subjected to external loading are reported. The experiments consisted on three-point bending tests of concrete beams, where a polyamide-coated optical fiber sensor was bonded directly onto the surface of an unaltered reinforcement bar and protected by a layer of silicone. The strain measurements obtained by the DOFS system exhibited an accuracy equivalent to that provided by traditional electrical foil gauges. Moreover, the analysis of the high spatial resolution strain profiles provided by the DOFS enabled the effective detection of crack formation. Furthermore, the comparison of the reinforcement strain profiles with measurements from a digital image correlation system revealed that determining the location of cracks and tracking the evolution of the crack width over time were both feasible, with most errors being below ±3 cm and ±20 mm, for the crack location and crack width, respectively.
Content may be subject to copyright.
Full Terms & Conditions of access and use can be found at
https://www.tandfonline.com/action/journalInformation?journalCode=nsie20
Structure and Infrastructure Engineering
Maintenance, Management, Life-Cycle Design and Performance
ISSN: 1573-2479 (Print) 1744-8980 (Online) Journal homepage: https://www.tandfonline.com/loi/nsie20
Crack monitoring in reinforced concrete beams by
distributed optical fiber sensors
Carlos G. Berrocal, Ignasi Fernandez & Rasmus Rempling
To cite this article: Carlos G. Berrocal, Ignasi Fernandez & Rasmus Rempling (2020): Crack
monitoring in reinforced concrete beams by distributed optical fiber sensors, Structure and
Infrastructure Engineering, DOI: 10.1080/15732479.2020.1731558
To link to this article: https://doi.org/10.1080/15732479.2020.1731558
© 2020 Informa UK Limited, trading as
Taylor & Francis Group
Published online: 26 Feb 2020.
Submit your article to this journal
View related articles
View Crossmark data
Crack monitoring in reinforced concrete beams by distributed optical
fiber sensors
Carlos G. Berrocal
a,b
, Ignasi Fernandez
a
and Rasmus Rempling
a,c
a
Division of Structural Engineering, Chalmers University of Technology, G
oteborg, Sweden;
b
Thomas Concrete Group AB, G
oteborg, Sweden;
c
NCC Sverige AB, G
oteborg, Sweden
ABSTRACT
This paper investigates the use of distributed optical fiber sensors (DOFS) based on Optical Frequency
Domain Reflectometry of Rayleigh backscattering for Structural Health Monitoring purposes in civil
engineering structures. More specifically, the results of a series of laboratory experiments aimed at
assessing the suitability and accuracy of DOFS for crack monitoring in reinforced concrete members
subjected to external loading are reported. The experiments consisted on three-point bending tests of
concrete beams, where a polyamide-coated optical fiber sensor was bonded directly onto the surface
of an unaltered reinforcement bar and protected by a layer of silicone. The strain measurements
obtained by the DOFS system exhibited an accuracy equivalent to that provided by traditional elec-
trical foil gauges. Moreover, the analysis of the high spatial resolution strain profiles provided by the
DOFS enabled the effective detection of crack formation. Furthermore, the comparison of the
reinforcement strain profiles with measurements from a digital image correlation system revealed that
determining the location of cracks and tracking the evolution of the crack width over time were both
feasible, with most errors being below ±3 cm and ±20 mm, for the crack location and crack width,
respectively.
ARTICLE HISTORY
Received 21 July 2019
Revised 26 October 2019
Accepted 22 November 2019
KEYWORDS
Concrete beams; crack
monitoring; damage
assessment; distributed
optical fiber sensors;
reinforced concrete;
structural health monitoring
1. Introduction
The use of an effective Structural Health Monitoring (SHM)
system could enable the early detection of deficiencies in
civil engineering structures and aid engineers and infrastruc-
ture owners in making informed decisions, thereby leading
to well-planned, timely maintenance operations with min-
imal disruption to the users and the consequent substantial
savings in terms of both money and time. However, despite
the great efforts made in the research and development of
SHM systems, see e.g. (Cawley, 2018; Seo, Hu, & Lee, 2016),
due to the lack of reliable, scalable and affordable monitor-
ing solutions, SHM is not yet implemented as a standard
practice in most civil engineering structures (Gli
si
c,
Hubbell, Sigurdardottir, & Yao, 2013).
One of the main elements in SHM systems is the sensing
devices. Traditional SHM systems have often relied on rela-
tively large, electrically powered sensors, such as displace-
ment transducers, inclinometers, accelerometers, etc.
However, over the last decades, the use of optical fiber sen-
sors in the field of SHM for strain and temperature moni-
toring has gained popularity owing to several features that
makes them very suitable for SHM applications. Optical
fibers can be easily bonded or embedded into a structure
thanks to their reduced dimensions (often <200 mmin
diameter), they are lightweight which facilitates their trans-
port and handling, they are chemically inert, corrosion
resistant and able to operate over a wide range of
temperatures, thus being suitable for a variety of environ-
ments and unlike electrically powered sensors, optical fibers
are not affected by electromagnetic interference (EMI)
caused by nearby electromagnetic field sources. More
importantly, a single optical fiber can accommodate multiple
sensing points along its length, thereby saving tremendous
amounts of time and money in wiring and installation
(Casas & Cruz, 2003).
Among the existing types of fiber optical measurement,
those based on Fiber Bragg Grating (FBG) have received a
great deal of attention in the past, see e.g. (Davis,
Bellemore, & Kersey, 1997; Morey, Dunphy, & Meltz, 1992)
and to date they are the most widely used system (Barrias,
Casas, & Villalba, 2016). The principle of FBG systems is
based on the modification of the refractive index inside the
fiber core by creating a periodic pattern that selectively
reflects a specific wavelength (Todd, Johnson, & Vohra,
2001). Using wavelength division multiplexing, multiple
points along the fiber can be measured, yet this system
presents certain limitations with respect to spatial resolution
and number of sensing points per fiber, hence it is regarded
as a discrete or quasi-distributed system (Majumder,
Gangopadhyay, Chakraborty, Dasgupta, & Bhattacharya,
2008). Consequently, even though FBG systems can provide
useful information about the global behaviour of a structure
and the local behaviour at specific locations, they fall short
of providing a truly continuous description of the strain
CONTACT Carlos G. Berrocal carlos.gil@chalmers.se Division of Structural Engineering, Chalmers University of Technology, G
oteborg, Sweden
ß2020 Informa UK Limited, trading as Taylor & Francis Group
STRUCTURE AND INFRASTRUCTURE ENGINEERING
https://doi.org/10.1080/15732479.2020.1731558
variation along the structure, which can be a critical issue in
certain cases, e.g. for crack monitoring in concrete struc-
tures, where the exact location of cracks cannot be a pri-
ori determined.
Distributed Optical Fiber Sensors (DOFS), on the other
hand, measure the return loss of the emitted light caused by
the backscattering that occurs along the fiber due to differ-
ent phenomena. Accordingly, DOFS do not require the
alteration of the fiber core since every segment in the fiber
acts as a sensor, thereby achieving a significant improve-
ment in spatial resolution compared to FBG systems. There
are three different scattering phenomena occurring concur-
rently that can be used to measure variations of temperature
and/or strain along the fiber: Raman, Brillouin and Rayleigh
scattering (Soga & Luo, 2018).
Raman scattering arises from the thermal vibration of the
glass molecules in the fiber core as light travels through the
fiber and is highly sensitive to temperature variations
(Rodriguez, Casas, & Villalba, 2015b). Brillouin scattering is
produced by the interaction of backscattered light and
acoustic waves generated when changes in the density of the
material occur as a result of thermal effects. Brillouin scat-
tering is sensitive to external changes of both mechanical
strain and temperature and despite its measuring range can
reach lengths of up to more than 300 km (Gyger, Rochat,
Chin, Nikl
es, & Th
evenaz, 2014), the spatial resolution that
can be achieved is often limited to several centimetres
(G
uemes, Fern
andez-L
opez, & Soller, 2010). Rayleigh scat-
tering, on the other hand, refers to the elastic distribution of
light in all directions that happens when light interferes
with local inhomogeneities in the fiber core that are smaller
than the wavelength of the light itself. These inhomogene-
ities are caused by fluctuations in the density and compos-
ition of the fiber core, which makes Rayleigh scattering
sensitive to both mechanical strain and temperature changes
(Palmieri, 2013). Systems based on Rayleigh scattering are
currently limited to a measuring range of up to 2 km, but in
exchange they provide an unprecedented spatial resolution
that can go down to the the sub-millimetric scale, thereby
offering new possibilities for the development of damage
detection systems (Rodriguez et al., 2015b).
Rayleigh scattering analysis is based on Optical
Frequency Domain Reflectometry (OFDR) where the
Rayleigh scattering pattern occurring along the fiber is ini-
tially measured and stored as a fingerprint or signature of
the fiber in a reference state. The Rayleigh scattering profile
is measured again when the fiber is subjected to mechanical
strain or temperature perturbations and subsequently both
data sets are divided into small segments that are Fourier
transformed into the frequency domain. By performing a
cross-correlation operation between the reference and per-
turbed states, a spectral shift in the correlation peak can be
found which can be then calibrated to strain or temperature
changes (Ding et al., 2018). The principle of OFDR of
Rayleigh backscattering is illustrated in Figure 1.
The clear advantages of optical fiber sensors have
attracted the interest of many researchers over the years. As
a result, the suitability of optical fiber systems in SHM has
already been tested for various experiments in the labora-
tory, see e.g. (Bado, Casas, & Barrias, 2018; Bao, Meng,
Chen, Chen, & Khayat, 2015; Sienko, Bednarski, &
Howiacki, 2019; Zeng et al., 2002), as well as field applica-
tions, see e.g. (Barrias, Rodriguez, Casas, & Villalba, 2018;
Brault, Hoult, Greenough, & Trudeau, 2019; Gli
si
c,
Posenato, & Inaudi, 2007; Matta, Bastianini, Galati, Casadei,
& Nanni, 2008). In the case of reinforced concrete struc-
tures, several authors have reported successful results in
detecting the onset and the location of load-induced crack-
ing, both for fibers with protective coatings commonly
embedded in the concrete, see e.g. (Bao & Chen, 2015;
Henault et al., 2012; Imai et al., 2010) as well as for fibers
without any intermediate layer between the fiber and the
substrate, frequently attached to the surface of the hardened
concrete, see e.g. (Davis, Hoult, Bajaj, & Bentz, 2017; Regier
& Hoult, 2015; Rodriguez, Casas, & Villalba, 2015a).
