Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
Vol.
4, No. 6, Decem
ber
2014, pp. 882~
892
I
S
SN
: 208
8-8
7
0
8
8
82
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Wind Turbine F
a
ult Det
ecti
o
n S
y
st
em in Real Ti
me Rem
o
t
e
Monitoring
Sa
ad Ch
ak
ko
r
*
, Mo
staf
a Ba
ghouri
*
, Abderrahmane
Hajraoui
*
*
Univers
i
t
y
of Abdelm
alek Es
s
aâdi
,
F
a
cult
y
of
Sciences, Dep
a
rtment of Ph
y
s
ics,
Communication
and Detection S
y
stem
s Laborato
r
y
,
Tetou
a
n, Morocco
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
J
u
l 15, 2014
Rev
i
sed
Sep
15
, 20
14
Accepted Oct 10, 2014
In new energ
y
development,
wind pow
er has boomed. It is due to the
prolifer
ation of
wind parks and their oper
a
tion
in supply
i
ng
the nation
a
l
ele
c
tri
c
gr
id with low cos
t
and c
l
ean res
ourc
e
s
.
Hence, th
ere is
an incre
a
s
e
d
need to
estab
lis
h a proa
ctiv
e m
a
int
e
nanc
e for
wind turbine m
a
chines base
d
on remote con
t
r
o
l and monitorin
g
. That
is n
ecess
a
r
y
with
a real-time wireless
connection in of
fshore or inaccessible
locations
while th
e wired
method has
man
y
flaws. The objective of th
is stra
teg
y
is to pr
olong wind turbine lifetime
and to
incr
eas
e productivity
.
The h
a
rdware of a r
e
mote
control
and
monitoring s
y
stem for wind turb
ine parks
is designed. It takes ad
vantag
e of
GPRS or Wi-
M
ax wireless module to
collect data m
easu
r
ements from
differen
t
wind m
achine sensors through IP based m
u
lti-hop com
m
unication
.
Computer simul
a
tions with Proteus IS
IS and
OPNET software tools hav
e
been condu
cted
to evalu
a
te the performance of
the studied s
y
s
t
em. Stu
d
y
findings show that
the d
e
signed
device
is
suit
ab
le for
appl
ic
atio
n in
a wind
park.
Keyword:
Em
bedded
Sy
s
t
em
Fau
lts Diagn
o
s
is
Mo
n
itoring
Real Tim
e
TC
P/
IP Pr
ot
oc
ol
Web
Wi
n
d
T
u
r
b
i
n
e
Copyright ©
201
4 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Saad C
h
a
k
kor,
Uni
v
ersity of
Abdelm
alek E
ssaâdi, Faculty of
Sciences
, Departm
e
nt
of
Physics,
C
o
m
m
uni
cat
i
o
n a
n
d
Det
ect
i
o
n Sy
st
em
s Lab
o
rat
o
ry
,
Tet
o
u
a
n, M
o
r
o
cc
o
Em
a
il: saad
ch
ak
kor@g
m
a
i
l
.co
m
1.
INTRODUCTION
In
fro
n
t
o
f
th
e
h
u
g
e
in
crease
d
e
m
a
n
d
in
en
erg
y
ov
er th
e wo
rl
d
,
and
in
o
r
d
e
r to
search
a
su
bstitu
tio
n
a
l
kind of ene
r
gy
against the
prices ri
se of the
energy fossil fuels re
source
s an
d
its exh
a
ustio
n
reserv
es
in
th
e
l
o
n
g
t
e
rm
. Fu
rt
herm
ore,
t
h
e
com
m
itm
ent
of
t
h
e
g
ove
r
n
m
e
nt
s t
o
red
u
ce g
r
een
h
o
u
s
e
gases
em
i
ssi
ons
has
favore
d
the res
earch of othe
rs
energy
so
ur
ce
s. The
rec
o
u
r
s
e
t
o
re
newa
bl
e
energy bec
o
mes a societal choice.
Th
e d
e
v
e
l
o
p
m
en
t o
f
t
h
is alter
n
ativ
e is enco
ur
ag
ed
b
ecau
s
e it o
f
f
e
r
s
n
a
tur
a
l, economic, clean
an
d
saf
e
resources. Am
ong the re
ne
wable energi
es,
wind ene
r
gy has been
progre
sse
d
in
a rem
a
rk
ab
le way in
th
ese
recent years
.
It provi
d
es a c
onsi
d
era
b
le electrical
energy production with less
expe
nse a pa
rt from the
co
nstru
c
tion
and
m
a
in
ten
a
n
ce
b
udg
et.
Nowadays, wind
en
erg
y
inv
e
stm
e
n
t
h
a
s in
creased
by th
e m
u
ltip
lic
atio
n
o
f
t
h
e
wind
p
a
rk
s cap
acities.
Th
is co
n
t
ribu
tes greatly to
th
e exp
a
n
s
i
o
n
of t
e
rrestrial an
d offsh
o
re
wind
parks
whic
h are
us
ually installed in far l
o
cations
, di
ffi
c
u
lt to access and subject to
e
x
trem
e environm
ental
conditions [1]
.
W
i
nd
turbines contain a com
p
lex elect
rom
ecanical
s
y
ste
m
which is prone to defects.
There
f
ore, t
h
e
i
r
m
oni
t
o
ri
ng
and di
a
g
nosi
s
becom
e
essent
i
a
l
t
o
redu
ce
m
a
i
n
t
e
nanc
e cost
s an
d ensu
re
co
n
tinu
ity of
pr
odu
ctio
n b
ecau
s
e st
o
p
p
i
ng
a
w
i
nd
in
sta
llatio
n fo
r un
ex
p
e
cted
failures cou
l
d
lead
to expen
s
iv
e
repai
r
a
n
d t
o
l
o
st
p
r
o
d
u
ct
i
on
[2]
.
