Journal of Signal and Information Processing, 2013, 4, 430-438
Published Online November 2013 (http://www.scirp.org/journal/jsip)
http://dx.doi.org/10.4236/jsip.2013.44055
Open Access JSIP
Construction of Wind Turbine Bearing Vibration
Monitoring and Performance Assessment System
Feng-Tai Wu1*, Chun-Chieh Wang1, Jui-Hung Liu 2, Chia-Ming Chang2, Ya-Ping Lee1
1Mechanical and Systems Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan; 2Green Energy and
Environment Research Laboratories, Industrial Technology Research Institute, Hsinchu, Taiwan.
Received September 23rd, 2013; revised October 21st, 2013; accepted October 27th, 2013
Copyright © 2013 Feng-Tai Wu et al. This is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
This study is primary to develop relevant techniques for the bearing of wind turbin e, such as the intelligent monitoring
system, the performance assessment, future trend prediction and possible fault classification etc. The main technique of
system monitoring and diagnosis is divided into three algorithms, such as the performance assessment, performance
prediction and fault diagno sis, respectively. Among them, the Logistic Regression (LR) is adopted to assess the bearing
performance condition, the Autoregressive Moving Averag e (ARMA) is adopted to predict the future variation trend of
bearing, and the Support Vector Machine (SVM) is adopted to classify and diagnose the possible fault of bearing.
Through testing, this intelligent monitoring system can achieve real-time vibration monitoring, current performance
assessment, future performance trend prediction and possible fault classification for the bearing of wind turbine. The
monitor and analysis data and knowledge not only can be used as the basis of predictive maintenance, but also can be
stored in the database for follow-up off-line analysis and used as the reference for improvement of operation parameter
and wind turbine system design.
Keywords: Signal Processing; Feature Extraction; Performance Assessment; Performance Prediction; Fault Diagnosis
1. Introduction
At present, the comparatively attractive wind power gen-
erator can be mainly divided into onshore wind turbine
and offshore wind turbine in accordance with the setup
location. The maintenance time of wind turbine can be
divided into the predictable periodic maintenance and the
unpredictable breakdown repair. When the unpredictable
breakdown is occurred, such as the breakdown of gear-
box, several months may be required for the repair, lift-
ing, or waiting for material, which may cause large loss
of windfarm. In order to prevent the unpredictable break-
down, the intelligent monitoring and diagnosis analytical
technique can be used to assure the normal condition of
whole wind turbine, and the preventive maintenance
schedule can be arranged for key components [1-3].
The base for intelligent monitoring and diagnosis ana-
lytical technique of wind power generator system is the
setup of Condition Monitoring System (CMS). The pur-
pose of CMS is to monitor the operatio n condition of ke y
parts in each subsystem of wind turbine, such as the vi-
bration for the parts of transmission system and generator.
The physical signals are analyzed statistically through
breakdown diag no sis, in ord er to assess th e cond ition and
probable failure type of monitoring subject. Finally, the
alarm is provided in accordance with the analytical con-
clusion to prevent unpredictable breakdown and major
failure, in order to raise the usability and reliability of
whole wind turbine system. The CMS can grasp the con-
dition of wind turbine system to reduce the u npredictable
maintenance cost effectively, especially to the offshore
wind turbine system. Because the maintenance of off-
shore wind turbine system must be matched with the
schedule of fleet, and the uncertainty of weather shall
also be considered, so when the unpredictable breakdown
is occurred, extra maintenance expenses, schedule delay
and substitute power generation will cause large loss of
relevant firm. Through the advanced alarm of intelligent
maintenance system, not only the unpredictable break-
down can be reduced, but also the arrangement and de-
ployment of operation and maintenance schedule can be
*Corresponding a uthor.

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 431
conducted, in order to reach the requirements of fastest
maintenance achievement and lowest maintenance cost
by matching the maintenance facility and the manpower
deployment [4-7].