Despite the advances in crack monitoring techniques
based on DOFS measurements, one of the remaining chal-
lenges is to achieve accurate predictions of crack widths in
reinforced concrete members. For optical fiber sensors with
a series of intermediate layers (outer claddings, coatings,
sheaths, jacketing, etc.) between the fiber core and the host
material, due to the lower stiffness of the intermediate
layers, the transfer of localized strains in the substrate
results in fiber strains smeared over a certain distance of up
to several centimetres. Feng, Zhou, Sun, Zhang, and Ansari
(2013), formulated a mechanical model based on shear lag
theory to evaluate the relationship between strain disconti-
nuities, such as cracks, and the measured strain distribution
in optical fiber sensors. This model has been later applied by
others to predict the surface crack width of concrete structures
by DOFS bonded to the surface of the concrete (Billon et al.,
2015) as well as embedded in the concrete (Bassil et al., 2019).
While promising results have been reported, the proposed
model requires the calibration of the shear lag factor, which
depends on the material and geometrical properties of the
cables, and it is not directly applicable to fiber sensors bonded
to the reinforcement, where the assumption of perfect bond
between all layers is not fulfilled.
Recent work carried out by the groups of Prof. Casas and
Prof. Hoult, have both shown great progress in the quantifi-
cation of crack widths in RC structures using Rayleigh-based
DOFS bonded without intermediate layers between optical
fiber sensor and the host material. In (Rodriguez et al.,
2015a), a methodology for the location and quantification of
crack widths is presented based on strain measurements
from a DOFS bonded to the tension surface of a concrete
element. Their method was evaluated experimentally using
crack measurements obtained by magnetic transducers from
a RC slab tested in bending up to failure and further vali-
dated with nonlinear finite element modelling. A similar
approach was used in (Rodriguez, Casas, & Villalba, 2019)
to quantify the width of shear cracks, where the authors
bonded DOFS on the surface of the web of three concrete
beams to create a two-dimensional grid of fiber sensors able
to measure strain profiles in two orthogonal directions. In
both studies the crack width estimations obtained by their
2 C. G. BERROCAL ET AL.
method based on the strain profiles measured by the DOFS
compared well with the crack width measurements from
transducers. However, one of the limitations of their
method is that it only provides an average crack width over
the cracked region.
In a recent piece of work by (Brault & Hoult, 2019b), the
authors investigated the feasibility of using of DOFS bonded
to the surface of concrete to evaluate the deflection and
cracking of reinforced concrete beams subjected to three-
point bending. Unlike in previous studies, their method
enabled the quantification of individual cracks, which
showed good agreement with experimental crack width
measurements obtained by Digital Image Correlation (DIC)
technique. In another study by (Brault & Hoult, 2019a), the
authors tested the ability of two different types of fiber sen-
sors, namely nylon-coated and polyamide-coated, to locate
cracks when the DOFS are bonded to the reinforcement and
investigated the relationship between surface crack widths,
as measured by DIC, and the reinforcement strain at the
cracks as measured by the DOFS. In both studies, the
authors showed very promising results and the great poten-
tial of DOFS for crack monitoring in RC structures.
The aim of the present study was to explore the feasibil-
ity of obtaining quantitative crack-damage information
through the monitoring of strain distributions by optical
fiber sensors directly bonded to the reinforcement. This art-
icle reports the findings from laboratory experiments aimed
at validating the suitability of Rayleigh scattering based
DOFS glued to steel reinforcement to detect crack formation
and determine crack location as reported by others (cf.
Brault & Hoult, 2019a). Additionally, a new method to esti-
mate crack widths in reinforced concrete beams subjected to
external loading is proposed. To that end, electrical strain
gauges and Digital Image Correlation (DIC) were used in
combination with optical fiber sensors to explore the
accuracy and reliability of DOFS under monotonic and cyc-
lic loading as well as to test the ability of the proposed
method to estimate crack widths.
2. Description of experiments
The main objective of this study was to explore whether
strain measurements from Rayleigh scattering DOFS, which
have been rendered very promising for the detection of
crack formation in reinforced concrete structures, could be
also used to obtain quantitative information about the
extent of cracking when bonded to the reinforcement. For
that purpose, an experimental program comprising a total
of eight reinforced concrete beams was devised. Two of the
beams, which were used as reference specimens to validate
the predicted failure mode and ultimate load, were not
instrumented with optical fiber sensors, hence their results
are not presented. The remaining six beams were measured
using electrical strain foil gauges, optical fiber sensors and
DIC. Two of the measured beams were loaded monotonic-
ally up to failure whereas four were subjected to cyclic load-
ing. In the following, the most relevant details about the
experimental programme are summarised.
2.1. Specimen geometry
In the present study, beam specimens with dimensions of
900 mm in length and a cross-section of 100150 mm were
designed to monitor the cracking process. Each beam was
reinforced with two ;10 mm rebar of normal ductility car-
bon-steel (B500B). The rebars were supported on spacers
placed at the corners of the forms to ensure a clear concrete
cover of 25 mm between rebars and the bottom and side
surfaces of the forms. The ends of the bars were bent
upwards to provide better anchorage as well as to create a
Figure 1. Working principle of Rayleigh Backscattering optical fiber sensors based on OFDR.
STRUCTURE AND INFRASTRUCTURE ENGINEERING 3
safe way out for the optical fiber sensors. Moreover, no stir-
rups were placed to ensure the beams underwent shear fail-
ure, based on the requirements of a parallel study where the
same beams were used. The geometry and reinforcement
layout of the beam specimens are depicted in Figure 2.
2.2. Materials and sample preparation
All beams were cast in playwood forms with a concrete mix
featuring a water-to-cement ratio (w/c) of 0.45 and prepared
using a Portland cement with low C
3
A content and moderate
heat development. In order to minimise the risk of accidentally
damaging the sensors during casting, a self-compacting mix
was designed to remove the need of compaction and vibration
of the concrete. The mix proportions are given in Table 1.
After casting, the beam specimens were covered with a poly-
ethylene sheet to prevent moisture evaporation and they were
stored to cure in the forms in an indoor climate (20± 2 C
and 60 ± 5% RH) for 40 days prior to testing.
Material tests were also carried out to assess the mechan-
ical properties of the concrete and the steel reinforcement.
The concrete compressive strength was assessed in accord-
ance with EN 12390-3,3 (2009) using three 100 mm cubes
while splitting tensile strength tests were conducted on three
150 mm cubes in accordance with EN 12390-6,6 (2001). The
mechanical properties of the steel reinforcement were deter-
mined through tensile tests according to EN ISO 15630-1
(2019). The results from the material tests are summarised
in Table 2.
2.3. Sensor installation and strain monitoring
Three different systems were used to measure strains in the
beam during the bending tests, namely electrical foil strain
gauges, distributed optical fiber sensors and DIC. The elec-
trical strain gauges were 3 mm foil gauges from HBM used
in a quarter bridge configuration. The foil gauges were
placed at the centre of one of the rebars in each of the
beams using a two-component epoxy resin adhesive (see
Figure 2). It must be noted that to provide a sufficiently
large flat surface suitable to accommodate the foil gauge in
the rebar, one of the transverse ribs was ground down. The
sampling rate for the foil gauges was set to obtain 1000
measurements per hour, hence 3.6 Hz.
Different configurations to deploy optical fiber sensors in
reinforced concrete have been described in the literature,
such as attaching the fiber to the concrete surface (Brault &
Hoult, 2019b; Grzymski, Trapko, & Musiał,2019), embed-
ding a jacketed fiber into the concrete (Henault et al., 2012)
or attaching the fiber to the reinforcement either by bond-
ing it to the surface (Barrias, Casas, & Villalba, 2017; Brault
& Hoult, 2019b) or inserting it into a previously etched
groove (Du et al., 2018; Kaklauskas, Sokolov, Ramanauskas,
& Jakubovskis, 2019). For crack detection purposes, bonding
the fiber directly onto the concrete surface is intuitively the
best choice, since large strain concentrations induced by the
crack formation can be easily detectable. Loss of information
after cracking, however, may occur if the strains in the fiber
exceed its measuring range when uncoated fibers are used
in combination with stiff adhesives, which highlights the
importance of choosing and correctly applying the right
adhesive so that stresses are properly transferred from the
substrate material to the fiber core, cf. (Barrias, Casas, &
Villalba, 2019a). Attaching the fiber to the reinforcement,
on the other hand, may not only be used to assess the
cracking state of the structures but it can also provide valu-
able information about the loading state. However, attaching
the optical fiber sensors to the reinforcement limits their
application to newly built structures.
In this study, the optical fiber sensors used (polyimide
coated low-bend loss fibers with a 155 mm diameter) were
attached directly onto the rebar along one of the longitu-
dinal ridges (see Figure 2). The rebars were sand-blasted
and degreased with acetone prior to the installation of the
fiber. The bonding of the fiber was achieved by applying a
thin layer of cyanoacrylate adhesive, which according to the
literature, exhibits a better performance than two-compo-
nent epoxy adhesives commonly used to attach foil strain
gauges (Barrias, Casas, & Villalba, 2018). Another important
documented issue related to the use of unprotected DOFS
embedded in concrete is the appearance of anomalous strain
readings after concrete cracking, see e.g. (Barrias, Rodriguez,
et al., 2018; Regier & Hoult, 2015). Consequently, an abun-
dant layer of a one-component water-proof silicone rubber
material was applied for protection based on the findings by
(Davis et al., 2017). It should be noted that foil gauges and
optical fibers were installed in different rebars. The Optical
Distributed Sensor Interrogator (ODiSI) 6000 series from
Luna Inc. was used as interrogation unit. The spatial reso-
lution in the fiber sensor was 0.625 mm between measuring
points while the sample rate was 5 Hz.