T
h
i
s
o
p
e
r
a
t
i
ng st
o
p
p
i
n
g b
ecom
e
s cri
t
i
cal and ca
uses
ve
ry
si
gni
fi
cant
l
o
sses
.
For these
reas
ons
, the
r
e is a
n
inc
r
ease
nee
d
to im
ple
m
en
t a rob
u
s
t efficien
t rem
o
te
main
ten
a
nce strateg
y
to
g
u
a
ran
t
ee un
interrup
t
ed power in
t
h
e m
o
d
e
rn
wind
system
s [3
]. Th
is on
lin
e su
rv
eillan
ce allows an early
det
ect
i
on
of
m
echani
cal
an
d el
ect
ri
cal
faul
t
s
. It
m
u
st
be abl
e
t
o
pre
v
ent
m
a
jor c
o
m
ponent
fai
l
u
r
e
s. That
facilitates a p
r
o
activ
e respo
n
se, an
ticip
ates
th
e fin
a
l sh
u
t
do
wn
o
f
wind
gen
e
rat
o
rs, m
i
n
i
mizes d
o
w
n
time an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE Vol. 4, No. 6, D
ecem
ber 2014
:
882 – 892
8
83
m
a
xim
i
zes pro
duct
i
v
i
t
y
by
anal
y
z
i
ng c
ont
i
n
u
o
u
sl
y
t
h
e
m
easure
d
p
h
y
s
i
c
al
si
gnal
s
col
l
ect
ed fr
om
di
ffere
nt
typ
e
s of sen
s
o
r
s [4
],
[5
],
[6
].
Th
is is
wh
y rel
i
ab
ility o
f
wind
turb
in
es b
e
co
m
e
s an
im
p
o
r
tan
t
top
i
c in scien
tific
researc
h
a
n
d in industry. Mos
t
of the
rece
nt
researc
h
es
ha
v
e
bee
n
ori
e
nt
e
d
t
o
wa
rd
el
ect
r
i
cal
m
oni
t
o
ri
n
g
, a
s
i
t
wo
ul
d
be t
h
e
m
o
st
pract
i
cal t
echni
q
u
e an
d l
e
ss cost
l
y
. An
ot
he
r p
o
w
e
r
f
u
l
t
ool
use
d
for
di
ag
no
si
s of an
in
du
ctio
n
m
o
tor or g
e
n
e
rat
o
r i
s
cu
rren
t stator an
alysis (CSA
) [1
], [4
],
[5
],
[6
], [7
].
It u
tilizes th
e resu
lt of th
e
spectral a
n
alysis of the
stator
curren
t t
o
indi
cate an e
x
isting
or incipie
n
t
failure.
M
o
re
over, with recent
digital
si
gnal
pr
ocess
o
r
(D
SP
) an
d
wi
rel
e
ss c
o
m
m
uni
cat
i
o
n t
ech
nol
ogy
de
vel
o
pm
ent
s
, i
t
i
s
possi
bl
e t
o
det
e
ct
el
ect
ri
c
mach
in
e fau
lts p
r
ior to
p
o
s
sib
l
e catastrop
h
i
c failu
re in
re
al
-t
im
e based on t
h
e st
at
or l
i
ne cur
r
e
n
t
al
l
o
wi
ng
preci
se an
d l
o
w-c
o
st
[
7
]
.
Th
e
m
a
i
n
object
i
v
e o
f
t
h
i
s
pape
r i
s
t
o
st
udy
t
h
e desi
g
n
of a r
eal
t
i
m
e
m
oni
tori
ng a
n
d
co
n
t
ro
lling
syste
m
fo
r state su
p
e
rv
isio
n
o
f
wind
g
e
n
e
rato
r m
ach
in
es wh
ich
in
tegrates in
tellig
en
ce and
r
obu
stn
e
ss functio
n
s
.
2.
RELATED WORK
In th
e literatu
re rev
i
ew
, few of
research stu
d
i
es
h
a
v
e
b
e
en
d
e
v
e
lop
e
d to
an
alyze
th
e
th
eo
retical
asp
ects
of the app
licatio
n part in co
nd
ition
m
o
n
itoring
o
f
wi
n
d
turb
ines op
eratin
g
[1
],
[8
], [9
],
[16
]
.
As
kn
o
w
n
,
t
h
ese f
a
ul
t
s
cause a
m
odul
at
i
on i
m
pact
i
n
t
h
e m
a
gnet
i
c
fi
el
d
of
t
h
e wi
n
d
ge
ne
r
a
t
o
r
,
w
h
i
c
h i
s
r
e
fl
ect
ed
by the appeara
n
ce of a signifi
cant ha
rm
onics
(pea
ks) i
n
the
stator curre
nt
s
p
ectrum
[3]. Som
e
research
works
are ap
pl
y
i
ng
e
nha
nce
d
si
g
n
al
pr
ocessi
ng t
e
c
hni
que
s l
i
k
e F
a
st
Fo
uri
e
r
T
r
a
n
sf
orm
m
e
t
hod (
FFT
), S
h
ort
T
i
m
e
Fou
r
ier
T
r
an
sfo
r
m
(STFT),
period
og
ram
,
Discrete
W
a
v
e
let
T
r
an
sform
(DWT),
W
i
gn
er
-V
ille representatio
n
,
C
onc
or
di
a
T
r
a
n
sf
orm
(C
T)
and t
h
e Hi
l
b
e
r
t
-
H
u
a
ng t
r
ans
f
o
r
m
and ot
h
e
r ad
va
nced t
ool
s
base
d o
n
wi
n
d
g
e
n
e
rator stator curren
t
to
d
i
ag
no
se
p
r
o
s
p
e
ctiv
e electro
mech
an
ical fau
l
t
s
un
d
e
r tran
sien
t con
d
ition
s
[3
], [6
],
[1
4]
, [
1
5]
. B
e
si
des, i
n
[2]
a
st
at
i
s
t
i
cal
di
agn
o
si
s a
p
pr
oa
ch i
s
pr
o
p
o
s
e
d
base
d
on
re
si
dues
anal
y
s
i
s
o
f
t
h
e
electrical
m
a
c
h
ine
state va
riables
by the
us
e of the Pr
in
ci
p
a
l Co
m
p
on
ents
An
alysis m
e
th
od
(PCA)
for fau
lts
det
ect
i
o
n
i
n
o
f
fs
ho
re
wi
nd
t
u
r
b
i
n
e
gene
rat
o
r
.
