energies

Review
Review on the Damage and Fault Diagnosis of Wind Turbine
Blades in the Germination Stage
Xiaowen Song 1,2 , Zhitai Xing 3,4 , Yan Jia 3,4, *, Xiaojuan Song 1,2 , Chang Cai 5 , Yinan Zhang 5 , Zekun Wang 5 ,
Jicai Guo 1,2 and Qingan Li 5, *

1 College of Mechanical Engineering, Inner Mongolia University of Technology, Hohhot 010051, China
2 Inner Mongolia Key Laboratory of Special Service Intelligent Robotics, Hohhot 010051, China
3 College of Energy and Power Engineering, Inner Mongolia University of Technology, Hohhot 010051, China
4 Key Laboratory of Wind Energy and Solar Energy Utilization Technology, Ministry of Education,
Inner Mongolia University of Technology, Hohhot 010051, China
5 Institute of Engineering Thermophysics, Chinese Academy of Sciences, Beijing 100190, China
* Correspondence: [email protected] (Y.J.); [email protected] (Q.L.)

Abstract: In recent years, wind turbines have shown a maximization trend. However, most of the
wind turbine blades operate in areas with a relatively poor natural environment. The stability, safety,
and reliability of blade operation are facing many challenges. Therefore, it is of great significance to
monitor the structural health of wind turbine blades to avoid the failure of wind turbine outages and
reduce maintenance costs. This paper reviews the commonly observed types of damage and damage
detection methods of wind turbine blades. First of all, a comprehensive summary of the common
embryonic damage, leading edge erosion, micro-cracking, fiber defects, and coating defects damage.
Secondly, three fault diagnosis methods of wind turbine blades, including nondestructive testing
(NDT), supervisory control and data acquisition (SCADA), and vibration signal-based fault diagnosis,
are introduced. The working principles, advantages, disadvantages, and development status of
nondestructive testing methods are analyzed and summarized. Finally, the future development trend
Citation: Song, X.; Xing, Z.; Jia, Y.; of wind turbine blade detection and diagnosis technology is discussed. This paper can guide the use
Song, X.; Cai, C.; Zhang, Y.; Wang, Z.; of technical means in the actual detection of wind turbine blades. In addition, the research prospect
Guo, J.; Li, Q. Review on the Damage of fault diagnosis technology can be understood.
and Fault Diagnosis of Wind Turbine
Blades in the Germination Stage. Keywords: wind turbine blades; the germination stage of damage; fault diagnosis; non-destructive in-
Energies 2022, 15, 7492. https:// spection
doi.org/10.3390/en15207492

Academic Editor: Davide Astolfi

Received: 6 September 2022

1. Introduction
Accepted: 5 October 2022
Published: 12 October 2022
At present, as primary energy such as oil, coal, and natural gas are gradually consumed,
the energy crisis and environmental problems are becoming increasingly serious [1]. To
Publisher’s Note: MDPI stays neutral
balance and eliminate carbon dioxide emissions, many economies have set targets for
with regard to jurisdictional claims in
carbon peaking and carbon neutrality, including 124 countries committing to achieving
published maps and institutional affil-
carbon neutrality by 2050 or 2060 as of February 2021 [2]. As one of the most commonly
iations.
used renewable and clean energies, the development and utilization of wind energy play a
key role in carbon neutralization [3–5]. By 2021, the world’s newly installed capacity of
wind energy reached 88 GW, and by 2050, wind and solar power are expected to account
Copyright: © 2022 by the authors.
for about 70 percent of the global power supply [6,7]. As the installed capacity of wind
Licensee MDPI, Basel, Switzerland. power increases, the number of wind turbine failures also shows an increasing trend. As
This article is an open access article shown in Figure 1, by September 2021, there had been a total of 3093 accidents reported [8].
distributed under the terms and As one of the key components of the wind turbine, the wind turbine blade (WTB) has
conditions of the Creative Commons growing requirements for a safe, stable and sustainable operation environment with its size
Attribution (CC BY) license (https:// increasing [9,10]. Onshore wind turbine operational and maintenance costs account for
creativecommons.org/licenses/by/ 5–10% of power generation costs, even 25–30% at sea [11]. Therefore, reducing operation
4.0/). and maintenance costs are very important for the healthy development of wind turbines.

Energies 2022, 15, 7492. https://doi.org/10.3390/en15207492 https://www.mdpi.com/journal/energies

Energies 2022, 15, x FOR PEER REVIEW 2 of 16

operational and maintenance costs account for 5–10% of power generation costs, even 25–30%
Energies 2022, 15, 7492 at sea [11]. Therefore, reducing operation and maintenance costs are very important2 of for
17
the healthy development of wind turbines.

Figure Statisticsofofwind
Figure1.1.Statistics windturbine
turbineaccidents
accidents recorded
recorded since
since 1996
1996 [8].[8].

Themain
The mainreasons
reasonsfor forthe
thedamage
damagefailure
failureof ofwind
windturbines
turbinescome comefrom from twotwo aspects.
aspects. On On
theone
the onehand,
hand,ititisisaapoor
poornatural
naturalenvironmental
environmentalfactor,factor,and andon onthetheother
otherhand,
hand,ititisisaahuman
human
factor. For
factor. For example,
example,improper
improper operation
operationin in the
the manufacture,
manufacture, transportation,
transportation,process, process,and and
maintenance of wind turbines causes damage to them. Identically,
maintenance of wind turbines causes damage to them. Identically, wind turbine blades are wind turbine blades
are usually
usually locatedlocated
in areasin areas
with awith a relatively
relatively poor natural
poor natural environment,
environment, which iswhich
prone to is prone
damage to
damage and failure. The germination stage of damage needs to
and failure. The germination stage of damage needs to be paid enough attention to. The slow be paid enough attention
to. The slow of
accumulation accumulation
micro-damage of micro-damage
may cause serious maydamage,
cause serious
which damage, which the
directly affects directly
effi-
ciency of wind turbine power generation. More seriously, it will lead to wind turbinelead
affects the efficiency of wind turbine power generation. More seriously, it will shut-to
wind or
down turbine shutdown
even collapse, or evenincollapse,
resulting resulting
catastrophic in catastrophic
consequences [12]. Theconsequences
detection and[12]. The
diagno-
detection and diagnosis of wind turbine blades conducted in time
sis of wind turbine blades conducted in time at the germination stage of microscopic damage at the germination stage
of microscopic damage can lower the probability of failure and prolong the service life of
can lower the probability of failure and prolong the service life of wind turbines. Therefore, in
wind turbines. Therefore, in order to ensure the normal operation of wind turbine blades,
order to ensure the normal operation of wind turbine blades, it is of great significance to realize
it is of great significance to realize the structural health monitoring (SHM) of the wind
the structural health monitoring (SHM) of the wind turbine blade monitoring system. The
turbine blade monitoring system. The most successful application of SHM technology is
most successful application of SHM technology is condition monitoring (CM) for rotating ma-
condition monitoring (CM) for rotating machinery [13]. CM has become an important part
chinery [13]. CM has become an important part of the wind turbine system. However, SHM
of the wind turbine system. However, SHM technology for blade health monitoring is still
technology for blade health monitoring is still in the development stage.
in the development stage.
There are three types of diagnosis methods of wind turbine blade damage: the non-de-
There are three types of diagnosis methods of wind turbine blade damage: the non-
structive testing (NDT) technology of blades, the supervisory control and data acquisition
destructive testing (NDT) technology of blades, the supervisory control and data acquisition
(SCADA)-based method, and the vibration signal-based method. In recent years, NDT has
(SCADA)-based method, and the vibration signal-based method. In recent years, NDT has
been widely used in wind turbine blade fault diagnosis [14,15]. Li et al. [16] reviewed the latest
been widely used in wind turbine blade fault diagnosis [14,15]. Li et al. [16] reviewed the
research progress
latest research on windonturbine
progress blade damage
wind turbine detection,detection,
blade damage mainly including four detection
mainly including four
methods, transmittance function, wave propagation, impedance,
detection methods, transmittance function, wave propagation, impedance, and and vibration method. In
vibration
further research, Yang et al. [17] comprehensively analyzed the
method. In further research, Yang et al. [17] comprehensively analyzed the advantages andadvantages and limitations
of common damage
limitations of common types and thetypes
damage NDTand of wind
the NDTturbine blades
of wind and predicted
turbine blades and the predicted
develop-
ment trend of WTB detection. Furthermore, Fausto et al. [18]
the development trend of WTB detection. Furthermore, Fausto et al. [18] summarized summarized the application of
nondestructive testing methods and monitoring technology in
the application of nondestructive testing methods and monitoring technology in SHM. In SHM. In order to avoid the
failure
order to of avoid
wind turbine
the failureblades, in addition
of wind turbinetoblades,
the application
in addition of NDT,
to the the conditionofmoni-
application NDT,
toring system (CMS)
the condition is alsosystem
monitoring installed on the
(CMS) windinstalled
is also turbine on blades.
the windThe real-time signal col-
turbine blades. The
lected by sensors
real-time and electronic
signal collected equipment
by sensors can be combined
and electronic equipment with canthebeSCADA
combined parameters
with the
to accurately
SCADA monitor the
parameters operation state
to accurately of thethe
monitor wind turbinestate
operation bladeof andtherealize
wind the intelligent
turbine blade
and digital monitoring of the wind turbine blade [19]. Igba
and realize the intelligent and digital monitoring of the wind turbine blade [19]. et al. [20] proposed a commonIgba
approach
et al. [20]toproposed
CMS for wind turbines
a common (WTs) toto
approach meet
CMS theforrequirements
wind turbines of lifetime
(WTs) engineering
to meet the
services (TES). of
requirements While the intelligent
lifetime engineering monitoring of CMS
services (TES). is realized,
While accurate monitoring
the intelligent fault identifi- of
cation
CMS isbecomes
realized,the research
accurate focus.
fault Koitz et al.
identification [21] developed
becomes the research an efficient
focus. Koitzand et accurate
al. [21]
developed an efficient and accurate fault identification program for improving WTs fault
detection, location, and maintenance to solve the problems of monitoring error and false
alarms in CMS. The fault diagnosis of wind turbine blades based on vibration signals can
Energies 2022, 15, 7492 3 of 17

