23. Citation 4

Construction and Building Materials 271 (2021) 121551
Contents lists available at ScienceDirect
Construction and Building Materials

journal homepage: www.elsevier.com/locate/conbuildmat
Review
Damage, nondestructive evaluation and rehabilitation

of FRP composite-RC structure: A review
Dongsheng Li a,b,⇑, Junlong Zhou a, Jinping Ou a,b
a
School of Civil Engineering, Dalian University of Technology, Dalian 116024, China
b
State Key Laboratory of Coastal and Offshore Engineering, Dalian University of Technology, Dalian 116024, China
h i g h l i g h t s
Nondestructive detection techniques (NDTs) that are applicable to RC structures reinforced with FRP composites are reviewed.
The damage levels of FRP composite-RC structures are summarized.
The detection capacities and characteristics of NDTs on FRP composite-RC structure damage are discussed.
Suggestions for selecting potential NDTs and remedial measures are presented for FRP composite-RC structure damage.
Future studies are recommended to better develop and apply NDTs to inspect FRP composite-RC structure damage.
a r t i c l e i n f o a b s t r a c t
Article history: Fiber reinforced polymer (FRP) composites have been widely used to retrofit RC structures, such as with
Received 26 May 2020 externally bonded FRP laminates (EB), externally bonded reinforcement on grooves (EBROG), embedded
Received in revised form 13 September FRP bars (embedded), near surface mounted FRP bars/strips (NSM), and mechanical anchorage (MA).
2020
Damage to retrofitted structures can be evaluated in advance or in real-time by nondestructive detection
Accepted 29 October 2020
Available online 30 December 2020
techniques (NDTs) to assess the health conditions, whether in the construction stage, service stage, or
repair stage. This present work reviews valuable studies on the damage detection of existing FRP
composite-RC structures conducted by some well-known NDTs, including piezoelectric transducer
Keywords:
FRP composite-RC structure
(PZT) methods, acoustic emission (AE) monitoring, ultrasonic testing, fiber optics sensor methods, micro-
Nondestructive detection technique wave detection, acoustic laser detection, infrared thermography (IR), and digital image correlation (DIC).
Damage level First, the damage levels of FRP composite-RC structures are presented to determine the inspection
Repair objects. Then, the capacities, advantages, and disadvantages of the NDTs are discussed by inspecting
the damage levels. Moreover, considering the damage levels, the suggestions for selecting potential
NDTs and repair measures are presented for different FRP composite-RC structures. Finally, the research
needs in this field are discussed to better develop and apply NDTs for detecting FRP composite-RC struc-
ture damage.
Ó 2020 Elsevier Ltd. All rights reserved.
Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Typical applications and mechanical behavior of FRP composites in RC structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.1. Material properties and failure modes of FRP composites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2.2. Typical applications and failure modes of FRP composites in RC structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. Damage detection of FRP composite-RC structures conducted by NDTs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
3.1. Piezoelectric transducer methods. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.2. Acoustic emission monitoring . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.3. Ultrasonic testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.4. Fiber optics sensor methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
⇑ Corresponding author at: School of Civil Engineering, Dalian University of Technology, Dalian 116024, China.
E-mail addresses: lidongsheng@dlut.edu.cn (D. Li), zhoujunlong@mail.dlut.edu.cn (J. Zhou).
https://doi.org/10.1016/j.conbuildmat.2020.121551
0950-0618/Ó 2020 Elsevier Ltd. All rights reserved.
D. Li, J. Zhou and J. Ou Construction and Building Materials 271 (2021) 121551
3.5. Microwave detection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

3.6. Acoustic laser detection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.7. Infrared thermography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.8. Digital image correlation technology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.9. Mixed nondestructive detection techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
4. Capacities and characteristics of NDTs on FRP composite-RC structures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
5. Selections of NDTs and repair measures on FRP composite-RC structures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
6. Future research needs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
7. Conclusion remarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23
Declaration of Competing Interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
1. Introduction microwave detection [45–49], acoustic laser detection [50–52],

infrared thermography (IR) [8,9,11,53–59], and digital image corre-
Over the last few decades, due to upgraded bearing require- lation (DIC) [12,13,60,61]. These techniques have been well devel-
ments, concrete aging, mistakes in design, chloride exposure, water oped in recent years. However, there is a lack of a comprehensive
ingress, steel rebar corrosion, and thermal shock, existing RC struc- evaluation of the capacity and selection of NDTs as well as the
tures have been subjected to engineering issues with concrete potential repair measures for damaged FRP composite-RC
cracking and even insufficient bearing capacity worldwide. The structures.
development of strengthening and repairing RC structures has In the present paper, NDTs effective for the damage detection of
therefore received much attention in recent years. RC structures reinforced with FRP composites are reviewed in the
Fiber reinforced polymers (FRPs) are widely used for retrofitting following aspects. First, based on the failure modes of FRP compos-
RC structures because of their excellent performance, including ites under tensile loadings and typical FRP composite-RC struc-
great mechanical properties, light weight, low thermal conductiv- tures, the damage levels are presented for the FRP composite-RC
ity, corrosion resistance, high strength-to-weight ratio, high structure in Section 2. The damage levels involve the material
stiffness-to-weight ratio, durability, high tensile strength, excellent defects or construction defects, and the damage in service resulted
fatigue resistance, versatility, low relaxation losses, minimal dis- from loadings or harsh environments. In Section 3, the methodol-
ruption, easy handling, ease of installation, chemical attack resis- ogy and valuable works of NDTs are highlighted for detecting
tance, insect and fungal resistance, and cost-effectiveness [1–5]. FRP composite-RC structure damage. Furthermore, the detection
The typical applications of FRPs on retrofitting RC structures capacity of NDTs covers the evaluation of the damage location,
cover externally bonded FRP laminates (EB), externally bonded type, size, and depth or degree for different damage levels. In addi-
reinforcement on grooves (EBROG), embedded FRP bars (embed- tion, in Section 4, the capacities, advantages, and disadvantages of
ded), near surface mounted FRP bars/strips (NSM), and mechanical potential NDTs are discussed to detect the different damage levels
anchorage (MA) techniques. According to the design guidelines of of FRP composite-RC structures. Moreover, aiming at detecting dif-
ACI 440.2R-17 [6] and existing works, different FRP composite-RC ferent damage levels, the selection of the NDTs and potential repair
structures could fail in different ways. The EB technique could fail measures are presented for the FRP composite-RC structures in
due to interface debonding between the FRP composites and con- Section 5, providing a reference for engineers and researchers.
crete, concrete cover peeling [7–13], or FRP rupture that results Finally, some limitations and future research needs for the devel-
in the yielding of the steel bars in the concrete. In the embedded opment and improved application of NDTs for detecting FRP
and NSM techniques, potential failures include bond-slip [14], con- composite-RC structure damage are discussed.
crete cracking [15], and FRP rupture [16]. As a result of the
improved bonding strength, the failure modes of the EBROG tech-
2. Typical applications and mechanical behavior of FRP
nique are represented by FRP rupture [13] and some debonding.
composites in RC structures
Additionally, for a high bonding strength, the MA technique often
fails due to anchor rupture [12], FRP strip longitudinal shear failure
2.1. Material properties and failure modes of FRP composites
[17,18], and fastener pull-out failure [19].
Nevertheless, two characteristics of the failure modes for the
In the last few decades, FRPs have aroused the interest of many
FRP composite-RC structure can be summarized as (1) invisible
researchers worldwide. The existing FRPs include carbon/glass/ara-
bonding damage and covered concrete substrate cracking and (2)
mid/basalt FRPs (C/G/A/BFRP). FRPs combine the matrix phase
obvious brittleness. FRP composites exhibit small deformations
(epoxy/vinyl esters/polyesters resin) and the reinforcement phase
and brittleness, and advanced debonding failure results in the
(C/G/A/B fibers). The combination of these two phases has an obvi-
material capacity of FRP composites not being fully used. There-
ous contact interface. The performance of the composite material is
fore, the damage needs to be detected in advance or in real time
different from the individual components and highly depends on
to ensure structural safety and to predict the residual life of FRP
the bonding conditions between the resin and fibers [62]. The resin
composite-RC structures in service.
system bonding the fibers together is required to have adverse-
Due to the features of nondestructive, comprehensive, compat-
environment resistance, workability, filling ability, and compatibil-
ible, and real-time, nondestructive detection techniques (NDTs) are
ity with and adhesion to the fibers, concrete substrate, and FRP
generally used in the health monitoring of FRP composite-RC struc-
composite system [6].
tures. To date, commonly used NDTs for inspecting FRP composite-
Some mechanical properties of C/G/A/BFRPs, including the elas-
RC structure damage include piezoelectric transducer (PZT) meth-
tic modulus, tensile strength, and ultimate tensile strain, are sum-
ods [20–30], acoustic emission (AE) monitoring [7,14,31,32], ultra-
marized by this review based on experimental data [2,5,13,14,32,
sonic testing [15,33–40], fiber optics sensor methods [41–44],
34,54,63–84]. The discussed mechanical properties of FRPs might
2
Fig. 1. Material properties of the FRP composites: (a) elastic modulus; (b) tensile strength; (c) rupture strain.
be suitable for whether the retrofitted or strengthened structures or cyclic tensile loadings. The damage process mainly exhibits
by FRP composites. Fig. 1 reveals that CFRPs have the highest ten- three stages [85,86]. In the first stage, there is almost no damage,
sile strength and elastic modulus, but their ductility is low. The or only microcracks appear in the resin, as shown in Fig. 2(a). With
high elastic modulus and high tensile strength ensure sufficient the increase in tensile strain, in the second stage, the microcracks
stiffness for the retrofitted RC structure, which helps to control gradually develop into transverse cracks, and debonding between
the strain of retrofitted RC structures. The tensile strength and the fibers and the resin is observed, as shown in Fig. 2(b). In the last
elastic modulus of AFRPs are inferior to those of CFRPs, but the stage, in addition to transverse cracks and debonding, longitudinal
ductility of AFRPs is more excellent than that of CFRPs. It is worth splitting, delamination, and fiber rupture occur successively, as
noting that BFRPs and GFRPs have similar tensile strengths and shown in Fig. 2(c) and (d). In summary, the potential failure modes
elastic modulus values, but BFRPs have a more considerable ulti- of FRP composites involve matrix resin damage, fiber-to-matrix
mate tensile strain than that of the GFRPs. This contributes to sub- delamination, fiber pull-out, and fiber rupture.
stantial fatigue resistance for BFRPs. Given this background, FRP
composites have excellent mechanical properties and outstanding 2.2. Typical applications and failure modes of FRP composites in RC
durability for seawater sea-sand concrete (SSC) structures. structures
In addition, FRP composites should not be relied on to resist
compressive forces; that is, the compressive strength of the FRP Up to the present, the application of FRP composites in retrofit-
reinforcement should be neglected [6]. Thus the potential damage ting RC structures includes embedded, EB, EBROG, NSM, and MA
is shown in Fig. 2 for FRP composites when subjected to monotonic techniques. There are differences in the mechanism and failure
3
Fig. 2. SEM images of the damage process in the BFRP plate subjected to tensile loadings [85]: (a) resin matrix micro-cracks; (b) fiber-resin interface peeling; (c) longitudinal
splitting; (d) delamination and fiber rupture.
modes for different FRP applications. This difference depends on concrete. For this reason, the failure modes of FRP reinforcing con-
the bonding conditions between FRP composites and the RC struc- crete structures are also different from those of conventional RC
ture (FRP-resin and resin-concrete substrate), concrete substrate structures. Potential failure modes of FRP bar-concrete structures
strength, and FRP strength. The polymeric resin, with varying prop- include FRP pull-out [92], concrete splitting [93], and FRP rupture
erties represented by viscosity, is an essential condition for con- [94], which are dominated by pull-out and concrete splitting fail-
necting FRP composites and the RC structure to ensure their ure [90]. These failure modes mainly depend on the concrete
work together. The resins include primers, putty fillers, and satu- strength, diameter, and spacing of the FRP bars, concrete cover,
rants [6]. The primer penetrates the concrete substrate, the putty and stirrup ratio. The bonding stress is nonlinearly distributed on
fills holes on the substrate to provide a smooth surface, and the the FRP bar from the ‘‘loading end” to the ‘‘free end” [14,95,96].
saturating resin impregnates the FRP composites and fixes them The ‘‘loading end” means the beginning of effective bonded length
on the concrete to give an expected shear load path. The damage in concrete near the loading force; this explanation also applies to
mechanisms associated with different FRP applications are ana- the same concept ‘‘loading end” that appears later. The ‘‘free end”
lyzed in advance to summarize the damage levels in this section. means the end of effective bonded length in concrete far away
When subjected to harsh environments, steel bar corrosion and from the loading force, also applied to the same concept ‘‘free
bonding damage between the steel bars and concrete can easily end ” that appears later.
shorten the service life of conventional RC structures. Therefore, For strengthening or repairing, the RC structure adhered FRP
FRP bars with excellent corrosion resistance have been widely used laminates (plates/strips/sheets) to improve the bearing capacity
in recent years, especially for SSC structures [87–89]. As is well and rigidity. The EB technique involves side bonding, U-
known, the cohesive stress between the embedded bars and con- wrapping, and full wrapping. For flexural specimens, FRP laminates
crete is mainly derived from chemical bonds, friction properties are arranged on the bottom surface instead of the top surface of the
of the contact surfaces, and mechanical interlock. Because different RC structure as they should not be counted on to resist compres-
materials and surface characteristics cause different surface rough- sive forces. However, it is acceptable for FRP tension laminates to
nesses and mechanisms [90,91], the bonding stress between FRP share a little compressive force due to the negative moment [6].
bars and concrete is different from that between steel bars and Possible failure modes include debonding of the FRP-concrete
4
Fig. 3. Configuration details of the FRP anchors technique [100,103]: (a) anchor spikes; (b) U-anchors; (c) plate anchors.
interface, concrete cover peeling, and even FRP rupture resulting in failure. Debonding usually begins with concrete shear cracks. Fur-
steel bar yielding. The potential debonding of FRP-concrete thermore, the stress concentration at the shear crack region
includes cohesive debonding and adhesive debonding. For the for- reduces the bonding area and eventually leads to debonding failure
mer, as the concrete substrate cannot bear the axial force in the as the debonding area increases. In contrast, for full wrapping, FRP
FRP laminates, a thin layer of concrete is attached to the FRP lam- rupture often occurs at the intersection region between the critical
inates. In contrast, the latter stems from concrete bending cracks or shear cracks and FRP laminates [98].
shear-flexural cracks and occurs at the FRP laminate-adhesive Aiming at improving the bonding strength of the EB technique,
interface or the adhesive-concrete interface. The failure mode of especially for side bonding and U-wrapping, the EBROG technique,
concrete cover peeling is due to the stress concentration at the which can avoid the debonding failure of the EB technique, has
end of the FRPs. The stress concentration causes concrete cracking, been developed. The failure mechanism is that the stress is concen-
which propagates along the tensile longitudinal bars [66]. In addi- trated at the loading end and does not extend towards the free end
tion, when the FRP strain exceeds the ultimate tensile strain, FRP to form a stress redistribution. Eventually, the FRP laminates rup-
rupture will occur, then the steel bars will yield, and finally, the ture near the loading end [13]. This failure is due to the stiff
concrete will be crushed in the compression regions. This is an epoxy-filled deep grooves result in higher bond strength, although
ideal failure mode for the EB technique [97]. For shear specimens, they provide a relatively smaller bond length than those in the EB
especially for side bonding and U-wrapping, scattered concrete technique. The optimized groove depth was recommended as
cracks can accelerate FRP debonding and cause premature shear 10 mm to ensure the expected ultimate bond strength.
Fig. 4. Configuration details of the MF-FRPs technique and the HB-FRP technique [17,19].
5
Alternatively, the MA technique can also improve the bonding cover [19] to enhance the bonding strength further. The failure
strength, involving FRP anchors, mechanically fastened FRPs (MF- mode of bending specimens with the HB-FRP technique is that
FRPs), and hybrid bonding FRPs (HB-FRPs). As commonly used the fasteners are pulled out at the end of the beam to result in
forms of anchors, FRP anchors include anchor spikes, U-anchors, debonding [19], and the FRP strips broken in the middle of the
and plate anchors, shown in Fig. 3(a)–(c). The failure modes of beam span [105].
anchor spikes can be summarized as anchor pull-out and FRP Finally, the NSM technique has been widely used in RC speci-
debonding [99,100] and anchor rupture [19,100]. Due to the mens subjected to moment and shear to improve the bond
increased bonding area, the failure of U-anchor reinforcing speci- strength of FRP-concrete and to protect the FRP bars/strips from
mens is FRP rupture in the groove [101]. Simultaneously, the slot the adverse effects of harsh environments. The failure of RC struc-
will trigger concrete cracking, resulting in the interface debonding tures with the NSM technique can be divided into the structural
of FRP laminate-concrete [102]. In the plate anchor reinforcing level and local level [106]. At the structural level, concrete crush-
specimens, the shear stress is transferred at the interface of the ing, FRP debonding, and FRP fracture occur in areas such as the
FRP laminate plate and then transmitted to the concrete substrate FRP ends, shear cracks, and flexural cracks [6,107]. At the local
through bolts. The ultimate failure mode is debonding or FRP rup- level, potential failure modes include FRP-adhesive debonding,
ture at the anchoring location [101]. adhesive-concrete debonding, adhesive splitting, and concrete
Similarly, the application of MF-FRPs [17,104] involves fixing splitting. Moreover, at the local level, one or more potential dam-
FRP strips to the RC structure surface with steel expansion anchors age areas may be present simultaneously in the NSM technique.
(EA), steel power actuated (PA) fasteners, and screw anchors, as Therefore, it is difficult to determine the specific damage mecha-
shown in Fig. 4(a). The MF-FRP technique utilizes the dowel action nism for the NSM technique. However, the following potential
of the PA/EA in the concrete to transfer the stress from the RC trend is still present [108,109]: if the adhesive is damaged, the
structure to the FRP strips to improve the strength and ductility. nearby concrete may not crack or split, but if the concrete is dam-
Different from the EB technique, the MF-FRP technique does not aged, the resin is likely to be damaged. Furthermore, the existing
require excessive surface treatment of the concrete and needs less works [109–111] summarized the potential damage mechanism
construction time. The failure mechanism of the MF-FRPs indicates of the NSM technique as follows: first, interface failure of the FRP
that FRP laminates experience ductile bearing failure or brittle lon- adhesive interface precedes that of the adhesives-concrete inter-
gitudinal shear failure, and the fasteners are pulled out, which face, then splitting failure occurs, and finally, the FRPs rupture.
results in FRP debonding [17,18]. The statistics of the failure modes of FRP composite-RC struc-
The HB-FRP technique (shown in Fig. 4(b)), which presents a tures mentioned above [7,8,12,13,17,18,21,60,61,92–94,99–101,1
substantial advantage, minimizes the damage to FRP materials 05] are presented in Fig. 5. In Fig. 5, AD represents adhesive
caused by anchors compared to that of the MF-FRP technique, debonding, occurring at the FRP-adhesive interface or adhesives-
especially for longitudinal shear failure. The stress on the RC struc- concrete interface, where concrete does not detach from the FRP
ture is predominantly transmitted to the FRPs through the FRP- composites; CD represents cohesive debonding, accompanied by
concrete bonding stress, steel plate-FRP bonding stress, and anchor detached concrete; SF represents shear failure, occurring in the
dowel action, especially the vertical compressive stress provided EB technique and accompanied by debonding failure, but in full
by the anchors. This increases the friction capacity of the concrete wrapping, shear failure is often accompanied by FRP rupture at
Fig. 5. Statistics on damage modes of referred FRP composite-concrete structures.
6
Table 1
Damage levels of FRP composite-RC structures.
FRP application Pre-existing defect Damage stage Failure mode