The
m
a
i
n
dra
w
back
of
t
h
i
s
a
p
pr
oach
i
s
t
h
at
t
h
e
det
ect
i
on
ef
ficiency
re
quires
a
good
choice
of the
pri
n
cipal
co
mp
on
en
ts
nu
mb
e
r
. So
me
r
e
s
e
a
r
c
h
er
s a
r
e
p
r
o
p
o
s
ed
f
a
ilu
r
e
s
d
i
agnosis an
d
m
o
n
itor
i
ng
of
w
i
n
d
tu
rb
in
es
g
e
n
e
r
a
to
r
s
u
s
ing
im
p
e
d
a
n
ce
sp
ect
r
o
scop
y (
I
S)
[
17-
21
].
Due
m
a
i
n
l
y
t
o
t
h
ei
r a
d
vant
a
g
es,
hi
g
h
res
o
l
u
t
i
on
m
e
t
hods
(
H
R
M
)
suc
h
a
s
M
U
SIC
,
E
SPR
IT
a
n
d t
h
ei
r z
o
om
i
n
g
t
echni
q
u
es
hav
i
ng
hi
g
h
acc
u
r
acy
can
det
ect
fre
que
nci
e
s
wi
t
h
l
o
w
S
N
R
[
2
1]
,
[2
2]
,
[
2
4]
,
[
30]
.
T
h
ey
ha
ve
b
een
recently introduced in t
h
e a
r
ea of induction m
o
tors a
n
d
wind
gene
ra
tors fa
ults
diagnosis to im
prove the
det
ect
i
on
of a l
a
r
g
e
n
u
m
b
er of
fre
que
nci
e
s i
n
a gi
ve
n ba
n
d
w
i
dt
h.
Ot
he
r p
r
a
c
t
i
cal
research
t
echni
q
u
es a
r
e
use
d
to
so
lve on
line
m
a
in
ten
a
n
c
e p
r
o
b
l
em
s with
th
e
h
e
lp
o
f
real ti
m
e
b
y
i
n
tegratio
n
o
f
t
h
e
W
i
reless
Sen
s
o
r
Net
w
or
k a
n
d t
h
e i
n
d
u
st
ri
al
c
o
m
m
uni
cat
i
on pr
ot
o
c
ol
s i
n
wi
n
d
t
u
r
b
i
n
es
di
ag
no
si
s [
1
0-
13]
,
[
20]
,
[
23]
.
T
h
i
s
p
a
p
e
r
p
r
esen
ts an
in
tellig
en
t
rem
o
te
m
o
n
ito
ring
and
fau
lt
d
i
agn
o
sis system fo
r
wind
turb
in
e
u
s
ing
real
ti
me
har
d
ware base
d on
a wi
rel
e
ss
com
m
uni
cat
i
on
m
odul
e.
3.
PROBLEM FORMUL
ATION
The c
u
r
r
e
n
t
m
a
i
n
t
e
nan
ce
pl
anni
ng
ba
sed
o
n
l
i
n
e m
oni
t
o
ri
ng
sy
st
em
s used i
n
wi
nd
pa
rks
i
s
n
o
t
o
p
tim
ized
. It h
a
s sev
e
ral limi
t
atio
n
s
bo
th
in ter
m
s o
f
p
e
rform
a
n
ce an
d
installatio
n
co
st.
Th
is is th
e resu
lt of
vari
ous
reas
o
n
s
:
C
a
bl
e st
ruct
ure
m
oni
t
o
ri
ng
wi
t
h
i
t
s
pr
obl
em
s (cut
s,
noi
se
, a
nd c
o
nfi
g
u
r
at
i
o
n
)
, m
a
i
n
l
y
t
h
rou
g
h
fi
be
r
opt
i
c
,
is no
t ef
f
ective and no
t app
r
o
p
r
i
ate sin
c
e
w
i
nd
p
a
r
k
s ar
e of
ten
d
e
p
l
o
y
ed
ov
er a lar
g
e g
e
og
r
a
ph
ic area
havi
ng
en
vi
r
o
nm
ent
a
l
obst
a
cl
es suc
h
as
o
f
fs
ho
re,
de
sert
,
m
ount
ai
ns,
ri
vers
,
fo
rest
s a
n
d
pl
ai
n
s
t
h
at
ar
e
located
far awa
y
from
the cont
rol ce
nter
.
W
i
n
d
turb
in
es
m
o
n
ito
rin
g
n
e
ed
t
o
im
p
l
e
m
en
t a
pro
activ
e main
ten
a
nce syst
e
m
b
a
sed
on
an
early
,
fast and
real
t
i
m
e
faul
t
det
ect
i
o
n
an
d
di
ag
no
si
s, a
l
l
o
wi
n
g
a
sec
u
re
an
d
rel
i
a
b
l
e com
m
uni
cat
i
on
fo
r
bet
t
e
r
m
a
i
n
t
e
nance m
a
nagem
e
nt
.
T
h
i
s
st
rat
e
gy
a
v
oi
ds
fai
l
u
re
s l
ead
i
n
g
t
o
seri
o
u
s
dam
a
ges, ex
pe
nsi
v
e
re
pai
r
a
n
d
pr
o
duct
i
o
n l
o
ss
.