not only be applied to wind turbine blades but also the entire wind turbine. Studying noise
removal is also the focus of scholars’ research. Aiming at the problem of noise interference
in vibration signal acquisition, Liu et al. [22] reviewed methods of noise removal by wind
turbine health condition monitoring (HCM). Based on the vibration signal analysis method,
the wind turbine was monitored, and the noise interference was eliminated to protect
the safe operation of the wind turbine. With the continuous development of science and
technology, new fault diagnosis technology is continuously applied to the damage detection
of wind turbine blades, which is of great significance to the development trend and prospect
of WTB fault diagnosis technology.
The content of this article is given as follows. Section 2 introduces the material struc-
ture of the wind turbine blade and the damage of leading edge erosion, micro-cracking,
fiber defects, and coating defects in the germination stage of the blade. In Section 3,
three fault diagnosis and detection methods of wind turbine blades are reviewed, among
which the nondestructive detection methods include acoustic emission, ultrasonic, strain
measurement, thermal imaging, machine vision, and other methods. The basic princi-
ples, advantages, and disadvantages of each method are summarized. Sections 4 and 5
discuss and summarize the development trend of future wind turbine blade damage
detection methods.

2. Microscopic Damage of Blades

The blade structure is constructed by the composite structure, forming six different
carbon fiber laminate regions [23], and modern wind turbine blades are usually com-
posed of fiber-reinforced composites. Most wind turbine blades are composed of glass
fiber/epoxy resin, glass fiber/polyester, wood/epoxy resin, or carbon fiber/epoxy resin
composites [24]. The material and structural performance of the blade directly affect the
stiffness and strength of the blade. With the rapid development of material technology
and the optimization of the structure, the probability of blade damage gradually decreases.
The risk of wind turbine blade damage largely depends on the surrounding environment
and climatic conditions. Lightning stroke, fatigue load, surface ice on the blades, and the
exposure of the blades to air particles, may cause the germination stage of damage such as
leading edge erosion and micro-cracking of blades [9]. In addition, offshore wind turbines
are faced with the influence of wave climate change and marine corrosion on the blade
tension distribution range and fatigue damage. The corrosion environment will reduce the
performance of the blade composite material, resulting in the fatigue failure of the blade
and accelerated damage [25,26]. Therefore, choosing the appropriate installation site and
timely detection in the germination stage of damage are the most effective measures to
reduce damage. Damage tolerance design is a modern fatigue fracture control method
developed and gradually applied in the 1970 s to ensure the safety of essential components
with cracks or possible crack damage [27]. The blade design currently considers a certain
margin and adopts the damage tolerance design. However, once installed, there will be
many uncontrollable factors that will still lead to damage initiation. In addition, there
is surface corrosion damage (due to bad weather), which is very difficult to avoid, so
regular testing is needed to detect damage in time. The specific causes and hazards of the
germination stage of damage are shown in the following sections.

2.1. Leading Edge Erosion

The leading edge erosion (as shown in Figure 2) of wind turbine blades is mainly
caused by the impacts of small particles in the air, such as dust, rain, hail, etc. Leading edge
erosion occurs mostly at the tip of the blade, and over time, erosion also occurs inside the
blade structure [28]. The leading edge erosion will lead to an increase in blade roughness,
thereby increasing the resistance of blade rotation. As a result of that, it will reduce the
efficiency of wind turbine power generation and affect economic benefits.
Energies 2022,15,
Energies2022, 15,7492
x FOR PEER REVIEW 44 of
of 17
16

Energies 2022, 15, x FOR PEER REVIEW 4 of 16

Energies 2022, 15, x FOR PEER REVIEW 4 of 16

Figure2.2.Photo
Figure Photoof
ofthe
theleading
leadingedge
edgeerosion
erosion[29].
[29].
Figure 2. Photo of the leading edge erosion [29].
2.2.Micro-Cracking
2.2. Micro-Cracking
2.2. Micro-Cracking
Figure 2. Photonormal
Under of the leading edge conditions,
working erosion [29]. a long time of wind load will cause fatigue degra-
Under normal working conditions, a long time of wind load will cause fatigue degrada-
dation
tion Under
of the
of the normal
wind wind working
turbine
turbine conditions,
bladeblade
duedue to acyclic
long
to time
cyclic of wind
andload
loading,
loading, the will
and the cause
bladeblade fatigue
material
material degra-
will will experi-
experience
2.2. Micro-Cracking
dation of the wind turbine blade due to cyclic loading, and the blade material will experi-
ence micro-damage
micro-damage cracks,cracks, as shown
as shown in Figure
in Figure 3. The
3. The slow
slow accumulationofofmicro-cracking
accumulation micro-cracking
enceUnder
will micro-damage
easily normal
causeblade cracks,
working
blade asconditions,
shown
fracture and in hit
Figure
a long 3. Theofslow
time
thenearby
nearby wind
windaccumulation
load of micro-cracking
will cause
turbine, fatigue
causing degra-accidents.
serious
will easily cause fracture and hit the wind turbine, causing serious accidents.
will easily
dation cause
of the wind blade fracture
turbine bladeandduehittothe nearby
cyclic windand
loading, turbine, causing
the blade serious
material accidents.
will experi-
Inaddition,
In addition, wind turbine turbine blades
blades operating
operating inin
extremely
extremely cold weather
cold weather maymay have brittle frac-
In addition,
ence wind turbine
micro-damage cracks,blades
as shownoperating
in Figurein extremely
3. The slowcold weather may
accumulation ofhave brittle frac- brittle
micro-cracking
have
tures
fractures inside
inside thethematerials andand thetheicing of blades andandsensors. Under worse conditions, mi-
tureseasily
will inside cause bladematerials
the materials and the
fracture andicing oficing
hit the blades
nearbyofand
blades
windsensors.
turbine, sensors.
Under Under
worse
causing worse
conditions,
serious conditions,
mi-
accidents.
cro-cracking
micro-cracking
cro-cracking
In addition, wind will develop
will
will developdeveloprapidly,
rapidly,
turbine blades rapidly, leading
leading to
operating leadingto
inblade blade
to failure
blade
failure
extremely and
failure
andweather
cold even
and
even themay the
even
collapse collapse
have ofthe of
collapse
the entire
brittle frac- the entire
of the
wind
entire turbine
wind
windinside
tures turbine the [16].
turbine
[16]. [16].and the icing of blades and sensors. Under worse conditions, mi-
materials
cro-cracking will develop rapidly, leading to blade failure and even the collapse of the entire
wind turbine [16].

Figure 3. Photo of the micro-cracking on the blade [30].

Figure 3. Photo of the micro-cracking on the blade [30].
Figure 3. Photo of the micro-cracking on the blade [30].
2.3. Fiber
Figure Defects
3. Photo of the micro-cracking on the blade [30].
2.3. Fiber Defects
The fiber
2.3. Fiber defect damage (see Figure 4) of wind turbine blades may occur in the process
Defects
2.3. Themanufacturing
Fiber
of blade fiber defect damage
Defects or may (see Figure
be caused by4)load
of wind
[31]. Inturbine
addition, blades may occur
fiber defects exist in
in the
the process
The
ofinterior
blade fiber defect
manufacturing damage
or(see (see
may Figure
be caused 4) of
by wind
load turbine
[31]. blades
In may
addition, occur
fiber in the process
defects exist
Theoffiber
the defect
composites,
damagewhich may cause
Figure cracks
4) of wind and other
turbine damage
blades may over time.
occur in Some
the schol-
process
inof
of
blade
arsthe
have
blade
manufacturing
interior
studied of
thethe
manufacturing
or may
composites,
mechanism
or may be
bewhich
of fiber caused
causeddefects. by [31].
may load
cause
Elhajjar
by load et[31].
Inal.
Infound
cracks
[32]
addition,
addition,
and that
fiber
fiberdamage
other
tensile
defects
defects
loadin
exist could exist in
over
the
the
time.
interior
interior
Some of
cause scholars
common the composites,
fiber
havedefects
of the composites, which
studied which
in carbon may
maymechanism
the cause cause
epoxycracks cracks
laminates
and
of and
through
other
fiber other
damage
defects. damage
the test of carbon
over time.
Elhajjar over time.
fibers.
etSome
al. Some
schol-
[32] schol-
found that
arshave
ars
tensilehaveloadstudied
studied thethe
could mechanism
mechanism
cause common offiber
of fiberfiber defects.
defects. Elhajjar
Elhajjar
defects in et et al.
al. [32]
carbon [32]that
found
epoxy found that
tensile
laminates tensile
load couldload
through thecould
test
cause
cause common
common
of carbon fibers. fiber
fiber defects
defects in in carbon
carbon epoxyepoxy laminates
laminates through through
the test the
of test
carbon of carbon
fibers. fibers.

Figure 4. Photo of the blade fiber defects [33].

Figure 4. Photo of the blade fiber defects [33].