Furthermore, the measurement of full-field deformation
and strains of one of the lateral sides of the beams was per-
formed on all the elements tested. For that purpose, the
commercially available optical deformation measurement
system from GOM, ARAMISV
R, was employed. The system
relies on a non-contact measurement technique based on
Digital Image Correlation (DIC) with an adjustable stereo-
camera setup, consisting of two CCD cameras with 12.0
Megapixel resolution (4000 3000 pixels). The cameras
were mounted on a rigid bar specifically designed for the
purpose with an angle of 25.5while the separation of the
cameras and the distance to the beam were adjusted accord-
ing to the prescribed measuring volume. In this study the
system was calibrated for a measurement volume of
980 795 795 mm
3
, hence covering the full length of the
specimen. The calibration information indicated a deviation
of 0.035 pixel which, considering the camera resolution and
measured length, provided an accuracy of 8.5 mm.
The main purpose of the use of DIC was to obtain accur-
ate and reliable information about the formation and devel-
opment of cracks. Additionally, DIC measurements were
also used to compute the deflection of the beams, where the
settlement over the supports was subtracted from the verti-
cal displacement at the mid-span. The sample rate used for
the DIC was lower than for the embedded sensors at only
4 C. G. BERROCAL ET AL.
0.2 Hz. It is worthwhile mentioning that relative position of
the rebars with the foil gauge and the fiber sensor was var-
ied, so that the rebar with the optical fiber sensor was in
some cases closest to the surface of the beam monitored
with DIC and sometimes further away.
2.4. Loading setup
In order to induce cracking, the beam specimens were sim-
ply supported on rollers and loaded under a three-point
bending setup. The distance between the centre of the sup-
ports was equal to 800 mm and the point load was applied
at mid-span, thus leaving a shear span of 400 mm. The load-
ing setup is schematically illustrated in Figure 2. Loading
was applied under displacement control using a closed-loop
feedback system at a displacement rate of 0.5 mm/min. Two
different loading schemes were used in the tests, monotonic
loading up until failure and cyclic loading with gradually
increasing load levels.
3. Results and discussion
All of the tested beams presented very similar results in
terms of structural performance with consistent values of
the maximum load and deflection achieved prior to failure
as well as a similar number of cracks formed, see Table 3.
Moreover, the beams also performed according to the
expected structural behaviour where the first crack initiated
directly under or close to the loading point, yet all the
beams exhibited an identical failure mechanism where
an inclined crack would suddenly propagate along the
reinforcement causing an abrupt and complete loss of the
load bearing capacity (see Figure 3). The agreement of
the structural results among the different beam specimens is
indicative of a good level of uniformity of the concrete
properties and reinforcement position, which is very favour-
able to test the repeatability of the DOFS for the purpose of
crack monitoring.
Figure 2. (a) Beam geometry, reinforcement layout and loading setup (all measurements in mm); (b) foil strain gauge installed a rebar with one of the ribs previ-
ously removed; (c) polyamide coated DOFS bonded along the longitudinal ridge of a rebar with cyanoacrylate adhesive; (d) Installed DOFS after the application of
a protective layer of silicone.
Table 1. Concrete mix proportions, in kg/m
3
.
Component Dosage
Cement (CEM I 42.5N BV/SR/LA) 415
Limestone filler (Limus 40) 75
Fine aggregate (sand 0/4) 868,3
Coarse aggregate (crushed 5/8) 862
Effective water 186,8
Superplasticizer Glenium 51/18 6,64
Table 2. Material properties.
Concrete
Property Average COV (%)
Compressive strength at 28 days [MPa] 59 1
Compressive strength at testing (40 days) [MPa] 61 2
Splitting tensile strength at 28 days [MPa] 4 5.5
Steel reinforcement
Property Average COV (%)
Yield strength [MPa] 518 3.3
Ultimate strength [MPa] 628 4.5
Elastic Modulus [GPa] 200 6.2
STRUCTURE AND INFRASTRUCTURE ENGINEERING 5
In the following sections, the results of investigating dif-
ferent aspects related to the ability of DOFS to provide
meaningful quantitative information about the cracking pro-
cess of RC elements are presented and discussed. Firstly, the
accuracy of the strain measurements provided by the optical
fiber sensors are tested against the local measurements of
electrical foil gauges. Subsequently, the suitability of DOFS
to fulfil the different levels of a SHM system, namely detec-
tion, location and quantification of crack damage,
are addressed.
3.1. Correlation between DOFS and strain gauges
The evolution of the strain measurements obtained by the
foil strain gauge over the duration of the tests is presented
in Figure 4 together with the corresponding strain measured
by the optical fiber at the mid-span for all the beam speci-
mens tested.
As observed, an excellent agreement between the optical
fiber and the foil gauge was generally obtained regarding
both the shape and the value of the strain profiles. Beam 3
and beam 5 presented small discrepancies between both sen-
sors at the end and beginning of the test, respectively.
Several factors may have contributed to such discrepancies,
the most obvious being the fact that each sensor was
attached to a different reinforcement bar of the same beam,
causing that cracks propagating in an arbitrary direction
which is not completely straight nor perpendicular to the
reinforcement might have crossed the rebars at different
locations. Other possible reasons for the discrepancies found
for beams 3 and 5 could be the removal of a rib in one of
the bars to accommodate the foil gauge or even a difference
in the relative position of the foil gauge and the DOFS on
the rebar, as local bending of the bar has been shown to
yield a noticeable variation in the strain measurements
(Davis et al., 2017). As a result, an actual difference in
strains between the mid-span of each rebar may
be expected.
Moreover, during the execution of the loading test for
beam 6, the misplacement of the hinge between the jack
and the beam caused a loss of contact due to excessive rota-
tion that resulted in the complete unloading of the beam.
This is apparent in Figure 4 by a plateau in the strain curves
of beam 6 of approximately 5 min. It is noteworthy that
even after the complete unloading of the beam, both sensors
displayed a consistent behaviour revealing a significant
amount of inelastic strain at the reinforcement. Those
results confirm that bonding an optical fiber sensor directly
onto the reinforcement with cyanoacrylate adhesive and
protected by a layer of silicone can provide results as accur-
ate as those provided by conventional foil strain gauges,
without the need of altering the reinforcement by either
grinding down ribs or edging a groove.
3.2. Damage detection: formation of the first
flexural crack
The most basic level of a SHM system, sometimes referred
to as anomaly detection, is the detection of damage, which
is often based on the identification of states that deviate sig-
nificantly from what is considered a normal state. Cracking
of concrete cannot be strictly considered an anomaly since
cracking is an inherent process to the structural behaviour
of reinforced concrete structures. Nevertheless, excessively
large cracks are regarded as a potential source of durability
problems and cracks, in general, induce changes in the
Table 3. Summary of structural performance results of the three-point bend-
ing tests.
Specimen
Loading
type
Max.
load [kN]
Max.
deflection [mm]
Number of
cracks
Beam 1 Monotonic 39.11 2.27 6
Beam 2 Monotonic 40.90 2.34 6
Beam 3 Cyclic 40.52 2.33 7
Beam 4 Cyclic 37.45 1.98 6
Beam 5 Cyclic 38.91 2.11 7
Beam 6 Cyclic 41.04 2.17 7
Figure 3. Beam specimens after testing displaying the developed shear failure mechanism.
6 C. G. BERROCAL ET AL.
structure that clearly differentiate the uncracked and cracked
states.
In the cracks, the contribution of the concrete to the
transfer of tensile stresses is negligible. As a result, the
moment of inertia of the cracked sections is greatly reduced,
thus leading to increased local strain in the reinforcement
and greater curvature of the section, which in turn reduces
the bending stiffness of the structure. Whereas the detection
of cracks is straight forward when the fiber sensor is bonded
to the surface of the concrete due to appearance of large
strain discontinuities, the strain gradient in the reinforce-
ment is less apparent. One way to determine the onset of
cracking is by finding deviations between the measured
strain profiles and the theoretical ones given by classical
beam theory assuming elastic behaviour of the materials and
perfect bond between the concrete and the reinforcement,
see e.g. (Carrera, Giunta, & Petrolo, 2011). The correspond-
ing longitudinal strain at the reinforcement for each pos-
ition e(x), in the uncracked state, can be calculated
according to:
ex
ðÞ
¼MyðxÞ
EcI
y
z(1)
where M
y
(x) is the in-plane bending moment, zis the dis-
tance to the neutral axis of the transformed section and E
c
is the modulus of elasticity of the concrete, which is taken
Figure 4. Time evolution of the reinforcement strain during the three-point loading tests measured by the electrical foil gauges and DOFS for all the tested
beam specimens.
STRUCTURE AND INFRASTRUCTURE ENGINEERING 7
as 35 GPa, and was obtained from the formulation proposed
in the Eurocode 2 (EN 1992-1-1 Eurocode 2, 2, 2004) using
the value of the mean cube compressive strength included
in Table 2.I
y
is the moment of inertia of the transformed
concrete section defined as:
I
y¼IcþAcd2
cþa1
ðÞ
Asd2
s(2)
where I
c
is the moment of inertia of the brut concrete sec-
tion, A
c
is the concrete brut area, A
s
is the reinforcement
area, ais the modular ratio of steel to concrete E
s
/E
c
and d
c
and d
s
are the distance between the neutral axis of the trans-
formed section and the centroids of the concrete brut sec-
tion and reinforcement respectively.
In Figure 5, a comparison between the theoretical and
measured strain profiles along the length of the beam is pre-
sented for all the beams at two different load levels, namely
for a point load of P¼2 kN and for the cracking load, P
cr
.
Owing to the small strain levels attained at the reinforce-
ment before the formation of cracks and the signal noise of
the optical fiber sensors which ranged between ±10 me, the
noise to signal ratio of the DOFS measurements made it dif-
ficult to have a clear comparison. It should be noted that
the signal noise in this study was similar to that reported by
Barrias, Casas, and Villalba (2019b) but significantly higher
than that obtained by others, see e.g. (Henault et al., 2012)
who reported values as small as 2 me. This increased signal
noise might be attributed to the settings of the analyser,
namely the small gauge length and high sampling frequency,
and could have been potentially mitigated by choosing a
larger gauge length. Nevertheless, as suggested by Barrias
et al. (2019b), a filtering operation consisting on a moving
average over a length of 10 mm was performed to remove
part of the noise.
As observed in Figure 5, a fair agreement exists between
the theoretical and the measured strain profiles for the
lower load level, despite the latter being still somewhat noisy
after filtering. Conversely, the theoretical and measured
strain profiles are visibly different at the cracking load,
where the latter presents apparent strain localization near
the mid-span. It should be highlighted that at the load level
of the cracking load, i.e. about 5 kN, the measured crack
width was approximately 40 mm as measured by the DIC,
hence imperceptible to the naked eye. However, in this
study the detection of cracks below that value based on the
strain profiles obtained by the DOFS bonded to the
reinforcement was not obvious, whereas a DOFS bonded to
the concrete surface could have clearly detected cracks as
small as 10 mm, see e.g. (Brault & Hoult, 2019b).