The storage of the m
onitore
d pa
ram
e
ters in the c
ontrol
center
databas
e
is a need,
because it is an
essential ope
ra
tion which he
lps
t
o
m
onitor
accurate
ly the lifetim
e of t
h
e
wind turbi
n
e c
o
m
p
onent
s
.
Th
erefo
r
e, it al
lo
ws ex
p
l
o
itin
g
co
llected d
a
t
a
fo
r st
u
d
y
ing
statistical
ly th
e m
o
st o
c
cu
rring
fau
lts and their
t
i
m
i
ng.
To ove
r
com
e
these im
posed application c
onst
r
aint
s, a
n
interactive embedde
d
system has bee
n
desi
g
n
e
d
t
o
pr
ovi
de an ef
fi
ci
ent
,
rel
i
a
bl
e an
d eco
nom
i
cal
li
nk bet
w
een
v
a
ri
o
u
s wi
n
d
t
u
rbi
n
e sens
ors f
o
r a
n
accurate
rem
o
te controlling a
n
d m
onitoring
schem
e
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Wind Turbi
n
e
Fault Detection
System
in Re
al Time
Re
mot
e
Monitoring
(
S
aad
C
h
a
k
ko
r)
88
4
4.
MONITORE
D PARAMETERS
IN WIND
TURBINE
Since the
wind turbine m
ach
ine is a com
p
lex system
, it co
ntains a lar
g
e
num
ber
of
para
m
e
ters to be
m
o
n
ito
red
.
Hen
ce, to
pu
rsu
e
an
d
t
o
con
t
ro
l
all fu
n
c
tion
s
of th
e wi
n
d
t
u
rbin
e in
ord
e
r to en
sure its op
tim
u
m
ope
ration at a
ny wind
velocit
y
, there is
a strong
nee
d
t
o
install a set of se
nsors t
h
at c
o
llect perform
a
nce data
o
f
th
e wi
n
d
turb
i
n
e co
n
t
i
n
uou
sly and th
en
tran
sm
its it to
an
i
n
tern
al
con
t
ro
ller
wh
ich
is d
e
si
g
n
e
d
t
o
allow
local supervision a
n
d comm
a
nd i
n
case the
machine functi
ons a
r
e re
quire
d
. In case
of a
n
om
alies or errors, the
dat
a
i
s
st
ore
d
i
n
a dat
a
base
or
l
og al
arm
,
al
l
o
wi
n
g
er
ro
r a
n
a
l
y
s
i
s
of wi
n
d
m
achi
n
e.
T
h
e
m
oni
t
o
red
para
m
e
t
e
rs
are ac
q
u
i
r
e
d
t
h
r
o
ug
h
A
D
C
devi
ces
c
o
n
n
e
c
t
e
d
by
m
ean
s of th
e d
i
r
ect
co
nn
ection or
b
y
RS23
2 an
d USB
interfaces
of t
h
e c
o
m
puter
or
by a
wireless
connecti
o
n.
Whe
n
a
failure occ
u
rs
insi
de the
wind t
u
rbine, the
cont
rol
l
e
r
reac
t
s
qui
c
k
l
y
t
o
a
voi
d t
h
e e
v
ol
u
t
i
on
of a
n
y
t
y
pe
of
defect
s
.
In t
h
i
s
o
p
t
i
c
, a
ro
b
u
st
an
d
re
l
i
a
bl
e
syste
m
m
u
st be
devel
o
ped
leading to the
possibility
of real tim
e and detecting fa
ults accuracy i
n
thei
r
in
cip
i
en
t stag
es b
e
fore th
ey
tran
sform
in
to
m
o
re
seri
o
u
s
fai
l
u
res
ca
usi
n
g
un
desi
ra
bl
e d
o
w
nt
i
m
es an
d
dam
a
ges.
T
a
bl
e I e
n
cl
ose
s
t
h
e
l
i
s
t
of
sens
o
r
s
and
t
h
e
pa
ram
e
t
e
rs m
oni
t
o
re
d
i
n
wi
nd
t
u
rbi
n
e
.
Table
1.
Mon
i
t
o
r
e
d
pa
ram
e
t
e
r
s
Para
m
e
ter
Sensor
Bear
ing,
gear
box lubr
ication
SKF
W
i
ndlub
W
i
ndvane I
N
V
-
4
0
A
Ane
m
o
m
et
er
I
N
A
-
4
6
A
T
e
m
p
eratu
r
e, gea
r
bearing
P
T
1
0
0
T
e
m
p
eratu
r
e, gene
rator
P
T
1
0
0
External Air
T
e
m
p
erature
P
T
1
0
0
T
e
m
p
eratu
r
e N
acel
le
P
T
1
0
0
T
e
m
p
eratu
r
e gear
oil
P
T
1
0
0
Rotor speed (gear t
ooth)
E
I
3
0
1
0
P
P
o
s
,
P
N
P
Gen
e
rato
r
sp
eed
D
U
1
0
,
E
P
,
P
N
P
Pitch position (linear actuator)
V
e
r
t
-
X
2
2
Atm
o
spher
i
c Pr
ess
u
r
e
NRG #BP20,
W
X
T
520
Hu
m
i
dity P14
SM
D
V
i
br
ation W
L
NJ-
S
2-
G,
HS-100
Hy
dr
aulic oil level
Gem
s
L
S
600-
150-
NO-
10
Fi
gu
re
1
gi
ves
a det
a
i
l
e
d
des
c
ri
pt
i
o
n o
f
t
h
e
l
o
cat
i
on
o
f
v
a
ri
o
u
s se
nso
r
s
use
d
i
n
t
h
e
m
oni
t
o
ri
ng
o
f
a wi
n
d
mach
in
e.