Figure 1 shows the annual failure rate and downtime
per failure for every subsystem in accordance with LWK
and WMEP wind turbine statistical database specified in
[5], wherein the statistical data for the operation status of
about 20,000 wind turbines operated in 13 years are
adopted. As shown in Figure 1, it is observed that there
is higher failure rate for the electric system and electric
control. Although th e failure rate is higher for these elec-
tric devices, the replacement cost is not high, the spare
parts are easy to be obtained, and the repair time of fail-
ure is pretty short, then the influence on downtime of
whole wind turbine system is not significant. On the con-
trary, the failure rate of large-scale devices such as gear-
box, drive train, generator, rotor blades is very low, less
than one time per year, but once these devices break
down, if there are no spare parts ready in advance, the
wind turbine must shut down and wait for the material,
which will elongate the downtime. Figure 2 shows the
comparison for the failure location of electric machine in
wind turbine. It is found that there is very high failure
rate at the part of bearing. So the condition monitoring
and fault diagnosis techniques have become very impor-
tant and been the center of attention [1-7].
The maintenance way of mechanical equipment can be
divided into the breakdown maintenance, time-based
maintenance, preventive maintenance, and predictive
maintenance in accordance with the maintenance timing.
Figure 3 shows the difference for comparison of periodic
maintenance and predictive maintenance. If the CMS is
added in relevant equipment of wind turbine, the main-
tenance period can be lengthened greatly compared to
traditional periodic maintenance. Not only many unnec-
essary maintenance times can be reduced, but the unpre-
dictable breakdown due to severe damage of components
can also be avoided. So, if the optimal predictive main-
tenance can be arranged in accordance with the monitor-
Figure 1. Annual failure rate and downtime per failure for
every subsystem of wind turbine [5].
Figure 2. Ratio for failure location of electric machine in
wind turbine [5].
Figure 3. Difference for comparison of periodic mainte-
nance and predictive maintenanc e [8-10].
ing and diagnosing condition, not only the downtime can
be reduced, but the optimal maintenance schedule can
also be planned, in order to shorten the failure time
[8-10].
2. Intelligent Monitoring System Structure
of Wind Turbine
The goal of intelligent monitoring system developed by
this study is to integrate the information and communica-
tion technique and equipment condition monitoring tech-
nique, and apply it to the intelligent maintenance for key
components of wind turbine. This intelligent system is
featured in the intelligent, determined and web-based
real-time monitoring and detecting function, which can
effectively raise the operation efficiency and mainte-
nance service ability of wind turbine.
The intelligent monitoring system of wind turbine
mainly provides the monitoring, predicting and prevent-
ing services, in order to avoid unpredictable breakdown
of wind turbine, and raises the usability of wind turbine
through predictive maintenance mechanism to maintain
stable power generation ability of wind turbine. The op-
eration mechanism of this system is to summarize rele-
vant operation data through the parameter collection de-
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 432
vice of wind turbine first, and transmit these data to
monitoring database of cloud computing platform throu gh
internet. Then, these data are analyzed, assessed, and
predicted by the computing modules of cloud computing
platform, and the analysis data are displayed by graph on
the operation interface of client end through internet. If
there is any abnormal condition, relevant operation per-
sonnel will be noticed through message or email, in order
to achieve the purpose of quickest troubleshooting.
The simple structure for intelligent monitoring system
of wind turbine is shown in Figure 4, which mainly in-
cludes the cloud computing platform (including Web-
based server and application server) and Data Acquisi-
tion (DAQ) equipment. The user can operate the system
by the mobile device, notebook or personal computer
through internet. In addition, the system can provide the
alarm function for the user to deal with abnormal condi-
tion of wind turbine in time. The user can also use real-
time graph to display the an alysis and co mparison d ata of
cloud computing, in order to grasp the performance con-
dition and relevant index of wind turbine totally. The
monitoring hardware setup of system is shown in Figure
5. An anemometer, a tachometer and a triaxial acceler-
ometer are installed at the top of tower. Two seismic ac-
celerometers are installed at the 1st floor and 2nd floor of
tower column, respectively. The triaxial accelerometer is
used to measure the vibration signal of main bearing, and
the seismic accelerometers are used to measure the vibra-
tion signal of tower column.
As shown in Figure 6, the monitoring system for per-
formance assessment and diagnosis of wind turbine
bearing is mainly composed of three computing modules.
First, the LR algorithm is adopted to assess the bearing
performance conditio n (confidence value, CV). Then , the
ARMA algorithm is adopted to predict the future per-
formance variation trend of bearing in accordance with
known bearing performance record. When possible fault
Figure 4. Structure for intelligent monitoring system of
wind turbine.