Nevertheless, the method used in this study might perform
better in cases where the expected strain profile is constant,
such as uniform bending moment regions or member in
tension, since the concentration of strain would be more
easily discernible.
3.3. Determination of crack position
As previously mentioned, one of the main advantages of
DOFS based on OFDR of Rayleigh scattering is the
unprecedented spatial resolution achieved, reaching the sub-
millimetre scale. It is precisely that spatial resolution what
enables the obtention of truly distributed strain profiles
along the reinforcement which can be analysed to detect
variations that may be associated to the existence of cracks
in the concrete.
The procedure used to compare the crack position
obtained from the DIC measurements and the crack pos-
ition as determined from the strain profiles measured with
the DOFS is presented in Figure 6. As observed, the cracks
formed at the surface of the concrete are clearly distinguish-
able from the 2 D strain field plot provided by the DIC.
Note, however, that the image is noisy towards the edges of
the beam where some strain information is even lost at spe-
cific regions. This was attributed to the images being slightly
out of focus in those regions, thus leading to the inability of
the software to accurately track the displacement of
image subsets.
Consequently, the large strain values that can be ran-
domly spotted near the edges were ignored in the outermost
100 mm of each end. Moreover, since cracks were generally
somewhat inclined or discontinuous (see mid-span crack in
Figure 6(a), in order to automatically extract the crack pos-
ition from DIC results in a consistent way, the following
procedure was implemented: (i) three longitudinal profiles
of the surface strain at the level of the reinforcement and at
±;/2 were obtained; (ii) for each longitudinal profile, an
initial thresholding was applied to remove all the peaks with
a strain lower than 1% (i.e. to remove noise and cracks with
a width smaller than approximately 50 mm); (iii) each profile
was then transformed by fitting a gaussian distribution with
maximum height of 1 and standard deviation of 7 to every
peak remaining after the thresholding and adding them
together; (iv) the harmonic mean of the three transformed
profiles was performed to obtain the profile use to deter-
mine the position of the cracks, which is illustrated in
Figure 6(b) as a black line. It should be noted that only fully
formed active cracks were tracked, i.e. red regions in Figure
6(a), whereas incipient and closing cracks which displayed
low strain values in the DIC results were not considered in
the comparison.
In Figure 6(c), a series of curves show the evolution of
the strain profiles measured by the DOFS during the cyclic
loading test of beam 4. Following the same rationale intro-
duced in the previous section, cracks can be associated to
local peaks in the strain profile of the reinforcement. Away
from the crack, the strain at the reinforcement decreases as
the load is partially transferred to the concrete due to bond
action until compatibility of strains is reached or until the
concrete reaches its tensile strength, where a new crack
would form. However, for a new crack to form, a certain
length is required to build up sufficient stress in the con-
crete, noted as the minimum transmission length, l
t,min
.As
a result, new cracks can continue forming until the separ-
ation between them is no greater than the minimum crack
spacing s
r,min
¼2l
t,min
. The crack spacing can generally
range from just a few centimetres to several tens of centi-
metres depending on the bond properties between the
8 C. G. BERROCAL ET AL.
concrete and the reinforcement, the reinforcement ratio and
the rebar diameter.
Consequently, using the above requirement as a basis to
discriminate the peaks associated to cracks in the strain pro-
file, a smoothing spline was fit to the strain profile corre-
sponding to the maximum load, where a stabilized crack
pattern had already been reached, to get rid of all the peaks
associated to the spatial resolution noise. Subsequently, the
derivative of the fit curve was analysed to select only those
peaks where the sum of the absolute value of the maximum
gradient before and after the peak was equal or larger than
10 me/mm. This threshold was found suitable to discern
between fully formed cracks, as detected by the DIC, and
other potential types of crack or defects, such as secondary
internal cracks that might not be visible at the surface
(Goto, 1971). Finally, the position of the cracks determined
by the optical fiber are represented by the red triangular
markers in Figure 6(c), whereas the grey shaded areas cor-
respond to the crack positions determined from the DIC.
Figure 7 illustrates the accuracy of the DOFS to locate
the position of cracks taking the measurements of the DIC
as the correct crack position. Figure 7(a), shows, qualita-
tively, that a fair agreement is attained between the results
of the optical fiber and the DIC, whereas Figure 7(b) dis-
plays the error difference between both measurements, e¼
xDIC xDOFS, as a function of the crack position. As
Figure 5. Comparison of theoretical and measured strain profiles obtained with the DOFS along the beam length for a load level of 2 kN and at the cracking load,
Pcr. A moving average filter over a 10 mm length is applied to the measured strain profiles to decrease the noise to signal ratio.
STRUCTURE AND INFRASTRUCTURE ENGINEERING 9
observed, all the errors were within the range ±30 mm with
the exception of three cracks whose position was mispre-
dicted by almost 50 mm. It should be noted, however, that
these results compare measurements of the same cracks at
the surface and the reinforcement level. Since those cracks
might not propagate perfectly perpendicular to the longitu-
dinal axis of the beam, part of the observed error could be
potentially attributed to an actual difference between the
position of the cracks at different depths. Moreover, it is
interesting to note that the amount of positive and negative
errors is rather evenly distributed but a slight tendency is
apparent where the crack position was underestimated by
the DOFS at low xcoordinates and, conversely, it was over-
estimated at high coordinate xcoordinates.
A potential explanation for this behaviour could be
attributed to the existence of secondary inclined cracks
growing from the cracks closer to the supports, which may
induce a small shift in the position of the strain peak, as for
the crack near the coordinate x¼200 mm in Figure 6(c).
This is in line with the findings by (Brault & Hoult, 2019a)
Figure 6. Example of the procedure used to obtain and compare the crack location from DIC and DOFS for beam 4: (a) two dimensional strain field on the concrete
surface from DIC; (b) combined profiles of the concrete surface strain at the level of the reinforcement from the DIC, where the position of the cracks has been
highlighted by grey shaded areas; (c) multiple strain profiles from the DOFS where the red triangular makers indicate the determined crack position based on the
strain profile at maximum load and the grey shaded area correspond to position determined from the DIC.
10 C. G. BERROCAL ET AL.
who observed that the strain in the reinforcement crossing
cracks with predominant in-plane shear stresses displaying
loading mode-II deviated from strains predicted by
beam theory.
3.4. Crack width evaluation based on reinforcement
strain DOFS measurements
The last aspect investigated in the present study is the feasi-
bility of using DOFS strain measurements to provide quan-
titative information regarding the crack width of existing
cracks. As briefly discussed in Section 1, for coated optical
fiber sensors bonded directly onto the concrete, Feng et al.
(Feng et al., 2013) proposed a mechanical model based on
shear lag theory to relate the crack opening displacement to
the strain in the fibre core. However, one of the assump-
tions in that model is the perfect bond between all the inter-
mediate layers. Since the opening of cracks is the result of
the relative displacement between concrete and the
reinforcement, that model is, in principle, not applicable to
optical fiber sensors bonded to the reinforcement.
Referring to current structural design codes, e.g. Eurocode 2
(EN 1992-1-1 Eurocode 2, 2, 2004), the calculation of the crack
width in reinforced concrete structures is based on mechanical
models derived from the study of thin members subjected to
direct tension. Those models state that the characteristic crack
width, w
k
, can be calculated as:
wk¼sr,max ðesm ecmÞ(3)
where s
r,max
is the maximum crack spacing and e
sm
and e
cm
are the mean strain at the steel and concrete, respectively.
In such models, the parameters in Equation (3) are obtained
from the equilibrium of tensile forces between the crack sec-
tion and a section located at 0.5s
r,max
from the previous
one when the stress in the concrete reaches its tensile cap-
acity, f
ct
. The expressions used, however, often involve sev-
eral empirical parameters to characterize the bond
behaviour between steel and concrete, the type of reinforce-
ment, the duration of the load, etc., which are the factors
governing the variation of strain along the reinforcement.
In the present investigation, a similar approach to that
included in structural design codes, expressed by Equation (3),
andusedbyotherresearchersforDOFSbondedtothecon-
crete surface (cf. Brault & Hoult, 2019b), was adopted with
appropriate modifications. The proposed method assumes that
the variation of the strain along the reinforcement is known
and that the position of the cracks, and therefore the crack
spacing, can be obtained from the DOFS measurements fol-
lowing the procedure described in the previous section.
Furthermore, it is assumed that in the absence of interaction
between the reinforcement and the surrounding concrete, the
variation of stress and strain in the reinforcement should fol-
low the moment distribution, namely a linear variation for
three-point loading. Consequently, it can be inferred that the
non-linear variation of strain between cracks measured by the
DOFS must include the effect of the stress transfer between
the reinforcement and the concrete due to bond action.
Accordingly, the following expression is suggested to calculate
the crack width:
wcr,i¼ðlþ
t,i
l
t,i
eDOFSðxÞdx qa ðlþ
t,i
lt,i
^
ex
ðÞ
eDOFS x
ðÞ
dx
"#
(4)
where e
DOFS
(x) is the strain along the reinforcement
measured by the DOFS, ^
eðxÞis the assumed linear strain
variation between cracks neglecting the steel-concrete inter-
action, q¼As=Ac,ef and a¼Es=Ecare the reinforcement
ratio and the modular ratio, respectively and l
t,iand lþ
t,iare
the transmission length to the left and right sides of the i-th
crack, w
cr,i
, along the beam. The proposed method to evalu-
ate crack widths based on DOFS measurements is graphic-
ally illustrated in Figure 8, where the two integrals in
Equation (4) are represented by the shaded areas in the
zoomed region.
It should be noted that, although the position of the
cracks is known, the actual boundaries of the integral given
by Equation (4) are not clearly defined. In a member with
varying moment, the local minima in the strain profile only
indicate the points where the rate of change of the strain in
the reinforcement due to bond is larger than that due to the
moment variation. Nevertheless, the valleys in the strain
profile between cracks were taken as a first approximation
for the limits of the transmission length, which for the case
Figure 7. Comparison of the crack position determined by the DOFS and the DIC system (left) and quantitative error difference between both measurements for
every crack of all the beams tested (right).