Embedded technique – (1) FRP bar damage (1) FRP bar pull-out
(2) Concrete cracking (2) FRP bar rupture
(3) Local slip between FRP bar and concrete
NSM technique – (1) FRP bar/strip damage (1) FRP debonding
(2) Adhesive material splitting (2) FRP bar/strip rupture
(3) Concrete cracking/splitting
(4) Debonding between FRP bar/strip and adhesive layer
(5) Debonding between adhesive layer and concrete
EB/EBROG techniques (1) Bonding defect (1) FRP laminate damage (1) FRP laminate rupture(2) FPR laminate peeling(3)
Delamination FRP laminate strain Concrete cover peeling(4) Serious debonding between
Air void FPR laminate cracking FRP laminate and concrete
FPR laminate abrasion
(2) Concrete substrate cracking
(3) Slight debonding between FRP laminate and concrete
MA technique (1) Bonding defect (1) FRP laminate damage (1) Anchor pull-out
Delamination FRP laminate strain (2) Anchor rupture
Air void FPR laminate cracking (3) FRP laminate shear failure
FPR laminate abrasion (4) FRP laminate bearing failure
(2) Anchoring defect (2) Concrete substrate cracking (5) RP laminate rupture
(3) Slight debonding between FRP laminate and concrete
(4) Anchor damage
the critical shear cracks; DF represents concrete cover peeling, structures reinforced with FRP composites in service resulted from
occurring in RC beams subjected to bending and strengthened by loads or adverse environments.
the EB technique. Noticeably, the number in Fig. 5 cannot indicate
the probability of various failures but covers almost all possible
3. Damage detection of FRP composite-RC structures conducted
failure modes.
by NDTs
Most of the failure modes in Fig. 5 characterizes by severe brit-
tleness. Compared to conventional steel bars, the failure of FRP
The debonding between FRP composites and concrete can cause
composite materials is obviously brittle, and no apparent warnings
the loss of stress transmission, thereby reducing the bearing capac-
occur before failure. Given this background, it is necessary to
ity of the reinforced structure. In addition, the bonding defects can
detect or monitor the damage of FRP composite-RC structures
cause a stress concentration at the FRP-concrete interface and then
through NDTs. Based on the previous discussion, the damage levels
promote the development of defects. To ensure the reliability of
of the FRP composite-RC structures are given in Table 1 before
FRP composite-RC structures, not only should the bonding behav-
reviewing the existing works of the NDTs. In Table 1, ‘‘Pre-
ior be improved, but the bonding conditions also need to be
existing defect” refers to the material defects or construction
detected in advance. The bonding condition that needs evaluation
defects in the RC structures retrofitted by the EB/EBROG/MA tech-
includes the bonding behavior of FRP bars/strips and concrete in
niques, and ‘‘damage stage” indicates that the damage of the RC
the embedded/NSM techniques, the bonding behavior of FRP
Fig. 6. Experimental setup of the WP and EMI techniques.
7

laminates and concrete in the EB/EBROG/MA techniques, and the
DI ¼ Hn Dn = Hn ð1Þ
bonding quality of the anchors and FRP laminates in the MA tech-
nique. The inspection and evaluation of these bonding conditions
where Hn is the average signal amplitude at a healthy state. Dn is the
can be achieved through NDTs due to the features of nondestruc-
average signal amplitude under different damage conditions.
tive, comprehensive, compatible, and real-time. Besides, damage
In the embedded technique, specimens in flexure are often
detection can be divided into five objectives: determining the pres-
accompanied by the larger deflection and the lower post-
ence of damage, locating the damage, identifying the type of dam-
cracking bending stiffness. Zheng et al. [26] placed SAs on the con-
age, estimating the degree of damage, calculating the residual
crete surface and proposed piezoelectric SA transducers based
bearing capacity of the structure after damage, and predicting
bending stiffness index to predict the bending stiffness of the
the remaining life of the structure [112].
beam. Compared with the surface-pasted PZT sheet, smart aggre-
Currently, NDTs used for inspecting FRP composite-RC structure
gates (SAs) can better reflect the interface damage of FRP laminate
damage involve piezoelectric transducer (PZT) methods [20–
concrete [21]. Therefore, only the SAs were embedded into con-
30,113], acoustic emission (AE) monitoring [7,14,31,32], ultrasonic
crete beams; the concrete cracking of the FRP bar-concrete struc-
testing [15,33–40,114], fiber optics sensor methods [41–44],
ture was analyzed in the time and frequency domains [113].
microwave detection [45–48], acoustic-laser [50–52,115], infrared
Results showed that with the increase of the crack number and
thermography (IR) [8,9,11,53–59,116], and digital image correla-
width, sensor voltage amplitude and wavelet packet energy
tion (DIC) [12,13,60,61]. PZT methods, AE monitoring, ultrasonic
decreased. Similarly, the PZT sheet was installed on the FRP bars,
testing, and fiber optics sensor methods are contact techniques,
and SAs were embedded in the concrete to detect the pull-out
and the rest are noncontact techniques. In addition, AE monitoring
debonding failure in the embedded technique based on the WP
and DIC cannot identify old damage. In this section, the methodol-
technique [22]. The results revealed that bonding damage could
ogy, the existing works, and the capabilities of the NDTs are dis-
be determined by the attenuation of the stress waves in the time
cussed. The detection capacity involves the type, location, size,
and frequency domains [22,23]. Meanwhile, the debonding degree
and depth or degree of damage in the FRP composite-RC structures.
could be qualitatively evaluated based on the wavelet packet dam-
age index [22,24], presented in Eq. (2). The results showed that the
3.1. Piezoelectric transducer methods pull-out process accompanied by bonding damage, and an increase
in the DI indicated the growth of bonding damage severity. Even-
The application of smart piezoelectric materials provides an tually, the structure fully debonded when DI was 1.0.
innovative method for the structural health monitoring (SHM) of vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u 2n
FRP composite-RC structures. At present, there are mainly two uX 2 X 2n
detection techniques based on the piezoelectric principle, includ- DI ¼ t Ei;j E1;j = E21;j ð2Þ
j¼1 j¼1
ing the electromechanical impedance (EMI) technique and the
wave propagation (WP) technique [20]. For the EMI technique where E1,j represents the signal energy at the initial state. Ei,j repre-
(firstly developed by Liang et al. [117]), one PZT is simultaneously sents the damaged signal energy.
used as an actuator and a sensor, where one PZT excites the host Similarly, Li et al. [27] used the EMI technique to monitor the
structure using the converse piezoelectric effect, and the mechan- pull-out damage of the CFRP bar-concrete structure bonded PZT
ical vibratory response of the host structure interacts with the sheet on the concrete surface. They used statistical damage met-
same PZT sheet. The interacted mechanical vibratory response rics, root-mean-square deviation (RMSD) and mean absolute per-
changes the electrical current across the PZT due to the direct centage deviation (MAPD), given by Eqs. (3) and (4), to quantify
piezoelectric effect. As a result, the electrical impedance is mea- the changes in the spectra of impedance and admittance. These
sured by an impedance analyzer to evaluate the damage. Instead, indexes can explain the debonding evolution with three stages.
for the WP technique, one PZT is used as the transmitter that gen- Firstly, the RMSD value was low, and no bonding damage was
erates the stress wave using the converse piezoelectric effect. After detected at this stage; then, the RMSD index rose suddenly, which
propagating in the FRP composite-RC structures, the stress wave is was attributed to the beginning of debonding. Finally, the RMSD
received by the other PZTs and is converted into electrical signals index tended to be relatively stable.
using the piezoelectric effect. Finally, the attenuation of the ampli- vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u N
tude, phase, and energy of the wave signal (such as the frequency, uX 2 X N
2
amplitude, and energy) after passing through the damage regions RMSD ¼ t Re Y i;1 Re Y i;0 = Re Y i;0 ð3Þ
is analyzed to detect the damage [21,25,118–120]. The experimen- i¼1 i¼1
tal setup is shown in Fig. 6 for the WP and EMI techniques. The
wave signal analysis includes time-domain analysis, frequency 1X N

Re Y i;1 Re Y i;0 =Re Y i;0
domain analysis, and time–frequency analysis. Note that the signal MAPD ¼ ð4Þ
N i¼1
must be filtered before analysis [121].
Wang et al. [21] analyzed the bonding damage of FRP plate- where N is the sampling point number of EMI spectra. Re(Y) is the
concrete beams in flexure. The damage location was determined electrical admittance, and the subscripts 0 and 1 represent the base-
by amplitude analysis, and the relative damage index DI was used line measurement and concurrent measurement, respectively.
to reflect the damage degree qualitatively, as presented in Eq. (1). Noticeably, the detection accuracy of bonding damage in the
The DI, taken from 0 to 1, depends on the damage degree, and a FRP strip-RC structures was susceptible to environmental temper-
more considerable value of DI represents more serious the bonding ature and noise. Therefore, Sun et al. [25] utilized an ensemble par-
damage. The results showed that the DI at the FRP ends increased ticle swarm optimization (EPSO) algorithm with a bagging
faster than the others. This result indicated that the debonding (bootstrap aggregating) algorithm to improve the accuracy of the
process initiated at the FRP ends and developed towards the damage detection. They pointed out that the proposed adaptive
mid-span in the flexural member. Results also pointed out that algorithm can effectively identify the location of the artificial
the concrete cracked when DI was greater than 0.3, and the con- defects in the FRP strip-RC structure. Wandowskia et al. [28]
crete cover began to peel when DI was greater than 0.45. detected artificial delaminations with different sizes based on the
8
EMI technique. The application of the temperature compensation

algorithm using signal-cross-correlation improved the sensitivity
of the EMI technique to the damage when subjected to tempera-
ture change. The algorithm is to shift the temperature-affected sig-
nal relative to the baseline signal, thereby maximizing their
correlation. It showed that the sensitivity depended on the investi-
gated frequency sub-band. The visualization of damage size can
realize with root mean square energy map for guided wave propa-
gation when PZT was excited at 100 kHz. Also, based on the EMI
technique, Park et al. [29] monitored the artificial bonding damage
of the CFRP plate-concrete structure. They quantified the changes
in impedance measured at the PZT sheet due to the debonding con-
ditions based on cross-correlation (CC) analysis. Results showed
that the CC increased with an increase in the bonding damage
levels (2 cm, 4 cm, and 6 cm). The CC index used as linear damage
is defined as Eq. (5), where the closer the CC is to 1.0, the healthier
the FRP-concrete structure.
N
1X
CC ¼ xi x yi y = rx ry ð5Þ
N i¼1 Fig. 8. Locating damage in the EB technique conducted by the AE monitoring [122]:
(a) front view; (b) side view.