Figure 5. Monitoring hardware setup of wind turbine.
of bearing is occurred, the SVM algorithm is adopted to
classify and diagnose the possible fault of bearing. The
relevant technique for setting up intelligent monitoring
system is relatively mature and there are many successful
application examples, the required technique, such as
assessment, diagnosis, prediction, classification and data
mining can be found in [10-18]. The monitoring system
developed by this study will select suitable and stable
algorithms to conduct the development, integration, and
test analysis. The rules and preliminary test results of the
abovementioned three computing modules will be de-
scribed as follows.
3. Performance Assessment for Wind
Turbine Bearing
3.1. LR Algorithm [19]
The Logistic Regression (LR) algorithm is mainly used
to assess dichotomous problems. If the performance as-
sessment result of wind turbine lies between normal and
abnormal condition, this algorithm may be adopted. The
purpose of LR algorithm is to obtain the optimal model
for representing output dependent variable and
input independent variable x. Then, this model is repre-
sented by the logistic model to reveal the probability
of event represented by the output dependent
variable. This probability is ranged between 0 and 1.
is a logistic function, which is shown as Equation
(1).
()
gx
()Px
()Px
()
()
()
1e
() 1e 1e
g
g
P− ==
++
x
x
xgx . (1)
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 433
Figure 6. Module structure for performance assessment and
diagnosis of wind turbine bearing.
The logistic model can be shown as Equation (2).
()
()
112 2
log1
K
K
P
x
x
P
αβ ββ = +++⋅⋅⋅+
−
x
xx. (2)
From Equation (2) it is known that the logistic model
is , which is the linear combination of input inde-
pendent variable x and its corresponding coefficient β.
()
gx
This study will adopt the abovementioned logistic
function as the Confidence Value (CV) for as-
sessing the performance index of wind turbine bearing.
CV can be shown as Equation (3).
()Px
()
()
112 2
1
1e KK
xx x
CV
αβ ββ − +++⋅⋅⋅+
=+ x. (3)
where the variables x1 to xK represent the statistical fea-
ture of time domain and relevant bearing feature of fre-
quency domain calculated in accordance with the bearing
vibration signal. The corresponding coefficients α and β1
to βK should be obtained from the training in advance.
First, sufficient bearing vibration signals at normal con-
dition are selected to calculate the representative feature
group of time domain and frequency domain. The CV of
these feature groups is defined as 0.95 (95%). Then, suf-
ficient bearing vibration signals at abnormal condition
are selected to calculate the representative feature group
of time domain and frequency domain. Then the CV of
these feature groups is defined as 0.05 (5%). The corre-
sponding coefficien ts of logistic model are obtained fro m
these known feature groups at normal condition and ab-
normal condition. After the corresponding coefficients α
and β1 to βK are obtained by the maximum likelihood
estimator, the CV value of current bearing performance
index can be calculated by Equation (3).
3.2. Performance Assessment Test of Wind
Turbine Bearing
Figure 7 shows the monitoring database of wind turbines
tested in this study, and the record period was from
2012/11 2012/12 2013/01 2013/022013/03 2013/042013/05 2013/06 2013/07
0
10
20
30
Wind Speed
2012/11 2012/12 2013/01 2013/022013/03 2013/042013/05 2013/06 2013/07
0
10
20
30
RPM
2012/11 2012/12 2013/01 2013/022013/03 2013/042013/05 2013/06 2013/07
Date
0
0.5
1
CV
Figure 7. Wind turbine database of original monitoring and
assessment data.
2012/11 to 2013/06 . Among them, the wind speed ( m/sec)
and revolution (RPM) are direct recording values, and
CV values are calculated in according with the bearing
vibration signal at the corresponding time specified in
database. The assessment standard is based on β calcu-
lated by LR algorithm in accordance with the vibration
signal at given normal and abnormal condition and se-
lected CV index.
Upon observing the record variance in Figure 7, the
values are not recorded successfully in many intervals.
This is because the abnormal connection between field
monitoring equipment and database. These abnormal
intervals will cause the difficulty of off-line signal proc-
essing and analysis. So, the data in successful record in-
tervals are integrated to continuous time series, as 122-
day records shown in Figure 8.
Because the initial continuous data contain many
spikes in Figure 8, these spikes may be generated by the
disturbance of sensor or data acquisition equipment, the
median filter can be used to remove these spikes effec-
tively to obtain smooth monitoring and assessment data
of wind turbine, and the result is shown as Figure 9.