Th
e
cu
rren
t/vo
ltag
e
in
tellig
en
t senso
r
s,
b
a
sed
on g
e
n
e
rator cu
rren
t
stato
r
sp
ectral an
alysis (CSA),
are
use
d
i
n
c
o
ope
rat
i
o
n
wi
t
h
o
n
e
bi
t
vi
b
r
a
t
i
on s
e
ns
or
s
f
o
r
an
ea
rl
y
i
d
ent
i
f
y
i
ng
of
p
r
os
pect
i
v
e
occ
u
r
r
i
n
g
mechanical fa
ults.
Fi
gu
re
1.
Se
ns
ors
l
o
cal
i
zat
i
o
n
schem
e
5.
DESIG
N
OF SYSTE
M
ST
RUCTURE AND
NETW
O
R
K
Th
e propo
sed system
is an In
tellig
en
t
Em
b
e
d
d
e
d Syst
e
m
for Con
t
ro
l an
d Rem
o
te Mon
ito
ri
ng
(IESCRM) t
h
at h
a
s th
e cap
a
b
ility o
f
pro
c
essin
g
, m
o
n
itoring
and
con
t
ro
llin
g [29
]
, this system
fin
d
s its
ap
p
lication
in
man
y
field
areas esp
ecially re
m
o
te elect
ro
mech
an
ical con
d
ition
m
o
n
itoring
of wi
n
d
t
u
rb
i
n
es in
a par
k
.
It
com
p
ri
ses se
ns
ors
,
m
i
croco
n
t
r
ol
l
e
r
,
m
e
m
o
ri
es, etc. It typically has
a specia
lized function with
p
r
og
ram
s
sto
r
ed
on
ROM.
An
add
e
d
feature in
th
is em
b
e
d
d
e
d
system
i
s
its ab
ility
to
co
mm
u
n
i
cate.
The
com
m
uni
cat
i
on ca
n
be
vi
a
W
i
-Fi
,
GPR
S
,
W
i
-M
ax
o
r
Et
her
n
et
ca
bl
es.
The
TC
P/
IP
p
r
ot
oc
ol
i
s
a
wi
d
e
l
y
used
st
anda
rd
f
o
r
m
ode
r
n
di
gi
t
a
l
com
m
uni
cat
i
on.
It
pr
ovi
des
rea
l
t
i
m
e
dat
a
t
r
af
fi
c.
T
h
e m
a
i
n
goal
o
f
t
h
i
s
ha
r
d
wa
re
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE Vol. 4, No. 6, D
ecem
ber 2014
:
882 – 892
8
85
i
s
t
o
m
a
ke i
n
t
e
rnet
/
I
P
e
n
abl
e
d f
o
r t
h
e
de
ve
l
ope
d em
bedd
ed
devi
ce
w
h
i
c
h se
rves
as a
dva
nce
d
rem
o
te dat
a
lo
gg
er to
b
e
accessed rem
o
tely v
i
a wo
rk
statio
n.
Th
e
m
e
n
tio
n
e
d
d
e
v
i
ce
will b
e
i
n
terfaced
with v
a
riou
s sen
s
ors
as illu
strated i
n
tab
l
e1.
Th
e
measu
r
em
en
ts an
d co
n
t
ro
l
d
a
ta are th
en
commu
n
i
cated
to th
e cen
t
ral serv
er
,
wh
ich
ad
op
ts
clien
t/serv
e
r web
an
d datab
a
se fram
e
work
s,
t
h
r
o
ug
h a
n
Et
her
n
et
or
wi
rel
e
ss co
n
n
ect
i
o
n
.
T
h
i
s
serv
er
ho
ld
s all
th
e
p
a
st data tran
sm
it
ted
b
y
t
h
e system
u
s
ing
web p
a
g
e
s. Fig
u
re2 illu
strate th
e m
o
n
ito
ri
ng
an
d
co
n
t
ro
l
system arch
itectu
r
e allo
wing
on
lin
e d
a
ta tran
sf
er
b
e
tw
e
e
n
th
e win
d
tur
b
in
e
s
an
d th
e co
n
t
r
o
l
c
e
n
te
r
.
The designe
d
interactive
embedde
d
ha
rdware provides a
n
ef
ficient,
rel
i
able and econom
i
ca
l link be
tween
vari
ous
sy
st
em
s,
se
ns
ors
a
n
d
act
uat
o
rs
.
The
choi
ce
o
f
t
h
e GPR
S
an
d
W
i
-
M
ax pr
ot
oc
ol
s i
s
j
u
st
i
f
i
e
d
by
severa
l
reasons: first,
wind pa
rks are im
pl
em
ent
e
d on a l
a
r
g
e
ge
og
ra
phi
cal
are
a
t
h
at
can rea
c
h som
e
ki
l
o
m
e
t
e
rs.
Therea
ft
er
, t
h
e
i
r rem
o
t
e
m
oni
t
o
ri
n
g
re
q
u
i
r
e
s
a wi
de
co
ve
r wi
rel
e
ss
com
m
uni
cat
i
on pr
ot
oc
ol
, a
nd t
h
en t
h
e
i
m
p
l
e
m
en
tatio
n
of th
ese commu
n
i
catio
n
tech
no
log
i
es
do
es
n
o
t
requ
ire an
y n
e
w infrastru
c
tu
re i
n
st
allatio
n
because it
uses the m
obile
cellular telephony
network 2.5G,
3G
or
4G. Som
e
dif
f
erences
existing
between
t
h
ese t
w
o
pr
ot
ocol
s
are
dat
a
r
a
t
e
and
t
r
a
n
sm
issi
on
t
i
m
e
[2
5]
.
Fi
gu
re
2.