STRUCTURE AND INFRASTRUCTURE ENGINEERING 11
of constant bending moment and pure tension state would
be an accurate choice.
Based on the described method, the crack widths of each
individual crack in every beam were calculated for all the
load steps. Figure 9 shows a comparison of the calculated
crack width and the crack width measurements from the
DIC for the widest crack of each beam. It can be observed
that, in general, the crack width estimation of the widest
crack based on the DOFS correlated well with the DIC
measurements, even for the cyclically loaded beams, where
the predicted crack widths followed the closing and re-open-
ing of cracks upon unloading and reloading.
However, it can be also seen that for beams 1 and 3, the
crack width estimated by the DOFS started overestimating
the DIC measurements for crack widths larger than 0.2 mm.
As discussed by (Brault & Hoult, 2019b), this could be
attributed to a reduction of the DOFS accuracy for large
strain where more anomalous readings may be present in
the strain profile. The crack width of Beam 6 agreed well
with the DIC measurements up to the point where the load-
ing was accidentally halted. Thereafter, the calculated crack
width consistently underestimated the DIC crack width.
Even though the DOFS and DIC measurements were not
interrupted, a slight misplacement of the loading plate might
have caused a subtle change in the load position and conse-
quently a change in the moment and strain distribution.
Furthermore, the calculated crack width for beam 2 dis-
played a weak correlation with DIC crack width measure-
ments. Considering that beam 2 exhibited the largest
difference between the predicted crack location and the
measured by DIC, as shown in Figure 7, one possible
explanation for the weak performance of the proposed
method could be the low accuracy in the determination of
the crack locations, which subsequently might have led to
erroneous crack width calculations.
In Figure 10, the distribution of crack width errors, e¼
wDOFS
cr wDIC
cr , for each individual crack in every beam is
presented. Figure 10 reveals several interesting aspects of the
proposed crack width measurement technique. The first is
that, except for beam 2, the crack width errors seem to be
symmetrically distributed around zero, i.e. some cracks are
overestimated while other are underestimated. Small under-
estimations could be explained by the fact that the crack
width calculated is at the reinforcement level whereas the
crack width measured by DIC is at the surface, which may
be potentially larger. However, for a concrete cover of only
25 mm as in this study, this effect was most likely very
small. Crack width overestimations, on the other hand,
could be attributed to the existence of incipient or internal
cracks forming in the vicinity of fully formed cracks, which
were not considered in this work. Another interesting aspect
is that, unlike what it was anticipated, the combination of
mode-I and mode-II loading for cracks formed in predom-
inantly shear loaded regions did not seem to affect nega-
tively the accuracy of the proposed method, i.e. cracks
closest to the supports did not show larger errors compared
to other cracks.
It is also interesting to note that the largest crack width
errors were concentrated in half of the beams, namely beam
2, 5 and 6, whereas beams 1, 3 and 4 showed systematically
better correlations with the DIC measurements. However,
the crack width errors did not seem to be correlated with
the crack location errors. A simple explanation for this
observation is that, in fact, the most relevant points are the
valleys, which define the integration limits. As long as the
position of the cracks are determined to be between the two
adjacent valleys, the actual position of the crack will not
affect the first integral in Equation (4), which is the main
contribution to the crack width. In the light of these results,
the larger crack width error in some of the beams can be
indicative that the accuracy of the DOFS measurements was
lower for some of the tests, suggesting the appearance of
anomalous readings.
It should be highlighted that, even though the accuracy
of the proposed method could be potentially improved by
increasing the quality of the DOFS measurements (adjusting
the analyser settings to use longer gauge length or using
other types of fiber sensors, such as nylon-coated fibers), in
most cases, the calculated crack widths differed in less than
20 mm with the crack width measurements from the DIC,
which for crack widths between 0.1 and 0.2 mm represent
an error of less than 20%. This error in crack width may be
Figure 8. Graphical representation of the method used to estimate the crack width. The integrals in Equation (4) correspond to the shaded areas in the zoom
region, where the light blue area represents the contribution of the reinforcement and the dark blue area represents the contribution of the concrete
between cracks.
12 C. G. BERROCAL ET AL.
considered as acceptable in the absence of more accurate
measurements. Particularly, it offers obvious advantages
with respect to other techniques as it provides an estimation
of the crack width for each individual crack detected by the
optical fiber sensor. Moreover, unlike for displacement
transducers, the location of cracks is not needed in
advanced, which means that the proposed method based on
DOFS can be used to measure the entire crack width his-
tory. Nevertheless, remaining questions to be explored are
related to the ability of DOFS to maintain the same level of
accuracy under long-term loading or in structures subjected
to expansive deterioration mechanisms that might affect the
strain at the reinforcement such as corrosion, sulphate
attack or alkali-silica reaction.
4. Conclusions
This paper explored the viability of using a relatively recent
technology, namely distributed optical fiber sensors (DOFS)
based on Optical Frequency Domain Reflectometry (OFDR)
of Rayleigh backscattering, for crack monitoring purposes in
concrete structures. Through a series of laboratory experi-
ments consisting of three-point bending tests of reinforced
Figure 9. Comparison between the crack width measured by the DIC and calculated from the DOFS measurements based on the Equation (4), for the widest crack
in each beam specimen.
STRUCTURE AND INFRASTRUCTURE ENGINEERING 13
concrete beams, the authors investigated the suitability of using
DOFS to provide meaningful information about the formation,
location and quantification of crack width. Moreover, in the
present work the DOFS were deployed directly onto the
reinforcement using a protective layer of silicone, without any
previous mechanical modification of the reinforcement to
accommodate the optical fibers. Electrical foil gauges and
Digital Image Correlation were used to validate the DOFS
measurements on monotonic and cyclic loading tests.
Overall, the DOFS system provided strain measurements
that agreed well with those measured by conventional elec-
trical foil gauges throughout the entire test up to the failure
of the beams, both for monotonic and cyclic tests. Even
though the noise to signal ratio of the DOFS measurements
was high within the elastic range of the bending tests, the
strains measured experimentally correlated well with those cal-
culated based on traditional beam theory. Moreover, excellent
repeatability was obtained among the six beams tested.
Unlike traditional strain gauges and other quasi-distrib-
uted optical fiber sensing systems which offer discrete meas-
urements with a gauge pitch that can be down to a few
centimetres, the system used in the present study delivered
an unprecedented spatial resolution of only 0.65 mm.
Nevertheless, the use of such a small gauge length led to
Figure 10. Error difference between the calculated crack width based DOFS measurements and the crack width measured by the DIC for each crack in each of the
tested beam specimens. Central markers indicate median values, the boxes indicate the 25th and 75th percentiles and the whiskers indicate the extreme data val-
ues not considered as outliers. Outliers are indicated by peripherical individual markers.
14 C. G. BERROCAL ET AL.
significant spatial variability of the measurements, especially
at low strains. The application of an averaging filter over a
1 cm region to the DOFS strain measurements was benefi-
cial to visualise and compare the results.
The fiber sensor deployment methodology tested enabled
the successful detection of crack formation at an early stage,
i.e. for cracks as small as 40 mm. Even though at that level
cracks are still not perceptible to the human eye, a fiber sen-
sor bonded to the concrete surface or embedded in the con-
crete would provided even earlier detection capabilities.
However, bonding the fiber to the reinforcement can reduce
the risk of fiber rupture, as well as idle readings caused by
exceeding the strain range of the sensor, since no strain dis-
continuities occur in the reinforcement.
It was found that based on the strain distribution at the
reinforcement provided by the DOFS, it is feasible to locate
the position of active fully-formed cracks which can be asso-
ciated to local peaks of strain. However, the determination
of the crack position is less apparent than for sensors
bonded to the surface or embedded in the concrete and it
required a certain post-processing of the strain data to
remove the noise associated to the spatial variability.
Nevertheless, the comparison of the crack location obtained
from the DIC and from the DOFS measurements revealed
that most cracks were located within an acceptable range of
±3 cm. Considering that cracks are measured at different
depths, i.e. at the surface and at the reinforcement level, and
that the crack planes might not be perfectly perpendicular
to the direction of the reinforcement, that difference might
not be entirely attributable to an error of the crack location
based on the DOFS measurements.
Lastly, the capabilities of the DOFS to provide quantita-
tive information about the crack width were demonstrated.
By integrating the strain profile over a certain length adja-
cent to the crack and removing the contribution of the ten-
sion stiffening between cracks, good agreement was
generally achieved with the DIC results, being the error for
most estimated crack widths within ±20 mm. While the pro-
posed method to estimate crack widths was not found to be
very sensitive to the exactness of the crack positions, a
higher quality of DOFS strain measurements could poten-
tially increase the accuracy of the method, thereby reducing
the crack width errors for cracks beyond 0.2 mm.
Disclosure statement
No potential conflict of interest was reported by the author(s).
Funding
This work was supported by the Swedish Transport Administration
(Trafikverket) under the grant TRV/BBT 2017-028.
ORCID
Carlos G. Berrocal http://orcid.org/0000-0003-4654-5498
Ignasi Fernandez http://orcid.org/0000-0003-4847-2894
Rasmus Rempling http://orcid.org/0000-0002-1122-7855
References
Bado, M., Casas, J., & Barrias, A. (2018). Performance of Rayleigh-
based distributed optical fiber sensors bonded to reinforcing bars in
bending. Sensors,18(9), 3125. doi:10.3390/s18093125
Bao, Y., & Chen, G. (2015). Strain distribution and crack detection in
thin unbonded concrete pavement overlays with fully distributed
fiber optic sensors. Optical Engineering,55(1), 11008. doi:10.1117/1.
OE.55.1.011008
Bao, Y., Meng, W., Chen, Y., Chen, G., & Khayat, K. H. (2015).
Measuring mortar shrinkage and cracking by pulse pre-pump
Brillouin optical time domain analysis with a single optical fiber.
Materials Letters,145, 344346. doi:10.1016/j.matlet.2015.01.140
Barrias, A., Casas, J. R., & Villalba, S. (2016). A review of distributed
optical fiber sensors for civil engineering applications. Sensors,
16(5), 748. doi:10.3390/s16050748
Barrias, A., Casas, J. R., & Villalba, S. (2017). Application study of
embedded Rayleigh based distributed optical fiber sensors in con-
crete beams. Procedia Engineering,199(2017), 20142019. doi:10.