The fine solid line shown at upper part of Figure 10 is
the amplification result of CV ordinate of Figure 9. Upon
observing the CV variance trend, it is found that many
CV values assessed by LR will be shifted to around 0.8.
The reason for CV sh ift is that the original bearing vibra-
tion signals used for training LR coefficients do not rep-
resent normal operation condition of these intervals, in-
cluding constant speed at different revolution, accelerat-
ing and decelerating operation, and stationary condition
etc. According to the wind speed and revolution of these
intervals, the LR algorithm can complete the representa-
tion of the coefficients; then, it can be used to modify
previous CV values off-line, and the results are shown as
bold dotted line at upper part of Figure 10 and fine solid
line shown at lower part of Figure 10. The CV values
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 434
010 20 30 40 50 60 70 80 90100110120
0
10
20
30
Wind Speed
010 20 30 40 50 60 70 80 90100110120
0
10
20
30
RPM
010 20 30 40 50 60 70 80 90100110120
Day
0
0.5
1
CV
Figure 8. Wind turbine time series of continuous monitor-
ing and assessment data.
will not shift significantly after modification. Upon ob-
serving the variance of CV, it is composed of gentle deg-
radation trend and high frequency fluctuation. The itera-
tive Gaussian filter proposed in [20 ] is adopted to smooth
the historical records of CV to obtain the medium and
highly smooth CV trend shown as bold dotted line and
bold solid line at lower part of Figure 10 for the further
prediction step. The iterative Gaussian filter is used as a
low-pass filter to extract the non-sinusoidal part of the
historical records of CV. The cutoff frequencies are se-
lected as 0.2 cycle/day in medium smooth case and as
0.02 cycle/day in highly smooth case, respectively.
4. Performance Prediction of Wind
Turbine Bearing
4.1. ARMA Algorithm [21,22]
The Autoregressive Moving Average (ARMA) algorithm
is a system identification model. According to the col-
lected historical performance data of wind turbine bear-
ing, it is able to identify relevant parameters sufficient to
represent the performance behavior, and predict the fu-
ture performance trend in accordance with ARMA model.
ARMA model is composed of AR model and MA model,
which can be expressed in Equation (4), where y repre-
sents the output of system, x represents the input of sys-
tem, a and p represent the coefficient and order of AR
model, and b and q represent the coefficient and order of
MA model.
11
pq
ijijj
jj
yay bx
−
==
=+

ij−
(4)
In Equation (4), it is found that ARMA model is a
general equation of linear differential equation, which
can be used to represent common linear systems. A gen-
eral mechanical system can be simplified to the combi-
nation of many multi-degree-of-freedom systems. So,
010 20 3040 50 60 7080 90100110120
0
10
20
30
W ind Speed
010 20 3040 50 60 7080 90100110120
0
10
20
30
RPM
010 20 3040 50 60 7080 90100110120
Day
0
0.5
1
CV
Figure 9. Wind turbine time series of smoothed monitoring
and assessment data.
010 20 30 40 50 60 708090100 110120
Day
0.6
0.7
0.8
0.9
1
CV
010 20 30 40 50 60 708090100 110120
Day
0.86
0.88
0.9
0.92
0.94
0.96
0.98
1
CV
Figure 10. Off-line processing for assessment data of wind
turbine bearing.
ARMA model is often used to represent the feature of
mechanical system at time domain and frequency do-
main.
4.2. Performance Trend Prediction Test of Wind
Turbine Bearing
Figure 11 shows the use of 90-day historical CV data for
training ARMA model and future performance trend
prediction result of next 32 days. Figure 12 shows the
use of 100-day historical CV data for training ARMA
model and future performance trend prediction result of
next 22 days. The historical CV data are divided into two
sections. The first section is used for training and shown
as blue solid line, and the seco nd section is used for test-
ing and shown as green dashed line. The performance
trend forecasted by ARMA model is shown as red dotted
line. Upon observing CV trend of actual assessment, it is
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 435
Figure 11. Performance trend prediction result of wind
turbine bearing: 90 days of training data.
Figure 12. Performance trend prediction result of wind
turbine bearing: 100 days of training data.
found that CV has significan t degradation tr end about 80
days after. So, th e prediction deviation is larg er by using
90-day historical CV data for training ARMA model.