B
l
oc
k
di
ag
ram
of m
oni
t
o
ri
n
g
a
n
d c
ont
rol
sy
st
em
Arc
h
i
t
ect
ure
The se
ns
ors listed in ta
ble
1,
are c
o
nnected
to the
m
i
croco
n
t
r
ol
l
e
r w
h
i
c
h
i
s
use
d
as
a c
o
m
put
er an
d
t
h
e Et
he
r
n
et
c
h
i
p
ENC
2
8
J
6
0
[
28]
i
s
use
d
t
o
c
o
n
n
ect
t
h
e
micro
c
on
tro
ller to LAN
with Eth
e
rn
et cab
l
e b
y
an
R
J
45
p
o
r
t
.
Fi
g
u
re
3 s
h
ow
s t
h
e i
n
t
e
rc
on
nect
i
o
n
o
f
ENC
28J
60
wi
t
h
M
C
U
.
The
el
ect
ri
cal
p
o
we
r s
u
ppl
y
o
f
t
h
e
IESCRM
m
o
dule does
not pose a problem
because it can
be powe
red
by a recharge
able
battery conne
cted to
an
AC/DC c
o
nversi
on circ
uit of
the c
u
rre
n
t received
from
wind
turbine
duri
ng operati
o
n.
Figure 3.
Ethern
et Ch
ip and
MC
U inter
c
onnectio
n
Whe
r
eas,
W
i
-
M
ax or
G
P
RS
m
odule (like ZTE IX
256-
RJ
45 Modem
[26]
for
W
i
-Max or SIM900A
-
RS232
M
odem
[2
7]
f
o
r
GPR
S
) a
r
e
use
d
t
o
est
a
bl
i
s
h a
n
o
n
l
i
n
e wireless
in
ternet
co
nn
ection
with
th
e b
a
se statio
n
serve
r
. T
h
e sens
ors are c
o
nnecte
d
to the AN
0-7
pi
ns
.
The M
C
U i
s
chose
n
fr
o
m
PIC
24 fa
m
i
ly
of
microcontrolle
rs
because
it has m
a
ny
advantages: large
num
ber
of I/O
pins,
high m
e
m
o
ry space,
built i
n
ADC
c
h
an
nel
s
whi
c
h m
a
ke i
t
easy
for i
n
t
e
rfaci
n
g
t
h
e se
n
s
ors
,
de
vel
o
pe
d usi
ng
Nan
o
W
at
t
t
echn
o
l
o
gy
t
h
at
red
u
ces
po
wer
cons
um
pt
i
on
du
ri
n
g
o
p
e
r
at
i
o
n
.
The LC
D
di
spl
a
y
s
t
h
e cu
rre
nt
an
d any
new
IP ad
d
r
es
s of t
h
e
comm
unication. T
h
e
serial c
o
nnection RS232 is used to
co
nf
igu
r
e th
e
mo
du
le
(
t
o
ch
ang
e
th
e I
P
a
ddre
s
s
)
.
I
t
can al
so be
us
ed f
o
r
deb
u
ggi
ng
pu
r
pose
s
. T
h
e IESC
R
M
m
o
d
u
l
e
i
s
an aut
o
m
a
t
on wi
t
h
i
n
t
e
g
r
at
ed we
b
serve
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Wind Turbi
n
e
Fault Detection
System
in Re
al Time
Re
mot
e
Monitoring
(
S
aad
C
h
a
k
ko
r)
88
6
Once
p
r
o
g
r
a
m
m
e
d, i
t
con
t
ai
ns so
ft
wa
re
co
nsi
s
t
i
ng
o
f
we
b
pa
ges
fo
r c
o
n
f
i
g
ur
i
ng,
m
oni
t
o
ri
ng
an
d
vi
sual
i
zat
i
o
n
o
f
dat
a
m
easure
m
ent
s
of
t
h
e
d
i
ffere
nt
se
ns
or
s co
nnect
e
d
t
o
t
h
i
s
m
odul
e.
B
o
t
h
GPR
S
a
n
d
W
i
-
Max are
two IP base
d
service prot
ocol
s
which can be
use
d
to access
a ra
nge
of IP servic
es
by running TCP/IP
pr
ot
oc
ol
st
ac
k,
suc
h
a
s
F
T
P,
Web
B
r
o
w
si
n
g
,
M
ach
i
n
e-t
o
-Machine M
2
M
and
Sim
p
le Mail Trans
f
er
E-m
a
i
l
SM
TP.
TC
P/
I
P
St
ac
k i
s
di
vi
ded
i
n
t
o
m
u
l
t
i
pl
e l
a
y
e
rs.
T
h
e
co
de i
m
pl
em
ent
i
n
g
eac
h l
a
y
e
r r
e
si
des
i
n
a s
e
parat
e
source
file, while the s
e
rvices
and
APIs (Application Programm
i
ng Interfaces)
are de
fine
d t
h
rough
heade
r
/
i
n
cl
u
d
e
fi
l
e
s. A
not
her
pr
ot
oc
ol
use
d
i
n
IE
SC
R
M
m
odul
e i
s
H
TTP
whi
c
h si
m
p
l
y
inv
o
l
v
e
s
an e
x
c
h
an
ge
of t
e
xt
m
e
ssages fol
l
o
wed
b
y
t
h
e t
r
ansfer
of
Web
dat
a
v
i
a a TC
P connect
i
o
n
.
