Optical Fiber Technology 24 (2015) 7076

Contents lists available at ScienceDirect

Optical Fiber Technology

www.elsevier.com/locate/yofte

Regular Articles

Crack monitoring capability of plastic optical bers for concrete

structures
Jinlei Zhao a,b, Tengfei Bao a,b,, Rui Chen c
a

State Key Laboratory of Hydrology-Water Resources and Hydraulic Engineering, Hohai University, Nanjing 210098, China
College of Water-conservancy and Hydropower, Hohai University, Nanjing 210098, China
c
Department of Civil Engineering and Engineering Mechanics, University of Arizona, Tucson, AZ 85721, United States
b

a r t i c l e

i n f o

Article history:
Received 15 January 2015
Revised 16 April 2015
Available online 28 May 2015
Keywords:
Plastic optical bers
Crack monitoring
Structural health monitoring

a b s t r a c t
Optical bers have been widely used in structural health monitoring. Traditional silica bers are easy to
break in eld applications due to their brittleness. Thus, silica bers are proposed to be replaced by plastic
optical bers (POFs) in crack monitoring in this study. Moreover, considering the uncertainty of crack
propagation direction in composite materials, the inuence of the angles between bers and cracks on
the monitoring capability of plastic optical bers is studied. A POF sensing device was designed and
the relationship between light intensity loss and crack width under different ber/crack angles was rst
measured through the device. Then, three-point bend tests were conducted on concrete beams. POFs
were glued to the bottom surfaces of the beams and light intensity loss with crack width was measured.
Experimental results showed that light intensity loss in plastic optical bers increased with crack width
increase. Therefore, application of plastic optical bers in crack monitoring is feasible. Moreover, the
results also showed that the sensitivity of the POF crack sensor decreased with the increase of angles
between bers and cracks.
2015 Elsevier Inc. All rights reserved.

1. Introduction
Cracks are signs of aging and damage of structures, which
shorten the integrity and service life of structures. For example,
cracks increase the permeability of concrete structures, permitting
the inltration of erosive materials such as chloride and sulfate
into the structures and thereby accelerating the aging process.
Therefore, crack monitoring is important to guarantee the safety
of structures. Currently, a number of techniques have been proposed for crack monitoring, such as ultrasonic method [1], acoustic
emission [2], impact-echo [3], infrared thermograph [4], ground
penetrating radar [5], etc. However, all have restrictions when
applied in real projects. All the above-mentioned techniques can
only detect the pre-existing cracks in a structure and cannot provide information about the development of the cracks.
In recent years, ber optic sensors have been widely used in the
health monitoring of large structures because of their abilities to
perform distributed and continuous monitoring of cracks, light
weight and small size, resistance to electromagnetic interference,
anti-lightning, corrosion resistance, etc. For example, Rossi et al.
Corresponding author at: State Key Laboratory of Hydrology-Water Resources
and Hydraulic Engineering, Hohai University, Nanjing 210098, China.
E-mail address: baotf@hhu.edu.cn (T. Bao).
http://dx.doi.org/10.1016/j.yofte.2015.05.008
1068-5200/ 2015 Elsevier Inc. All rights reserved.

[6] embedded optical bers in concrete to detect cracks based on

the idea that signal can be interrupted by ber breakage due to
advance of the cracks. Ansari et al. [7] developed a ring ber optic
sensor to monitor cracks based on the nding that macro-bend
optical loss in bers increases due to the increase of crack width.
Li et al. [8] developed a ber optic sensor based on Michelson
white light interferometer to monitor the crack tip width displacement. Shi et al. [9] measured the distribution of strain on the surface of concrete structures by Brillouin optical ber both on
transverse and vertical direction to monitor damage elds. Bao
et al. [10] proved the ber Bragg grating (FBG) is feasible in crack
monitoring and FBG can serve as a quasi-distributed crack sensor.
Leung et al. [11] worked on a distributed crack sensor which can
detect and monitor a large number of cracks with a small number
of bers and a priori knowledge of crack locations is not required.
The researches above applied optical bers in crack monitoring and
results showed that optical bers have potential in crack monitoring. However, all the bers in the above researches are silica bers
which are brittle (their elongation rates only about 1%) [12], so
they are easy to break in real applications and only small cracks
can be monitored. Therefore, the brittleness of silica bers limits
the application of above researches.
To overcome this disadvantage, researchers attempted to use
plastic optical bers (POFs) in place of silica bers [1315]. POFs