1016/j.proeng.2017.09.466
Barrias, A., Casas, J. R., & Villalba, S. (2019a). Distributed optical fibre
sensors in concrete structures: Performance of bonding adhesives
and influence of spatial resolution. Structural Control and Health
Monitoring,26(3), e2310. doi:10.1002/stc.2310
Barrias, A., Casas, J. R., & Villalba, S. (2019b). SHM of reinforced con-
crete elements by Rayleigh backscattering DOFS. Frontiers in Built
Environment,5(March), 114. doi:10.3389/fbuil.2019.00030
Barrias, A., Casas, J., & Villalba, S. (2018). Embedded distributed
optical fiber sensors in reinforced concrete structuresA case study.
Sensors,18(4), 980. doi:10.3390/s18040980
Barrias, A., Rodriguez, G., Casas, J. R., & Villalba, S. (2018).
Application of distributed optical fiber sensors for the health moni-
toring of two real structures in Barcelona. Structure and
Infrastructure Engineering,14(7), 967985. doi:10.1080/15732479.
2018.1438479
Bassil, A., Wang, X., Chapeleau, X., Niederleithinger, E., Abraham, O.,
& Leduc, D. (2019). Distributed fiber optics sensing and coda wave
interferometry techniques for damage monitoring in concrete struc-
tures. Sensors,19(2), 356. doi:10.3390/s19020356
Billon, A., H
enault, J.-M., Quiertant, M., Taillade, F., Khadour, A.,
Martin, R.-P., & Benzarti, K. (2015). Qualification of a distributed
optical fiber sensor bonded to the surface of a concrete structure: a
methodology to obtain quantitative strain measurements. Smart
Materials and Structures,24(11), 115001. doi:10.1088/0964-1726/24/
11/115001
Brault, A., & Hoult, N. (2019a). Distributed reinforcement strains:
Measurement and application. ACI Structural Journal,116(4),
115127. doi:10.14359/51714483
Brault, A., & Hoult, N. (2019b). Monitoring reinforced concrete ser-
viceability performance using fiber-optic sensors. ACI Structural
Journal,116(1), 5770. doi:10.14359/51710870
Brault, A., Hoult, N. A., Greenough, T., & Trudeau, I. (2019).
Monitoring of beams in an RC building during a load test using
distributed sensors. Journal of Performance of Constructed Facilities,
33(1), 4018096. doi:10.1061/(ASCE)CF.1943-5509.0001250
Carrera, E., Giunta, G., & Petrolo, M. (2011). The Euler-Bernoulli and
Timoshenko theories. In Beam structures (pp. 922). NJ, USA:
Wiley. doi:10.1002/9781119978565.ch2
Casas, J. R., & Cruz, P. J. S. (2003). Fiber optic sensors for bridge
monitoring. Journal of Bridge Engineering,8(6), 362373. doi:10.
1061/(ASCE)1084-0702(2003)8:6(362)
Cawley, P. (2018). Structural health monitoring: Closing the gap
between research and industrial deployment. Structural Health
Monitoring,17(5), 12251244. doi:10.1177/1475921717750047
Davis, M. A., Bellemore, D. G., & Kersey, A. D. (1997). Distributed
fiber Bragg grating strain sensing in reinforced concrete structural
components. Cement and Concrete Composites,19(1), 4557. doi:10.
1016/S0958-9465(96)00042-X
STRUCTURE AND INFRASTRUCTURE ENGINEERING 15
Davis, M. B., Hoult, N. A., Bajaj, S., & Bentz, E. C. (2017). Distributed
sensing for shrinkage and tension stiffening measurement. ACI
Structural Journal,114(3), 753764. doi:10.14359/51689463
Ding, Z., Wang, C., Liu, K., Jiang, J., Yang, D., Pan, G., Liu, T.
(2018). Distributed optical fiber sensors based on optical frequency
domain reflectometry: A review. Sensors,18(4), 1072. doi:10.3390/
s18041072
Du, Y., Chen, Y., Zhuang, Y., Zhu, C., Gerald, R., & Huang, J. (2018).
A uniform strain transfer scheme for accurate distributed optical
fiber strain measurements in civil structures. Inventions,3(2), 30.
doi:10.3390/inventions3020030
EN 12390-3. (2009). Testing hardened concrete. Part 3: Compressive
strength of test specimens.
EN 12390-6. (2001). Testing hardened concrete. Tensile splitting strength
of test specimens.
EN 1992-1-1 Eurocode 2. (2004). EN 1992-1-1 Eurocode 2: Design of
concrete structuresPart 1-1: General rules and rules for buildings.
Brussels, Belgium: CEN.
EN ISO 15630-1. (2019). Steel for reinforced and prestresed concrete
Testing methods. Part 1: Bars and wires for reinforced concrete.
Feng, X., Zhou, J., Sun, C., Zhang, X., & Ansari, F. (2013). Theoretical
and experimental investigations into crack detection with BOTDR-
distributed fiber optic sensors. Journal of Engineering Mechanics,
139(12), 17971807. doi:10.1061/(ASCE)EM.1943-7889.0000622
Gli
si
c, B., Hubbell, D., Sigurdardottir, D. H., & Yao, Y. (2013).
Damage detection and characterization using long-gauge and dis-
tributed fiber optic sensors. Optical Engineering,52(8), 87101. doi:
10.1117/1.OE.52.8.087101
Gli
si
c, B., Posenato, D., & Inaudi, D. (2007). Integrity monitoring of
old steel bridge using fiber optic distributed sensors based on
Brillouin scattering. In H. F. Wu, A. A. Diaz, & P. J. Shull (Eds.),
Nondestructive characterization for composite materials, aerospace
engineering, civil infrastructure, and homeland security 2007 (Vol.
6531, p. 65310P). San Diego, CA: SPIE. doi:10.1117/12.716055
Goto, Y. (1971). Cracks formed in concrete around deformed tension
bars. ACI Journal Proceedings,68(4), 244251. doi:10.14359/11325
Grzymski, F., Trapko, T., & Musiał, M. (2019). Practical use of distrib-
uted fibre optic sensors in research on FRCM composites. In E3S
Web of Conferences (Vol. 97, p. 2019). France: EDP Sciences. doi:10.
1051/e3sconf/20199702019
G
uemes, A., Fern
andez-L
opez, A., & Soller, B. (2010). Optical fiber dis-
tributed sensingPhysical principles and applications. Structural
Health Monitoring: An International Journal,9(3), 233245. doi:10.
1177/1475921710365263
Gyger, F., Rochat, E., Chin, S., Nikl
es, M., & Th
evenaz, L. (2014).
Extending the sensing range of Brillouin optical time-domain ana-
lysis up to 325 km combining four optical repeaters. In J. M. L
opez-
Higuera, J. D. C. Jones, M. L
opez-Amo, & J. L. Santos (Eds.), 23rd
International Conference on Optical Fibre Sensors (p. 91576Q).
Santander, Spain: SPIE. doi:10.1117/12.2059590
Henault, J.-M., Quiertant, M., Delepine-Lesoille, S., Salin, J., Moreau,
G., Taillade, F., & Benzarti, K. (2012). Quantitative strain measure-
ment and crack detection in RC structures using a truly distributed
fiber optic sensing system. Construction and Building Materials,37,
916923. doi:10.1016/j.conbuildmat.2012.05.029
Imai, M., Nakano, R., Kono, T., Ichinomiya, T., Miura, S., & Mure, M.
(2010). Crack detection application for fiber reinforced concrete
using BOCDA-based optical fiber strain sensor. Journal of Structural
Engineering,136(8), 10011008. doi:10.1061/(ASCE)ST.1943-541X.
0000195
Kaklauskas, G., Sokolov, A., Ramanauskas, R., & Jakubovskis, R.
(2019). Reinforcement strains in reinforced concrete tensile mem-
bers recorded by strain gauges and FBG sensors: Experimental and
numerical analysis. Sensors,19(1), 200. doi:10.3390/s19010200
Majumder, M., Gangopadhyay, T. K., Chakraborty, A. K., Dasgupta,
K., & Bhattacharya, D. K. (2008). Fibre Bragg gratings in structural
health monitoring-Present status and applications. Sensors and
Actuators A: Physical,147(1), 150164. doi:10.1016/j.sna.2008.04.008
Matta, F., Bastianini, F., Galati, N., Casadei, P., & Nanni, A. (2008).
Distributed strain measurement in steel bridge with fiber optic sen-
sors: Validation through diagnostic load test. Journal of Performance
of Constructed Facilities,22(4), 264273. doi:10.1061/(ASCE)0887-
3828(2008)22:4(264)
Morey, W. W., Dunphy, J., & Meltz, G. (1992). Multiplexing fiber
Bragg grating sensors. In A. D. Kersey & J. P. Dakin (Eds.),
Distributed and multiplexed fiber optic sensors (Vol. 1586, pp.
216224). Boston, MA: SPIE. doi:10.1117/12.56520
Palmieri, L. (2013). Distributed optical fiber sensing based on Rayleigh
scattering. The Open Optics Journal,7(1), 104127. doi:10.2174/
1874328501307010104
Regier, R., & Hoult, N. A. (2015). Concrete deterioration detection
using distributed sensors. Proceedings of the Institution of Civil
EngineersStructures and Buildings,168(2), 118126. doi:10.1680/
stbu.13.00070
Rodriguez, G., Casas, J. R., & Villalba, S. (2015a). Cracking assessment
in concrete structures by distributed optical fiber. Smart Materials
and Structures,24(3), 35005. doi:10.1088/0964-1726/24/3/035005
Rodriguez, G., Casas, J. R., & Villalba, S. (2015b). SHM by DOFS in
civil engineering: a review. Structural Monitoring and Maintenance,
2(4), 357382. doi:10.12989/smm.2015.2.4.357
Rodriguez, G., Casas, J. R., & Villalba, S. (2019). Shear crack width
assessment in concrete structures by 2D distributed optical fiber.
Engineering Structures,195(May), 508523. doi:10.1016/j.engstruct.