To f
e
t
c
h a W
e
b pa
ge, t
h
e
br
ow
ser o
p
e
n
s
a TC
P con
n
ect
i
on t
o
ser
v
er
port 80, and the
n
uses HTTP to
send
a r
e
qu
est. Th
e b
a
sic pr
og
r
a
m
whic
h is l
o
calized i
n
MCU EPROM
runs
as a se
rve
r
pr
oviding m
u
lti-use
r
access to a
num
b
er
of databases. A
M
y
SQL/
PH
P
Web/
Dat
a
base
serve
r
base
d
on
AP
Is a
r
e u
s
ed t
o
i
m
pl
em
ent
ade
q
uat
e
s
o
ft
ware t
h
r
o
u
g
h
C
G
I
pr
ot
oc
ol
fo
r i
n
t
e
rfaci
n
g
ext
e
r
n
al
appl
i
cat
i
o
n
soft
wa
re w
ith
an
in
fo
rm
atio
n
web
serv
er. By th
is tech
n
i
q
u
e, th
e
inform
ation re
garding physic
al param
e
ters
is sensed
fr
om the sens
ors
.
This is receive
d by m
i
crocontroller
and is stored in CGI
varia
b
les. T
h
ese C
G
I
variables
ca
n
be
accessed by C
G
I
scripts
running i
n
web/
databas
e
serve
r
. T
h
e
data are stored i
n
the MySQL
da
tabase. Fi
na
lly th
e requ
ired
graph
i
cal represen
tatio
n
o
f
t
h
e d
a
ta
can be
di
spl
a
y
e
d i
n
t
e
ract
i
v
el
y
i
n
t
h
e dy
nam
i
c websi
t
e
as i
l
l
u
st
rat
e
d i
n
fi
g
u
re
4. T
h
e de
v
e
l
ope
d m
odul
e
offe
rs
several feature
s
that
re
quire a
SIM ca
rd:
Access
t
o
web
page
s of
t
h
e
m
o
d
u
l
e
vi
a
Et
he
rnet connection, GPRS or
Wi-Max
Sending alarm
messages
by e-mail via GPRS or
W
i
-Max
Sen
d
i
n
g al
arm
m
e
ssages
by
S
M
S
Fi
gu
re 4.
CGI
and W
e
b Serve
r
co
nn
ection
with
Datab
a
se
The
IESC
R
M
m
odul
e p
r
o
v
i
d
es an
earl
y
da
m
a
ge det
ect
i
o
n
p
r
oce
d
ure
t
h
at
ena
b
l
e
s
rem
o
t
e
,
aut
o
m
a
t
i
c
maintenance a
nd c
o
ntrol of a
wind turb
i
n
e
hard t
o
access
,
e
l
iminating the
n
eed for a
n
on-site service e
n
ginee
r
to
p
e
rfo
r
m
extra m
a
n
u
a
l in
terv
en
tio
n. Reactin
g
to
t
h
e
p
r
ob
lem
s
d
e
te
cted
b
y
th
is syste
m
.
A con
d
itio
n
m
o
n
ito
rin
g
sp
ecialist can
set t
h
e pro
p
e
r alarm settin
g
s
to
t
r
i
gge
r ap
pr
o
p
ri
at
e decision. If failures a
r
e
de
tected
su
ch
as em
p
t
y
o
r
b
l
o
c
k
e
d
lub
r
ication
pu
mp
s or torn
feed
lin
es
o
r
o
il/grease le
v
e
ls,
op
erat
o
r
s are no
tified
im
m
e
di
at
el
y
.
I
n
t
h
i
s
case, de
pen
d
i
n
g o
n
t
h
e
t
y
pe of fa
u
lt an
d
its sev
e
rity, it wil
l
b
e
n
ecessary th
at th
e wind
tu
rb
in
e con
t
ro
ller tak
e
s an
operatin
g strategy o
f
aro
und
fou
r
op
eration
a
l
states:
On
Pause
Of
f
Em
ergency
6.
CIRCUIT DE
SIGN
USI
N
G PROTE
U
S
S
O
FTWARE
T
o
e
v
al
uat
e
t
h
e o
p
erat
i
o
n
o
f
t
h
e
desi
gne
d
m
odul
e, i
t
s
c
o
m
put
er si
m
u
l
a
t
i
on
has
bee
n
m
a
de wi
t
h
Prote
u
s IS
IS s
o
ft
ware ve
rsio
n 7.
7, fig
u
r
e 5
sh
o
w
el
ectro
nic sch
e
m
a
tics
circu
its
o
f
IESCRM with
a
GPR
S
m
odul
e. F
u
rt
h
e
rm
ore, fi
g
u
r
e
6
prese
n
t
el
ec
t
r
o
n
i
c
schem
a
ti
c ci
rcui
t
s
o
f
t
h
e I
E
SC
R
M
u
s
i
ng
W
i
-M
ax
m
odem
whe
r
eas
t
h
e
ar
chi
t
ect
ure
o
f
t
h
e ce
nt
ral
i
zed
net
w
or
k
f
o
r
su
per
v
i
s
i
o
n a
n
d
rem
o
t
e
m
a
i
n
t
e
nance
of
a
wi
n
d
pa
r
k
can
b
e
d
e
scr
i
bed
as show
n in f
i
gu
r
e
7.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE Vol. 4, No. 6, D
ecem
ber 2014
:
882 – 892
8
87
Fi
gu
re 5.
Electronic
sc
hem
a
tics
of IESCRM with GPRS
m
o
dule
The we
b pa
ges
can be vi
ewe
d
by
br
o
w
si
n
g
di
rect
l
y
t
o
t
h
e speci
fi
c IP ad
dress
devi
ce f
r
o
m
t
h
e base
station or from any other
c
o
mpute
r
on the ne
twork. To m
a
xim
i
ze pr
ogram
m
e
m
o
ry space (and be a
b
le to fit
t
h
e co
de i
n
a
PIC
2
4FJ
1
28
GA
0
1
0
)
an e
x
t
e
rnal
2
5
LC
25
6 Seri
al
EE
P
R
OM
i
s
adde
d w
h
i
c
h
uses
t
h
e S
P
I
in
terface th
at can
b
e
sh
ared
with
th
e ENC28J6
0
a
n
d the
de
vice m
u
ch fast
er tha
n
t
h
e I2C
pa
rt.