71

J. Zhao et al. / Optical Fiber Technology 24 (2015) 7076

have a high elongation rate (more than 40%) because they are
made of high performance polymer material [16,17]. POFs have
many excellent properties, such as uneasy to break, low cost, easy
to be cut and connected, simple to be applied, etc. POFs were rst
widely used in local-area-network and automotive electronics. In
recent years, POFs have been used in engineering elds. For example, Babchenko et al. [18] developed a POF sensor with
multi-structural imperfections on it to measure the displacement
of structures. Takeda et al. [19] embedded POFs in FRP laminates
to detect transverse cracks. Husdi et al. [16] and Nakamura et al.
[20] studied the sensing characteristics of POFs by optical
time-domain reectometry technique. Krebber et al. [21] and
Liehr et al. [22] combined POFs with textiles and developed a type
of geotextile with distributed strain measurement capability,
which can be used for health monitoring of earthwork structures.
Liehr et al. [22] also studied the sensing characteristics of POFs
when crack increased. Kuang et al. [23] developed a POF sensor
by abrading the surface of POFs and the sensor was attached to
concrete specimens and the concrete specimens were conducted
in three- and four-point bend tests. The possibility of POF sensors
for detecting crack initiation and post-crack vertical deection was
evaluated. Kuang et al. [24] also developed a surface-attached optical ber sensor using POFs and applied the sensor in structure
health monitoring.
In previous works, emphasis is put on FRP laminates (Takeda
et al. [19]) and earthwork structures (Krebber et al. [21], Liehr
et al. [22]) using POFs for structural health monitoring. To the best
of our knowledge, only a few works about application of POFs in
structural health monitoring of concrete structures have been
reported, and researchers mainly used POFs with structural imperfections for structural health monitoring, for example, Kuang et al.
[23] and Babchenko et al. [18] created the structural imperfections
by abrading the surface of POFs as sensing area. The imperfections
can increase the amount of light loss when bent in the sensors, and
thus increase the sensibility of the sensors. However, the damage
detection will be limited by the size of the imperfections.
In this paper, POFs without structural imperfections are used in
crack monitoring for concrete structures. A simple and low-cost
POF sensing device is designed and the sensing principle of the
device is introduced. Then, the capability of bare POFs in crack
monitoring will be investigated by two experiments. In the rst
experiment, the capability of POFs in crack monitoring is studied
using a crack simulator that can provide well-controlled bending
of the bers. In the second experiment, the POFs are glued to the
bottom of a series of concrete beams and three-point bend tests
are conducted on the beams to assess the capability of POFs in
crack monitoring. Moreover, POFs of different ber diameters are
used to study the crack monitoring capability for concrete structures under different ber inclinations.
2. Structure of POFs and sensing principle

methacrylate or polystyrene. Compared to the silica bers, the

diameter of the ber core is bigger (up to 1 mm or even bigger)
but the thickness of cladding is thinner, therefore the coupling efciency of optical energy in POFs is higher.
The POFs used in this study are Mitsubishi Rayon ESKA SK10
and SK20. Details of the POFs used are given in Table 1.
2.2. Sensing principle
As shown in Fig. 2, two pieces of plexiglass plates (one xed and
the other can be moved freely) are placed together, the gap
between which can be viewed as a crack. A POF is glued at the surfaces of the two plexiglass plates, crossing the crack. When the
crack becomes bigger, the ber intersecting the crack has to be
bended to stay continuous, thus two micro bends will be formed
on both sides of the crack. Bending loss in the ber will occur at
the curved portion, which can be monitored by an optical power
meter on the left. Thereby, based on the change of bending loss,
the increase of the crack can be continuously monitored.
The micro bend loss is mainly caused by spatial ltering, mode
leakage, and mode coupling. The mode coupling plays a dominant
role among them and there are four kinds of mode coupling: coupling from low order modes to high order modes, coupling from
high order modes to low order modes, coupling from dissipative
modes to transmission modes, and coupling from transmission
modes to dissipative modes. Among the four, the coupling between
low order modes and high order modes has no effect on the intensity change of optical in ber. There is little coupling from dissipative modes to transmission modes. The coupling from transmission
modes to dissipative modes is the main reason which leads to
micro bend loss.
Assume the deformation function of micro bend as [25]:

f z Dt sin qz

where D(t) is the amplitude of micro bend; q is the spatial frequency

of micro bend; z is the distance from deformation points to input
terminal.
The period of the deformation function of micro bend is:

K 2p=q

From the wave theory, the rst order approximation of coefcient a of micro bend loss can be expressed as [25]:

1
4



sinq  DbL=22

q  DbL=2 

a KD2 tL

where K is the proportional coefcient; L is the length of micro

bend; Db is the constant difference of wave propagation in bers.
From Eq. (3), a is proportional to D2(t), indicating that the bigger
the micro bend, the more bend loss is.