Figure 4. Photo of the blade fiber defects [33].
Figure 4. Photo of the blade fiber defects [33].
Energies 2022, 15, x FOR PEER REVIEW 5 of 1

Energies 2022, 15, 7492 5 of 17

2.4. Coating Defects

2.4. Coating Defects
The wind turbine blade coating can protect the composites from environmental factor
suchThe
as wind turbineradiation,
ultraviolet blade coating can protect
moisture, the composites
and heat [34]. Therefrom environmental
are mainly factors
two application meth
such as ultraviolet radiation, moisture, and heat [34]. There are mainly two application
ods for blade surface coating: the in-mold application and the post-mold application [35]. Be
methods for blade surface coating: the in-mold application and the post-mold applica-
cause of the long-term and high-speed operation of wind turbine blades, as well as the erosio
tion [35]. Because of the long-term and high-speed operation of wind turbine blades, as well
of solid particles and rainwater in the air, the blade coating will inevitably suffer from dam
as the erosion of solid particles and rainwater in the air, the blade coating will inevitably
age defects
suffer [36], as
from damage shown
defects inas
[36], Figure
shown5.inAny type
Figure of defect
5. Any type ofdamage to windtoturbine
defect damage wind blade
will affect
turbine itswill
blades aerodynamic efficiency and
affect its aerodynamic reduce
efficiency andpower
reducegeneration.
power generation.

Figure
Figure 5. 5. Photo
Photo of the
of the blade
blade coating
coating defects
defects [33]. [33].

3.3.Wind
WindTurbine
Turbine Blade
Blade Fault
Fault Diagnosis
Diagnosis Method
Method
Researchers
Researchers usually
usuallyneed
needto to
spend
spenda lot of time
a lot andand
of time energy
energycollecting andand
collecting collating
collating lite
literature in different research fields. With the development of information
ature in different research fields. With the development of information technology, technology, thethe appl
application of bibliometric analysis tools dramatically facilitates the work of researchers.
cation of bibliometric analysis tools dramatically facilitates the work of researchers. The bibl
The bibliometric analysis supports linking documents by keywords to form visual charts
ometric analysis supports linking documents by keywords to form visual charts based on spe
based on specific databases (for example, Web of Science, WoS, CNKI, etc.), articles in
cific databases
specific fields, or(for example, Web
a combination of methods
of all Science, WoS, CNKI,
[37,38]. etc.), articles
Bibliometric in specific
analysis can help fields, or
combination
researchers of all methods
understand [37,38].
research Bibliometric
hotspots analysis can
and development helpThe
trends. researchers
five mostunderstand
com- re
searchused
monly hotspots and development
bibliometric analysis toolstrends. TheVoSviewer,
include five most commonly
Biblioshiny,usedGephi,bibliometric
HistCite, analys
and CiteSpace
tools include [39]. This article
VoSviewer, mainly selects
Biblioshiny, Gephi, 53 articles onand
HistCite, wind turbine blade
CiteSpace [39]. fault
This diag-
article mainl
nosis through
selects the Web
53 articles of Science
on wind database
turbine blade and
faultuses VoSviewer
diagnosis software
through to visualize
the Web the databas
of Science
literature, and the cluster diagram composed of literature keywords is obtained,
and uses VoSviewer software to visualize the literature, and the cluster diagram composed o as shown
in Figure 6. The literature on the fault diagnosis and condition monitoring of wind turbine
literature keywords is obtained, as shown in Figure 6. The literature on the fault diagnosis an
blades mainly includes three aspects, namely non-destructive testing, fault diagnosis based
condition monitoring of wind turbine blades mainly includes three aspects, namely non-de
on operating data, and fault diagnosis based on vibration signals.
structive testing, fault diagnosis based on operating data, and fault diagnosis based on vibra
tion signals.

3.1. Non-Destructive Techniques

NDTs can conduct SHM on wind turbine blades to avoid serious accidents and ensur
the safe operation of wind turbines. In addition, NDTs can also determine the cause of th
damage. Some detection methods can detect the location and size of the blade damage to us
for later maintenance and repair [40]. At present, the non-destructive testing methods of win
turbine blades mainly include strain measurement, acoustic emission, ultrasonic, vibration
thermal imaging, machine vision, etc. Although these detection methods tend to be perfec
and mature, few combine multiple detection methods for detection [41]. Muñoz et al. [42] be
lieve that an NDT is applied to SHM systems to detect the internal performance of the materia
structure, which can reduce maintenance costs and prolong the service life of wind turbine
Gholizadeh et al. [43] classified NDTs into contact and non-contact. This section focuses on th
principle, working methods, advantages, and disadvantages of damage detection methods.
Energies
Energies2022,
2022,15,
15,x7492
FOR PEER REVIEW 6 6ofof16
17

Keywordscluster
Figure6.6.Keywords
Figure clusterof
ofwind
windturbine
turbineblade
bladefault
faultdiagnosis
diagnosismethod.
method.

3.1. Non-Destructive
3.1.1. Strain DetectionTechniques
Method
NDTs can conduct
The detection method SHM on wind
of strain turbine blades
measurement to avoid
detects serious accidents
the micro-damage changeand ofensure
wind
the safeblade
turbine operation
lengthoforwind turbines.by
deformation Inusing
addition, NDTssensor
the strain can also determine
[44]. the cause
The advantage of the
of strain
damage. Some
measurement is detection
that it canmethods can detect
continuously monitorthe wind
location and size
turbine of the
blades for blade
a longdamage
time, but to
useaccuracy
the for laterand
maintenance
sensitivity and repair
of strain [40]. At present,
measurement the non-destructive
are dependent testing
on the distance methods
between the
of wind
sensor andturbine
damage blades mainly
[45]. In recentinclude strain
years, strain measurement,
measurement acoustic
based on theemission,
Fiber Bragg ultrasonic,
Grating
vibration, thermal imaging, machine vision, etc. Although these detection
(FBG) sensor has been widely used. The core component of FBG is the grating structure methods tenden-to
be perfect and mature, few combine multiple detection methods for detection
graved in the fiber core, and it usually changes the refractive index, making the refractive in- [41]. Muñoz
et al.
dex of[42] believe
the fiber that
core an NDT isdistribute
periodically applied toalong
SHMthe systems to detect[46].
axial direction the internal
When the performance
surround-
of the
ing material
stress changes, structure, which
the reflection can reduceλBmaintenance
wavelength will move to costs
realizeand
theprolong the service
strain detection, as
life of wind turbines. Gholizadeh et al. [43] classified NDTs into contact
shown in Figure 7. The overall performance of FBG can be tested by experimental comparison. and non-contact.
Thiset
Guo section
al. [47]focuses on the
used FBG to principle,
monitor theworking
internal methods, advantages,
stress fatigue and disadvantages
of composite wind turbine of
damage detection methods.
blades. Through experimental comparison, they concluded that the durability of FBG is
stronger than other electronic strain sensors. In addition, FBG can effectively monitor the
3.1.1. Strain Detection Method
operational status of wind turbines in practical applications. Bang et al. [48] applied FBG to
onshore Theanddetection
offshoremethod
MW wind of strain
turbinesmeasurement detects the
for health monitoring. micro-damage
Briefly, 41 FBG onchange
the wind of
wind turbine
turbine supportblade length
structure wasorinstalled,
deformation
whichby hadusing the strain
a sampling ratesensor
of more[44].
than The
40 advantage
Hz, and it
of strain measurement is that it can continuously monitor wind turbine blades for a long
showed good stability and accuracy in dynamic strain monitoring. It is also feasible to apply
time, but the accuracy and sensitivity of strain measurement are dependent on the distance
FBG to wind turbine blade strain monitoring. Schroeder et al. [49] installed the FBG on the
between the sensor and damage [45]. In recent years, strain measurement based on the
wind turbine blade to monitor the blade strain data in real time and used the accumulated
Fiber Bragg Grating (FBG) sensor has been widely used. The core component of FBG
data to determine the fatigue load distribution. Compared with the FBG, the strain sensor
is the grating structure engraved in the fiber core, and it usually changes the refractive
has no obvious advantages. To improve it, Wu et al. [50] proposed a new strain sensor for
index, making the refractive index of the fiber core periodically distribute along the axial
surface strain measurement. It was verified that the sensor could be used to generate two-
direction [46]. When the surrounding stress changes, the reflection wavelength λ will
dimensional strain images. In addition, the combination with an intelligent algorithmB can
move to realize the strain detection, as shown in Figure 7. The overall performance of FBG
improve the measurement accuracy, Lee et al. [51] proposed a new strain estimation method
can be tested by experimental comparison. Guo et al. [47] used FBG to monitor the internal
and objective function to obtain the optimal arrangement of strain sensors on the wind tur-
stress fatigue of composite wind turbine blades. Through experimental comparison, they
bine blade.
concluded that the durability of FBG is stronger than other electronic strain sensors. In
addition, FBG can effectively monitor the operational status of wind turbines in practical
applications. Bang et al. [48] applied FBG to onshore and offshore MW wind turbines for
health monitoring. Briefly, 41 FBG on the wind turbine support structure was installed,
which had a sampling rate of more than 40 Hz, and it showed good stability and accuracy
in dynamic strain monitoring. It is also feasible to apply FBG to wind turbine blade strain
monitoring. Schroeder et al. [49] installed the FBG on the wind turbine blade to monitor
the blade strain data in real time and used the accumulated data to determine the fatigue
load distribution. Compared with the FBG, the strain sensor has no obvious advantages.
Figure 7. Structure and principle of FBG [52].
 operational status of wind turbines in practical applications. Bang et al. [48] applied FBG to
onshore and offshore MW wind turbines for health monitoring. Briefly, 41 FBG on the wind
turbine support structure was installed, which had a sampling rate of more than 40 Hz, and it
showed good stability and accuracy in dynamic strain monitoring. It is also feasible to apply
FBG to wind turbine blade strain monitoring. Schroeder et al. [49] installed the FBG on the
Energies 2022, 15, 7492 wind turbine blade to monitor the blade strain data in real time and used the accumulated 7 of 17
data to determine the fatigue load distribution. Compared with the FBG, the strain sensor
has no obvious advantages. To improve it, Wu et al. [50] proposed a new strain sensor for
surface strainit,measurement.
To improve Wu et al. [50] It was verified
proposed a newthat the sensor
strain sensor for
could be used
surface to generate
strain two-
measurement.
dimensional
It was verifiedstrain
thatimages. In addition,
the sensor could bethe usedcombination
to generatewith an intelligent algorithm
two-dimensional can
strain images.
improve the measurement
In addition, the combination accuracy,
with an Lee et al. [51] algorithm
intelligent proposed acannew strain estimation
improve method
the measurement
and objective
accuracy, Lee function
et al. [51]toproposed
obtain thea optimal arrangement
new strain estimationof strain sensors
method on thefunction
and objective wind tur- to
bine blade.
obtain the optimal arrangement of strain sensors on the wind turbine blade.