After comparison, the prediction accuracy is higher by
using 100-day historical CV data for training ARMA
model. In Figure 11, the performance predictions are
overestimated by using the original and medium smooth
historical CV data, and the prediction is underestimated
by using the highly smooth historical CV data. In Figure
12, the performance predictions are correlated well to the
testing data by using the original and highly smooth his-
torical CV data, but the prediction is still overestimated
by using the medium smooth data.
The historical CV smoothing degree will also influ-
ence the prediction result of ARMA model. So, the sys-
tem developed by this study will re-train the prediction
ARMA model regularly and determine the influence of
different CV historical data length and smoothing degree
on prediction ARMA model. Then, the system uses the
rule as the basis of dynamic ARMA parameter adjust-
ment and verifies the reliability of prediction model.
5. Fault Classification of Wind
Turbine Bearing
5.1. SVM Algorithm [23-25]
The Support Vector Machine (SVM) algorithm is a su-
pervised learning classifier derived from the statistical
learning theory. Unlike other learning network, the goal
of SVM is to minimize structural risk not to minimize
empirical risk. So, it is guaranteed as the optimal classi-
fier after learning process. SVM mainly uses binary clas-
sification way to classify the samples. As shown in Fig-
ure 13, SVM transfers the samples from original space to
the feature space for really classifying the samples by a
linear hyperplane, and guarantees that two classes of
samples have equal distance from the plane. Finally, this
hyperplane is transferred back to original space to obtain
optimal nonlinear classification curve for two classes of
samples. If more than two classes of classification are
required, the binary SVM classifier can be expanded to
multiple SVM classifiers. Figure 14 shows the structure
of Decision Directed Acyclic Graph (DDAG) SVM clas-
sifier, which can be used for classifying 4 classes of fault.
In Figure 14, represents the binary classifier of
i class and j class. According to the directed decision
flow to conduct several binary classification, the sample
x will finally be able to be classified successfully.
()
ij
Dx
5.2. Bearing Fault Classification Test
Figure 15 shows the bearing vibration signal and its short-
time Fourier spectrogram. In the figure, the 6-second
Figure 13. Binary SVM classifier [23].
D12(x)
D14(x) D32(x)
D43(x) D42(x)D13(x)
Class 1Class 3Cl ass 4Class 2
1, 2, 3, 4
1, 3, 42, 3, 4
1, 32 , 4
3, 4
NOT 2NOT 1
NOT 4NOT 1NOT 3
NOT 2
NOT 1NOT4 NOT 3NOT3 NOT 4NOT 2
Figure 14. Multiple DDAG SVM classifier [24].
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 436
00.5 11.5 22.5 33.5 44.5 55.5 6
Time [sec]
-0.5
0
0.5
Vibration
00.5 11.5 22.5 33.5 44.5 55.5
Time [sec]
0
1000
2000
3000
4000
5000
6000
Frequency [Hz]
Figure 15. Bearing vibration signal and short-time Fourier
analysis.
vibration signal is composed of three 2-second segments,
in order to compare the bearing vibration waveform and
frequency component between normal and abnormal con-
dition. From the vibration signal, the first segment repre-
sents normal bearing, the middle one slightly abnormal
bearing, and the last one significant abnormal bearing.
Upon observing the amplitude variance of waveform
and spectrogram, it is found when the bearing is abnor-
mal, the amplitude modulation will be occurred, and the
modulated frequency will be correlated to bearing char-
acteristic frequency. Figure 16 shows the comparison
result of vibration envelope spectrum and characteristic
frequencies for bearing vibration signal specified in Fig-
ure 15. In the figure, BPFO represents the ballpass fre-
quency at outer race, BPFI represents the ballpass fre-
quency at inner race, BSF represents the ballspin fre-
quency, and FTF represents the fundamental frequency.
The calculation is shown in Equations (5)-(8) [26].
BPFO1 cos
2r
nd
f
D
φ

=−




. (5)
BPFI1 cos
2r
nd
f
D
φ

=+




. (6)
2
BSF1 cos
2r
Dd
f
dD
φ


=−






. (7)
FTF1cos
2r
nd f
D
φ

=−




. (8)
where n is ball number, d is ball diameter, D is bearing
pitch diameter,
φ is contact angle, and fr is rotation fre-
quency of inner race.