2.1. Structure of POFs

The POFs mainly consist of three layers, the protective layer,
cladding and ber core, as shown in Fig. 1. They are respectively
made of polyurethane, uorinated Polymer polymethyl

Protective
layer

Cladding

Fiber
core

Fig. 1. Schematic diagram of plastic optical ber structure.

Table 1
Specication of SK10 and SK20.
Item

Unit

SK10

SK20

Core material
Cladding material

Core refractive index

Refractive index prole
Numerical aperture
Approximate weight
Max operating
temperature
Max transmission loss

g/m
C

PMMA resin
Fluorinated
polymer
1.49
Step-index
0.5
0.1
70 Max

PMMA resin
Fluorinated
polymer
1.49
Step-index
0.5
0.2
70 Max

dB/
m

0.3 Max (at

650 nm)

0.18 Max (at

650 nm)

72

J. Zhao et al. / Optical Fiber Technology 24 (2015) 7076

Table 3
Specication of power meter.

Bare POF
A spool of POF

(a)

Power meter

Light source

JW3233

Power range (dBm)

Wavelength range (nm)
Output interface
Calibration wavelength (nm)
Working temperature (C)
Power
Dimension (mm)

60+6
6001000
SI, FC
635,650,780,850,980
10+60
3pcs AAA battery
150  74  26

used in experiments, which have diameters of 0.25 mm and

0.5 mm, respectively.
The light generated from the light source enters a spool of ber
that is connected with the light source and the bare ber. The light
travels successively through the bare ber and another spool of
ber, nally reaches the light power meter, being detected.

Model

3.2. Tests with crack simulator specimens

(b)

(c)

Fig. 2. Schematic diagram of POF sensing device.

3. Experimental approach
In applications, the micro bend caused by cracks is hard to be
measured, but the micro bend is related to crack width.
Therefore, the relationship between crack width and light
intensity loss has been studied to verify the sensing principle of
POFs. In addition, the crack propagation direction in composite
materials is uncertain, so the angles between cracks and bers
pre-arranged are also uncertain. To study the monitoring capability
of POFs under different crack/ber angles, multiple sets of different
crack/ber angles are set in the following experiments.

3.1. POF crack sensing device

The schematic diagram of the POF crack sensing device used in
this paper is illustrated in Fig. 2, which consists of a light source, a
light power meter, and POFs. Specication of the light source and
the power meter is shown in Tables 2 and 3, respectively. The light
source is the input device and the light power meter is used for
monitoring light output power. As shown in Fig. 2, there are a certain length of bare POF (without protective layer) and two spools of
POF (with protective layer). The length of the bare POF is 0.6 m in
the tests with crack simulator specimens and 4 m in the tests with
concrete beams, respectively. The length of the two spools of POF
(with protective layer) is 5 m in total in the tests. The bare ber
(without protective layer) is the key part of the device, which is
the sensing region of the device. ESKA SK10 and SK20 POFs were

The capability of POFs in crack monitoring is rst studied with a

crack simulator that can provide well-controlled bending of the
bers. The crack simulator used in the tests consists of a pair of
plexiglass plates. One of the plexiglass plates is xed on the surface
of a table and the other can be moved on the surface of the table
along the horizontal direction. A number of blades were inserted
into the gap between the plexiglass plates, and then the development of cracking can be simulated by the relative displacement
of the plexiglass plates. As shown in Fig. 2, the bare ber is glued
to the plexiglass plates, with an angle h, of 30, 45, and 60 respectively to the crack width direction.
Both ESKA SK10 and SK20 POFs under the angles of 30, 45, and
60 were tested. Therefore, six sets of tests were conducted. For
each set, the tests were repeated three times, which are denoted
as Test 1, Test 2 and Test 3, respectively. In the tests, the light
source BML-206-10 was used for both bers. The deformation of
the POF SK20 under the angle of 30 in the test is shown in
Fig. 3. It can be seen that the POF presents red light with the
advance of the simulated crack as a result of the loss of light
intensity.
3.3. Tests in concrete beams with POFs
Based on the tests with the above crack simulator, the capability of POFs in crack monitoring for concrete structures was studied
through three-point bend tests of concrete beams. The concrete
beams had a dimension of 150 mm  150 mm  550 mm and the
mix proportion by weight was designed to be 0.44 : 1 : 1.42 :
3.17, corresponding to water : cement: sand: aggregate. In preparation of the specimens, 6 mm diameter steel bars were placed
on the tensile sides to avoid brittle failure of the beams. After being