where x and y are the mean of signature two data sets x and y,
where rx and ry are standard deviations of the two data sets,
respectively. density, and cumulative energy to determine the damage of the
structure. This method is suitable for structures with little damage.
Ghiassi et al. [31] studied the cumulative AE energy and cumula-
3.2. Acoustic emission monitoring
tive AE energy to hit ratio of masonry with GFRP strips in the
single-lap shear bond test. Combined with the final failure modes
As a contact detection technique, the principle of acoustic emis-
of the test specimens, they identified potential debonding failure
sion (AE) monitoring is that once the FRP-concrete structure in the
modes, which included cohesive debonding, pure adhesive
stressed state is damaged with new fracture, crack, and debonding,
debonding, and cohesive/adhesive debonding. Different debonding
the structure will be in another balanced state of stress. In this pro-
failure modes have different AE characteristics. However, it is dif-
cess, the structure will release strain energy through the full or
ficult to determine the damage evolution without locating the
partial recovery of deformation [31]. This strain energy will prop-
damage. In contrast, Di et al. [14] conducted a direct pull-out test
agate from the damaged location to the surroundings inside the
on self-compacting concrete with B/GFRP bars and analyzed the
structure in the form of transient elastic waves. Sensors such as
collected AE signals. They found that the AE signal intensity was
piezoelectric transducers can receive the weak elastic wave and
highly consistent with the bonding stress. The fluctuation of the
convert it into an AE signal. Finally, the location and degree of
signal represented damage with a low energy release. Moreover,
damage can be evaluated according to wave theory. However, if
they located the damage, as shown in Fig. 7, where each point rep-
the internal damage of the structure does not expand or change,
resented an AE event, and different colors of the point represented
no AE signal will be generated.
the signal intensity. As a result, it can be determined that the dam-
Conventionally, the AE monitoring uses the direct analysis of
age evolution was as follows. Bonding damage started at the free
damage signals such as with the amplitude, AE evens, AE evens
Fig. 7. Locating damage in the embedded technique conducted by the AE monitoring [14]: (a) SCC-0.0%-GFRP-d12-80; (b) SCC-0.0%-BFRP-d12-80.
9
end of the GFRP bar and developed towards the middle, but the with the increase in water-immersion aging, the acoustic parame-
opposite result appeared in the specimen with the BFRP bar. In ters decreased substantially and had a correlation with the fracture
addition, according to the time and propagation speed of the energy attenuation. This result indicated that the specimen was
received AE signals and various triangulation techniques based subject to debonding or delamination. Similarly, the combination
on the mathematical analysis of acoustic wave trajectories, Grazz- of the laser ultrasonic technique and a time–frequency representa-
ini et al. [122] realized the locating of AE hits in the EB technique, tion could develop a single lamb wave into a unique divergence
as shown in Fig. 8. It was mentioned that the damage located by AE curve, thereby reflecting the presence of bonding damage [39].
monitoring depends on the complexity of the structural form and Moreover, to avoid the influence of the coupling conditions
the signal attenuation caused by the heterogeneity of the between the probe and the structure, an equivalent time length
materials. (ETL) method, presented in Eq. (6), was proposed to quantify the
Inconveniently, multiple damage mechanisms could occur con- energy of the ultrasonic signal that propagated along with the
currently, which resulted in AE signals overlapping in the relatively interface in the EB technique [33], where ETL trends decreased
complex FRP composite-RC structures [7]. A feasible method to with more severe defects.
classify the damage mechanisms is to perform the multiparameter pffiffiffi rffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi
ETL ¼ 1= 2 6p ð3 þ 8n2 p2 Þ= f
2
analysis on the AE signal, where the mathematical processing of ð6Þ
the AE signal data is critical. Nair et al. [7] used pattern recognition
to perform the cluster analysis on the collected AE data in the where f is the frequency, and n is the number of cycles.
bending test of the EB FRP strip-RC beams. They found that the fail- Previous studies have indicated that GUWs can detect the dam-
ure modes were concrete cover peeling or a mixed failure of shear age of EB FRP composite-RC structures, but the location and size of
failure and concrete cover peeling. Cluster 1 first appeared and the damage need to be discussed further. In homogeneous materi-
penetrated during the entire loading process; hence, it corre- als, as the propagation speed of echo ultrasonic monitoring is con-
sponded to concrete substrate cracking. Cluster 2 was a dominat- stant and known, it is easy to identify the location of the defects by
ing cluster with a high density and steady growth and was evaluating the delay of the echo ultrasonic. However, concrete is a
specified to correspond to the bonding damage. Cluster 0 was the heterogeneous material that results in echo ultrasonic attenuation,
sparsest, so it could have originated from early microcrack devel- making it difficult for GUWs to detect the damage of FRP
opment and flexural cracks. Additionally, by processing the AE sig- composite-RC structures. For this reason, Kundu et al. [38]
nal based on principal component analysis (PCA), Degala et al. [32] employed the longitudinal wave or P-wave (C-scan) and Lamb
classified the bonding damage mechanisms and concrete cracking wave (L-scan) to visualize the artificial bonding defects of the EB
in EB FRP strip reinforcing concrete slabs. GFRP plate-concrete, as shown in Fig. 9. They determined that
the flaws could be detected using both C-scan and L-scan correctly,
3.3. Ultrasonic testing but the former was more sensitive to changes in the structure
behavior. In contrast, the L-scan provided better visual images
Damage detection of FRP composite-RC structures conducted by for the defects. Similarly, using the first echo peak amplitude and
ultrasonic testing is predominantly detected by guided ultrasonic C-scan techniques, Bastianini et al. [40] detected the bonding dam-
waves (GUWs) with a strong penetrability and high sensitivity. ages of C/GFRP-coated concrete columns. They were able to deter-
GUWs use an ultrasonic transducer as a transmitter (Tx) to trans- mine the location and size of the defects. Moreover, the changes in
mit ultrasonic signals such as SH waves, SV waves, Lamb waves, resin thickness did not affect the technique. The ratio of the
and Rayleigh waves. Then, the ultrasonic signals are reflected and refracted energy to the reflected energy depended on the mis-
scattered on the FRP-concrete interface. This process results in match degree of the acoustic impedance in the different materials.
complex interference effects to form GUWs. Finally, after a receiver Trovillion et al. [37] successfully detected the position and size of
(Rx) received GUWs, the signal difference of the GUWs before and damage related to delamination, debonding, and concrete cracking
after damage to the interface is analyzed to evaluate the damage for EB CFRP composite-RC structures using radio frequency (RF)
information. waveforms and C-scans, as shown in Fig. 10.
Some vital parameters of the GUWs are related to the bonding In addition, ultrasonic testing has been used to detect bonding
damage of FRP composite-RC structures [33,39,114]. Mahmoud damage between steel bars and concrete [15,34–36]. These studies
et al. [114] analyzed the parameters of surface acoustic waves might demonstrate the possibility of ultrasonic detection of the
(SAWs) in EB CFRP composite-RC specimens. They claimed that bonding damages of FRP bars and concrete in the embedded tech-
Fig. 9. Transmitter (T) and receiver (R) layout in the C-scan and L-scan imaging [38].
10
Fig. 10. Damage visualization by C-scan image of the ultrasonic testing [37].
nique. Du et al. [36] embedded the protected piezoelectric trans- ously, due to the compatibility with FRP materials, fiber optic sen-
ducer in concrete as the ultrasonic transducer and analyzed the sors are superior to other strain sensors. The analysis of the strain
acoustic parameters to detect the presence of damage. Simultane- measured by fiber optic sensors can determine the presence of CD
ously, through the pulse-echo (PE) method, Sharma et al. [34] damage [42,43]. In addition, because of the effects of the adhesives
observed the steel bar corrosion process of FRP-wrapped RC col- and fiber-protected layers, part of the energy will be converted into
umns in a chloride environment. They correlated the PE signal with shear strain. This process will affect the accuracy of the strain mea-
the damage type to distinguish between the pitting corrosion and surement. Given this background, Lau et al. [123] proposed a the-
surface corrosion of the steel bar. However, corrosion of steel bars oretical model to evaluate the previous influencing factors. As the
in concrete can also trigger concrete cracking. thickness of the adhesives increased, the measured axial strain was
For this reason, Climent et al. [15] used nonlinear ultrasonic lower than the real strain; the thick bonds and the fiber-protected
(NLU) techniques to detect cracks in concrete caused by steel bar layers with a lower elastic modulus rose the area where there was
corrosion. They found that when using ultrasonic excitation with a more concentrated shear stress at the end of the bonding region.
a stable center frequency, concrete cracking was accompanied by Indeed, previous studies indicate that the optical fiber sensor can
the high harmonics of the output signal. Furthermore, Sriramadasu detect damage, but they do not involve the location and degree
et al. [35] utilized the arrival time of the scattered wave model to of damage. Kesavan et al. [44] arranged five FBGs at different posi-
calculate the location of pitting corrosion on a bare steel bar. After tions along the longitudinal direction of the beam in the EB CFRP
the steel bar was divided into n elements with n + 1 nodes, they strip-concrete beam. The results showed that the initial FBG strain
proposed a damage index DI(x) to monitor the corrosion process at the beam end was higher than that at the beam middle. This
using the reflection pattern that occurs due to the pitting corrosion, result indicated that the stress concentration first occurred in the
given by Eq. (7). Typically, there was no corrosion when the DI(x) CFRPs at the beam end, consistent with the destructive test results.
was equal to 0, and the increase in the corrosion degree increased In addition, the damage type information depended on the
the value of DI(x). arranged location of the sensors; that is, to detect the damage in
sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Z t2 ffi the FRP composites, the sensor needed to be embedded into the
2 FRP composites.
DIðxÞ ¼ ðS BÞ dt ð7Þ
t1 In the embedded/NSM techniques, pultrusion technology, FBG
sensors and glass fibers can be synthesized into a self-sensing GFRP
where t1=(L + 2x)/v; t2 = t1 + D. v is the propagation speed of the bar. Wang and Kalarnkarov et al. [16,124] compared the test
scattered wave model. L is the steel bar length. is the distance results of conventional strain gages with those of the smart GFRP
between the node and the sensor. D is the input pulse bandwidth bar. They concluded that the results from the FBG sensor were
(50 ms for 100 kHz). S represents a damaged signal. B represents a more reliable because strain gages were susceptible to damage
healthy signal. due to epoxy resins and concrete cracking. Unless the sensor expe-
riences delamination due to GFRP bar rupture or rupture due to
3.4. Fiber optics sensor methods shear and axial forces, the FBG sensor can continue to obtain accu-
rate data [124] and accurately access the local slip. Eventually, the
Due to their light weight, small size, high-pressure resistance, FBG sensor is damaged when it reaches its maximum range, so an
corrosion resistance, electromagnetic field resistance, and compat- FBG sensor with a more extensive measuring range is needed to
ibility with composite materials, fiber optics sensors are commonly monitor the strain of RC structures with a larger opening of con-
used for damage detection of FRP composite-RC structures. Fiber crete cracking. Tang et al. [125] used self-sensing BFRP bars, as
Bragg grating (FBG) is the representative fiber optic sensor. It is shown in Fig. 11, to accurately measure the macrostrain of BFRP
based on the principle that the strain transmitted from the host bars. When concrete cracking occurred near the sensor, the
material to the fiber core grating region will change the grating macrostrain of the bar unit increased substantially. Therefore, the
length and refractive index of the fiber core. The changes reflect monitoring accuracy in locating concrete cracking depends on
the changes in the strain of the host material. Existing techniques the division degree of the units and the number of FBG sensors.
include optical low-coherence reflectometry (OLCR), optical time-
domain reflectometry (OTDR), and optical frequency domain
reflectometry (OFDR). The most suitable technique is OFDR for civil 3.5. Microwave detection
engineering applications. OFDR can conduct millimeter-level mea-
surements in the range of tens to hundreds of meters [77]. An electromagnetic wave (EM) with a wavelength range of
Fiber optics sensors were installed in the adhesives of the CFRP 1 mm–1 m and a frequency range of 300 M to 300 GM is a micro-
sheet concrete, and the beginning and failure of the AD damage can wave. As the absorption of microwaves by detected materials is
be determined by a sudden increase in the signal [41]. Simultane- proportional to the dielectric constant of the material, microwaves
11
Fig. 11. Manufacture of smart FRP bar [125]: (a) the smart FRP bar structure; (b) the unit of the long-gauge FBG sensing core; (c) the product.
can be used to detect the damage in FRP composite-RC structures

due to the characteristics of reflection, scattering, and projection
at discontinuous interfaces, as shown in Fig. 12. The available elec-
tromagnetic wave signals involve continuous waves, pulse waves,
and modulated waves. The primary detection methods include
constant wave/sweep wave/pulse modulated wave penetration,
continuous-wave/swept wave/pulse modulated wave reflection,
standing wave interference, and scattering waves.
By transmitting EM waves onto the GFRP sheet-concrete col-
umn, Feng et al. [45,47] analyzed the reflection characteristics in
the time and frequency domains to detect artificial bonding
defects. The location and size of the damage could be further iden-
tified by the EM imaging technique [49], but it was necessary to
use the dielectric lens to focus the EM wave on the bonding inter-
face. Similarly, open-ended rectangular waveguides [46] and ultra-
wideband (UWB) noise waveforms [48] could also detect the
location and size of artificial defects in CFRP sheet-concrete speci-
Fig. 12. Reflection measure of the microwave detection in the EB FRP-concrete mens. Moreover, a subsurface imaging technology using a focused
specimen [45].
EM wave was applied to detect the depth of artificial defects [47]. A
Fig. 13. Detection system and flow of the acoustic laser detection in the EB technique [51,127].
12
comprehensive analysis of microwave applications can be found in 3.7. Infrared thermography