Figure 17 shows multiple SVM classification test re-
sult of bearing vibration fault features, in which 8 kinds
Figure 16. Comparison of vibration envelope spectrum and
characteristic frequencies for bearing vibration signal.
Figure 17. Multiple SVM classification test result of bearing
vibration fault features.
of known fault feature are used, including normal condi-
tion and other fault combination. In the figure, the ab-
scissa is the first principle component PC1 of signal fea-
ture, and the ordinate is the second principle component
PC2 of signal feature. The center lines represent the clas-
sification curves of fault. The results show that multiple
SVM classifiers can distinguish different bearing fault
classification effectively. The feature series can be ob-
tained after comparing and analyzing the bearing vibra-
tion behaviors at time domain and frequency domain. In
order to represent the bearing fault sufficiently, at least
10 kinds of feature are required in general. If the original
feature series is used to conduct the classification directly,
not only the SVM classification computing efficiency
will be poor due to high dimension of feature series, but
also the distribution of feature space may be overlapped
to cause difficult SVM classification. So, the Principle
Component Analysis (PCA) will be matched with SVM
classifier to reduce the dimension of feature series, where
only the most key components will be used for faster and
more accurate classification [16-18].
6. Conclusion
The final purpose of this study is to set up the intelligent
monitoring system for predictive maintenance of wind
turbine, particularly the monitoring, assessment, and di-
agnosis for key components, such as bearing and gearbox,
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System 437
so that the user can predict the best replacement timing
of the abnormal component in advance, and grasp the
usability and maintenance cost. The cond ition monitoring
and prediction system of wind turbine bearing integrates
three functional modules. Firstly, the real-time bearing
performance assessment module retrieves current bearing
vibration signal from wind turbine database and adopts
the LR algorithm to access current CV value. Then, it
transfers CV value back to the database. Secondly, when
the CV historical data of bearing performance are accu-
mulated to a sufficient number, the ARMA algorithm is
adopted to predict the future variation trend of bearing
performance in accordance with known CV historical
data of bearing performance and assure if the predicted
performance is lower than the threshold value. Then, it
transfers relevant predicted data back to the database.
Thirdly, when the identification for reason of bearing
performance redu ction is required, the SVM algorithm is
adopted to determine the fault classification through re-
trieving bearing vibration signal at fault interval from
wind turbine database. Then, it transfers fault condition
data and corresponding records back to the database in
order to strengthen the completeness of bearing fault
condition database, which not only can modify the coef-
ficient correctness of LR algorithm, but also can be used
as the reference for operation parameter adjustment and
modification of follow-up design.
7. Acknowledgements
The financial support provided by Bureau of Energy
(Grant No.102 -D 01 0 5) is gratefully acknowledged.
REFERENCES
[1] P. Tavner, “Offshore Wind Turbines: Reliability, Avail-
ability and Maintenance,” The Institution of Engineering
and Technology, London, 2012.
[2] P. A. Lynn, “Onshore and Offshore Wind Energy: An
Introduction,” John Wiley & Sons Ltd., Chichester, 2012.
[3] R. Billinton, R. Karki and A. K. Verma, “Reliability and
Risk Evaluation of Wind Integrated Power Systems,”
Springer, London, 2013.
http://dx.doi.org/10.1007/978-81-322-0987-4
[4] C. R. Farrar and K. Worden, “Structural Health Monitor-
ing: A Machine Learning Perspective,” John Wiley &
Sons Ltd., Chichester, 2013.
[5] C. J. Crabtree, “Condition Monitoring Techniques for
Wind Turbines,” Ph.D. Thesis, Durham University, Dur-
ham, 2011.
[6] F. P. G. Márquez, A. M. Tobias, J. M. P. Pérez and M.
Papaelias, “Condition Monitoring of Wind Turbines:
Techniques and Methods,” Renewable Energy, Vol. 46,
2012, pp. 169-178.
http://dx.doi.org/10.1016/j.renene.2012.03.003
[7] K. Tracht, G. T. Goch, P. Schu h, M. Sor g and J. F. W e s t e r-
kamp, “Failure Probability Prediction Based on Condition
Monitoring Data of Wind Energy Systems for Spare Parts
Supply,” Manufacturing Technology, Vol. 62, 2013, pp.
127-130.