Table 2
Specication of light source.
Model

BML-206-10

BML-206-20

Output power (mW)

Wavelength (nm)
Output interface

>10
650
2.5 mm universal
connector
Both CW and Pulse
available
0+60

>20
650
2.5 mm universal
connector
Both CW and Pulse
available
0+60

2pcs AA battery
175  26  26

2pcs AA battery
175  26  26

Working mode
Working temperature
(C)
Power
Dimension (mm)

(a)

(b)

Fig. 3. Deformation diagram of POF caused by cracking.

73

J. Zhao et al. / Optical Fiber Technology 24 (2015) 7076

cured under laboratory conditions for 28 days, a notch was cut on

the tensile side of each concrete beam to simulate a crack. Before
testing, a pair of POFs (both ESKA SK10 and SK20 POFs) was glued

to the tensile sides of concrete beams, with an angle of 30, 45,

and 60 to the notches, as shown in Fig. 4.
In the tests with concrete beams, the sensing device is the same
as that shown in Fig. 2. The light source BML-206-20 was selected
for SK10 POF and the light source BML-206-10 was used for SK20
POF to improve the readability of the power meter. Then
three-point tests on the concrete beams were conducted using a
universal testing machine. A loading rate of 1.2 mm min1 was
used for all tests and the light loss was measured simultaneously
with crack width.

(a)

4. Results and discussion

4.1. Test results with crack simulator specimens

According to the test results, the relationship between light

power and crack width for the two kinds of POFs with different
angles to the crack are shown in Figs. 5 and 6.
From Figs. 5 and 6, the test results are quite different in the
same test set. The reason is that for each new test the bare POF
was replaced by a new one because the bare POFs in the previous
tests were stretched and might be debonded from the plexiglass
plates, while the two spools of POF (with protective layer) were
unaffected and reused. Due to the different cleanness and rmness
at the ber interfaces, the light intensity loss at the points where
the bare POF and two spools of POFs were connected was different
in each test, so the light intensity is different for different tests in
the same test set.
It can be seen that the light power decreases gradually both in
SK10 and SK20 POF and the light power of SK20 POF was stronger
than that in SK10 POF due to larger diameter of the SK20 POF.
Moreover, the change rate of light power versus crack width of
SK20 was more obvious than the SK10. For example, when the

(b)
SIDE VIEW
Notch

(c)

BOTTOM

SK10
SK20

Fig. 4. Test set-up for concrete beams with POFs.

(b) 45

(a) 30
45

60

Test 1
Test 2
Test 3
Fit line 1
Fit line 2
Fit line 3

50
45

Test 1
Test 2
Test 3
Fit line 1
Fit line 2
Fit line 3

40

Light power ( w )

Light power ( w )

55

40
35
30

35
30
25
20
15

25
10

Crack width (mm)

Crack width (mm)

(c) 60
39
Test 1
Test 2
Test 3
Fit line 1
Fit line 2
Fit line 3

Light power ( w )

36
33
30
27
24
21
18
0

Crack width (mm)

Fig. 5. Relationships between the light power and crack width in SK10 (The ber/crack angle is 30, 45, and 60, respectively).

74

J. Zhao et al. / Optical Fiber Technology 24 (2015) 7076

(b) 45

(a) 30
160
Test 1
Test 2
Test 3
Fit line 1
Fit line 2
Fit line 3

120
100

Test 1
Test 2
Test 3
Fit line 1
Fit line 2
Fit line 3

100
90

Light power ( w )

140

Light power ( w )

110

80
60

80
70
60
50
40
30

40

20
0

Crack width (mm)

Crack width (mm)

(c) 60

Light power ( w )

80

Test 1
Test 2
Test 3
Fit line 1
Fit line 2
Fit line 3

70
60
50
40
0

Crack width (mm)

Fig. 6. Relationship between the light power and crack width in SK20 (The ber/crack angle of is 30, 45, and 60, respectively).