the work of Feng et al. [126].
As a noncontact technique, the basic principle of infrared ther-
3.6. Acoustic laser detection mography (IR) is that when the FRP composites in the EB technique
have a temperature difference from the environment, heat will
Sound sources, such as loudspeakers, generate sound waves flow inside the structure. During the diffusion and transfer of heat
(represented by white noise whose power spectral density is con- flow in FRP composite-RC structures, the effect of bonding defects
stant throughout the frequency domain) to excite and vibrate FRP on the heat conduction reflects the difference in the surface tem-
composite-RC structures. The amplitude and vibration frequency of perature of FRP composites, which forms hot regions and cold
the structure with and without bonding damage are significantly regions. The infrared radiation energy is measured by an infrared
different. Immediately after the structures are excited, a laser camera and is then focused on with a detector that converts the
beam is used to shoot the structure surface, and a photoreceiver infrared radiation energy into an electrical signal. The electrical
received the reflected laser beam. Therefore, different electrical signal after processing is further displayed as an infrared thermal
signals converted from optical signals will exist in the structures image by a monitor. In reflection IR observation, the defects are
with and without bonding damage. Finally, these electrical signals shown as hot regions, while in transmission IR observation, the
are collected by a digital acquisition system and analyzed by a same defects are portrayed as cold regions. Reflection IR observa-
computer to determine whether the structure is damaged, as tions are generally used for FRP composite-RC structures, as shown
shown in Fig. 13. in Fig. 15. According to different heat sources, IR consists of passive
The theoretical model and experimental results indicated that IR and active IR. The advantage of passive IR is the reduction in the
the acoustic laser could be used to evaluate the location and size detection time and cost, but the disadvantage is the low efficiency
of the bonding damage in EB FRP composite-RC structures [50– and high requirements at the scene. In addition, its infrared image
52,115,127,128]. To ensure effectiveness, Chen et al. [127], who intensity depends not only on the temperature of the object but
analyzed the effects of operational parameters, revealed that these also on the infrared source attenuation caused by environmental
parameters had a qualitative relationship with the signal-to-noise absorption. Instead, active IR is not easily affected by the environ-
ratio (SNR) and the resonance frequency corresponding to the ment but requires thermal excitation.
defect size. Qiu et al. [51] also showed that the distance of sound IR can effectively detect the location, type, and size of bonding
excitation, sound pressure level, sound excitation incident angle, damage mechanisms [9,11,57,116] and can also efficiently monitor
and laser beam incident angle had a more substantial impact on the cracking location and width in the invisible concrete substrate
the sensitivity of the results. They suggested that the distance of of the EB technique. The detection capacity is primarily determined
sound excitation should not exceed 40 cm. Sound pressure level by the heating method, heating position, heating time interval, and
(SPL) that states the level of acoustic excitation is at least 100 dB. the layers of FRP laminates [57,116]. Combining active reflection IR
The sound excitation incident angle is recommended as 90°. The and grayscale images, the location and size of the damage of CFRP
laser beam incident angle should range from 30° to 60°. Simultane- strip-RC structures in a heated water environment can be output as
ously, detecting artificial defects in EB CFRP composite-concrete visible images [9]. Further, combined with theoretical analysis
slabs showed that the vibration amplitude at the defect regions from Silva et al. [130], the damage evolution shows that in the
was 100 times larger than that in the healthy areas [52]. Further- thermal environment, water molecules destroy the bonds in the
more, through the grid measurements of the debonding zones, polymer. The swelling stress causes matrix cracking and hydroly-
vibration amplitude images at different resonance frequencies sis, as well as the debonding of the fiber matrix, resulting in bond-
were generated, showing the resonance mode and boundary of ing surface damage [9]. This result indicates that NDTs are great
debonding. The results could be interpreted as follows: the healthy auxiliary tools for revealing the damage mechanism of FRP
regions were dark blue to represent a lower vibration amplitude. In composite-RC structures. Ghosh et al. [11] showed that the ther-
contrast, the areas with a higher vibration amplitude were sub- mograms of the active pulsed IR could characterize the invisible
jected to bonding damage, as shown in Fig. 14. Based on the plate damage of CFRP strip/sheet reinforcing RC bridges subjected to
vibration theory, an analytical model of the FRP laminate surface punching shear failure. However, when punching shear failure
vibration at the defect regions with acoustic excitation was estab- occurred at the edge of the beam, most cracks exceeded the region
lished to determine the damage size [52], as presented in Eq. (8). in the thermal imaging inspection. For this reason, it is essential to
The recorded time-domain waveform was mapped to the choose an appropriate field of view (FOV). In addition, attention
frequency-domain spectrum by fast Fourier transform. The peak should be paid to the matching of the minimum size of the damage
frequency f could be obtained by the acoustic-laser method, given and the spatial resolution of the infrared camera.
Eq. (9), and the frequency f could be substituted into Eq. (10) to cal- It is worth mentioning that some progress has been made in
culate the defect area A further. detecting the bonding damage depth in EB FRP composite-RC
structures. The damage severity can be quantified based on the
Dr4 w þ qh@ 2 w=@t2 ¼ 0 ð8Þ thermal intensity [11]. The depth of the bonding damage can be
evaluated using the thermal diffusion time [53], presented in Eq.
where the flexural rigidity of the FRP laminate is D = Eh3/[12(1v2)].
(11), and the results are shown in Fig. 16.
E, q, h, h are the elastic modulus, density, thickness and Poisson
ratio of the FRP laminate, respectively. The lateral displacement w pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
is a function of coordinates (x, y) and time t [129]. The resonance d¼ atmax =F 0 ð11Þ
frequency f of the FRP laminate is calculated according to Eq. (9)
where the thermal diffusivity of FRP in thickness direction a = k/qc.
[52].
q, k and c are the density, thermal conductivity, and heat capacity of
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
f ¼ k= 2pl
2
D=ðqhÞ ð9Þ the FRP, respectively. tmax is the time associated with the maximum
thermal contrast. The Fourier number F0 is calculated by a function
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi related to the thermal resistance of the bonding damage R*=Rk/
A ¼ k=ð2pf Þ D=ðqhÞ ð10Þ (atmax)0.5 [53] and the pulse Fourier number F0s = s/tmax. R is the dis-
continuous resistance of the bonding damage, and s is the pulse
where l is the side length of the FRP laminate. duration.
13
Fig. 14. Measure at the center of the GFRP-concrete specimen defect [52]: (a) specimen picture (in cm); (b) image constructed of the specimen at 3200 Hz.
Fig. 15. Reflection observation of the active IR in the EB technique.
Similarly, the normalization technique is used in the relation- where C0 1 is the coupled coefficient. C2 is the offset constant. l0 is
ship between temperature and time originating from each pixel the modified diffusion length, and fb is the blind frequency. The con-
for all thermal images to determine the depth of the damage and stants C0 1 = 8.255 and C0 1=2.988 can be obtained through the fb in
can effectively reduce the effects of uneven heating [54]. Pulse the known deep defect [59]. Finally, according to the blind fre-
phase thermography (PPT) provides reliable detection [58] and quency determined by all types of the defect, the defect depth can
presents the overall trend between the defect depth and blind fre- be determined.
quency. However, a longer heating time makes it challenging to
identify the blind frequency accurately. Hence, a model related to 3.8. Digital image correlation technology
the blind frequency and defect depth needs to be developed for a
specific thermal pulse duration. After the thermal excitation on As an optical-based noncontact technique, digital image corre-
the EB technique conducted by PPT, a calibration method is pro- lation (DIC) uses a camera to capture digital speckle images of
posed to modify the linear model of the blind frequency and defect the FRP composite-RC structure surface before and after deforma-
depth [59] to further obtain their relationship, given by Eq. (12). tion. DIC obtains the displacement of each point on the surface by
This contribution provides a calibration method suitable for the matching the corresponding image subregions in the digital
detection of the defect depth in the EB technique. In addition, speckle images before and after deformation, as shown in Fig. 17.
Tashan [57] reported that pulsed active IR can effectively detect Finally, the full strain field of the surface is measured. The function
the location and width of invisible concrete substrate cracking in C defines the relationship between the deformed image and the
the EB technique but cannot determine the depth of the cracks. reference image [131], presented in Eq. (13).
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
d / C 01 1=ðpf b Þ C 2 ¼ C 01 l0 C 2 ð12Þ X
n=2
Cðx; y; u; v Þ ¼ ðIðx þ i; y þ jÞ I ðx þ u þ i; y þ v þ jÞÞ
2
ð13Þ
i:j¼n=2
14
Fig. 16. Identification of the bonding defect depth of IR in the EB technique [53]: (a) the specimen; (b) thermal image at 40 s after heating; (c) depth estimation of the defect.
Fig. 17. Detection system of the DIC technology in the EB technique [60].
Fig. 18. DIC results and failure modes of the narrow straight anchor [12]: (a) el map; (b) et map; (c) failure picture.
15
where u and v are the displacements in the and y directions, when subjected to pulling force. (2) The longitudinal strain of a
respectively. The reference image I is a function of the pixel values narrow straight anchor was characterized by the concentration of
x + i and y + j, and the deformed image I* is a function of x + u + i and tensile strain at the bottom of the anchor. The strain level was close
y + v + j. The function C is the sum of the square differences between to the rupture strain of 1.3%. However, the longitudinal strain of
the deformed image and the reference image. n is the number of FRP sheets was close to 0 because tensile forces were transferred
tracking points. by the straight anchors, as shown in Fig. 18(a). The lateral strain
In the EB technique, obtaining the full strain field of the FRP sur- in Fig. 18(b) demonstrates the expected failure mode and location,
face through DIC is an important method to study its bonding and the final failure mode is shown in Fig. 18(c). This phenomenon
behavior and to determine the location of damage. DIC can accu- reveals that the lateral strain has a substantial effect on the fiber
rately measure the surface slip on the EB FRP laminates rupture of the narrow straight anchor [12] in the MA specimen
[61,74,132] and obtain an accurate bond-slip relationship. This ‘‘A100 when subjected to pulling force. (3) In the conventional EB
capacity contributes to the explanation of the failure mechanism. technique, debonding cracks developed from the loading end to
However, to detect or analyze the bonding damage information, the free end, eventually causing CD damage. During the initial
the full strain field on the FRP surface needs to be characterized loading, the area involved in deformation was minimal. Neverthe-
and interpreted. The multidirectional strain field on the surface less, when the debonding load was reached, the deformed area
of EB FRPs can be obtained by DIC [12], but it is difficult to achieve increased rapidly, as shown in the EB region in Fig. 19. This phe-
with conventional strain gages. nomenon illustrates the brittleness of debonding failure.
The following mechanisms were revealed based on the full However, after being reinforced with the EBROG technique, the
strain field from the DIC: (1) The strain of the overlapping joint bonding strength was substantially improved, and there was no
of the FRP sheets was relatively small, as there was more FRP mate- phenomenon of strain redistribution. As shown in the EBROG
rial in this area. However, the stress concentration at the end of region in Fig. 19, the FRP eventually broke near the loading end,
this area caused debonding failure [12] for the EB specimen ‘‘L” consistent with the strain field results obtained by DIC [13].
Fig. 19. Strain field corresponding to different load levels in the specimen [13]: (a) 0.25Pu; (b) 0.50Pu; (c) 0.70Pu; (d) 0.80Pu; (e) 0.98Pu; (f) Pu, where the left half represents the
EB specimen, and the right represents the EBROG specimen.
16
Fig. 20. Strain field in the loading direction of the specimen (in mm) [60].
Fig. 21. Fish-spine cracking mode of the NSM technique by DIC [61]: strain field cxy.
In addition, the strain field provided by DIC can characterize the the damage, including AD and CD. After qualitatively determining
location and width of the tensile cracks of the invisible concrete the degree of damage by the AE signal intensity, the size was fur-
substrate covered with FRP laminates [60], as shown in Fig. 20. ther evaluated by IR. Likewise, based on the AE hits, frequency, and
The sudden increase of strain indicated the location of each crack cluster analysis, all the damage modes with different types in the
in the concrete, and the crack width was continuously measured U-wrapped FRP sheet-RC structure during the flexural test could
through two tracking points near the crack. The detailed analytical be determined, including concrete shear cracking, fiber rupture,
calculations were found in [60]. and debonding on the beam side [76]. The damage information
Using the strain field measured by DIC, the stress mechanism of of the FRP material and FRP-concrete debonding in the FRP strip-
the NSM technique was also revealed [61]. The maximum strain RC structure were obtained by the duration and energy of the AE
appeared near the free end, and the adhesive created a stress trans- signals [134]. Finally, the full strain field of the FRP laminates
fer path in the diagonal compressive struts to transfer the stress to was determined by DIC to detect the cracking process, crack distri-
the concrete. As a result, a fish-spine cracking mode was generated, bution, and crack width of the invisible concrete substrate [76].
as shown in Fig. 21, eventually causing the debonding failure of the Besides, IR scanned the specimen to indicate the crack’s position.
adhesive concrete. Consequently, ultrasonic sensors were placed on the specimen to
make a more detailed assessment for the crack ’s depth [135].
3.9. Mixed nondestructive detection techniques However, in the previous application of the mixed techniques,
at least one kind of NDT is required to be in contact with the
The mixed technique of NDTs guarantees the advantages of detected structure. In particular, it is challenging to achieve contact
each NDT and minimizes the disadvantages. This hybrid technique detection for some unique structures, so remote detection is
can more comprehensively and accurately detect the location, needed. Yu et al. [115] compared and analyzed the artificial defect
type, size, and degree of damage of FRP composite-RC structures. inspection of EB FRP composite-concrete columns by acoustic laser
Verstrynge et al. [133] initially used AE monitoring to determine and imaging radar techniques. The qualitative suggestion was that
if the EB technique is subjected to bonding damage and to classify the location of the potential defects could be found by scanning
17
Table 2
D. Li, J. Zhou and J. Ou

Detection capacities of NDTs for the FRP composite-RC structure damage.
EB technique Embedded technique NSM technique MA EBROG Artificial defect

technique technique
AD CD Concrete DF SF Bar corrosion/ AD/adhesive material Anchor FRP Scattered debonding Concrete
cracking debond ing/concrete splitting/concrete rupture rupture cracking
cracking splitting