Energies 2022, 15, x FOR PEER REVIEW 7 of 16

Figure 7. Structure and principle of FBG [52].
Figure 7. Structure and principle of FBG [52].
The strain detection method can be applied to the monitoring of both onshore and
The strain detection method can be applied to the monitoring of both onshore and off-
offshore wind turbine blades. In the future, further exploration is still needed to reduce the
shore wind turbine blades. In the future, further exploration is still needed to reduce the
development cost and improve detection accuracy.
development cost and improve detection accuracy.
3.1.2. Acoustic Emission Detection Method
3.1.2. Acoustic Emission Detection Method
The acoustic emission detection method focuses on the detection of electrical signals
fromThe acousticelastic
transient emission detection
wave method
conversion focuses
caused by on the detection
damage of electrical
initiation, signals fromor
crack propagation,
transient elastic wave conversion caused by damage initiation,
plastic deformation release energy [53]. The working principle is shown in Figure crack propagation, or plastic
8; the
deformation release energy [53]. The working principle is shown in Figure
sensor converts the sound signal from damage into an electrical signal. Acoustic emission 8; the sensor con-
verts the sound
detection signal from
can identify thedamage
increaseinto an electrical
in blade fatiguesignal.
damage Acoustic
and the emission
locationdetection can
of that. Tang
identify the monitored
et al. [54] increase in theblade fatiguedamage
fatigue damageofand 2MWthe wind
location of that.
turbine Tang through
blades et al. [54]acoustic
moni-
tored the fatigue
emission damage
technology andof verified
2MW wind thatturbine
acoustic blades through
emission acoustic emission
technology technology
can effectively detect
and
the occurrence of cracks on blades through experiments. In addition, acousticofemission
verified that acoustic emission technology can effectively detect the occurrence cracks
ondetection
blades through experiments.
technology can alsoIndetect
addition, acoustic
damage emissionand
generation detection technology
processes. Zhou et canal.also
[55]
detect damage
monitored thegeneration
damage and andfailure
processes.
processZhou et al. turbine
of wind [55] monitored
blades in the
thedamage
tensile and failure
test through
process
acousticof emission
wind turbine blades inThe
technology. the tensile test through
test results showedacoustic
that theemission
damagetechnology.
started from The the
test
endresults
of theshowed that theand
shear plane, damage started
acoustic from thetechnology
emission end of the shearcouldplane,
monitorand the
acoustic emis-
occurrence
sion
andtechnology
propagation could monitor the
of damage. Theoccurrence and propagation
key to improving the accuracy of damage. The key to im-
of this technology is to
proving the accuracy of this technology is to eliminate noise interference
eliminate noise interference to acoustic emission signals. Liu et al. [56] applied acoustic to acoustic emission
signals.
emissionLiutechnology
et al. [56] applied
to the acoustic
monitoring emission technology
and fault diagnosis to the monitoring
of wind turbine and faultbearing.
blade diag-
nosis
Theyofproposed
wind turbinea newblade bearing.
cepstrum They proposed
editing method toa perform
new cepstrum editing
the noise method
reduction to per-
processing
of the
form theacoustic emission processing
noise reduction signal source. of the acoustic emission signal source.

Principleofofacoustic
Figure8.8.Principle
Figure acousticemission
emissiondetection
detectiontechnology
technology[53].
[53].

Theacoustic
The acoustic emission
emission detection
detection method
method hashas a good
a good effect
effect on crack
on crack damage
damage detection
detection and
and can also locate internal structural damage. However, there is often noise interference
can also locate internal structural damage. However, there is often noise interference in the
in the process of signal acquisition; eliminating noise interference will also increase the cost
process of signal acquisition; eliminating noise interference will also increase the cost of the
detection system, and it requires a data acquisition system with a high sampling frequency
[48,49].

3.1.3. Ultrasonic Testing Method

 Energies 2022, 15, 7492 8 of 17

of the detection system, and it requires a data acquisition system with a high sampling
frequency [48,49].

3.1.3. Ultrasonic Testing Method

The ultrasonic testing method is where an ultrasonic wave first propagates through
the internal material, and then the sensor detects the reflected wave [57]. As shown in
Figure 9, ultrasonic sensors are installed on the blades and tower of the wind turbine for
signal acquisition. Different reflection, attenuation, resonance, and transmission modes
can be distinguished depending on the material or structure [58]. Therefore, this method
can be used to detect microdamage and judge the size, location, and other information of
Energies 2022, 15, x FOR PEER REVIEWdamage [59]. The ultrasonic detection method can be used to detect both the internal 8 of and
16
surface damages of wind turbine blades. Tiwari et al. [60] used ultrasonic technology to
detect the surface debonding damage of wind turbine blades and used discrete wavelet
change, variational modal decomposition, and Hilbert transforms to process the signal, to
and location
estimate ofsize
the blade damage.
and Lamarre
location of bladeet damage.
al. [53] used ultrasonic
Lamarre et al.phased array
[53] used technology
ultrasonic to
phased
detect
arraythe internal delamination,
technology to detect thewrinkles, and adhesive thickness
internal delamination, wrinkles,ofand
wind turbinethickness
adhesive blades. Inof
addition, the ultrasonic detection method can also be applied to the damage
wind turbine blades. In addition, the ultrasonic detection method can also be applied detection of off-to
shore wind turbines. Brett et al. [61] used the swept frequency ultrasonic technology
the damage detection of offshore wind turbines. Brett et al. [61] used the swept frequency below
100 kHz to detect
ultrasonic the foundation
technology below 100 of kHz
offshore wind the
to detect turbines. This method
foundation is also
of offshore applicable
wind turbines.
toThis
the fault
methoddiagnosis
is also of wind turbine
applicable to theblades.
fault diagnosis of wind turbine blades.

Figure 9. 9.
Figure Sketch of of
Sketch thethe
ultrasonic sensors
ultrasonic placed
sensors onon
placed wind turbines
wind [18].
turbines [18].

The
The ultrasonic
ultrasonic detection
detection method
method cancan continuously
continuously monitor
monitor the internal
the internal and and surface
surface of
wind turbine blades. However, ultrasonic testing requires a long time to collect signals,signals,
of wind turbine blades. However, ultrasonic testing requires a long time to collect and
theand the data
signal signal data processing
processing is also complex,
is also complex, which
which leads leads
to the to of
delay the delay of
damage damage[62].
judgment judg-
ment [62].
Therefore, Therefore,
future future
research researchintelligence
on artificial on artificialalgorithms
intelligencecan
algorithms
improve thecan processing
improve the
processing
capacity capacity of data.
of data.
3.1.4. Thermal Imaging Detection Method
3.1.4. Thermal Imaging Detection Method
The thermal imaging detection method is mainly applied to detect the change of
The thermal imaging detection method is mainly applied to detect the change of ther-
thermodynamic properties of wind turbine blades by scanning the surface of that. When
modynamic properties of wind turbine blades by scanning the surface of that. When the
the micro-damage fault occurs, the temperature anomaly will occur, which can be utilized
micro-damage fault occurs, the temperature anomaly will occur, which can be utilized to
to detect and judge the fault [57]. This technology requires accurate image processing. In
detect and judge the fault [57]. This technology requires accurate image processing. In real
real applications, it is difficult to eliminate the influence of blade damage on temperature
applications, it is difficult to eliminate the influence of blade damage on temperature and
and other factors [57], making ambient temperature interference the key to accurately
other factors [57], making ambient temperature interference the key to accurately identify-
identifying damage. Doroshtnasir et al. [63] used thermal imaging technology to carry out
ing damage. Doroshtnasir
nondestructive testing onetlong-distance
al. [63] used thermal imaging
wind turbine technology
blades to carrythe
and calculated out nonde-
differential
structive testing on long-distance wind turbine blades and calculated the differential
temperature of blades to eliminate signal interference reflection, to ensure the accuracy tem-of
perature of blades to eliminate signal interference reflection, to ensure the accuracy
thermal imaging technology in blade damage diagnosis. It concluded that the temperature of ther-
mal imaging near
difference technology
the hub inisblade
large,damage diagnosis.
and there It concluded
is the largest that the
possibility temperature
of damage. dif-
Thermal
ference near the hub is large, and there is the largest possibility of damage. Thermal imaging
detection technology can identify the fault of wind turbine blades and extract the damage
characteristics. Hwang et al. [64] proposed thermal imaging technology using the continu-
ous line later to visualize the damage of wind turbine blades under rotating conditions and
extract the characteristics of damage. Avdelidis et al. [65] applied infrared thermal imaging
Energies 2022, 15, 7492 9 of 17

imaging detection technology can identify the fault of wind turbine blades and extract the
damage characteristics. Hwang et al. [64] proposed thermal imaging technology using
the continuous line later to visualize the damage of wind turbine blades under rotating
conditions and extract the characteristics of damage. Avdelidis et al. [65] applied infrared
thermal imaging technology to wind turbine blade damage detection and summarized the
advantages and disadvantages of this technology.
The thermal imaging detection method can detect the internal structure of the wind
turbine blade without contact. However, the temperature change caused by damage is
delayed, and it is easy for the environmental temperature to cause interference in the
detection process. In future research, the influence of environmental temperature has to be
reduced to improve the reliability and accuracy of this method.