(a) SK10

Light intensity loss ( w )

25
20
15
10

30
30
45
45
60
60

5
0
0

1st
3rd
1st
3rd
1st
3rd

30
30
45
45
60
60

2nd
Avg
2nd
Avg
2nd
Avg

Crack width (mm)

(b) SK20
30
30
45
45
60
60

100

Light intensity loss ( w )

crack width increases from 0 to 6 mm (the angle between the crack

and the POF is 45), the light power in the SK20 decreases from
103.17 lW to 40.47 lW. On the contrary, for the SK10, the light
power decreases from 39.42 lW to 17.73 lW.
To further analyze the test results, the light power is subtracted
by the light power at the zero crack width, so light loss can be
obtained. The relationship between the light loss and crack width
for the two kinds of POFs is shown in Fig. 7. Here, the light loss
is dened as the change of light intensity.
It can be seen that for both POFs the light intensity loss
increased with the increase of crack width, smaller intersection
angle inducing larger light intensity loss. Specically, when the
crack width reached 6 mm, the maximum light intensity loss of
SK10 was 24.28 lW (with the angle of 30 to the crack) and the
minimum light intensity loss was 15.59 lW (with the angle of
60 to the crack). For SK20, the maximum light intensity loss was
78.76 lW (with the angle of 30 to the crack) and the minimum
light intensity loss was 33.52 lW (with the angle of 60 to the
crack). Same as the discussion above, the POFs with a larger diameter had a bigger bending loss due to the larger cross section.
From the relationship between the average light intensity loss
and crack width both for SK10 and SK20 POFs, when a POF intersects a crack at an angle of 30, the curve slope is bigger.
However, the curve slope is smaller when the angle is 60. In other
words, the average light intensity loss shows a smaller increase
trend when the angle between the POF and the crack increased.
This is because the bending degree is smaller when the angle is
bigger. Therefore, the bending loss would be less in this case. In
addition, the sensitivity of POF sensors declines with the increase
of the angle between the POF and the crack.

80

1st
3rd
1st
3rd
1st
3rd

30
30
45
45
60
60

2nd
avg
2nd
avg
2nd
avg

60
40
20
0
0

Crack width (mm)

Fig. 7. Relationship between the light intensity loss and crack width in tests with
crack simulator specimens.

75

J. Zhao et al. / Optical Fiber Technology 24 (2015) 7076

(b) SK20

(a) SK10
35

25
20

Light intensity loss ( w )

Test data 30
Test data 45
Test data 60
Fit line 30
Fit line 45
Fit line 60

30

Light intensity loss ( w )

35

15
10
5

Test data 30
Test data 45
Test data 60
Fit line 30
Fit line 45
Fit line 60

30
25
20
15
10
5
0

0
0

Crack width (mm)

Crack width (mm)

Fig. 8. Relationship between the light intensity loss and crack width in tests with concrete beams.