Ref.: [8, 9, 12, 31, Ref.:[13, Ref.: [9, Ref.: [7, 11, Ref.: Ref.: [14, 15, 22–24, Ref.: [61] Ref.: [12] Ref.: [13] Ref.: [25, 28–30, 33, 37– Ref.: [37,
41, 55, 56, 58, 31, 114, 60, 122] 21, 32, 42– [76, 27, 34–36] 40, 45–48, 50–54, 59, 57]
133] 133] 44] 135] 115]
PZT methods Ref: w w
[21–25,
27–30]
AE monitoring Ref: [7, 14, w w
31, 32,
122]
Ultrasonic testing Ref: [15, w w w
33–40]
Fiber optics sensor Ref: w
methods [41–44]
Microwave Ref: w
detection [45–48]
Acoustic laser Ref: w
detection [50–52]
IR Ref: [8, 9, w w w w w
11, 53–59]
DIC technology Ref: [12, w w w w w w
13, 60, 61]
Mixed AE and IR Ref: [133] w w
Mixed AE and DIC Ref: [76] w w
18
Mixed IR and Ref: [135] w

ultrasonic
Mixed acoustic Ref: [115] w
laser and
imaging radar
Mixed acoustic Ref: [50] w
laser and laser
reflection
where w represents NDTs could detect the damage location, represents the damage type, represents the damage size/shape, and represents the damage depth.
Construction and Building Materials 271 (2021) 121551

Table 3
Positive and negative characters of introduced NDTs to detect the FRP-concrete structure damage.
NDT Advantages Limitations

PZT methods (1) Low costs (1) Moisture-proof treatment required
(2) High voltage electrical conversion (2) Low output direct current response
(3) Great stability
(4) High-frequency bandwidth
(5) High SNR
(6) Easy operation
AE monitoring (1) Real-time (1) No ability to identify old damage
(2) No requirement on structural type (2) Vulnerable to electromechanical noise
Ultrasonic testing (1) Highly mobile (1) Calibration required for thickness and FRP materials change
(2) Damage visualization through L-scan and C-
scan
Fiber optics sensor (1) Better than conventional strain gauges (1) Accessible to damage in construction
methods
Microwave detection (1) Noncontact technique (1) High costs
(2) Damage visualization through L-scan and C- (2) Qualified inspectors required
scan
Acoustic laser detection (1) Noncontact technique (1) High costs
(2) Damage visualization (2) High-temperature degradation of the adhesives due to improper
operation
IR (1) Noncontact technique (1) High costs
(2) High efficiency (2) Vulnerable to environmental temperature
(3) Vast detection areas (3) Limited FRP thickness required
(4) Damage visualization (4) Thermal properties of FRP required
(5) Only surface/sub-surface detection
DIC technology (1) Noncontact technique (1) No ability to identify old damage
(2) Vast detection areas (2) Equipment location and speckle pattern required
(3) Damage visualization (3) Only surface/sub-surface detection
with imaging radar techniques. Then, the specific location and size degree can be initially determined. Then, the qualified inspector
of the defects could be inspected by analyzing the vibration of the can further utilize PZT methods/ultrasonic testing/microwave
bonding FRP system with an acoustic laser. detection/acoustic laser detection/IR/DIC techniques to locate the
Similarly, artificial defect inspection of the concrete slabs in the damage and to evaluate the size of the damage to determine
EB technique was achieved by using acoustic laser and laser reflec- whether the structure requires maintenance.
tion techniques [50]. A large grid size would cause the interaction In addition to damage similar to that of the EB/EBROG tech-
of the defects to be weakened during laser heating, and a small grid niques, the MA technique has the most potential anchor damage.
size would result in low efficiency. The structure was divided with An effective detection method is to obtain the strain field of
a 65 mm 65 mm grid, and the laser reflection technique was anchors through DIC [12]. For the embedded/NSM techniques,
used to locate single or multiple damage regions. In the single the potential damage can be initially detected [14,22,124,125],
damage region, an acoustic laser could be used to detect the dam- located [14,25,124,125], and classified [14] in real-time by arrang-
aged area directly. Instead, in the areas of multiple damage mech- ing PZT/AE/FBG sensors in advance. These damage mechanisms
anisms, the mesh was initially subdivided, and the previous cover the FRP laminate damage, bonding damage, and concrete
process was repeated. cracking. Furthermore, ultrasonic testing can evaluate the damage
length in the embedded technique [35], and DIC can obtain the
4. Capacities and characteristics of NDTs on FRP composite-RC strain field of the adhesives and concrete in the NSM technique
structures [61]. These techniques contribute to evaluating the type, location,
and size of damage further.
Table 2 summarizes the detection capability of NDTs for differ- The nature of the NDTs determines their ability to detect differ-
ent damage levels of FRP composite-RC structures in existing ent damage levels for different FRP composite-RC structures. All
works. The detection capability involves the type, location, size, NDTs have advantages and disadvantages, as summarized in
and depth of damage. The EB/EBROG techniques can experience Table 3. Overall, PZT methods have the benefits of low costs, high
potential bonding damage, concrete substrate cracking, and FRP voltage electrical conversion, good stability, a high-frequency
laminate damage under loads or harsh environments. The PZT bandwidth, high SNR, and easy operation. However, PZT methods
methods/AE monitoring/FBG methods can detect the damage and require moisture-proof treatment, and the output direct current
damage locations for the EB/EBROG techniques in real time (DC) response is unsatisfactory. AE monitoring is in real time but
[25,28,44,122]. By mathematically processing the signals, such as cannot identify old damage mechanisms. Ultrasonic testing is
performing the cluster analysis and principal component analysis, highly mobile and can visualize damage through L-scan and C-
AE monitoring can achieve damage classification [7,32]. To more scan techniques. However, calibration is required when FRP mate-
accurately and comprehensively grasp the damage information, rials are different or when the thickness of the structure varies.
PZT methods/ultrasonic testing/microwave detection/acoustic Fiber optics sensors have better performance than conventional
laser detection/IR/DIC can be applied to evaluate the location, type, strain gauges but are easily damaged in construction. Microwaves
and size of the damage [13,37,47,49,52,53]. Even the detection of can also visualize damage through L-scan and C-scan techniques,
the damage depth can be determined by IR/microwave detection but the detection cost is high, and a qualified inspector is required.
[47,53]. In summary, with the PZT/AE/FBG sensors arranged in Similarly, acoustic lasers can realize damage visualization. Still,
advance for the EB/EBROG techniques, the presence of structural the improper operation can easily result in high-temperature
damage can be evaluated in real-time, and the damage type and degradation of the adhesives, and the cost is higher compared to
19
that of other techniques. IR has the advantages of high efficiency 440.2R-17 [6]. In addition, evidence has shown that the other NDTs
and vast detection areas. However, IR is susceptible to environ- also have great potential for evaluating the damage of EB/EBROG
mental temperature effects and requires a limited FRP thickness; techniques. AE/PZT/FBG sensors can be arranged in advance to con-
in addition, the thermal properties of FRP materials need to be duct real-time monitoring. When damage occurs in the FRP
measured in advance. DIC has vast detection areas but cannot composite-RC structures, the monitoring system can capture the
detect old damage mechanisms (similar to AE). In addition, it is damage signals and initially use the mathematical algorithm to
necessary to determine the location of the equipment and to pro- determine the damage type and location; refer to
duce digital speckle images on the FRP laminate surface with spray [7,25,32,139,139]. In detail, potential NDTs and corresponding
paint or airbrush in advance for DIC detection. remedial measures include (1) the location and size of the damage
in the FRP laminates and the interface debonding, which can be
determined by IR/DIC/PZT methods/ultrasonic testing/microwave
5. Selections of NDTs and repair measures on FRP composite-RC
detection/acoustic laser detection; refer to
structures
[8,12,28,37,38,47,49,52,59]. Available remedies include resin injec-
tion for minor FRP laminate delamination or replacement technol-
According to the requirements of ACI 437R-19 [136], NDTs can
ogy for serious FRP laminate peeling and debonding. The lap splice
be used for the SHM of FRP composite-RC structures during the
length L is required and is calculated by Eq. (14) (suggested by GB
total life cycle to guide regular maintenance. The requirements
50608-2010 [137]). (2) FRP laminate strain should be monitored in
cover quality control in new and retrofitted construction, assess-
real time during service. Feasible NDTs are fiber optic and DIC tech-
ment of existing structures, and quality assurance of repair work.
nologies, which allow the surface strain of FRP laminates to be
The schematic layout of the SHM system for EB/EBROG reinforc-
obtained in real-time; refer to [12,13,41,44,123]. As a result, exter-
ing techniques of RC structures during the total life cycle is shown
nal prestressing/EB steel plates/EB angle steel/NSM techniques can
in Fig. 22(a). It covers the quality inspection during construction,
be used for repair when the tensile strength of FRP (the design
damage detection during the service stage, and potential remedial
value is shown in Table 4) reaches its design value, or the ultimate
measures. The examination should be conducted by a qualified
tensile strain of FRP exceeds 0.01 [137]. However, it should be
inspector who should be knowledgeable of FRP systems and
noted that shear failure before the flexural failure of the reinforced
trained in the installation of FRP systems. After inspection, an
structure should be avoided, and the ultimate limit state should be
assessment should focus on the proposal for repairing deficiencies
checked. If needed, corresponding shear reinforcement can be
and preventing further degradation [6]. However, in this process,
adopted. (3) Excessive concrete substrate cracks cause local stress
the maximum utilization of the residual life for damaged struc-
concentrations of the FRP laminates. This process can result in FRP
tures should be ensured to balance the structural safety and repair
laminate delamination, FRP-concrete debonding, and concrete
costs.
cover peeling. Such damage is brittle and triggers the premature
For the EB/EBROG techniques, before being in service, the mate-
failure of the structure. The most effective techniques to detect
rial properties such as the adhesion strength, resin cure, and cured
invisible concrete substrate cracking are IR and DIC techniques,
thickness need laboratory tests. NDTs on-site can evaluate the
referring to [57,60]. The damage visualization can clearly show
potential bonding defects within the FRP laminates or the inter-
the location and width of cracks, and microwaves can detect the
faces between FRP and concrete. The bonding defects are referred
crack depth. According to the suggestion of GB 50010-2010
to as ‘‘pre-existing defects” that contain delamination and air
[140], the steel bars placed in concrete may be subjected to corro-
voids, as presented in Table 1. Suggested detection techniques,
sion when the concrete crack width exceeds 0.2 mm. Potential
required to be capable of inspecting defects of 1300 mm2, consist
repair methods include resin injection or cement paste injection
of hammer sounding, ultrasonic testing, and IR, suggested by ACI
for concrete cracking and FRP anchoring for severe cracking. It is
440.2R-17 [6]. In addition, the effect of the inspected deficiencies
worth mentioning that during the repair process, fiber-cement
on the integrity and durability of FRP retrofitting RC structures
microcapsule technology [141] can replace cement paste. This
should be evaluated. ACI 440.2R-17 [6] recommends that defects
microcapsule technology can repair the existing concrete cracks.
smaller than 16,000 mm2 can be repaired by resin injection or
Furthermore, the microcracks will break and release the fiber-
FRP laminate replacement. However, defects larger than
cement in the microcapsule to achieve the self-repair of the con-
16,000 mm2 are considered to affect the performance of FRP retro-
crete microcracks for the reinforced structure in the future.
fitting RC structures. Cutting away the affected FRP laminates and
It should be emphasized that the live and dead loads other than
installing an overlapping FRP (that with equivalent layers) are
the dead weight of the structure should be removed for repairing
required. After repair, the FRP laminates should be redetected until
FRP composite-RC structures. Besides, the serviceability limit state
they are determined to be acceptable. Note that for the replace-
(excessive deflections and cracking) and ultimate limit state (fail-
ment, the lap splice length L is required and is calculated by Eq.
ure, stress rupture, and fatigue) should be checked. Engineering
(14) (suggested by GB 50608-2010 [137]).
principles, such as transformed section calculations using modular
L ¼ maxf150mm; tf t =sav e g ð14Þ ratios, could be used in assessing the serviceability, and the possi-
ble failure modes and the changed strain and stress in all materials
where ft is the tensile strength of the FRP material (MPa). t is the should be evaluated for the ultimate limit state [6]. In general, the
thickness of the single-layer FRP laminate (mm). save is the average design and repair process of the EB/EBROG techniques should meet
shear strength of the FRP laminate overlap regions, equal to the requirements of ACI 318 [142] in strength and serviceability.
4.0 MPa. However, the reliability index b (proposed by Szerszen and Nowak
During service, the RC structure strengthened with EB/EBROG [143] and given by Eq. (15)) can be introduced to consider the
techniques should be periodically inspected in terms of the FRP uncertainty inherent of FRP composites. The reliability index of
laminate strain, FRP laminate cracking, FRP laminate abrasions, FRP-concrete structures are recommended above 3.5 [6], and lower
debonding, and concrete substrate cracking. These deficiencies reliability is allowed for repaired specimens than for the new ones.
are referred to as the ‘‘damage stage” and presented in Table 1. qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Hammer sounding, ultrasonic testing, and IR techniques can be b ¼ ðmR mQ Þ= r2R þ r2Q ð15Þ
used to detect damage according to the recommendations by ACI
20
(a)
(b)
Fig. 22. Schematic layout of the SHM system and potential maintaining measures for FRP composite-RC structures during the total life cycle: (a) EB/EBROG/MA techniques;
(b) NSM/embedded techniques.
Table 4 capacity until the damaged FRP is repaired. The strengthening limit
Main mechanical parameters of FRP composites in the design [137]. is calculated according to Eq. (16) [6].
Material type Tensile strength Elastic modulus Rupture
(MPa) (MPa) strain (%) ðURn Þexisting P ð1:1SDL þ 0:75SLL Þnew ð16Þ
High strength CFRP 2,500 210 1.3 where U is the strength reduction factor. Rn is the nominal
High strength CFRP 3,000 210 1.4
strength. SDL is the dead load effects. SDL is the live load effects.
High strength CFRP 3,500 230 1.5
High elastic 2,900 390 0.7 In general, the structure is designed to have sufficient shear
modulus CFRP strength. The limit conditions are discussed first for FRP-concrete
GFRP 1,500 75 2.0 structures that need to be repaired in flexure. Typically, for the
GFRP 2,500 80 2.3 EB/EBROG specimens in flexure, the expected failure process is
AFRP 2,000 110 2.0
BFRP 2,000 90 2.0
when the FRP strain exceeds the design tensile strain (given by
Table 4), FRP rupture will occur. The steel bars will yield, and
finally, the concrete will be crushed in the compression regions
(maximum usable compressive strain is 0.003). Whether this type
where mR is the average resistance value. mQ is the average total of damage needs to be repaired can be controlled by the effective
load effect. rR is the standard deviation of resistance. rQ is the strain of FRP, efe, obtained from Eq. (17). Moreover, the damage
standard deviation of the total load effect. that should be avoided is an intermediate crack-induced FRP
At the same time, to avoid the failure caused by the structure debonding, where a limited debonding strain of FRP composites,
collapse or other FRP systems damage, the strengthening limit (ex- efd, is required to determine the repair necessity for preventing
isting structure strength) is required to maintain sufficient bearing debonding, given by Eq. (18) [6,144].
21