3.1.5. Machine Vision Detection

At present, a large number of innovative sensing and monitoring systems based on
machine vision technology have been developed and applied in the field of SHM [66].
The full-scale condition monitoring of wind turbine blade surfaces can be carried out by
shooting images. Yang et al. [67] proposed a video measurement technology for monitoring
the deformation of large wind turbine blades in full-scale tests and operations. Figure 10a
shows the setup of the machine vision system on a wind turbine. This technology requires
two photos from different angles of three-dimensional samples through a parallel network
measurement method, which are prepared by applying a grid on the surface to be tested.
This method improves the measurement accuracy and reliability, but the measurement
accuracy is highly dependent on the measurement position. Poozesh et al. [68] used
a relatively new optical sensing technology for the early detection of the design and
manufacturing defects of wind turbine blades, and the images taken from a pair of stereo
cameras were used to determine the surface strain of the blade surface. This method has
high accuracy, but the calculation process is relatively complex and needs to be optimized
and improved. Compared with ordinary machine vision technology, vision performed by
unmanned aerial vehicles (UAVs) has better detection accuracy. QIU et al. [69] built an
image acquisition system; the system uses a latitude and longitude M600 UAV equipped
with a Zenmuse Z30 cloud camera. The images obtained by the system were of high
resolution, so the microscopic damage to the blade could be detected clearly. Moreover,
based on the image data collected by UAVs, Long et al. [70] automatically detect and
diagnose the surface cracking of wind turbine blades. By analyzing the image taken by the
UAV to quickly detect blade cracking, a data-driven framework is developed to process
blade images and obtain blade cracking. However, the time and effectiveness of image
processing need to be further optimized. In addition, efficient intelligent algorithms are
used to recognize images captured by UAVs. Wang et al. [71] collected a large number
of wind turbine blade data through UAVs, and they proposed an unsupervised learning
method combined with image data features to distinguish the normal and abnormal parts
of the blade, as shown in Figure 10b.The results show that this method is very useful for
detecting anomalies in the blade image. However, this method cannot effectively identify
and eliminate the fouling on the blade surface. Traditional wind farms need manual
periodic maintenance detection. In order to improve safety and detection efficiency, the
robot detection of wind turbine faults has been gradually developed. Kuang et al. [72]
proposed that the wind turbine blade could be crawled by magnetic adsorption force and
friction force. At the same time, infrared sensors and high-definition cameras were installed
on the robot to diagnose the micro-damage on the blade surface. However, it was difficult
to maintain the robot, and the equipment for blade detection needed to be further improved.
Similarly, Josef Franko et al. [73] proposed a light magnetic crawler with visual and laser
radar sensor surface contact to NDT the wind turbine, as shown in Figure 10c. In addition,
Zhang et al. [74] proposed a new method of using climbing robots with ground control
stations and high-resolution cameras to scan wind turbine blades.
 adhesive expansion [77,78]. X-ray technology can effectively detect the internal structure of
the blade. Evenly, X-ray tomography can present three-dimensional information inside the
blade of a wind turbine [18]. In summary, NDT is applied to the fault diagnosis of wind tur-
bine blades, which can examine the surface and internals of blades. However, due to the
Energies 2022, 15, 7492 10 of 17
limitation of technical means, cost, and complex signal processing, new technologies are
still needed to be developed and utilized to make up for defects in the future.

(a) (b)

(c)
Figure 10. (a)
Figure 10.Machine vision
(a) Machine [67].[67].
vision (b) Data collecting
(b) Data using
collecting the UAV
using [71].[71].
the UAV (c) Crawling robot
(c) Crawling [73].[73].
robot

At of
Table 1. List present, the research
the advantages on the SHM
and disadvantages of of wind detection
different turbine blades
methodsbased on the machine
[45,59,75,76,79–81].
vision detection method is still in its infancy; the machine vision technology will have
Test Methods Advantages Shortcomings
more extensive applications in the future. Although machine vision detection accuracy
The increase in blade fatigue damage degree Affected by the noise environment, this makes the data pro-
Acoustic emission highly depends on image processing and data acquisition, its advantages are still obvious.
and the determination of location were detected cessing task complex and the cost higher
The staff can remotely control the machine equipment to detect the wind turbine blades,
The size, location, and other information on Signal acquisition for a long time is also complex for signal pro-
Ultrasonic which can improve the detection efficiency and protect their safety [75,76]. In the future,
damage degree can be judged cessing, and there is a delay in judging the signal
the combination of the machine vision detection method and big data can realize earlier
Strain measure- Continuous monitoring of wind turbine blades The sensitivity depends on the distance between the sensor lo-
detection of the occurrence of damage, making it an important part of SHM.
ment for a long time cation and the damage
Although the aforementioned technologies have been widely developed and utilized,
The detection of all blades is more intuitive and The influence of the environment on temperature leads to devi-
there are certain deficiencies in each of them. A better choice for NDT is to combine various
Thermal imaging effective ation in the results
technologies into the SHM system. As shown in Table 1, the advantages and disadvantages
Machine vision Low Cost, High Safety, and Real-time Online Accuracy depends on image accuracy and cannot explain the
of each detection method are summarized. In addition, new NDT techniques have been
Detection of Personnel Wind Farm physical mechanism of damage
developed and applied. For instance, shear imaging can visualize the change of blade
surface deformation by interfering laser points and can effectively detect joint failure caused
by adhesive expansion [77,78]. X-ray technology can effectively detect the internal structure
of the blade. Evenly, X-ray tomography can present three-dimensional information inside
the blade of a wind turbine [18]. In summary, NDT is applied to the fault diagnosis of wind
turbine blades, which can examine the surface and internals of blades. However, due to the
limitation of technical means, cost, and complex signal processing, new technologies are
still needed to be developed and utilized to make up for defects in the future.
Energies 2022, 15, 7492 11 of 17

Table 1. List of the advantages and disadvantages of different detection methods [45,59,75,76,79–81].

Test Methods Advantages Shortcomings

The increase in blade fatigue damage degree and Affected by the noise environment, this makes the
Acoustic emission
the determination of location were detected data processing task complex and the cost higher
Signal acquisition for a long time is also complex
The size, location, and other information on
Ultrasonic for signal processing, and there is a delay in judging
damage degree can be judged
the signal
Continuous monitoring of wind turbine blades for a The sensitivity depends on the distance between
Strain measurement
long time the sensor location and the damage
The detection of all blades is more intuitive and The influence of the environment on temperature
Thermal imaging effective leads to deviation in the results
Low Cost, High Safety, and Real-time Online Accuracy depends on image accuracy and cannot
Machine vision
Detection of Personnel Wind Farm explain the physical mechanism of damage

3.2. Fault Diagnosis Method Based on Operation Data

In recent years, with the rapid development of artificial intelligence algorithms and
big data analysis, artificial intelligence can imitate the learning skills of the human brain. At
the same time, it combined with data analysis is widely used. The application of intelligent
algorithms such as a neural network in the fault diagnosis of wind turbine blades has been
well tested.
Xin et al. [82] extracted the feature vector of modal parameters by using a deep belief
network. Aiming at the problem of inaccurate modal parameters caused by environmental
noise, measurement error, and other factors during the operation of wind turbine blade
damage, the feature vector was used as the input signal of the error backpropagation
training (BP) neural network to reduce the influence of noise and improve the accuracy
of the microdamage diagnosis of wind turbine blades. However, this method only diag-
noses single microscopic damage and cannot be applied to other types of damage. Wang
et al. [83] introduced the depth automatic encoder (DAE) model. They combined the neural
network with SCADA data to monitor the damage to wind turbine blades and verified
the effectiveness of this method in the fault diagnosis of wind turbine blades in practice.
Sahir et al. [84] built a Convolutional Neural Networks (CNN) model to diagnose the
micro-damage of wind turbine blades. The model can diagnose a variety of damage types,
including radiation effects, wear, and fracture. The diagnostic accuracy could be as high as
81.25%. As shown in Figure 11, the CNN structure consists of three types: the convolution
layer, the pool layer, and the full connection layer. Different damage categories are classified
by full connection layer calculation. In addition, migration learning can also improve the
network training results. Liu et al. [85] used the transfer learning method of the Inception
v3 model to diagnose the damage to wind turbine blades. The experimental results show
that the calculation speed and accuracy of the model algorithm are better than R-CNN.
However, a more accurate diagnosis of micro-damage images of wind turbine blades has
always been the difficulty of deep learning optimization. Long Wang et al. [70] proposed
an extended Haar-like feature set classifier based on the optimization and improvement
of the original Haar cascade classifier for the crack damage of the wind turbine blade,
which was verified by being proved superior to the LogitBoost classifier. Xiao-Yi et al. [86]
proposed a deep learning framework, the Alexnet model, to solve the problem of the low
accuracy of the identification and detection of micro-damage on the surface of wind turbine
blades. The accuracy of the model for damage diagnosis reached 99.001%, which was
19.424% higher than that of the traditional BP neural network model. In order to improve
the diagnostic performance of the deep learning model for micro-damage of wind turbine
blades, Yang et al. [87] added transfer learning and ensemble learning classifiers based on
convolutional neural network, which enhanced the ability of abstract feature extraction
for the micro damage faults of wind turbine blades. Most of these studies focus on single
 classifier. Xiao-Yi et al. [86] proposed a deep learning framework, the Alexnet model, to solve
the problem of the low accuracy of the identification and detection of micro-damage on the
surface of wind turbine blades. The accuracy of the model for damage diagnosis reached
99.001%, which was 19.424% higher than that of the traditional BP neural network model. In
order to improve the diagnostic performance of the deep learning model for micro-damage of
Energies 2022, 15, 7492 12 of 17
wind turbine blades, Yang et al. [87] added transfer learning and ensemble learning classifiers
based on convolutional neural network, which enhanced the ability of abstract feature extrac-
tion for the micro damage faults of wind turbine blades. Most of these studies focus on single
blade microscopic
microscopic damage
damage diagnosis.
diagnosis. Whether
Whether it is
is feasible
feasible to apply
apply a deep
deep learning model
to other wind turbine blades for image diagnosis
diagnosis accuracy
accuracy still
still needs
needs to
to be
be discussed.
discussed.