4.2. Test results with concrete beams

The relationships between the light intensity loss and crack
width for the two kinds of POFs for different angle in the 3-point
bending tests on concrete beams are shown in Fig. 8. Similar
increase trend with the development of crack in the 3-point bend
tests can be observed in the tests. Moreover, the light intensity loss
shows an obvious increase trend in the initial stage of cracking,
which indicates the POFs have a good initial perception of cracking.
However, the results shown in Fig. 8 are more noisy than those in
Fig. 7 due to a number of oscillations in the curves, which may be
caused by the inhomogeneous of concrete and the unsmoothed
surface of specimens.
Compared to the test results using crack simulator specimens,
the monitoring range of POFs was larger. In the tests using crack
simulator specimens, the value of light loss tended to be stable
when the crack width reached 6 mm. However, the value of light
loss still showed an increase trend when the crack width was close
to 7.8 mm in the 3-point bend tests of concrete beams. The reason
is that the length of POF had been stretched at the cracks. In the
tests with crack simulator specimens, the two pieces of plexiglass
plates were rst pushed together and then the whole bare POF was
glued to the plates, while in the tests with concrete beams about
10 mm bare POF was not glued to the concrete beams due to the
notch presence. Therefore, the POF can be stretched longer in the
3-point bend tests of concrete beams due to the width of notch.
In addition, in the tests with crack simulator specimens, at large
crack width, the part of the bare ber glued to the plexiglass plates
slid due to the smoothness of the plexiglass plates. Therefore, it can
be seen from Fig. 7 that the loss of light intensity increases slowly
at large crack width. However, in the tests with concrete beams,
the bare POFs were glued rmly to the surfaces of the concrete
beams and no sliding occurred. In this case, there is a certain
amount of light intensity loss due to the big axial stress in addition
to the bend loss.
According to Eq. (3), the bend loss is proportional to the square
of the bend amplitude. From Fig. 2, when the inclination angle h is
small, the part of ber between the two crack faces will rotate signicantly. In other words, the bend amplitude will be big if the
inclination angle h is small, so the bend loss induced will be big
too. Both the test results in the tests with crack simulator specimens and with concrete beams, shown in Figs. 7 and 8, agree with
the theoretical analysis.
In the previous work of Leung et al. [11], the feasibility of traditional silica bers in crack monitoring has been studied. They
attempted to embed silica bers inside a sensing sheet. The sensing
sheet (with a ber in it) was glued to the bottom of a concrete
beam to study the feasibility of traditional silica bers in crack
monitoring. In the tests, the capability of silica bers in crack

monitoring under different ber inclinations was also studied.

Results showed that silica bers are sensitive in crack monitoring
and silica bers are more sensitive when the inclinations between
bers and cracks are bigger. However, only small cracks (crack
width below 2 mm) can be monitored by the silica bers in the
tests. In this paper, the crack width can be monitored with crack
width up to 7.8 mm by POFs. Moreover, the POFs can be glued to
the surfaces of concrete specimens easily by curing epoxy adhesive
without additional effort to embedding the ber into concrete.
Therefore, the advantage of POFs in crack monitoring is obvious.
5. Conclusions
The capability of POFs in crack monitoring at different angles
between POFs and cracks was studied through a simple sensing
device in this paper. The angles between POFs and cracks was set
as 30, 45, and 60 and tests were conducted both on crack simulator specimens and concrete beams. Both ESKA SK10 and SK20
POFs were tested, and comparison was made. The following conclusions can be drawn from the study results:
The light intensity loss in POFs shows an obvious change at the
beginning of cracking development, which means POFs have a
good initial perception of cracking. Moreover, the monitoring range
of POFs can reach 7.8 mm in the 3-point bend tests using concrete
beams, which far exceeds traditional silica bers in crack
monitoring.
Both the test results using crack simulator specimens and concrete beams have proved that the sensitivity of POFs in crack monitoring decrease with the increase of the angle between POFs and
cracks.
The POFs have low cost and distributed monitoring characteristics. Therefore, the POFs show potential for monitoring cracks in
concrete structures.
Acknowledgments
This study was supported by the National Natural Science
Foundation of China (Grant Nos. 51379068, 51139001,
51409227), the Jiangsu Natural Science Foundation (Grant No.
BK20140039), the Program for New Century Excellent Talents in
University (Grant No. NCET-11-0628), the Research Fund for the
Doctoral Program of Higher Education of China (Grant No.
20120094110005).
Reference
[1] A.A. Shah, Y. Ribakov, S. Hirose, Nondestructive evaluation of damaged
concrete using nonlinear ultrasonics, Mater. Des. 30 (2009) 775782.