2019.05.079
Seo, J., Hu, J. W., & Lee, J. (2016). Summary review of structural
health monitoring applications for highway bridges. Journal of
Performance of Constructed Facilities,30(4), 4015072. doi:10.1061/
(ASCE)CF.1943-5509.0000824
Sienko, R., Bednarski, L., & Howiacki, T. (2019). About distributed
internal and surface strain measurements within prestressed con-
crete truck scale platforms. In IOP Conference Series: Materials
Science and Engineering (Vol. 471, pp. 52074). Bristol, UK: IOP
publishing. doi:10.1088/1757-899X/471/5/052074
Soga, K., & Luo, L. (2018). Distributed fiber optics sensors for civil
engineering infrastructure sensing. Journal of Structural Integrity
and Maintenance,3(1), 121. doi:10.1080/24705314.2018.1426138
Todd, M. D., Johnson, G. A., & Vohra, S. T. (2001). Deployment of a
fiber Bragg grating-based measurement system in a structural health
monitoring application. Smart Materials and Structures,10(3),
534539. doi:10.1088/0964-1726/10/3/316
Zeng, X., Bao, X., Chhoa, C. Y., Bremner, T. W., Brown, A. W.,
DeMerchant, M. D., Georgiades, A. V. (2002). Strain measure-
ment in a concrete beam by use of the Brillouin-scattering-based
distributed fiber sensor with single-mode fibers embedded in glass
fiber reinforced polymer rods and bonded to steel reinforcing bars.
Applied Optics,41(24), 5105. doi:10.1364/AO.41.005105
16 C. G. BERROCAL ET AL.
... Moreover, distributed optical fiber sensors provide indirect information about structural damage. Anomalous strain concentrations in the vicinity of cracks indicate the cracking of structural elements [34][35][36][37][38][39][40]. This allows specialists to locate cracks and estimate their width using strain measurements and indirect numerical methods developed for this The optical-fiber sensor systems that are most commonly used for SHM employ Optical fiber Bragg gratings (FBGs), optical frequency domain reflectometry (OFDR), Brillouin scattering optical time-domain analysis (BOTDA), and optical fiber Fabry-Perot resonators (FPRs). ...
... Moreover, distributed optical fiber sensors provide indirect information about structural damage. Anomalous strain concentrations in the vicinity of cracks indicate the cracking of structural elements [34][35][36][37][38][39][40]. This allows specialists to locate cracks and estimate their width using strain measurements and indirect numerical methods developed for this purpose. ...
... This allows specialists to locate cracks and estimate their width using strain measurements and indirect numerical methods developed for this purpose. Such estimations have been carried out on reinforced concrete (RC) beams [34][35][36][37][38]. However, the maximum width of the cracks evaluated using this technique did not exceed 0.4 mm. ...
Article
Full-text available
This paper presents a study on the suitability and accuracy of detecting structural cracks in brick masonry by exploiting the breakage of ordinary silica optical fibers bonded to its surface with an epoxy adhesive. The deformations and cracking of the masonry specimen, and the behavior of pilot optical signals transmitted through the fibers upon loading of the test specimen were observed. For the first time, reliable detection of structural cracks with a given minimum value was achieved, despite the random nature of the ultimate strength of the optical fibers. This was achieved using arrays of several optical fibers placed on the structural element. The detection of such cracks allows the degree of structural danger of buildings affected by earthquake or other destructive phenomena to be determined. The implementation of this technique is simple and cost effective. For this reason, it may have a broad application in permanent damage-detection systems in buildings in seismic zones. It may also find application in automatic systems for the detection of structural damage to the load-bearing elements of land vehicles, aircraft, and ships.
... Instead of integrated Bragg gratings, natural variations of the refractive index, caused by geometric differences and imperfections in the fiber core, are used to determine the strain based on the Rayleigh backscattering. Distributed optical fiber sensors (DOFS) have already been successfully applied in reinforced concrete for crack detection and strain measurements in concrete, [15] and [16]. ...
... Berrocal et al. [15] carried out six 3-point bending tests on beams with dimensions 900x100x150 mm, where two reinforcing bars were equipped with strain gauges and DOFS. Furthermore, digital image correlation (DIC) was used as an additional measurement method. ...
Conference Paper
Full-text available
In the past decades, there has been an increasing effort to describe the local bond behavior with a bond stress-slip model. Such bond models are often used in studies of long bond lengths, numerical calculations and crack width determination. Local bond behavior is usually investigated experimentally in pull-out tests with short bond lengths equal to or less than 5 times the bar diameter ds, recording the pull-out force and bar slip. Therefore, existing bond models, like the approach of fib Model Code 2010, are mainly based on data determined outside the bond zone. Distributed optical fiber sensing opens up the possibility to get an insight of the processes within the bond zone during testing. Current investigations at Technische Universität Dresden focus on the bond behavior and the bond stress distribution for ribbed bars with bond lengths from 1·ds up to 60·ds under various loading conditions. The test program includes systematic investigations of the influence of the bond length on the ultimate bond stress testing pull-out, beam-end and tensile tie specimens as well as the targeted use of distributed optical fiber sensing. Therefore, the hair thin sensors are applied directly on the pull-out bar and enable quasi-continuously strain reading along the bond length. Based on the local resolution of the strain, conclusions about the local force transmission between reinforcement and concrete at different load levels or after a certain load duration can be drawn. Hence, already with a bond length of 2·ds, non-linear distributions of the bond stresses and local peaks in front of the ribs could be observed. This article deals with the experimental and instrumental setup, test results and the procedure of evaluation. Furthermore, it will be discussed how distributed optical fiber sensing can help to derive a local bond stress-slip model.
... Away from the crack, as the load is partly transferred to the concrete due to bond action, the strain at the reinforcement lowers until the stresses are compatible or the concrete reaches its tensile strength, at which point a new crack will form. Furthermore, the examination of high spatial resolution strain profiles of the DOFS allowed for the early detection of crack initiation [121]. All civil engineering infrastructures are vulnerable to the ravages of time and decay, as well as external factors that threaten their structural integrity, create significant economic losses, pollute the environment, and endanger the safety of their users. ...
... This could be due to strain redistribution in the element following cracking around the discontinuity that forms the crack, which is influenced by the stiffness of the bonding materials utilized. The results showed that the strains measured throughout the experiments were accurate and stable Fig. 7. DOFS in crack detection in rebar and surface, and crack width monitoring a) Multiple strain profiles from the DOFS, with the red triangular shape indicating the determined crack location based on the strain profile at maximum load, and the grey shaded region corresponding to the DIC position [121]. b) Detection and identification of cracks at the commencement of load cycles [122] c) Information about the position and width of cracks is included in the development of crack functions [86] d) Installation of optical fibre sensors [86] e) Sensors mounted to beams and adhesives used [122]. ...
Article
Structural health monitoring (SHM) systems in civil engineering structures have been a growing focus of research and practice. Over the last few decades, optical fibre sensor (OFS) technology has advanced rapidly, and various types of OFS technologies have found practical uses in civil engineering. Due to recent advances in optical sensors and data-driven solutions, the SHM systems are gaining prominence. Because of its superior ability to detect damage and flaws in civil engineering structures, deep learning (DL) gradually gained substantial attention among researchers in recent years. The main goal of this paper is to review the most recent publications in SHM related to bridges, buildings, and pipelines using emerging OFS and DL-based applications, and to provide readers with an overall knowledge and understanding of various SHM applications. Finally, current research trends and future research needs have been identified.
... At present, many crack detection methods based on sensors or detectors have been studied. For example, optical fiber sensors [10][11][12][13], flexible strain sensors [14], piezoelectric ceramic sensors [15,16], and acoustic emission sensors [17,18] have all been used to detect the development of cracks. However, these sensors require data acquisition equipment, which is complicated and expensive. ...
Article
Full-text available
A new type of concrete surface gel coating using thermosensitive fluorescent (TSF) microcapsules was proposed to monitor micro-cracks of cement-based materials. The gel materials can adhere other materials, and the incorporation of microcapsules into the gel coating can be cured on various structural surfaces. Zinc sulfide and phenyl acetate were encapsulated into a polymethyl methacrylate shell to prepare the TSF microcapsules by a solvent evaporation method. When micro-cracks are generated on the surface of the gel coating, the ruptured TSF microcapsules burst out, fill the damaged area, and then emit fluorescence after being excited at ambient temperature. It was found that the brightness of the fluorescence increased with increasing temperature from 80–110 °C. When the concentration of TSF microcapsules was 15% of the mass of the gel coating, the cement-based damage-sensing material had sufficient damage-indicating effects, and the fluorescence brightness of the crack location remained even after a long time. It is expected that this study will provide an effective and intuitive method for crack location detection of cement-based materials.
... (3) Structural health monitoring: Finally, our work is related to a large body of research on structural health monitoring [22,23,66]. SHM systems can broadly be divided into three categories: surface-mounted sensors, intrusive wired sensors, and wirelessconnected embedded sensors. ...
... It may provide insight into structural capacity through finite element analysis (FE). It may also act as a decision support tool [44,45]. ...
Article
Full-text available
In recent years, there has been a growing seismic demand for existing bridges and the final redesign of bridges, especially after a major earthquake One method to strengthen concrete frames on bridges is to use steel sheets or profiles to use the confining force. During this study, a sample at 30% scale under gravity and lateral cycle loading was examined within the laboratory. A finite element model is additionally used to compare the behavior of laboratory samples. The laboratory sample was a model of a typical bridge in iran that was generally designed with deficient detailing requirements in agreement with the typical regulations of the 1970s. A finite element analysis set was used to evaluate various parameters in improving the behavior of the laboratory sample. The finite element model correctly predicted the weakness of the model. Subsequently, a reinforced specimen was investigated by increasing the prestressing force within the concrete beam and the thickness of the frp sheets utilized in the bridge pier by the finite element method. The results show the energy absorbed within the hysteresis curves improved the propagation of the failure. The result also showed that a 100% increase in the prestressing load caused a 67% increase in resistance .
Chapter
Distributed fibre optic sensing (DFOS) technology is being widely exploited in many civil infrastructure monitoring applications due to its inherent advantages over conventional sensing technologies. Over the last decade, research and development have enhanced the performance of DFOS systems in SHM, which led to new applications in civil engineering. This paper presents a brief review of DFOS technology, the latest laboratory experiments utilising DFOS, and the applications of this system in monitoring piles, bridges, tunnels, buildings, and railways. Finally, promising future directions for DFOS technology are discussed. This review demonstrates that DFOS is a versatile monitoring technology with significant capabilities to support the development of smart infrastructure.