Figure 11. CNN structural model.

The
The combination
combination of of artificial
artificial intelligence algorithms and
intelligence algorithms and operation data can
operation data can improve
improve
the reliability and accuracy of the monitoring system. However, this method
the reliability and accuracy of the monitoring system. However, this method is still is still in
in the
the exploratory stage for various types of damage identification and the real-time
exploratory stage for various types of damage identification and the real-time online moni- online
monitoring
toring test oftest of large-scale
large-scale wind turbine
wind turbine blades,blades, and it be
and it cannot cannot be promoted
widely widely promoted
at present.at
present.

3.3. Fault Diagnosis Based on Vibration Signal

The fault diagnosis of wind turbine blades based on vibration signal is mainly through
the selection of damage index and modal parameters. In the process of signal acquisition,
it is necessary to eliminate the interference of environmental noise on the signal. The
interference of the signal will reduce the accuracy of the vibration signal and the error of
the modal parameters. Therefore, it is necessary to reduce the influence of the environment
in the study of vibration signal damage identification.
The damage can be effectively identified by comparing the modal parameters before
and after it occurs. Emilio Di Lorenzo et al. [88] installed accelerometers on wind turbine
blades to collect vibration data. By comparing the modal parameters before and after the
buckling test, the occurrence of damage can be successfully predicted. In addition, the
establishment of the finite element model can also be used to analyze the structural damage
of the blade. Moradi et al. [89] firstly installed intelligent sensors on wind turbine blades for
experiments to obtain strain and vibration data and then simulated the structural state, and
after the blade damage by finite element simulation, which can comprehensively detect the
blade damage, this method can achieve a reliable SHM system. To exclude the influence
of environmental noise, Abouhnik et al. [90] used the empirical mode decomposition
method to divide the vibration signal into basic components and built a model in the
finite element software ANSYS to simulate the vibration of the wind turbine with three
blades. At the same time, the crack damage was set on the wind turbine blade, and the
vibration characteristics of the blade at different speeds were tested. By comparing the
simulation and experimental results, the method can identify the location and extent of
the blade damage. Gómez et al. [91] proposed a supervised statistical method to solve the
interference of uncertainty in the vibration signal detection of wind turbine blade damage
under different environments, and they developed three specific methods to improve the
accuracy of damage detection. Furthermore, Wang et al. [92] proposed a finite element
method combined with dynamic analysis (modal analysis and response analysis) to obtain
modal shape difference curvature. The numerical results show that the method can detect
the blade damage location and improve detection accuracy.
The fault diagnosis method based on vibration signals can identify damage by compar-
ing modal parameters such as natural frequency, vibration mode, and damping. However,
it remains a huge challenge to distinguish between signals collected by damaged blades
and environmental and operational conditions.
Energies 2022, 15, 7492 13 of 17

4. Prospect of Wind Turbine Blade Damage Detection Method

The future development trend of wind turbine blade damage detection methods is
discussed as follows:
1. Intelligent detection methods and deep learning methods require high accuracy of
image processing. How to improve the accuracy and algorithm of collected data
is the focus of future research. In addition, the use of new sensors on CMS, with
SCADA-generated data analysis;
2. In the future, the combination of 5G technology, big data, artificial intelligence, and
detection technology has achieved real-time, accurate, and timely results. The opera-
tion of smart wind farms and the self-maintenance of some fault damage can reduce
the operation and maintenance costs and improve the detection efficiency;
3. With the development of NDT technology, new technologies continue to emerge, such
as electromagnetic detection, shear imaging, photogrammetry, spectroscopy, radar
imaging system, etc. Most NDT technologies need to stop wind turbines for detection.
The new development direction is towards real-time online detection without shut-
down, and different technologies are combined to improve the accuracy of detection.
The combination of UAVs and the nondestructive testing system can achieve this
purpose, such as UAVs equipped with image, thermal imaging, photogrammetry, and
other equipment.
Through the aforementioned analysis, it can be seen that it is of great significance to
timely diagnose the damage of wind turbine blades, and more and more experts have paid
attention to the fault diagnosis of wind turbine blades by using different methods. Although
many practical problems have not been solved, there is no doubt that the development
potential of future fault diagnosis technology is huge.

5. Summary
With the global installed capacity increasing, the risk of wind turbine blade damage
enlarges with the length of wind turbine blades. It is of great significance to make the early
warning and identification of wind turbine blade failure in advance through different fault
diagnosis methods to avoid serious damage. The development of wind turbines needs
to meet the requirements of SHM, ND, and CM. Therefore, damage detection technol-
ogy will develop towards real-time online, remote, global detection, and non-destructive
non-contact. This paper summarizes three common diagnostic methods, including nonde-
structive testing technology, data acquisition, monitoring-based detection technology, and
vibration signal detection technology. At present, most nondestructive testing technology
depends on the performance of the testing instrument, and the detection accuracy is greatly
affected by the environment. The research of intelligent detection methods for wind turbine
blade diagnosis is a new trend, and intelligent detection technology to some extent makes
up for the lack of the accuracy of nondestructive testing. In the future, in the era of the
rapid development of computers, artificial intelligence, and big data, the damage detection
method based on data fusion will continue to optimize and improve the diagnosis and
treatment of wind turbine blades with great potential.

Author Contributions: Conceptualization, X.S. (Xiaowen Song) and Q.L.; writing of the manuscript,
Z.X. and Y.J.; updating of the text, X.S. (Xiaowen Song) and Y.Z.; literature searches, X.S. (Xiaojuan
Song), Z.W. and J.G.; critical reviewing of the manuscript, Z.X., X.S. (Xiaojuan Song) and C.C.;
organization and editing of the manuscript, Z.X., X.S. (Xiaowen Song) and Y.Z.; Project administration:
C.C.; Supervision: Q.L. All authors have read and agreed to the published version of the manuscript.
Funding: This research was funded by the National Natural Science Foundation of China (Grant
No. 52176212 and Grant No. 11962020) and the National Engineering Research Center for Offshore
Windpower (Grant No. HSFD22001). We are also very grateful to the Program for Innovative Research
Team in Universities of Inner Mongolia Autonomous Region (NMGIRT2213) and Inner Mongolia
Science and Technology Program under Grant (2021GG0336).
Energies 2022, 15, 7492 14 of 17

Conflicts of Interest: The authors declare no conflict of interest.