76

J. Zhao et al. / Optical Fiber Technology 24 (2015) 7076

[2] S. Degala, P. Rizzo, K. Ramanathan, K.A. Harries, Acoustic emission

monitoring of CFRP reinforced concrete slabs, Constr. Build. Mater. 23
(2009) 20162026.
[3] K. Mori, A. Spagnoli, Y. Murakami, G. Kondo, I. Torigoe, A new non-contacting
non-destructive testing method for defect detection in concrete, NDT & E Int.
35 (2002) 399406.
[4] C.-C. Cheng, T.-M. Cheng, C.-H. Chiang, Defect detection of concrete structures
using both infrared thermography and elastic waves, Autom. Constr. 18 (2008)
8792.
[5] D. Huston, J.Q. Hu, K. Maser, W. Weedon, C. Adam, GIMA ground penetrating
radar system for monitoring concrete bridge decks, J. Appl. Geophys. 43 (2000)
139146.
[6] P. Rossi, F. Le Maou, New method for detecting cracks in concrete using ber
optics, Mater. Struct. 22 (1989) 437442.
[7] F. Ansari, R.K. Navalurkar, Kinematics of crack formation in cementitious
composites by ber optics, J. Eng. Mech. 119 (1993) 10481061.
[8] Q. Li, G. Li, G. Wang, L. Yuan, CTOD measurement for cracks in concrete by ber
optic sensors, Opt. Laser Eng. 42 (2004) 377388.
[9] W. Zhang, B. Shi, Y.-F. Zhang, J. Liu, Y.-Q. Zhu, The strain eld method for
structural damage identication using Brillouin optical ber sensing, Smart
Mater. Struct. 16 (2007) 843850.
[10] T. Bao, Distributed ber Bragg grating sensors for monitoring cracks in
concrete structures, in: Earth and Space 2012, 2012, pp. 13901399.
[11] C.K. Leung, N. Elvin, N. Olson, T.F. Morse, Y.-F. He, A novel distributed optical
crack sensor for concrete structures, Eng. Fract. Mech. 65 (2000) 133148.
[12] G. Ball, W. Morey, Compression-tuned single-frequency Bragg grating ber
laser, Opt. Lett. 19 (1994) 19791981.
[13] G. Stepniak, J. Siuzdak, Experimental investigation of PAM, CAP and DMT
modulations efciency over a double-step-index polymer optical ber, Opt.
Fiber Technol. 20 (2014) 369373.
[14] Y. Luo, B. Yan, M. Li, X. Zhang, W. Wu, Q. Zhang, G.-D. Peng, Analysis of
multimode POF gratings in stress and strain sensing applications, Opt. Fiber
Technol. 17 (2011) 201209.

[15] A. Kulkarni, J. Na, Y.J. Kim, S. Baik, T. Kim, An evaluation of the optical ber
beam as a force sensor, Opt. Fiber Technol. 15 (2009) 131135.
[16] I.R. Husdi, K. Nakamura, S. Ueha, Sensing characteristics of plastic optical bres
measured by optical time-domain reectometry, Meas. Sci. Technol. 15 (2004)
15531559.
[17] Y. Wong, P. Scully, R. Bartlett, K.S.C. Kuang, W. Cantwell, Plastic optical bre
sensors for environmental monitoring: biofouling and strain applications,
Strain 39 (2003) 115119.
[18] A. Babchenko, Z. Weinberger, N. Itzkovich, J. Maryles, Plastic optical bre with
structural imperfections as a displacement sensor, Meas. Sci. Technol. 17
(2006) 11571161.
[19] N. Takeda, T. Kosaka, T. Ichiyama, Detection of transverse cracks by embedded
plastic optical ber in FRP laminates, in: 1999 Symposium on Smart Structures
and Materials,, Int. Soc. Opt. Photonics 1999 (1999) 248255.
[20] K. Nakamura, I.R. Husdi, S. Ueha, Memory effect of POF distributed strain
sensor, in: Second European Workshop on Optical Fibre Sensors, International
Society for Optics and Photonics, 2004, pp. 144147.
[21] K. Krebber, P. Lenke, S. Liehr, J. Witt, M. Schukar, Smart technical textiles with
integrated POF sensors, in: The 15th International Symposium on: Smart
Structures and Materials & Nondestructive Evaluation and Health Monitoring,
International Society for Optics and Photonics, 2008, pp. 69330V69330V.
[22] S. Liehr, P. Lenke, M. Wendt, K. Krebber, M. Seeger, E. Thiele, H. Metschies, B.
Gebreselassie, J.C. Munich, Polymer optical ber sensors for distributed strain
measurement and application in structural health monitoring, Sensors J. IEEE 9
(2009) 13301338.
[23] K.S.C. Kuang, W. Cantwell, C. Thomas, Crack detection and vertical deection
monitoring in concrete beams, using plastic optical bre sensors, Meas. Sci.
Technol. 14 (2003) 205216.
[24] K.S.C. Kuang, S. Quek, M. Maalej, Assessment of an extrinsic polymer-based
optical bre sensor for structural health monitoring, Meas. Sci. Technol. 15
(2004) 21332141.
[25] M. Born, E. Wolf, Principles of Optics: Electromagnetic Theory of Propagation,
Interference and Diffraction of Light, 3d ed., Pergamon, New York, 1965.