Article
This paper is a review on distributed optic fiber sensing for structural health monitoring applications, with a deeper focus on technologies relying on the Rayleigh backscattering phenomenon. It addresses the basic physical principles which are involved, the implementation and instrumentation of the measurement techniques, as well as recent practical applications, current performance, and remaining challenges. Being written at an elementary level and integrating relevant theoretical and technical details, we hope the document can be useful for researchers and engineers looking for an up-to-date overview on a field which currently undergoes significant development and increasing attractiveness, in particular for damage tracking in complex mechanical structures.
Article
Full-text available
Fiber-reinforced polymer composites are an excellent choice for bridge decks due to high strength, lightweight, resistance to corrosion, and long-term durability with a 100-year design life. Structural health monitoring is useful for the long-term assessment of the condition of the bridge structure and obtaining a response to complex loads considering environmental conditions. Bridge structures have been studied primarily using distributed fiber optic sensing, such as Brillouin scattering; however, critical events, including damage detection, can be missed due to low spatial resolution. There is also a critical need to conduct a comprehensive study of static and dynamic loading simultaneously for fiber-reinforced composite bridge structures. In this study, a novel approach was implemented using two sensor technologies, optical frequency domain reflectometry and fiber Bragg grating-based sensors, embedded in a glass-fiber-reinforced composite bridge deck to simultaneously monitor the deformation response of the bridge structure. The optical frequency domain reflectometry sensor utilizing Rayleigh scattering provides high spatial strain resolution were positioned strategically based on expected stress distributions to measure strain in the longitudinal, transverse, and diagonal directions along the span of the composite bridge. Furthermore, fiber Bragg grating based sensors are used to monitor the response to dynamic vehicular loading and deformations from an automotive-crash-type event on the bridge structure. To monitor environmental variables such as temperature, a custom wireless configured sensor package was developed for the study and integrated with a composite bridge located in Morgan County, Tennessee. Additionally, a triaxial accelerometer was used to monitor the vehicular dynamic loading of the composite bridge deck in parallel with fiber Bragg grating sensors. When appropriate, mid-point displacements were compared with strain-distribution measurements from the fiber optic sensor-based data.
Article
Full-text available
This article describes research on FRCM (Fabric Reinforced Cementitious Matrix) composites, which unlike commonly used FRP (Fibre Reinforced Polymers) composites make use of a mineral matrix instead of epoxy resin, which allows to achieve much higher resistance to elevated temperatures. In the described studies, experimental measurement of deformations with the use of the DFOS (Distributed Fibre Optic Sensors) method was applied. This method allows for geometrically continuous measurement of deformations, which is its significant advantage compared to traditional electric resistance wire strain gauge, as it reduces the possibility of measuring deformations in a place where they are not representative. The tests were carried out using two reinforced concrete slab elements loaded to failure in the 4-point bending scheme. Fibre optic sensors were installed on an unstrengthened control element and on an element strengthened with FRCM composite. During the tests, deformations of the concrete under tension and the external surface of the FRCM reinforcing composite were determined. Measurements were carried out simultaneously in two manners: using the DFOS method, and strain gauges placed at the characteristic points of the element. The test results based on both methods were compared and analysed. The comparative analysis confirmed the usefulness and effectiveness of the DFOS method while measuring deformations in strengthening composites, and showed its significant advantages such as precise indication of the place of elements cracking as well as the possibility of conducting a global analysis of deformations.
Article
Full-text available
This document showcases the latest research conducted within UPC–BarcelonaTech on the performance of distributed optical fiber sensors (DOFS), more specifically the case of the optical backscattered reflectometry (OBR) system, in the structural health monitoring (SHM) of bridges, and large scale structures. This technology has demonstrated promising results for monitoring applications in a wide range of fields but due to its novelty, still presents several uncertainties which prevent its use in a more systematic and efficient way in civil engineering infrastructures, being this even more evident in the case of concrete structures. Therefore, different laboratory experimental campaigns were devised where multiple aspects of the instrumentation of DOFS technology in civil engineering applications were assessed and scrutinized. Such as the study of new implementation methods, comparison, and performance analysis of different bonding adhesives and spatial resolution. Additionally, the fatigue performance of this sensing typology was also assessed. Furthermore, the use of the OBR system technology was applied in a real world structure in Barcelona, Spain, where new challenging conditions had to be addressed. Consequently, with this work, different conclusions are obtained related to the proficiency and limitations on the use of this particular type of optical sensing system in concrete structures.
Article
Full-text available
Distributed optical fibre measurement technology provides new possibilities in structural technical condition assessment in comparison with traditional spot measurements. It is not only possible to analyse strains continuously over structural member length with spatial resolution starting from as fine as 5 mm, but also the cracks and displacements state within reinforced concrete structures. The article presents pilot studies regarding prestressed concrete track scale platforms with a length of 8 to 16 m. Optical fibres were glued within composite rods attached to the stirrups inside the cross section of the slab on the height of lower and upper prestressing tendons, as well as they were glued to the concrete surface after bonding. That allows for comprehensive analysis of concrete strains state including all local nonlinearities (cracks) starting from hydration process (thermal-shrinkage strains), through prestressing tendons activation (strains regarding the transfer of compression forces from the tendond to the concrete) and finally during laboratory tests, when slabs were mechanically loaded until destruction. The ways of installation and exemplary results from selected phases of the research were presented as well as data interpretation was described and discussed.
Article
Full-text available
The assessment of Coda Wave Interferometry (CWI) and Distributed Fiber Optics Sensing (DFOS) techniques for the detection of damages in a laboratory size reinforced concrete beam is presented in this paper. The sensitivity of these two novel techniques to micro cracks is discussed and compared to standard traditional sensors. Moreover, the capacity of a DFOS technique to localize cracks and quantify crack openings is also assessed. The results show that the implementation of CWI and DFOS techniques allow the detection of early subtle changes in reinforced concrete structures until crack formation. With their ability to quantify the crack opening, following early detection and localization, DFOS techniques can achieve more effective monitoring of reinforced concrete structures. Contrary to discrete sensors, CWI and DFOS techniques cover larger areas and thus provide more efficient infrastructures asset management and maintenance operations throughout the lifetime of the structure.
Article
Full-text available
Experimental and numerical studies have been carried out on reinforced concrete (RC) short tensile specimens. Double pull-out tests employed rectangular RC elements of a length determined not to yield any additional primary cracks. Tests were carried out with tensor strain gauges installed within a specially modified reinforcement bar and, alternatively, with fibre Bragg grating based optical sensors. The aim of this paper is to analyse the different experimental setups regarding obtaining more accurate and reliable reinforcement strain distribution data. Furthermore, reinforcement strain profiles obtained numerically using the stress transfer approach and the Model Code 2010 provided bond-slip model were compared against the experimental results. Accurate knowledge of the relation between the concrete and the embedded reinforcement is necessary and lacking to this day for less scattered and reliable prediction of cracking behaviour of RC elements. The presented experimental strain values enable future research on bond interaction. In addition, few double pull-out test results are published when compared to ordinary bond tests of single pull-out tests with embedded reinforcement. The authors summarize the comparison with observations on experimental setups and discuss the findings.
Article
This paper outlines the second part of an experimental study to show the capabilities of distributed optical fiber sensors (DOFS) in their application to the structural health monitoring (SHM) of the shear performance of concrete structures. SHM seeks to obtain the shear crack characteristics of concrete elements: detection, localization and quantification of shear damage. The two first were discussed in the first part of the experimental study. In the present paper, the quantification is dealt with by proposing a method to obtain the mean shear crack width in concrete beams. The method is based on the experimental data obtained by a DOFS bonded to the concrete surface. First, the basis of the methodology are presented and, later on, experimentally checked by testing of three partially pre-stressed concrete (PPC) beams subjected to a shear test with increasing level of load. The DOFS were deployed in the web of the beams to conform a 2D grid mesh to measure the strain profile along two orthogonal directions. The experimental data was obtained using an OBR (Optical Backscattered Reflectometer) system with high spatial resolution and sensitivity that allow a complete mapping of the cracking pattern and to obtain the required data for the calculation of the crack width. The results show the feasibility of the proposed method in calculating the shear crack width when compared to the results from traditional instrumentation.
Article
The understanding of the complex behavior of reinforced concrete elements in situ would be aided if distributed strain measurements were captured during load tests, potentially leading to more accurate assessments and load ratings, and optimized future designs. This research investigates the use of distributed fiber optic sensors (FOS) to monitor three beams in a newly constructed RC building during a load test. It is the first case where an FOS technique capable of monitoring distributed strains, distributed deflections, and crack widths simultaneously is implemented in the field. The FOS data captured inflection points, moment transfer at the supports, crack locations, and crack openings. The FOS also captured deflected shapes, enabling maximum deflections to be determined without prior knowledge of where they would occur. Lastly, the measured results from the load test were compared to design model predictions for each element.
Article
Monitoring the serviceability performance of reinforced concrete (RC) structures in the field provides critical data for informing RC modeling, analysis, and design. Currently, there is a lack of monitoring technology that can capture the complex behavior of RC elements in place, in detail, while also being feasible to implement. A technique of using fiber-optic sensors to simultaneously monitor both distributed RC beam deflections and crack widths is developed and described. Thirteen beam specimens with varying properties were tested in three-point bending to evaluate the technique against other sensor technologies. The results showed that distributed beam deflections can be accurately measured up until load levels approaching failure, while crack widths can be measured up until a width of 0.3 mm (0.0118 in.). The technique is practical to implement and provides robust data sets not achievable with other sensing technologies, making it an effective option for field-monitoring applications.
Article
In this paper, the authors conducted an experiment where a reinforced concrete beam was instrumented with a 5‐m long polyimide single distributed optical fibre sensors (DOFS) performing four equal segments externally bonded to the bottom surface of the element, using for each segment a different type of adhesive. Three strain gauges were also used for comparison purposes. This beam was then loaded, producing expected equal levels of strain in each of the fibre segments for a more direct comparison of the different adhesives performance. The effect of alternating the spatial resolution is also analysed. In this exercise, additionally to the comparison with the other instrumented sensors, it is also important the consideration and analysis of the associated spectral shift quality (SSQ) values of the DOFS measurements.