efe ¼ ecu df c =c ebi 6 efd ð17Þ life cycle is shown in Fig. 22(b). It also covers the quality inspection
during construction, damage detection during the service stage,
qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
ffi and potential remedial measures. AE/PZT/fiber optics sensors or
efd ¼ 0:41 f 0c = nEf tf 6 0:9efu ð18Þ
smart self-sensing FRP bars [124,125] can be arranged in the struc-
where efe is the effective strain of FRP at failure. ebi is the initial ture in advance to achieve real-time monitoring. The detection
strain of concrete substrate when installing FRP. ecu is the ultimate properties include the FRP strips/bars strain, concrete cracking,
axial strain of the concrete substrate. df is the effective depth of FRP and local slip, referred to as the ‘‘damage stage”, presented in
(mm). fc0 is the cylinder compressive strength (MPa). Ef is the tensile Table 1. In the embedded technique, damage can be monitored
elastic modulus of FRP (MPa). tf is the thickness of one-ply FRP by analyzing the damage signals, and the damage can be located
(mm). n is the plies number of FRP. efu is the design rupture strain of [14]. Simultaneously, smart self-sensing FRP bars can be used to
FRP. achieve real-time monitoring of the FRP bar strain [124,125].
In addition, in some cases, the EB/EBROG FRP-concrete struc- External prestressing/planting bars/NSM/EB steel plates/EB angle
tures may suffer from insufficient shear strength. The effective steel/EB FRP/EBROG/MA techniques can strengthen the structure
strain in FRP laminates can be used to determine whether the dam- when the FRP strips/bars reach the design strength. Noticeably,
aged EB/EBROG FRP-concrete structure subjected to the low shear the planting bars require that the original concrete grade is not
resistance has reached the level that requires repair [6]. The effec- lower than C20, and the adhesives are required to include modified
tive strain efe, that is, the maximum strain, represents the critical epoxy and modified vinyl ester, as suggested by GB 50367-2006
failure mode after considering all possible failure modes, and its [148]. For crack control, DIC/IR/microwaves can detect concrete
recommended value is as follows. For the full wrapping, the effec- cracking, and resin injection or cement paste injection are used
tive strain should be limited to 0.4% (less than 0.75efu) to avoid the for repair. Notably, unlike steel-reinforced concrete structures,
failure with loss of concrete aggregate interlock before the FRP the specimens reinforced with FRP bars/strips can tolerate larger
strain reach its ultimate strain [145]. Also, due to the observed crack widths due to more excellent corrosion resistance. In the
adhesive/cohesive debonding before the loss of concrete aggregate flexural design of the direct calculation for the crack width, an
interlock, the bond strength is used to control the effective strain allowable crack width has been suggested by many design criteria
recommended as Eq. (19) [146]. referring to different considerations. An allowable width of con-
crete cracking is 0.5 mm that only considers esthetics supported
efe ¼ jv efu 6 0:004 ð19Þ by the Japan Society of Civil Engineers [149]. For specimens sub-
where jv is the bond reduction coefficient, depending on the jected to adverse environments, an allowable width of 0.5 mm is
concrete strength, wrapping type, and the FRP stiffness, calculated suggested by CAN/CSA S6-06 [150]. Although the crack control in
by Eqs. (20)–(23) [147]. ACI 318 does not include FRP reinforcement specimens, an allow-
able width of 0.46 mm to 0.56 mm is required for steel bars, as sug-
jv ¼ k1 k2 Le = 11; 900efu 6 0:75 ð20Þ gested by ACI 224R-01 [151]. In addition to the treatment methods
similar to those of the embedded technique, adhesive material
0:58 splitting can be detected by IR/DIC techniques [61]. It can be
Le ¼ 23; 300= nt f Ef ð21Þ
repaired by resin injection for the NSM technique. Besides, the
0 2=3 presence and location of the debonding regions can be detected
k1 ¼ f c =27 ð22Þ by AE monitoring/PZT methods/fiber optics sensor methods. The
( repair measure is concrete replacement by removing the concrete
df v Le =df v ; for U - wrapping cover, reprocessing the adhesives, and using new concrete with a
k2 ¼ ð23Þ
df v 2Le =df v ; for side bonding higher grade. In addition, a limited debonding strain in FRP bars/
strips, efd, is required to prevent an intermediate crack-induced
where Le is the active bond length (mm). k1 and k2 are two modifi- FRP debonding for the embedded/NSM specimen in flexure. The
cation factors, respectively. dfv is the effective depth of FRP in shear efd is recommended less than 0.7 times the design rupture strain
(mm). of FRP, efu [6,152].
Simultaneously, the schematic layout of the SHM system in the
MA technique for reinforcing RC structures during the total life
cycle is shown in Fig. 22(a), covering quality inspection in con- 6. Future research needs
struction, damage detection in the service stage, and potential
remedial measures. Before the service stage, the detection methods By analyzing the existing works, it can be indicated that the dis-
and remedial measures of the bonding quality in the MA technique cussed NDTs can comprehensively detect the damage of FRP
can be compared to that of the EB/EBROG techniques. However, it composite-RC structures. However, it would seem that there is a
should be noted that the anchoring quality should be detected by need for further research in this area.
IR to ensure that the bonding strength between the anchors and At present, NDTs have focused on damage detection with EB/
FRP composites is adequate. During the service stage, the strength- EBROG/MA techniques. The damage can be fully detected in terms
ened RC structure should be periodically detected. The detection of location, type, size, and depth. Noticeably, an accurate grasp of
properties include the FRP laminate strain, FRP laminate cracking, the size and depth of damage will contribute to predicting the
FRP laminate abrasions, debonding, concrete substrate cracking, residual life of FRP composite-RC structures. However, the detec-
and anchor strain. These defects are referred to as the ‘‘damage tion of depth still needs to be improved. Meanwhile, detailed fea-
stage” and are presented in Table 1. Except for the anchor strain, sible solutions for the relationship between existing FRP
the detection methods and repair measures of the ‘‘damage stage” composite-RC structures and the selection of potential NDTs
can be compared to those of the EB/EBROG techniques. However, should be discussed in depth to achieve a balance between the eco-
when the anchors are unqualified after detection by DIC/fiber nomic benefits and detection requirements.
optics sensor methods, new anchors should be added for the MA
technique. (1) In the embedded/NSM techniques, the most useful NDTs
For the RC structures with embedded/NSM reinforcing tech- include AE monitoring, PZT methods, fiber optics sensor
niques, the schematic layout of the SHM system during the total methods, and ultrasonic testing. AE monitoring, the most
22
effective technique, can classify the damages by data pro- appropriate NDTs and potential maintenance measures are pre-
cessing, locate the damages based on the arrival time, and sented. The following conclusions are of importance in this review.
qualitatively determine the damage degree from the signal
intensity. However, the detection of the damage length (1) ‘‘Pre-existing defects” include bonding defects with the FRP
requires further research. The potential new technique that delamination and air voids in the RC structure retrofitted
combines smart FRP bars/strips with ferromagnetic materi- by the EB/EBROG/MA techniques and anchoring defects in
als and the self-magnetic flux leakage (SMFL) [153–156] the MA technique.
technique seems capable of determining the damage length. The pre-existing defect can be evaluated by hammer sounding/
(2) The correlation between the damage type, damage degree, ultrasonic testing/IR that require to be capable of inspecting
and signals obtained by AE monitoring/PZT methods are defects of 1,300 mm2.
worthy of further analysis and standardization for FRP Defects smaller than 16,000 mm2 can be repaired by resin injec-
composite-RC structures. After processing the signals, the tion/FRP laminate replacement, but defects larger than
preliminary conclusion on the damage type and degree 16,000 mm2 are repaired by cutting away the affected FRP lam-
determines whether the structure needs to be repaired. After inates and installing an overlapping FRP.
that, the NDTs with high efficiency and low costs can be (2) For the embedded technique, the ‘‘damage stage” covers
selected according to the preliminary damage information. local slip between FRP bar and concrete, FRP bar damage,
Finally, effective remedial measures can be chosen. and concrete cracking. However, the NSM technique in the
(3) Seawater sea-sand concrete (SSC) structures reinforced with ‘‘damage stage” could subject to adhesive material cracking
FRP bars have received much attention recently. More and other potential damage similar to that in the embedded
research on inspecting bonding damage in SSC structures technique.
enhanced with embedded and NSM techniques needs to be Smart self-sensing FRP bars/strips can be made of fiber optics
conducted to analyze the bonding behavior of SSC and FRP sensors to monitor the strain. A limited debonding strain, efd,
bars/strips. Fiber optics sensors and FRP bars/strips can be less than 0.7 times the design rupture strain of FRP, efu, in FRP
combined to produce smart self-sensing FRP bars/strips to bars/strips is required to prevent an intermediate crack-
self-monitor the displacement and local strain to determine induced FRP debonding for the embedded/NSM specimen in
the bonding mechanism further. Meanwhile, FBG sensors flexure. Suggested NDTs include DIC/IR/microwave detection
with a broader measuring range and the protection technol- for evaluating concrete cracking and adhesion splitting and AE
ogy of the sensors in the SSC environment need to be studied monitoring/PZT methods/fiber optics sensor methods for moni-
for SSC structures. toring the local slip (debonding).
(4) In acoustic laser detection, the laser power or acoustic exci- Repair measures include resin/cement paste injection for con-
tation intensity should be increased to improve the SNR. crete cracks (allowable width 0.5 mm). Resin injection repairs
However, excessive laser power could threaten the safety adhesion splitting. Concrete replacement deals with severe
of the inspectors and may cause the adhesive to deteriorate. local-slip areas. External prestressing/planting bars/NSM/EB/
Therefore, a laser vibrometer that balances high power and EBROG/MA/EB steel plates/EB angle steel techniques strengthen
inspector safety needs further development. Additionally, the structure when the FRP strength cannot meet the
the existing theoretical model is based on the assumption requirements.
that FRP laminates are isotropic. This assumption can result (3) For the EB technique retrofitting RC structures during ser-
in the difference between the actual resonance frequency vice, the side bonding and U-wrapping are prone to debond-
and analytical calculation. For this reason, the material ani- ing failure. The full wrapping is often subjected to FRP
sotropy of FRP laminates needs to be further considered. rupture accompanied by concrete substrate cracking.
(5) In addition, after evaluating the damage of FRP strengthen- Signals from the AE/PZT/FBG sensors can be mathematically
ing RC structures inspected by NDTs, the residual bearing processed to conduct the classification and preliminary locating
capacity of the structure should be further predicted to for the damage mechanisms. Furthermore, PZT methods/IR/DIC/
determine what remedial measures to implement. However, ultrasonic testing/acoustic laser detection/microwave detection
in practice, it is also essential to maximize the usable resid- are employed to evaluate the location and size of the damage.
ual life of damaged structures and to minimize the repair Microwaves and IR can detect damage depth.
costs. Furthermore, the serviceability limit state (excessive In general, a reliability index b of FRP-concrete structures are
deflections and cracking) and ultimate limit state (failure, recommended above 3.5 to consider the uncertainty inherent
stress rupture, and fatigue) should be checked for the of FRP composites. Remedial measures are resin/cement paste
repaired structure. These stages require more significant injection for concrete cracking (allowable width 0.2 mm) and
improvement and standardization. FRP laminate replacement for debonding damage larger than
16,000 mm2. Alternatively, the effective strain of FRP, efe, is
7. Conclusion remarks recommended as less than 0.9 times the design rupture strain
of FRP, efu, to prevent an intermediate crack-induced FRP
This review summarizes the damage levels of existing FRP debonding for specimens in flexure. For specimens in shear,
composite-RC structures based on the material performance of efe is recommended as less than 0.75efu to avoid the failure
FRP composites and the failure modes of FRP composite-RC struc- of concrete aggregate interlock loss before the FRP strain
tures. The damage levels of FRP composite-RC structures can be reaches its ultimate strain. The bond strength (bond reduction
divided into ‘‘pre-existing defect” and ‘‘damage stage”. Pre- coefficient jv) is used to control the effective strain to avoid
existing defects refer to the material defects or construction the adhesive/cohesive debonding.
defects, and the damage stage covers the damage of FRP (4) Due to the improved bonding strength for the EB technique,
composite-RC structures in service resulted from loadings or the MA and EBROG techniques mostly bear with FRP damage
adverse environments. Additionally, the detection methodology, in the ‘‘damage stage”, and the MA technique can also be
capabilities, advantages, and disadvantages of available NDTs are subjected to anchor damage. However, the debonding and
highlighted to inspect FRP composite-RC structure damage. More- concrete substrate cracking, whose inspection and repair
over, considering different damage levels, suggestions for selecting methods refer to those in the EB technique, may also occur.
23
FRP damage can be detected by monitoring the strain field of [13] Davood Mostofinejad, Mohsen Heydari Mofrad, Ardalan Hosseini, Hassan
Heydari Mofrad, Investigating the effects of concrete compressive strength,
the anchors and FRP composites by DIC/fiber optics sensor
CFRP thickness and groove depth on CFRP-concrete bond strength of EBROG
methods. joints, Constr. Build. Mater. 189 (2018) 323–337, https://doi.org/10.1016/
If the strength of FRP reaches its design rupture strain, external j.conbuildmat.2018.08.203.
prestressing/EB steel plates/EB angle steel/NSM techniques are [14] B. Di, J.K. Wang, H.T. Li, J.H. Zheng, Y. Zheng, G.B. Song, Investigation of
bonding behavior of FRP and steel bars in self-compacting concrete structures
used for rehabilitation. Adding new anchors can address the using acoustic emission method, Sensors 19 (1) (2019) 14, https://doi.org/
anchor damage issue. 10.3390/s19010159.
(5) Although the above indexes are recommended to determine [15] M.A. Climent, M. Miro, J. Carbajo, P. Poveda, G. de Vera, J. Ramis, Use of non-
linear ultrasonic techniques to detect cracks due to steel corrosion in
whether the FRP composite-RC structure needs repair after reinforced concrete structures, Materials 12 (5) (2019) 12, https://doi.org/
nondestructive evaluation, in practice, it is still worthy of 10.3390/ma12050813.
in-depth analysis on maximizing the usable residual life of [16] Bo Wang, J.G. Teng, Laura De Lorenzis, Li-Min Zhou, Jinping Ou, Wei Jin, K.T.
Lau, Strain monitoring of RC members strengthened with smart NSM FRP
damaged structures and minimizing the repair costs. bars, Constr. Build. Mater. 23 (4) (2009) 1698–1711, https://doi.org/10.1016/
j.conbuildmat.2008.07.027.
[17] Brad C. McCoy, Zakariya Bourara, Rudolf Seracino, Gregory W. Lucier, Anchor
bolt patterns for mechanically fastened FRP plates, J. Compos. Constr. 23 (4)
Declaration of Competing Interest (2019) 04019024, https://doi.org/10.1061/(ASCE)CC.1943-5614.0000951.
[18] Lawrence C. Bank, Dushyant Arora, Analysis of RC beams strengthened with
mechanically fastened FRP (MF-FRP) strips, Compos. Struct. 79 (2) (2007)
The authors declare that they have no known competing finan- 180–191, https://doi.org/10.1016/j.compstruct.2005.12.001.
cial interests or personal relationships that could have appeared [19] Lei Gao, Feng Zhang, Jiaqi Liu, Xiaorui Lu, Huarui Gao, Experimental and
numerical study on the interfacial bonding characteristics of FRP-to-concrete
to influence the work reported in this paper.
joints with mechanical fastening, Constr. Build. Mater. 199 (2019) 456–470,
https://doi.org/10.1016/j.conbuildmat.2018.12.033.
[20] Y.Y. Lim, Z.S. Tang, S.T. Smith, Piezoelectric based monitoring of structural
Acknowledgments adhesives curing: a novel experimental study, Smart Mater. Struct. 28 (2018),
https://doi.org/10.1088/1361-665X/aaeea4.
This work was supported by the financial support from the [21] Yi Wang, Xiaoda Li, Junhui Li, Qiang Wang, Bing Xu, Jun Deng, Debonding
damage detection of the CFRP-concrete interface based on piezoelectric
National Key Research and Development Program of China [grant ceramics by the wave-based method, Constr. Build. Mater. 210 (2019) 514–
numbers 2017YFC0703410], and the National Natural Science 524, https://doi.org/10.1016/j.conbuildmat.2019.03.042.
Foundation of China (NSFC) [grant numbers 51778104]. The [22] Tianyong Jiang, Qingzhao Kong, Devendra Patil, Zhoutao Luo, Linsheng Huo,
Gangbing Song, Detection of debonding between fiber reinforced polymer bar