[8] D. N. P. Murthy and K. A. H. Kobbacy, “Complex Sys-
tem Maintenance Handbook,” Springer, London, 2008.
[9] R. Manzini, A. Regattieri, H. Pham and E. Ferrari, “Main-
tenance for Industrial Systems,” Springer, London, 2010.
http://dx.doi.org/10.1007/978-1-84882-575-8
[10] H. Czichos, “Handbook of Technical Diagnostics: Fun-
damentals and Application to Structures and Systems,”
Springer, London, 2013.
http://dx.doi.org/10.1007/978-3-642-25850-3
[11] V. Palade, C. D. Bocaniala and L. Jain, “Computational
Intelligence in Fault Diagnosis,” Springer, London, 2006 .
http://dx.doi.org/10.1007/978-1-84628-631-5
[12] V S. Nandi, S. Choi and H. Meshgin-Kelk, “Electric Ma-
chines: Modeling, Condition Monitoring, and Fault Di-
agnosis,” CRC Press, New York, 2012.
[13] D. J. Inman, C. R. Farrar, V. L. Junior and V. S. Junior,
“Damage Prognosis: For Aerospace, Civil and Mechani-
cal Systems,” John Wiley & Sons Ltd., Chichester, 2005.
http://dx.doi.org/10.1002/0470869097
[14] B. L. Song and J. Lee, “Framework of Designing an Ada-
ptive and Multi-Regime Prognostics and Health Man-
agement for Wind Turbine Reliability and Efficiency Im-
provement,” International Journal of Advanced Com-
puter Science and Applications, Vol. 4, No. 2, 2013, pp.
142-149.
[15] G. Vachtsevanos, F. L. Lewis, M. Roemer, A. Hess and B.
Wu, “Intelligent Fault Diagnosis and Prognosis for Engi-
neering Systems,” John Wiley & Sons Ltd., New Jersey,
2006. http://dx.doi.org/10.1002/9780470117842
[16] M. Kantardzic, “Data Mining: Concepts, Models, Meth-
ods, and Algorithms,” 2nd Edition, John Wiley & Sons
Ltd., New Jersey, 2011.
http://dx.doi.org/10.1002/9781118029145
[17] P. Xanthopoulos, P. M. Pardalos and T. B. Trafalis, “Ro-
bust Data Mining,” Springer, London, 2013.
http://dx.doi.org/10.1007/978-1-4419-9878-1
[18] T. Hastie, R. Tibshirani and J. H. Friedman, “The Ele-
ments of Statistical Learning: Data Mining, Inference,
and Prediction,” 2nd Edition, Springer, London, 2009.
[19] J. Yan, “Degradation Assessment and Fault Modes Clas-
sification Using Logistic Regression,” Journal of Manu-
facturing Science and Engineering, Vol. 127, No. 4, 2005,
pp. 912-914. http://dx.doi.org/10.1115/1.1962019
[20] Y. N. Jeng, P. G. Huang and Y. C. Cheng, “Decomposi-
tion of One-Dimensional Waveform Using Iterative Gaus-
sian Diffusive Filtering Methods,” Proceedings of the
Royal Society A, Vol. 464, No. 2095, 2008, pp. 1673-
1695. http://dx.doi.org/10.1098/rspa.2007.0031
[21] P. C. Young, “Recursive Estimation and Ti me-Series Ana -
lysis: An Introduction for the Student and Practitioner,”
2nd Edition, Springer, London, 2011.
http://dx.doi.org/10.1007/978-3-642-21981-8
[22] M. Najim, “Modeling, Estimation and Optimal Filtration
in Signal Processing,” John Wiley & Sons Ltd., New Jer-
Open Access JSIP

Construction of Wind Turbine Bearing Vibration Monitoring and Performance Assessment System
Open Access JSIP
438
sey, 2010.
[23] L. Wang, “Support Vector Machines: Theory and Appli-
cations,” Springer, New York, 2010.
[24] S. Abe, “Support Vector Machines for Pattern Classifica-
tion,” 2nd Edition, Springer, New York, 2010.
http://dx.doi.org/10.1007/978-1-84996-098-4
[25] C. Campbell and Y. Ying, “Learning with Support Vector
Machines,” Morgan & Claypool Publishers, 2011.
[26] C. Scheffer and P. Girdhar, “Practical Machinery Vibra-
tion Analysis and Predictive Maintenance,” Elsevier,
2004.