References
1. Mooney, H.; Larigauderie, A.; Cesario, M.; Elmquist, T.; Hoegh-Guldberg, O.; Lavorel, S.; Mace, G.M.; Palmer, M.; Scholes, R.;
Yahara, T. Biodiversity, climate change, and ecosystem services. Curr. Opin. Environ. Sustain. 2009, 1, 46–54. [CrossRef]
2. Chen, J.M.J.T.I. Carbon neutrality: Toward a sustainable future. Innovation 2021, 2, 100127. [CrossRef]
3. Zhang, M.; Zhang, J.; Li, Y.; Yu, J.; Qin, J.; Wei, K.; Song, L. Multi-site measurement for energy application of small distributed
wind farm in complex mountainous areas. Energy Rep. 2020, 6, 1043–1056. [CrossRef]
4. Arutyunov, V.S.; Lisichkin, G.V. Energy resources of the 21st century: Problems and forecasts. Can renewable energy sources
replace fossil fuels? Russ. Chem. Rev. 2017, 86, 777–804. [CrossRef]
5. Wang, F.; Harindintwali, J.D.; Yuan, Z.; Wang, M.; Wang, F.; Li, S.; Yin, Z.; Huang, L.; Fu, Y.; Li, L.; et al. Technologies and
perspectives for achieving carbon neutrality. Innovation 2021, 2, 100180. [CrossRef]
6. Council, G.W.E. Floating Offshore Wind—A Global Opportunity. The Global Wind Energy Council. 2022. Available online:
https://www.abeeolica.org.br/wp-content/uploads/2019/07/GWEC2018-compactado.pdf (accessed on 3 October 2022).
7. GWEC. Global Wind Report 2021. Global Wind Energy Council: Brussels, Belgium, 2021. Available online: https://gwec.net/
global-wind-report-2021/ (accessed on 3 October 2022).
8. Caithness Windfarm Information Forum. Available online: http://www.caithnesswindfarms.co.uk (accessed on 3 October 2022).
9. Katsaprakakis, D.A.; Papadakis, N.; Ntintakis, I.J.E. A Comprehensive Analysis of Wind Turbine Blade Damage. Energies 2021, 14,
5974. [CrossRef]
10. Song, Y.-D.; Li, P.; Liu, W.; Qin, M.J.M. Control, An overview of renewable wind energy conversion system modeling and control.
Meas. Control. 2010, 43, 203–208. [CrossRef]
11. Karyotakis, A.; Bucknall, R.J.J. Planned intervention as a maintenance and repair strategy for offshore wind turbines. J. Mar. Eng.
Technol. 2010, 9, 27–35. [CrossRef]
12. Chou, J.-S.; Ou, Y.-C.; Lin, K.-Y. Collapse mechanism and risk management of wind turbine tower in strong wind. J. Wind. Eng.
Ind. Aerodyn. 2019, 193, 103962. [CrossRef]
13. Farrar, C.R.; Worden, K. An introduction to structural health monitoring. Philos. Trans. R. Soc. A Math. Phys. Eng. Sci. 2007, 365,
303–315. [CrossRef]
14. de la Hermosa González, R.R.; Márquez, F.P.G.; Dimlaye, V.; Ruiz-Hernández, D.J.M.S. Pattern recognition by wavelet transforms
using macro fibre composites transducers. Mech. Syst. Signal Process. 2014, 48, 339–350. [CrossRef]
15. Jiménez, A.A.; Zhang, L.; Muñoz, C.Q.G.; Márquez, F.P.G. Maintenance management based on machine learning and nonlinear
features in wind turbines. Renew. Energy 2020, 146, 316–328. [CrossRef]
16. Li, D.; Ho, S.-C.M.; Song, G.; Ren, L.; Li, H. Structures, A review of damage detection methods for wind turbine blades. Smart
Mater. Struct. 2015, 24, 033001. [CrossRef]
17. Yang, R.; He, Y.; Zhang, H. Progress and trends in nondestructive testing and evaluation for wind turbine composite blade. Renew.
Sustain. Energy Rev. 2016, 60, 1225–1250. [CrossRef]
18. Márquez, F.P.G.; Chacón, A.M.P. A review of non-destructive testing on wind turbines blades. Renew. Energy 2020, 161, 998–1010.
[CrossRef]
19. Asensio, S.E.; Pérez, J.; Márquez, F. Economic Viability Study for Offshore Wind Turbines Maintenance Management; Springer:
Berlin/Heidelberg, Germany, 2015.
20. Igba, J.; Alemzadeh, K.; Durugbo, C.; Eiriksson, E.T. Technology, Through-life engineering services of wind turbines. CIRP J.
Manuf. Sci. Technol. 2017, 17, 60–70. [CrossRef]
21. Koitz, R.; Wotawa, F.; Lüftenegger, J.; Gray, C.S.; Langmayr, F. Wind Turbine Fault Localization: A Practical Application of Model-
Based Diagnosis. In Diagnosability, Security and Safety of Hybrid Dynamic and Cyber-Physical Systems; Springer: Berlin/Heidelberg,
Germany, 2018; pp. 17–43.
22. Liu, W.Y. A review on wind turbine noise mechanism and de-noising techniques. Renew. Energy 2017, 108, 311–320. [CrossRef]
23. Herrera, C.; Correa, M.; Villada, V.; Vanegas, J.D.; Garcia, J.G.; Nieto-Londono, C.; Sierra-Perez, J. Structural design and
manufacturing process of a low scale bio-inspired wind turbine blades. Compos. Struct. 2019, 208, 1–12. [CrossRef]
24. Jureczko, M.P.M. Optimisation of wind turbine blades. J. Mater. Process. Technol 2005, 167, 463–471. [CrossRef]
25. Du, J.; Wang, H.; Wang, S.; Song, X.; Wang, J.; Chang, A.J.O.E. Fatigue damage assessment of mooring lines under the effect of
wave climate change and marine corrosion. Ocean. Eng. 2020, 206, 107303. [CrossRef]
26. Jimenez-Martinez, M.J.H. Harbor and coastal structures: A review of mechanical fatigue under random wave loading. Heliyon
2021, 7, e08241. [CrossRef] [PubMed]
27. Chang, J.B.; Rudd, J.L. Damage Tolerance of Metallic Structures: Analysis Methods and Applications; ASTM International: West
Conshohocken, PA, USA, 1984; Volume 842.
28. Slot, H.; Gelinck, E.; Rentrop, C.; Van Der Heide, E.J.R.E. Leading edge erosion of coated wind turbine blades: Review of coating
life models. Renew. Energy 2015, 80, 837–848. [CrossRef]
29. Gaudern, N. A Practical Study of the Aerodynamic Impact of Wind Turbine Blade Leading Edge Erosion. J. Phys. Conf. Ser. 2014,
524, 012031. [CrossRef]
Energies 2022, 15, 7492 15 of 17

30. Yu, Y.; Cao, H.; Yan, X.; Wang, T.; Ge, S.S.J.N. Defect Identification of Wind Turbine Blades based on Defect Semantic Features
with Transfer Feature Extractor. Neurocomputing 2019, 376, 1–9. [CrossRef]
31. Talreja, R. Fatigue damage mechanisms. Model. Damage Fatigue Fail. Compos. Mater. 2016, 4, 25–40.
32. Elhajjar, R.F.; Shams, S.S.J.C.S. Technology, A new method for limit point determination in composite materials containing defects
using image correlation. Compos. Sci. Technol. 2016, 122, 140–148. [CrossRef]
33. Guo, J.; Liu, C.; Cao, J.; Jiang, D.J.R.E. Damage identification of wind turbine blades with deep convolutional neural networks.
Renew. Energy 2021, 174, 122–133. [CrossRef]
34. Zhang, S.; Kiil, S.; Dam-Johansen, K.; Bernad, P.J., Jr. Accelerated Rain Erosion of Wind Turbine Blade Coatings; Danmarks Tekniske
Universitet (DTU): Copenhagen, Denmark, 2014.
35. Keegan, M.H.; Nash, D.H.; Stack, M.M. On erosion issues associated with the leading edge of wind turbine blades. J. Phys. D:
Appl. Phys. 2013, 46. [CrossRef]
36. Cripps, D. The future of blade repair. Reinf. Plast. 2011, 55, 28–32. [CrossRef]
37. Qiao, Q.; Yunusa-Kaltungo, A.; Edwards, R.E. Towards developing a systematic knowledge trend for building energy consumption
prediction. J. Build. Eng. 2021, 35, 101967. [CrossRef]
38. Van Raan, A. The use of bibliometric analysis in research performance assessment and monitoring of interdisciplinary scientific
developments. TATuP 2003, 12, 20–29. [CrossRef]
39. Rabbani, M.R.A.; Bashar, A.; Atif, M.; Jreisat, A.; Zulfikar, Z.; Naseem, Y. Text Mining and Visual Analytics in Research: Exploring
the Innovative Tools. In Proceedings of the 2021 International Conference on Decision Aid Sciences and Application (Dasa),
Online, 7–8 December 2021.
40. Kusiak, A.; Zhang, Z.; Verma, A. Prediction, operations, and condition monitoring in wind energy. Energy 2013, 60, 1–12.
[CrossRef]
41. Raišutis, R.; Jasiūnienė, E.; Šliteris, R.; Vladišauskas, A.J.U.U. The review of non-destructive testing techniques suitable for
inspection of the wind turbine blades. Ultrasound 2008, 63, 26–30.
42. Muñoz, C.Q.G.; Márquez, F.P.G. Future Maintenance Management in Renewable Energies. In Renewable Energies; Springer:
Berlin/Heidelberg, Germany, 2018; pp. 149–159.
43. Gholizadeh, S. A review of non-destructive testing methods of composite materials. Procedia Struct. Integr. 2016, 1, 50–57.
[CrossRef]
44. Schubel, P.J.; Crossley, R.J.; Boateng, E.K.B.; Hutchinson, J.R. Review of structural health and cure monitoring techniques for large
wind turbine blades. Renew. Energy 2013, 51, 113–123. [CrossRef]
45. Ozbek, M.; Rixen, D.J. Operational modal analysis of a 2.5 MW wind turbine using optical measurement techniques and strain
gauges. Wind Energy 2013, 16, 367–381. [CrossRef]
46. Hill, K.; Meltz, G. Fiber Bragg grating technology fundamentals and overview. J. Light. Technol. 1997, 15, 1263–1276. [CrossRef]
47. Guo, Z.-S.; Zhang, J.; Hu, H.; Guo, X. Structural health monitoring of composite wind blades by fiber bragg grating. In Proceedings
of the International Conference on Smart Materials and Nanotechnology in Engineering, International Society for Optics and
Photonics, Harbin, China, 1–4 July 2007; p. 64230I. [CrossRef]
48. Bang, H.-J.; Jang, M.; Shin, H. Structural health monitoring of wind turbines using fiber Bragg grating based sensing system. In
Sensors and Smart Structures Technologies for Civil, Mechanical, and Aerospace Systems; SPIE: Bellingham, DC, USA, 2011; Volume
7981, pp. 716–723. [CrossRef]
49. Schroeder, K.; Ecke, W.; Apitz, J.; Lembke, E.; Lenschow, G. A fibre Bragg grating sensor system monitors operational load in a
wind turbine rotor blade. Meas. Sci. Technol. 2005, 17, 1167–1172. [CrossRef]
50. Wu, J.; Song, C.; Saleem, H.S.; Downey, A.; Laflamme, S. Network of flexible capacitive strain gauges for the reconstruction of
surface strain. Meas. Sci. Technol. 2015, 26, 055103. [CrossRef]
51. Lee, K.; Aihara, A.; Puntsagdash, G.; Kawaguchi, T.; Sakamoto, H.; Okuma, M. Feasibility study on a strain based deflection
monitoring system for wind turbine blades. Mech. Syst. Signal Process. 2017, 82, 117–129. [CrossRef]
52. Wen, B.; Tian, X.; Jiang, Z.; Li, Z.; Dong, X.; Peng, Z. Monitoring blade loads for a floating wind turbine in wave basin model tests
using Fiber Bragg Grating sensors: A feasibility study. Mar. Struct. 2020, 71, 102729. [CrossRef]
53. Nair, A.; Cai, C.S. Acoustic emission monitoring of bridges: Review and case studies. Eng. Struct. 2010, 32, 1704–1714. [CrossRef]
54. Tang, J.; Soua, S.; Mares, C.; Gan, T.-H. An experimental study of acoustic emission methodology for in service condition
monitoring of wind turbine blades. Renew. Energy 2016, 99, 170–179. [CrossRef]
55. Zhou, W.; Li, Y.; Li, Z.; Liang, X.; Pang, Y.; Wang, F. Interlaminar Shear Properties and Acoustic Emission Monitoring of the
Delaminated Composites for Wind Turbine Blades. In Advances in Acoustic Emission Technology; Springer: Berlin/Heidelberg,
Germany, 2014; pp. 557–566. [CrossRef]
56. Liu, Z.; Wang, X.; Zhang, L. Fault Diagnosis of Industrial Wind Turbine Blade Bearing Using Acoustic Emission Analysis. IEEE
Trans. Instrum. Meas. 2020, 69, 6630–6639. [CrossRef]
57. Tchakoua, P.; Wamkeue, R.; Ouhrouche, M.; Slaoui-Hasnaoui, F.; Tameghe, T.A.; Ekemb, G. Wind Turbine Condition Moni-toring:
State-of-the-Art Review, New Trends, and Future Challenges. Energies 2014, 7, 2595–2630. [CrossRef]
58. Amenabar, I.; Mendikute, A.; López-Arraiza, A.; Lizaranzu, M.; Aurrekoetxea, J. Comparison and analysis of non-destructive
testing techniques suitable for delamination inspection in wind turbine blades. Compos. Part B: Eng. 2011, 42, 1298–1305.
[CrossRef]
Energies 2022, 15, 7492 16 of 17