authors are grateful to those who allowed us to use the valuable and concrete structure using piezoceramic transducers and wavelet packet
data, formulas, and figures. analysis, IEEE Sensors J. 17 (7) (2017) 1992–1998, https://doi.org/10.1109/
JSEN.2017.2660301.
[23] K. Xu, C.C. Ren, Q.S. Deng, Q.P. Jin, X.M. Chen, Real-time monitoring of bond
References slip between GFRP bar and concrete structure using piezoceramic transducer-
enabled active sensing, Sensors 18 (8) (2018) 2653, https://doi.org/10.3390/
s18082653.
[1] A. Siddika, M.A.A. Mamun, R. Alyousef, Y.H.M. Amran, Strengthening of
[24] Rachel Howser, Yashar Moslehy, Haichang Gu, Hemant Dhonde, Y L Mo,
reinforced concrete beams by using fiber-reinforced polymer composites: a
Ashraf Ayoub, Gangbing Song, Smart-aggregate-based damage detection of
review, J. Build. Eng. 25 (2019) 100798, https://doi.org/10.1016/j.
fiber-reinforced-polymer-strengthened columns under reversed cyclic
jobe.2019.100798.
loading, Smart Mater. Struct. 20 (7) (2011) 075014, https://doi.org/10.1088/
[2] C. Yuan, W. Chen, T.M. Pham, H. Hao, Bond behaviour between hybrid fiber
0964-1726/20/7/075014.
reinforced polymer sheets and concrete, Constr. Build. Mater. 210 (2019) 93–
[25] Rui Sun, Enrique Sevillano, Ricardo Perera, Debonding detection of FRP
110, https://doi.org/10.1016/j.conbuildmat.2019.03.082.
strengthened concrete beams by using impedance measurements and an
[3] L. Biolzi, C. Ghittoni, R. Fedele, G. Rosati, Experimental and theoretical issues
ensemble PSO adaptive spectral model, Compos. Struct. 125 (2015) 374–387,
in FRP-concrete bonding, Constr. Build. Mater. 41 (2013) 182–190, https://
https://doi.org/10.1016/j.compstruct.2015.02.011.
doi.org/10.1016/j.conbuildmat.2012.11.082.
[26] Y. Zheng, L.S. Huo, Y.X. Ye, D.D. Chen, H.W. Ma, G.B. Song, Monitoring of
[4] M.-W. Wei, J.-H. Xie, H. Zhang, J.-L. Li, Bond-slip behaviors of BFRP-to-
bending stiffness of BFRP reinforced concrete beams using piezoceramic
concrete interfaces exposed to wet/dry cycles in chloride environment,
transducer enabled active sensing, Smart Mater. Struct. (2020).
Compos. Struct. 219 (2019) 185–193, https://doi.org/10.1016/
[27] Weijie Li, Shuli Fan, Siu Chun Michael Ho, Jianchao Wu, Gangbing Song,
j.compstruct.2019.03.049.
Interfacial debonding detection in fiber-reinforced polymer rebar–reinforced
[5] S. Soares, J. Sena-Cruz, J.R. Cruz, P. Fernandes, Influence of surface preparation
concrete using electro-mechanical impedance technique, Struct. Health
method on the bond behavior of externally bonded CFRP reinforcements in
Monitor. 17 (3) (2018) 461–471, https://doi.org/10.1177/
concrete, Materials 12 (3) (2019) 20, https://doi.org/10.3390/ma12030414.
1475921717703053.
[6] ACI 440.2R-17, Guide for the design and construction of externally bonded
[28] T. Wandowski, P.H. Malinowski, W.M. Ostachowicz, Delamination detection
FRP systems for strengthening concrete structures, American Concrete
in CFRP panels using EMI method with temperature compensation, Compos.
Institute Report, Farmington Hills, Michigan, 2017.
Struct. 151 (9) (2016) 99–107, https://doi.org/10.1016/
[7] A. Nair, C.S. Cai, X. Kong, Acoustic emission pattern recognition in CFRP
retrofitted RC beams for failure mode identification, Compos. B Eng. 161
[29] Seunghee Park, Ju-Won Kim, Changgil Lee, Sun-Kyu Park, Impedance-based
(2019) 691–701, https://doi.org/10.1016/j.compositesb:2018.12.120.
wireless debonding condition monitoring of CFRP laminated concrete
[8] M.R. Valluzzi, E. Grinzato, C. Pellegrino, C. Modena, IR thermography for
structures, NDT & E Int. 44 (2) (2011) 232–238, https://doi.org/10.1016/j.
interface analysis of FRP laminates externally bonded to RC beams, Mater.
ndteint.2010.10.006.
Struct. 42 (1) (2009) 25–34.
[30] W. Ostachowicz, P. Kudela, M. Krawczuk, A. Zak, Guided Waves in Structures
[9] W.L. Lai, S.C. Kou, C.S. Poon, W.F. Tsang, C.C. Lai, Effects of elevated water
for SHM: The Time-Domain Spectral Element Method, John Wiley & Sons,
temperatures on interfacial delaminations, failure modes and shear strength
2011.
in externally-bonded CFRP-concrete beams using infrared thermography,
[31] Bahman Ghiassi, Els Verstrynge, Paulo B. Lourenço, Daniel V. Oliveira,
gray-scale images and direct shear test, Constr. Build. Mater. 23 (10) (2009)
Characterization of debonding in FRP-strengthened masonry using the
3152–3160, https://doi.org/10.1016/j.conbuildmat.2009.06.012.
acoustic emission technique, Eng. Struct. 66 (2014) 24–34, https://doi.org/
[10] F.S. Imani, A. Chen, J.F. Davalos, I. Ray, Model II fracture evaluation of CFRP-
10.1016/j.engstruct.2014.01.050.
concrete interface using accelerated aging test, earth and space, engineering,
[32] Sandeep Degala, Piervincenzo Rizzo, Karthik Ramanathan, Kent A. Harries,
science, construction, and operations in challenging, Environments 2010
Acoustic emission monitoring of CFRP reinforced concrete slabs, Constr.
(2010) 621–633, https://doi.org/10.1061/41096(366)62.
Build. Mater. 23 (5) (2009) 2016–2026, https://doi.org/10.1016/
[11] Kumar K. Ghosh, Vistasp M. Karbhari, Use of infrared thermography for
j.conbuildmat.2008.08.026.
quantitative non-destructive evaluation in FRP strengthened bridge systems,
[33] Emma La Malfa Ribolla, Mohsen Rezaee Hajidehi, Piervincenzo Rizzo,
Mater Struct. 44 (1) (2011) 169–185, https://doi.org/10.1617/s11527-010-
Giuseppe Fileccia Scimemi, Antonino Spada, Giuseppe Giambanco,
9617-5.
Ultrasonic inspection for the detection of debonding in CFRP-reinforced
[12] Enrique del Rey Castillo, Tom Allen, Richard Henry, Michael Griffith, Jason
concrete, Struct. Infrastruct. Eng. 14 (6) (2018) 807–816, https://doi.org/
Ingham, Digital image correlation (DIC) for measurement of strains and
10.1080/15732479.2017.1384843.
displacements in coarse, low volume-fraction FRP composites used in civil
[34] Ashutosh Sharma, Shruti Sharma, Sandeep Sharma, Abhijit Mukherjee,
infrastructure, Compos. Struct. 212 (2019) 43–57, https://doi.org/10.1016/
Ultrasonic guided waves for monitoring corrosion of FRP wrapped concrete
24
structures, Constr. Build. Mater. 96 (2015) 690–702, https://doi.org/10.1016/ [57] J. Tashan, R. Al-Mahaidi, Detection of cracks in concrete strengthened with
j.conbuildmat.2015.08.084. CFRP systems using infra-red thermography, Compos. B Eng. 64 (2014) 116–
[35] Rajeshwara Chary Sriramadasu, Sauvik Banerjee, Ye Lu, Detection and 125, https://doi.org/10.1016/j.compositesb:2014.04.011.
assessment of pitting corrosion in rebars using scattering of ultrasonic [58] J. Brown, S.H. Chittineni, Comparison of lock-in and pulse-phase
guided waves, NDT & E Int. 101 (2019) 53–61, https://doi.org/10.1016/j. thermography for defect characterization in FRP composites applied to
ndteint.2018.10.005. concrete, Int. Soc. Opt. Photon. (2015) 9485, https://doi.org/10.1117/
[36] Peng Du, Dongyu Xu, Shifeng Huang, Xin Cheng, Assessment of corrosion of 12.2177260.
reinforcing steel bars in concrete using embedded piezoelectric transducers [59] Nehemiah J. Mabry, Kara J. Peters, Rudolf Seracino, Depth detection of bond
based on ultrasonic wave, Constr. Build. Mater. 151 (2017) 925–930, https:// defects in multilayered externally bonded CFRP-to-concrete using pulse
doi.org/10.1016/j.conbuildmat.2017.06.153. phase thermography, J. Compos. Constr. 19 (6) (2015) 04015002, https://doi.
[37] J.C. Trovillion, V.F. Godinez-Azcuaga, R.D. Finlayson, A nondestructive org/10.1061/(ASCE)CC.1943-5614.0000551.
evaluation technique for fiber reinforced polymer (FRP) composites using [60] Mehdi Zomorodian, Guang Yang, Abdeldjelil Belarbi, Ashraf Ayoub, Cracking
acoustic guided waves (AGW), (2004). behavior and crack width predictions of FRP strengthened RC members under
[38] T. Kundu M. Ehsani K.I. Maslov D. Guo C-scan and L-scan generated images of tension, Eng. Struct. 125 (2016) 313–324, https://doi.org/10.1016/j.
the concrete/GFRP composite interface NDT & E International 32 2 1999 61 69 engstruct.2016.06.042.
10.1016/S0963-8695(98)00044-9 https://doi.org/10.1016/S0963-8695(98) [61] Julio Garzón-Roca, José M. Sena-Cruz, Pedro Fernandes, José Xavier, Effect of
00044-9. wet-dry cycles on the bond behaviour of concrete elements strengthened
[39] Kritsakorn Luangvilai, Wonsiri Punurai, Laurence J. Jacobs, Guided Lamb wave with NSM CFRP laminate strips, Compos. Struct. 132 (2015) 331–340, https://
propagation in composite plate/concrete component, J. Eng. Mech. 128 (12) doi.org/10.1016/j.compstruct.2015.05.053.
(2002) 1337–1341, https://doi.org/10.1061/(ASCE)0733-9399(2002)128:12 [62] Brahim Benmokrane, Fareed Elgabbas, Ehab A. Ahmed, Patrice Cousin,
(1337). Characterization and comparative durability study of glass/vinylester,
[40] F. Bastianini, A. Di Tommaso, G. Pascale, Ultrasonic non-destructive basalt/vinylester, and basalt/epoxy FRP bars, J. Compos. Constr. 19 (6)
assessment of bonding defects in composite structural strengthenings, (2015) 04015008, https://doi.org/10.1061/(ASCE)CC.1943-5614.0000564.
Compos. Struct. 53 (4) (2001) 463–467, https://doi.org/10.1016/S0263-8223 [63] Mônica Garcez, Leila Meneghetti, Luiz Carlos da Silva Filho, Structural
(01)00058-7. performance of rc beams poststrengthened with carbon, aramid, and glass
[41] K. Joshi, J.B. Frketic, D. Olawale, T. Dickens, Damage monitoring of CFRP FRP systems, J. Compos. Constr. 12 (5) (2008) 522–530, https://doi.org/
retrofit using optical fiber sensors, Sensors and Smart Structures Technologies 10.1061/(ASCE)1090-0268(2008)12:5(522).
for Civil, Mechanical, and Aerospace Systems 2015, International Society for [64] Ahmed El Refai, Mohamed-Amine Ammar, Radhouane Masmoudi, Bond
Optics and Photonics (2015) 9435. https://doi.org/10.1117/12.2086984. performance of basalt fiber-reinforced polymer bars to concrete, J. Compos.
[42] B. Arun Sundaram, K. Kesavan, S. Parivallal, A.K. Farvaze Ahmed, K. Constr. 19 (3) (2015) 04014050, https://doi.org/10.1061/(ASCE)CC.1943-
Ravisankar, Monitoring of FRP strengthened concrete structures using FBG 5614.0000487.
sensors, Procedia Eng. 14 (2011) 1549–1556, https://doi.org/10.1016/j. [65] Davood Mostofinejad, Khaled Sanginabadi, Mohammad Reza Eftekhar, Effects
proeng.2011.07.195. of coarse aggregate volume on CFRP-concrete bond strength and behavior,
[43] K. Kesavan, K. Ravisankar, R. Senthil, B. Arun Sundaram, S. Parivallal, FBG Constr. Build. Mater. 198 (2019) 42–57, https://doi.org/10.1016/
sensor technology to interfacial strain measurement in CFRP-strengthened j.conbuildmat.2018.11.188.
concrete beam, Exp. Tech. 39 (5) (2015) 21–29, https://doi.org/10.1111/ [66] A. Zhou, R.Y. Qin, L. Feo, R. Penna, D. Lau, Investigation on interfacial defect
j.1747-1567.2012.00858.x. criticality of FRP-bonded concrete beams, Compos. Pt. B-Eng. 113 (2017) 80-
[44] K. Kesavan, K. Ravisankar, R. Senthil, A.K. Farvaze Ahmed, Experimental 90. https://doi.org/0.1016/j.compositesb.2016.12.055.
studies on performance of reinforced concrete beam strengthened with CFRP [67] L. Koutas, T.C. Triantafillou, Use of anchors in shear strengthening of
under cyclic loading using FBG array, Measurement 46 (10) (2013) 3855– reinforced concrete T-beams with FRP, J. Compos. Constr. 17 (1) (2013)
3862, https://doi.org/10.1016/j.measurement.2013.07.031. 101–107, https://doi.org/10.1061/(ASCE)CC.1943-5614.0000316.
[45] Maria Q. Feng, Franco De Flaviis, Yoo Jin Kim, Use of microwaves for damage [68] Francesca Ceroni, Marisa Pecce, Evaluation of bond strength in concrete
detection of fiber reinforced polymer-wrapped concrete structures, J. Eng. elements externally reinforced with CFRP sheets and anchoring devices, J.
Mech. 128 (2) (2002) 172–183, https://doi.org/10.1061/(ASCE)0733-9399 Compos. Constr. 14 (5) (2010) 521–530, https://doi.org/10.1061/(ASCE)
(2002)128:2(172). CC.1943-5614.0000118.
[46] D. Hughes, M. Kazemi, K. Marler, R. Zoughi, J. Myers, A. Nanni, Microwave [69] A. Ghobarah, M.N. Ghorbel, S.E. Chidiac, Upgrading torsional resistance of
detection of delaminations between fiber reinforced polymer (FRP) reinforced concrete beams using fiber-reinforced polymer, J. Compos. Constr.
composite and hardened cement paste, Am. Inst. Phys. Melville 615 (1) 6 (4) (2002) 257-263. https://doi.org/0.1061/(ASCE)1090-0268(2002)6:4
(2002) 512–519, https://doi.org/10.1063/1.1472841. (257).
[47] Y.J. Kim, F. De Flaviis, L. Jofre, M.Q. Feng, Microwave-based NDE of FRP- [70] J.P. Firmo, M.R.T. Arruda, J.R. Correia, I.C. Rosa, Three-dimensional finite
jacketed concrete structures, Soc. Advan. Mater. Proce. Eng. (2001) 439–450. element modelling of the fire behaviour of insulated RC beams strengthened
[48] Marc D. Navagato, Ram M. Narayanan, Microwave imaging of multilayered with EBR and NSM CFRP strips, Compos. Struct. 183 (2018) 124-136.
structures using ultrawideband noise signals, NDT & E Int. 104 (2019) 19–33, https://doi.org/0.1016/j.compstruct.2017.01.082.
https://doi.org/10.1016/j.ndteint.2019.02.009. [71] Hui Huang, Wen-Wei Wang, Jian-Guo Dai, John C. Brigham, Fatigue behavior
[49] B. Akuthota, D. Hughes, R. Zoughi, J. Myers, A. Nanni, Near-field microwave of reinforced concrete beams strengthened with externally bonded
detection of disbond in carbon fiber reinforced polymer composites used for prestressed CFRP Sheets, J. Compos. Constr. 21 (3) (2017) 04016108,
strengthening cement-based structures and disbond repair verification, J. https://doi.org/10.1061/(ASCE)CC.1943-5614.0000766.
Mater. Civ. Eng. 16 (6) (2004) 540–546, https://doi.org/10.1061/(ASCE)0899- [72] Wensu Chen, Thong M. Pham, Henry Sichembe, Li Chen, Hong Hao,
1561(2004)16:6(540). Experimental study of flexural behaviour of RC beams strengthened by
[50] Qiwen Qiu, Denvid Lau, A novel approach for near-surface defect detection in longitudinal and U-shaped basalt FRP sheet, Compos. B Eng. 134 (2018) 114–
FRP-bonded concrete systems using laser reflection and acoustic-laser 126, https://doi.org/10.1016/j.compositesb:2017.09.053.
techniques, Constr. Build. Mater. 141 (2017) 553–564, https://doi.org/ [73] R.A. Hawileh, H.A. Rasheed, J.A. Abdalla, A.K. Al-Tamimi, Behavior of
10.1016/j.conbuildmat.2017.03.024. reinforced concrete beams strengthened with externally bonded hybrid
[51] Q.W. Qiu, D. Lau, The sensitivity of acoustic-laser technique for detecting the fiber reinforced polymer systems, Mater. Design 53 (1) (2014) 972–982,
defects in CFRP-bonded concrete systems, J. Nondestruct. Eval. 35 (2) (2016) https://doi.org/10.1016/j.matdes.2013.07.087.
10, https://doi.org/10.1007/s10921-016-0351-y. [74] Niloufar Moshiri, Amir Tajmir-Riahi, Davood Mostofinejad, Christoph
[52] J.G. Chen, R.W. Haupt, O. Buyukozturk, The acoustic-laser vibrometry Czaderski, Masoud Motavalli, Experimental and analytical study on CFRP
technique for the noncontact detection of discontinuities in fiber reinforced strips-to-concrete bonded joints using EBROG method, Compos. B Eng. 158
polymer-retrofitted concrete, Mater. Eval. 72 (10) (2014) 1305–1313. (2019) 437–447, https://doi.org/10.1016/j.compositesb:2018.09.046.
[53] Frédéric Taillade, Marc Quiertant, Karim Benzarti, Christophe Aubagnac, [75] C. Barris, L. Torres, I. Vilanova, C. Miàs, M. Llorens, Experimental study on
Shearography and pulsed stimulated infrared thermography applied to a crack width and crack spacing for Glass-FRP reinforced concrete beams, Eng.
nondestructive evaluation of FRP strengthening systems bonded on concrete Struct. 131 (2017) 231–242, https://doi.org/10.1016/j.engstruct.2016.11.007.
structures, Constr. Build. Mater. 25 (2) (2011) 568–574, https://doi.org/ [76] Daniel A. Pohoryles, Jose Melo, Tiziana Rossetto, Matthias Fabian, Colum
10.1016/j.conbuildmat.2010.02.019. McCague, Katerina Stavrianaki, Ben Lishman, Ben Sargeant, Use of DIC and AE
[54] Jeff R. Brown, H.R. Hamilton, Quantitative infrared thermography for monitoring effective strain and debonding in FRP and FRCM-retrofitted RC
inspection for FRP applied to concrete using single pixel analysis, Constr. beams, J. Compos. Constr. 21 (1) (2017) 04016057, https://doi.org/10.1061/
Build. Mater. 38 (2013) 1292–1302, https://doi.org/10.1016/ (ASCE)CC.1943-5614.0000715.
j.conbuildmat.2009.12.016. [77] J.E. Woods, D.T. Lau, X. Bao, W. Li, Measuring strain fields in FRP strengthened
[55] W.L. Lai, S.C. Kou, C.S. Poon, W.F. Tsang, K.K. Lee, A durability study of RC shear walls using a distributed fiber optic sensor, Eng. Struct. 152 (2017)
externally bonded FRP-concrete beams via full-field infrared thermography 359–369, https://doi.org/10.1016/j.engstruct.2017.09.034.
(IRT) and quasi-static shear test, Constr. Build. Mater. 40 (2013) 481–491, [78] G. Wu, Z.S. Wu, Z.T. Lu, Y.B. Ando, Structural performance of concrete
https://doi.org/10.1016/j.conbuildmat.2012.07.105. confined with hybrid FRP composites, J. Reinf. Plast. Compos. 27 (12) (2008)
[56] W.L. Lai, K.K. Lee, S.C. Kou, C.S. Poon, W.F. Tsang, A study of full-field debond 1323–1348, https://doi.org/10.1177/0731684407084989.
behaviour and durability of CFRP-concrete composite beams by pulsed [79] Jian-Guo Dai, Yu-Lei Bai, J.G. Teng, Behavior and modeling of concrete
infrared thermography (IRT), NDT & E Int. 52 (2012) 112–121, https://doi.org/ confined with FRP composites of large deformability, J. Compos. Constr. 15 (6)
10.1016/j.ndteint.2012.07.001. (2011) 963–973, https://doi.org/10.1061/(ASCE)CC.1943-5614.0000230.
25
[80] Paulo Silva Lobo, Pedro Faustino, Mariana Jesus, Rui Marreiros, Design model [103] Stephen V. Grelle, Lesley H. Sneed, Review of anchorage systems for
of concrete for circular columns confined with AFRP, Compos. Struct. 200 externally bonded FRP laminates, Int J Concr Struct Mater 7 (1) (2013) 17–
(2018) 69–78, https://doi.org/10.1016/j.compstruct.2018.05.094. 33, https://doi.org/10.1007/s40069-013-0029-0.
[81] Emmanuel Chinaka, Mehdi Shokouhian, Monique Head, Steve Efe, An [104] A.J. Lamanna, L.C. Bank, D.W. Scott, Flexural strengthening of reinforced
experimental investigation of bond strength of AFRP bars with self- concrete beams using fasteners and fiber-reinforced polymer strips, ACI
consolidating concrete, Civ. Eng. Des. 1 (5-6) (2019) 148–160, https://doi. Struct. J. 98 (3) (2001) 368–376.