59. Márquez, F.P.G.; Tobias, A.M.; Pérez, J.M.P.; Papaelias, M. Condition monitoring of wind turbines: Techniques and methods.
Renew. Energy 2012, 46, 169–178. [CrossRef]
60. Tiwari, K.A.; Raisutis, R. Post-processing of ultrasonic signals for the analysis of defects in wind turbine blade using guided
waves. J. Strain Anal. Eng. Des. 2018, 53, 546–555. [CrossRef]
61. Brett, C.R.; A Gunn, D.; Dashwood, B.A.J.; Holyoake, S.J.; Wilkinson, P.B. Development of a technique for inspecting the
foundations of offshore wind turbines. Insight Non-Destructive Test. Cond. Monit. 2018, 60, 19–27. [CrossRef]
62. Sarrafi, A.; Mao, Z.; Niezrecki, C.; Poozesh, P. Vibration-based damage detection in wind turbine blades using Phase-based
Motion Estimation and motion magnification. J. Sound Vib. 2018, 421, 300–318. [CrossRef]
63. Doroshtnasir, M.; Worzewski, T.; Krankenhagen, R.; Röllig, M. On-site inspection of potential defects in wind turbine rotor blades
with thermography. Wind Energy 2016, 19, 1407–1422. [CrossRef]
64. Hwang, S.; An, Y.-K.; Sohn, H. Continuous Line Laser Thermography for Damage Imaging of Rotating Wind Turbine Blades.
Procedia Eng. 2017, 188, 225–232. [CrossRef]
65. Avdelidis, N.P.; Gan, T.-H. Non-destructive evaluation (NDE) of Composites: Infrared (IR) thermography of wind turbine blades.
In Non-Destructive Evaluation (NDE) of Polymer Matrix Composites; Karbhari, V.M., Ed.; Woodhead Publishing: Sawston, UK, 2013;
pp. 634–650. [CrossRef]
66. Ye, X.W.; Dong, C.Z.; Liu, T. A Review of Machine Vision-Based Structural Health Monitoring: Methodologies and Applications.
J. Sens. 2016, 2016, 7103039. [CrossRef]
67. Yang, J.; Peng, C.; Xiao, J.; Zeng, J.; Yuan, Y. Application of videometric technique to deformation measurement for large-scale
composite wind turbine blade. Appl. Energy 2012, 98, 292–300. [CrossRef]
68. Baqersad, J.; Poozesh, P.; Niezrecki, C.; Harvey, E.; Yarala, R. Full Field Inspection of a Utility Scale Wind Turbine Blade Using
Digital Image Correlation. CAMX 2014, 10, 2891–2960. [CrossRef]
69. Qiu, Z.; Wang, S.; Li, M. Defect Detection of Wind Turbine Blade Based on Unmanned Aerial Vehicle-taken Images. Power Gener.
Technol. 2018, 39, 277–285.
70. Wang, L.; Zhang, Z. Automatic Detection of Wind Turbine Blade Surface Cracks Based on UAV-Taken Images. IEEE Trans. Ind.
Electron. 2017, 64, 7293–7303. [CrossRef]
71. Wang, Y.; Yoshihashi, R.; Kawakami, R.; You, S.; Harano, T.; Ito, M.; Komagome, K.; Iida, M.; Naemura, T. Unsupervised anomaly
detection with compact deep features for wind turbine blade images taken by a drone. IPSJ Trans. Comput. Vis. Appl. 2019, 11, 3.
[CrossRef]
72. Kuang, N. Blade Detection Robot. In Proceedings of the 7th International Conference on Mechatronics, Computer and Education
Informationization (MCEI 2017), Shenyang, China, 3–5 November 2017; Atlantis Press: Amsterdam, The Netherlands, 2017;
pp. 317–321.
73. Franko, J.; Du, S.; Kallweit, S.; Duelberg, E.; Engemann, H. Design of a Multi-Robot System for Wind Turbine Maintenance.
Energies 2020, 13, 2552. [CrossRef]
74. Huiyi, Z.; Jackman, J. A Feasibility Study of Wind Turbine Blade Surface Crack Detection Using an Optical Inspection Method. In
Proceedings of the 2013 International Conference on Renewable Energy Research and Applications (ICRERA), Madrid, Spain,
20–23 October 2013; pp. 847–852.
75. Beganovic, N.; Söffker, D. Structural health management utilization for lifetime prognosis and advanced control strategy
deployment of wind turbines: An overview and outlook concerning actual methods, tools, and obtained results. Renew. Sustain.
Energy Rev. 2016, 64, 68–83. [CrossRef]
76. Kim, D.Y.; Kim, H.-B.; Jung, W.S.; Lim, S.; Hwang, J.-H.; Park, C.-W. Visual testing system for the damaged area detection of wind
power plant blade. In Proceedings of the ISR 2013, Seoul, Korea, 24–26 October 2013; pp. 1–5. [CrossRef]
77. Francis, D. Non-Destructive Evaluation (NDE) of Composites: Introduction to Shearography. In Non-Destructive Evaluation (NDE)
of Polymer Matrix Composites; Elsevier: Amsterdam, The Netherlands, 2013; pp. 56–83. [CrossRef]
78. Macedo, F.J.; Benedet, M.E.; Fantin, A.V.; Willemann, D.P.; da Silva, F.A.A.; Albertazzi, A. Inspection of defects of composite
materials in inner cylindrical surfaces using endoscopic shearography. Opt. Lasers Eng. 2018, 104, 100–108. [CrossRef]
79. Qiao, W.; Lu, D. A Survey on Wind Turbine Condition Monitoring and Fault Diagnosis—Part II: Signals and Signal Processing
Methods. IEEE Trans. Ind. Electron. 2015, 62, 6546–6557. [CrossRef]
80. Amezquita-Sanchez, J.P.; Adeli, H. Signal Processing Techniques for Vibration-Based Health Monitoring of Smart Structures.
Arch. Comput. Methods Eng. 2016, 23, 1–15. [CrossRef]
81. Sanati, H.; Wood, D.; Sun, Q. Condition Monitoring of Wind Turbine Blades Using Active and Passive Thermography. Appl. Sci.
2018, 8, 2004. [CrossRef]
82. Xin, Z. Simulation Research on Deep Learning in Structural Damage Identification of Fan Blades; Lanzhou Jiaotong University: Lanzhou,
China, 2016.
83. Wang, L.; Zhang, Z.; Xu, J.; Liu, R. Wind Turbine Blade Breakage Monitoring with Deep Autoencoders. IEEE Trans. Smart Grid
2016, 9, 2824–2833. [CrossRef]
84. Moreno, S.; Pena, M.; Toledo, A.; Trevino, R.; Ponce, H. A New Vision-Based Method Using Deep Learning for Damage Inspection
in Wind Turbine Blades. In Proceedings of the 15th International Conference on Electrical Engineering, Computing Science and
Automatic Control (CCE), Mexico City, Mexico, 5–7 September 2018.
85. Yanxia, L. Wind turbine blade crack identification based on migration learning. Sens. Microsyst. 2019, 38, 107–110.
Energies 2022, 15, 7492 17 of 17

86. Zhao, X.-Y.; Dong, C.-Y.; Zhou, P.; Zhu, M.-J.; Ren, J.-W.; Chen, X.-Y. Detecting Surface Defects of Wind Turbine Blades Using an
Alexnet Deep Learning Algorithm. Trans. Fundam. Electron. Commun. Comput. Sci. 2019, 102, 1817–1824.
87. Yang, X.; Zhang, Y.; Lv, W.; Wang, D. Image recognition of wind turbine blade damage based on a deep learning model with
transfer learning and an ensemble learning classifier. Renew. Energy 2020, 163, 386–397. [CrossRef]
88. Di Lorenzo, E.; Petrone, G.; Manzato, S.; Peeters, B.; Desmet, W.; Marulo, F. Damage detection in wind turbine blades by using
operational modal analysis. Struct. Health Monit. 2016, 15, 289–301. [CrossRef]
89. Moradi, M.; Sivoththaman, S. MEMS Multisensor Intelligent Damage Detection for Wind Turbines. IEEE Sensors J. 2014, 15,
1437–1444. [CrossRef]
90. Abouhnik, A.; Albarbar, A. Wind turbine blades condition assessment based on vibration measurements and the level of an
empirically decomposed feature. Energy Convers. Manag. 2012, 64, 606–613. [CrossRef]
91. González, A.G.; Fassois, S. A supervised vibration-based statistical methodology for damage detection under varying environ-
mental conditions & its laboratory assessment with a scale wind turbine blade. J. Sound Vib. 2016, 366, 484–500. [CrossRef]
92. Wang, Y.; Liang, M.; Xiang, J. Damage detection method for wind turbine blades based on dynamics analysis and mode shape
difference curvature information. Mech. Syst. Signal Process. 2014, 48, 351–367. [CrossRef]