org/10.1002/cend.201900013. [105] Yu-Fei Wu, Yue Huang, Hybrid bonding of FRP to reinforced concrete
[82] Bo Wu, Shuo Zhang, Fentao Liu, Ting Gan, Effects of salt solution on structures, J. Compos. Constr. 12 (3) (2008) 266–273, https://doi.org/10.1061/
mechanical behaviors of aramid fiber–reinforced polymer (AFRP) sheets and (ASCE)1090-0268(2008)12:3(266).
AFRP-to-concrete joints, Adv. Struct. Eng. 19 (12) (2016) 1855–1872, https:// [106] Mário R.F. Coelho, José M. Sena-Cruz, Luís A.C. Neves, A review on the bond
doi.org/10.1177/1369433216649389. behavior of FRP NSM systems in concrete, Constr. Build. Mater. 93 (2015)
[83] Farid Abed, Abdul Rahman Alhafiz, Effect of basalt fibers on the flexural 1157–1169, https://doi.org/10.1016/j.conbuildmat.2015.05.010.
behavior of concrete beams reinforced with BFRP bars, Compos. Struct. 215 [107] Standards Australia, Design handbook for RC structures retrofitted with FRP
(2019) 23–34, https://doi.org/10.1016/j.compstruct.2019.02.050. and metal plates: beams and slabs, Standards Australia GPO Box 476, Sydney,
[84] Shao-Fei Jiang, Xinggui Zeng, Sheng Shen, Xincheng Xu, Experimental studies NSW 2001, Australia, 2008.
on the seismic behavior of earthquake-damaged circular bridge columns [108] Deric John Oehlers, Matthew Haskett, Chengqing Wu, Rudolf Seracino,
repaired by using combination of near-surface-mounted BFRP bars with Embedding NSM FRP plates for improved IC debonding resistance, J.
external BFRP sheets jacketing, Eng. Struct. 106 (2016) 317–331, https://doi. Compos. Constr. 12 (6) (2008) 635–642, https://doi.org/10.1061/(ASCE)
org/10.1016/j.engstruct.2015.10.037. 1090-0268(2008)12:6(635).
[85] Yanlei Wang, Yongshuai Wang, Baolin Wan, Baoguo Han, Gaochuang Cai, [109] Shehab M. Soliman, Ehab El-Salakawy, Brahim Benmokrane, Bond
Ruijuan Chang, Strain and damage self-sensing of basalt fiber reinforced performance of near-surface-mounted FRP bars, J. Compos. Constr. 15 (1)
polymer laminates fabricated with carbon nanofibers/epoxy composites (2011) 103–111, https://doi.org/10.1061/(ASCE)CC.1943-5614.0000150.
under tension, Compos. A Appl. Sci. Manuf. 113 (2018) 40–52, https://doi. [110] Rudolf Seracino, Nicola M. Jones, M.S. Ali, Mark W. Page, Deric J. Oehlers,
org/10.1016/j.compositesa:2018.07.017. Bond strength of near-surface mounted FRP strip-to-concrete joints, J.
[86] T Loutas, V Kostopoulos, Health monitoring of carbon/carbon, woven Compos. Constr. 11 (4) (2007) 401–409, https://doi.org/10.1061/(ASCE)
reinforced composites. Damage assessment by using advanced signal 1090-0268(2007)11:4(401).
processing techniques. Part I: acoustic emission monitoring and damage [111] A. Bilotta, F. Ceroni, M. Di Ludovico, E. Nigro, M. Pecce, G. Manfredi, Bond
mechanisms evolution, Compos. Sci. Technol. 69 (2) (2009) 265–272, https:// efficiency of EBR and NSM FRP systems for strengthening concrete members,
doi.org/10.1016/j.compscitech.2008.07.020. J. Compos. Constr. 15 (5) (2011) 757–772, https://doi.org/10.1061/(ASCE)
[87] Z.Q. Dong, G. Wu, J.L. Lian, Experimental study on the durability of FRP bars CC.1943-5614.0000204.
reinforced concrete beams in simulated ocean environment, Sci. Eng. [112] H. Sohn, C.R. Farrar, F.M. Hemez, J.J. Czarnecki, A review of structural health
Compos. Mater. 25 (6) (2018) 1123–1134, https://doi.org/10.1515/secm- review of structural health monitoring literature: 1996-2001, Los Alamos
2017-0237. National Laboratory (2003) 1-7.
[88] Zhiqiang Dong, Gang Wu, Xiao-Ling Zhao, Hong Zhu, Jin-Long Lian, Durability [113] T.Y. Jiang, Y. Hong, J.B. Zheng, L. Wang, H.C. Gu, Crack detection of FRP-
test on the flexural performance of seawater sea-sand concrete beams reinforced concrete beam using embedded piezoceramic smart aggregates,
completely reinforced with FRP bars, Constr. Build. Mater. 192 (2018) 671– Sensors 19 (9) (2019) 1979, https://doi.org/10.3390/s19091979.
682, https://doi.org/10.1016/j.conbuildmat.2018.10.166. [114] Ahmed M. Mahmoud, Hussein H. Ammar, Osama M. Mukdadi, Indrajit Ray,
[89] Zhiqiang Dong, Gang Wu, Xiao-Ling Zhao, Hong Zhu, Jinlong Lian, Bond Fatemeh S. Imani, An Chen, Julio F. Davalos, Non-destructive ultrasonic
durability of steel-FRP composite bars embedded in seawater sea-sand evaluation of CFRP–concrete specimens subjected to accelerated aging
concrete under constant bending and shearing stress, Constr. Build. Mater. conditions, NDT & E Int. 43 (7) (2010) 635–641, https://doi.org/10.1016/j.
192 (2018) 808–817, https://doi.org/10.1016/j.conbuildmat.2018.10.154. ndteint.2010.06.008.
[90] Fei Yan, Zhibin Lin, Mijia Yang, Bond mechanism and bond strength of GFRP [115] Tzuyang Yu, Tin Kei Cheng, Ao Zhou, Denvid Lau, Remote defect detection of
bars to concrete: a review, Compos. B Eng. 98 (2016) 56–69, https://doi.org/ FRP-bonded concrete system using acoustic-laser and imaging radar
10.1016/j.compositesb:2016.04.068. techniques, Constr. Build. Mater. 109 (2016) 146–155, https://doi.org/
[91] M. Pepe, H. Mazaheripour, J. Barros, J. Sena-Cruz, E. Martinelli, Numerical 10.1016/j.conbuildmat.2015.12.113.
calibration of bond law for GFRP bars embedded in steel fibre-reinforced self- [116] Jawdat Tashan, Riadh Al-mahaidi, Investigation of the parameters that
compacting concrete, Compos. B Eng. 50 (2013) 403–412, https://doi.org/ influence the accuracy of bond defect detection in CFRP bonded specimens
10.1016/j.compositesb:2013.03.006. using IR thermography, Compos. Struct. 94 (2) (2012) 519–531, https://doi.
[92] Julio F. Davalos, Yi Chen, Indrajit Ray, Effect of FRP bar degradation on org/10.1016/j.compstruct.2011.08.017.
interface bond with high strength concrete, Cem. Concr. Compos. 30 (8) [117] C. Liang, F.P. Sun, C.A. Rogers, Coupled electro-mechanical analysis of
(2008) 722–730, https://doi.org/10.1016/j.cemconcomp.2008.05.006. adaptive material systems-determination of the actuator power
[93] Najia Saleh, Ashraf Ashour, Dennis Lam, Therese Sheehan, Experimental consumption and system energy transfer, J. Intell. Mater. Syst. Struct. 8 (4)
investigation of bond behaviour of two common GFRP bar types in high – (1997) 335–343, https://doi.org/10.1177/1045389X9700800406.
Strength concrete, Constr. Build. Mater. 201 (2019) 610–622, https://doi.org/ [118] 7Xiao-Yan Sun, Jian-Guo Dai, Hai-Long Wang, Wei-Wei Dong, Jie Wang,
10.1016/j.conbuildmat.2018.12.175. Decision support system for optimizing the maintenance of RC girder bridge
[94] Esmaeil Pournamazian Najafabadi, Asghar Vatani Oskouei, Mohammad superstructures in consideration of the carbon footprint, J. Bridge Eng. 20
Houshmand Khaneghahi, Parham Shoaei, Togay Ozbakkaloglu, The tensile (12) (2015) 04015022, https://doi.org/10.1061/(ASCE)BE.1943-
performance of FRP bars embedded in concrete under elevated temperatures, 5592.0000774.
Constr. Build. Mater. 211 (2019) 1138–1152, https://doi.org/10.1016/ [119] Enrique Sevillano, Rui Sun, Alberto Gil, Ricardo Perera, Interfacial crack-
j.conbuildmat.2019.03.239. induced debonding identification in FRP-strengthened RC beams from PZT
[95] Xingyu Gu, Bin Yu, Ming Wu, Experimental study of the bond performance signatures using hierarchical clustering analysis, Compos. B Eng. 87 (2016)
and mechanical response of GFRP reinforced concrete, Constr. Build. Mater. 322–335, https://doi.org/10.1016/j.compositesb:2015.09.006.
114 (2016) 407–415, https://doi.org/10.1016/j.conbuildmat.2016.03.211. [120] Rui Sun, Enrique Sevillano, Ricardo Perera, Identification of intermediate
[96] Manjola Caro, Yaser Jemaa, Samir Dirar, Andrew Quinn, Bond performance of debonding damage in FRP-strengthened RC beams based on a multi-objective
deep embedment FRP bars epoxy-bonded into concrete, Eng. Struct. 147 updating approach and PZT sensors, Compos. B Eng. 109 (2017) 248–258,
(2017) 448–457, https://doi.org/10.1016/j.engstruct.2017.05.069. https://doi.org/10.1016/j.compositesb:2016.10.060.
[97] H.K. Kang, J. Howell, S. Kim, J.L. Dong, A state-of-the-art review on debonding [121] Bin Xu, Ting Zhang, Gangbing Song, Haichang Gu, Active interface debonding
failures of FRP laminates externally adhered to concrete, Int. J. Concr. Struct. detection of a concrete-filled steel tube with piezoelectric technologies using
Mater. 6 (2) (2012) 123–134. wavelet packet analysis, Mech. Syst. Sig. Process. 36 (1) (2013) 7–17, https://
[98] Amir Mofidi, Omar Chaallal, Shear strengthening of RC beams with EB FRP: doi.org/10.1016/j.ymssp.2011.07.029.
influencing factors and conceptual debonding model, J. Compos. Constr. 15 [122] A. Grazzini, G. Lacidogna, Fatigue analysis of FRP strengthened masonry by
(1) (2011) 62–74, https://doi.org/10.1061/(ASCE)CC.1943-5614.0000153. acoustic emission monitoring, Key Engineering Materials, Trans Tech Publ,
[99] Nagaraj Eshwar, Timothy James Ibell, Antonio Nanni, Effectiveness of CFRP 817 (2019) 594-601. https://doi.org/10.4028/www.scientific.net/
strengthening on curved soffit RC beams, Adv. Struct. Eng. 8 (1) (2005) 55–68, KEM.817.594.
https://doi.org/10.1260/1369433053749607. [123] Kin-tak Lau, Libo Yuan, Li-min Zhou, Jingshen Wu, Chung-ho Woo, Strain
[100] Dionysios A. Bournas, Alberto Pavese, Walid Tizani, Tensile capacity of FRP monitoring in FRP laminates and concrete beams using FBG sensors, Compos.
anchors in connecting FRP and TRM sheets to concrete, Eng. Struct. 82 (2015) Struct. 51 (1) (2001) 9–20, https://doi.org/10.1016/S0263-8223(00)00094-5.
72–81, https://doi.org/10.1016/j.engstruct.2014.10.031. [124] Alexander Kalamkarov, Gobinda Saha, Srujan Rokkam, John Newhook,
[101] Francesca Ceroni, Marisa Pecce, Stjin Matthys, Luc Taerwe, Debonding Anastasis Georgiades, Strain and deformation monitoring in infrastructure
strength and anchorage devices for reinforced concrete elements using embedded smart FRP reinforcements, Compos. B Eng. 36 (5) (2005)
strengthened with FRP sheets, Compos. B Eng. 39 (3) (2008) 429–441, 455–467, https://doi.org/10.1016/j.compositesb:2004.12.003.
https://doi.org/10.1016/j.compositesb:2007.05.002. [125] Y.S. Tang, Z.S. Wu, Distributed long-gauge optical fiber sensors based self-
[102] David A. Petty, Paul J. Barr, G. Parry Osborn, Marvin W. Halling, Travis R. sensing FRP bar for concrete structure, Sensors 16 (3) (2016) 18, https://doi.
Brackus, Carbon fiber shear retrofit of forty-two-year-Old AASHTO I-shaped org/10.3390/s16030286.
girders, J. Compos. Constr. 15 (5) (2011) 773–781, https://doi.org/10.1061/ [126] M. Feng, G. Roqueta, L. Jofre, Non-destructive evaluation (NDE) of
(ASCE)CC.1943-5614.0000208. composites: microwave techniques, non-destructive evaluation (NDE), of
26
Polymer Matrix Compos. (2013) 574–616, https://doi.org/10.1533/ [140] GB 50010-2010, Code for Design of Concrete Structures, Ministry of Housing
9780857093554.4.574. and Urban-Rural Development of the People’s Republic of China, Beijing,
[127] Justin G. Chen, Robert W. Haupt, Oral Buyukozturk, Operational and defect 2010.
parameters concerning the acoustic-laser vibrometry method for FRP- [141] F. Xing, Z. Zhou, Z. Huang, Mechanical properties of microcapsule-basalt
reinforced concrete, NDT & E Int. 71 (2015) 43–53, https://doi.org/10.1016/ fiber/cement composites, Acta. Mater. Compos. Sin. 31 (1) (2014) 133–139,
j.ndteint.2014.12.006. https://doi.org/10.1007/s12595-012-0057-9.
[128] D. Lau, One dimensional predictive model of interfacial stiffness in FRP- [142] ACI 318-11, Building Code Requirements for Structural Concrete and
bonded concrete system using acoustic-laser technique, Nondestructive Commentary, American Concrete Institute Report, Farmington Hills,
Characterization for Composite Materials, Aerospace Engineering, Civil Michigan, 2011.
Infrastructure, and Homeland Security 2013, International Society for [143] M.S. Maria, S.N. Andrzej, Calibration of design code for buildings (ACI 318):
Optics and Photonics (2013) 8694. https://doi.org/10.1117/12.2010251. part 2—reliability analysis and resistance factors, ACI Struct. J. 100 (3) (2003).
[129] A. Leissa, Vibration of Plates, Scientific and technical information division, [144] J.G. Teng, S.T. Smith, J. Yao, J.F. Chen, Intermediate crack-induced debonding
National Aeronautics and Space Administration, US GPO, Washington, DC in RC beams and slabs, Constr. Build. Mater. 17 (6-7) (2003) 447–462, https://
(1969). doi.org/10.1016/S0950-0618(03)00043-6.
[130] Manuel A.G. Silva, Hugo Biscaia, Degradation of bond between FRP and RC [145] F. Seible, M.J.N. Priestley, G.M. Calvi, Seismic Design and Retrofit of Bridges,
beams, Compos. Struct. 85 (2) (2008) 164–174, https://doi.org/10.1016/ 1996.
j.compstruct.2007.10.014. [146] T.C. Triantafillou, Shear strengthening of reinforced concrete beams using
[131] T. Jin, N.S. Goo, S.-C. Woo, H.C. Park, Use of a digital image correlation epoxy-bonded FRP composites, ACI Struct. J. 95 (2) (1998) 107–115.
technique for measuring the material properties of beetle wing, J. Bionic Eng. [147] Ahmed Khalifa, William J. Gold, Antonio Nanni, M.I. Abdel Aziz, Contribution
6 (3) (2009) 224–231. of externally bonded FRP to shear capacity of RC flexural members, J. Compos.
[132] Majid Ghorbani, Davood Mostofinejad, Ardalan Hosseini, Bond Behavior of Constr. 2 (4) (1998) 195–202, https://doi.org/10.1061/(ASCE)1090-0268
CFRP sheets attached to concrete through EBR and EBROG joints subject to (1998)2:4(195).
mixed-mode I/II loading, J. Compos. Constr. 21 (5) (2017) 04017034, https:// [148] GB 50367-2013, Design Code for Strengthening Concrete Structure, Ministry
doi.org/10.1061/(ASCE)CC.1943-5614.0000816. of Housing and Urban-Rural Development of the People’s Republic of China,
[133] E. Verstrynge, M. Wevers, B. Ghiassi, P.B. Lourenço, Debonding damage Beijing, 2013.
analysis in composite-masonry strengthening systems with polymer- and [149] Japan Society of Civil Engineers, Recommendation for Design and
mortar-based matrix by means of the acoustic emission technique, Smart Construction of Concrete Structures Using Continuous Fiber Reinforcing
Mater. Struct. 25 (1) (2016) 015009, https://doi.org/10.1088/0964-1726/25/ Materials, Concrete Engineering Series No. 23, Japan, 1997b.
1/015009. [150] CAN/CSA-S6-06, Canadian Highway Bridge Design Code, Canadian Standards
[134] Kolluru V.L. Subramaniam, Michel Ghosn, Mohamad Ali-Ahmad, Influence of Association, 2006.
variation in the local interface fracture properties on shear debonding of [151] ACI 224R-01, Control of Cracking in Concrete Structures, American Concrete
CFRP composite from concrete, J. Adhes. Sci. Technol. 31 (19-20) (2017) Institute Report, Farmington Hills, Michigan, 2002.
2202–2218, https://doi.org/10.1080/01694243.2016.1261505. [152] Vincenzo Bianco, Giorgio Monti, J.A.O. Barros, Design formula to evaluate the
[135] D.G. Aggelis, E.Z. Kordatos, D.V. Soulioti, T.E. Matikas, Combined use of NSM FRP strips shear strength contribution to a RC beam, Compos. B Eng. 56
thermography and ultrasound for the characterization of subsurface cracks in (2014) 960–971, https://doi.org/10.1016/j.compositesb:2013.09.001.
concrete, Constr. Build. Mater. 24 (10) (2010) 1888–1897, https://doi.org/ [153] J.T. Zhou, J.L. Qiu, Y.X. Zhou, Y. Zhou, R.C. Xia, Experimental study on residual
10.1016/j.conbuildmat.2010.04.014. bending strength of corroded reinforced concrete beam based on
[136] ACI 437R-19, Strength Evaluation of Existing Concrete Buildings, American micromagnetic sensor, Sensors 18 (2018) 2635. https://doi.org/10.3390/
Concrete Institute Report, Farmington Hills, Michigan, 2019. s18082635.
[137] GB 50608-2010, Technical Code for Infrastructure Application of FRP [154] R.C. Xia, J.T. Zhou, H. Zhang, L. Liao, R.Q. Zhao, Z.Y. Zhang, Quantitative study
Composites, Ministry of Housing and Urban-Rural Development of the on corrosion of steel strands based on self-magnetic flux leakage, Sensors 18
People’s Republic of China, Beijing, 2010. (5) (2018) 1396, https://doi.org/10.3390/s18051396.
[138] Qinghua Han, Jie Xu, Alberto Carpinteri, Giuseppe Lacidogna, Localization of [155] H. Zhang, J.T. Zhou, R.Q. Zhao, L. Liao, M. Yang, R.C. Xia, Experimental study on
acoustic emission sources in structural health monitoring of masonry bridge: detection of rebar corrosion in concrete based on metal magnetic memory,
localization of acoustic emission sources in masonry bridge, Struct. Control Int. J. Robot. Automat. 32 (2017), https://doi.org/10.2316/
Health Monit. 22 (2) (2015) 314–329, https://doi.org/10.1002/stc.1675. Journal.206.2017.5.206-5082.
[139] M.C. Ge, Analysis of source location algorithms part II: iterative methods, J. [156] H. Zhang, L. Liao, R.Q. Zhao, J.T. Zhou, M. Yang, R.C. Xia, The non-destructive
Acoustic Emission 21 (2003) 29–51. test of steel corrosion in reinforced concrete bridges using a micro-magnetic
sensor, Sensors 16 (9) (2016) 1439, https://doi.org/10.3390/s16091439.
27

23. Citation 4

Uploaded by

Copyright:

Available Formats

23. Citation 4

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

23. Citation 4

Uploaded by

Copyright:

Available Formats

Construction and Building Materials 271 (2021) 121551

Contents lists available at ScienceDirect

Construction and Building Materials

Damage, nondestructive evaluation and rehabilitation

3.5. Microwave detection. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

1. Introduction microwave detection [45–49], acoustic laser detection [50–52],

Fig. 5. Statistics on damage modes of referred FRP composite-concrete structures.

FRP application Pre-existing defect Damage stage Failure mode

Fig. 6. Experimental setup of the WP and EMI techniques.

EMI technique. The application of the temperature compensation

can be used to detect the damage in FRP composite-RC structures

comprehensive analysis of microwave applications can be found in 3.7. Infrared thermography

Fig. 15. Reflection observation of the active IR in the EB technique.

D. Li, J. Zhou and J. Ou

EB technique Embedded technique NSM technique MA EBROG Artificial defect

Mixed IR and Ref: [135] w

Construction and Building Materials 271 (2021) 121551

NDT Advantages Limitations

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.