The IESCRM physical interface
m
odel transmits and
receives data packets to base stat
ion through
GPRS
or
W
i
-Max
co
nn
ection
.
Meanwh
ile, th
e PIC24
ru
ns it ow
n TCP/IP stack and
sof
t
w
a
r
e
, co
m
p
r
i
sin
g
a
si
m
p
le web serv
er.
Th
is software en
ab
les th
e m
i
cro
c
ontro
ller t
o
b
e
hav
e
as a co
mm
u
n
i
catio
n
s
serv
er,
p
r
ov
id
ing
a sup
port fo
r a
web clien
t/serv
e
r.
Fi
gu
re 6.
El
ect
ro
ni
c
sc
hem
a
t
i
cs
o
f
IESC
R
M
wi
t
h
W
i
-M
ax m
odem
Wh
en
t
h
e simu
latio
n
is
started
,
t
h
e LC
D d
i
sp
lay th
e
IP add
r
ess allo
cated to
th
e
wireless m
o
d
e
m
b
y
t
h
e of
fi
ce net
w
o
r
k
DHC
P serve
r
. F
o
r
dem
onst
r
at
i
on
pu
r
pos
e, t
h
e IP a
d
d
r
ess
of di
ffe
rent
IE
SC
R
M
i
n
wi
n
d
par
k
net
w
or
k
are assi
g
n
e
d
t
o
be 1
9
2
.
1
6
8
.
1
1
.
X,
w
h
ere
X t
a
ke i
t
val
u
e
bet
w
een
2 a
nd
2
5
5
an
d t
h
e
IP
o
f
t
h
e
cen
tr
al
w
e
b
/
d
a
tab
a
se serv
er
was assign
ed
to
b
e
1
9
2
.
16
8.1.1, bo
th conn
ected
to th
e
sam
e
n
e
tw
or
k ro
u
t
er
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Wind Turbi
n
e
Fault Detection
System
in Re
al Time
Re
mot
e
Monitoring
(
S
aad
C
h
a
k
ko
r)
88
8
Fi
gu
re
7.
St
r
u
c
t
ure
of
co
nt
r
o
l
and
m
oni
t
o
ri
n
g
sy
st
em
of a
wi
nd
pa
r
k
In t
h
is architec
t
ure,
each wi
nd m
achine is e
qui
ppe
d
with a
n
IESCRM m
odule i
d
entifie
d by its own
IP address, t
h
is allows to c
o
llect and to st
ore
data
recei
ved
from
diffe
rent se
nsors installed on eac
h wi
nd
tu
rb
in
e.
Th
e IESCRM tak
e
s
care of th
e real
tim
e
regular t
r
ansm
ission of
all
m
easurem
e
n
ts via
GPRS
or
Wi-
M
a
x t
o
t
h
e ser
v
er o
f
su
pe
rvi
s
i
on an
d co
nt
r
o
l
st
at
i
on
which has a specific IP addres
s. Thi
s
comm
unication is
est
a
bl
i
s
hed
by
m
eans of a
dy
nam
i
c web pag
e
sho
w
n i
n
fi
g
u
re
8 i
n
di
cat
i
n
g
t
h
e t
i
m
e
and d
a
t
e
, t
h
e wi
n
d
t
u
r
b
i
n
e
ID and its IP addre
ss a
n
d t
h
e
receive
d
data. On t
h
e
ot
her side
,
IESCRM collaborates with the
central
autom
a
ton re
gardi
n
g the
operati
on m
ode
of
t
h
e
wi
n
d
m
achi
n
e.
Figu
re
8.
IE
SCRM
Serv
er
We
b/
Dat
a
ba
se Pa
ge m
oni
t
o
ri
ng
The
base
station serve
r
allows m
a
nagem
e
nt and proce
s
s
i
ng
of the
database sam
p
les received from
dif
f
e
r
e
n
t
wi
n
d
t
u
r
b
i
n
es i
n
or
der
t
o
m
a
ke a deci
si
on
o
r
t
o
ap
pl
y
a c
o
m
m
a
nd
usi
n
g
di
a
g
n
o
st
i
c
al
g
o
ri
t
h
m
s
.
7.
SIMULATION RESULTS
The sy
st
em
descri
be
d i
n
t
h
e
pre
v
i
o
us sect
i
on
f
o
r
wi
n
d
p
a
rk c
o
nt
rol
l
i
n
g
and m
oni
t
o
ri
ng
has
bee
n
i
m
p
l
e
m
en
ted
an
d sim
u
lated
based
on
a
realistic task
b
y
th
e O
P
N
E
T Mod
e
ler
14
.5
an
Proteu
s ISIS
v
i
a vir
t
u
a
l
seri
al
con
n
ect
i
on
p
o
rt
s
un
de
r
t
w
o sce
n
ari
o
s
and ca
rri
e
d
out the sim
u
lation t
o
eval
uate
and t
o
com
p
are the
per
f
o
r
m
a
nce o
f
t
h
e IE
SC
R
M
sy
st
em
i
n
t
e
rm
of real
t
i
m
e
dat
a
t
r
affi
c
fo
r
t
w
o di
ffe
rent
net
w
or
k t
ech
n
o
l
o
gi
es.
The sce
n
ari
o
s
devel
ope
d i
n
O
P
NET
f
o
r
Wi
-
M
ax an
d
GPR
S
o
v
er
UM
TS
are sh
o
w
n i
n
fi
gu
re
9 an
d
10
and t
h
e
syste
m
p
a
ram
e
ters fo
r bo
th n
e
twork
s
are d
e
tailed
in
tab
l
e 2
an
d 3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE Vol. 4, No. 6, D
ecem
ber 2014
:
882 – 892
8
89
Tabl
e 2. Wi
-M
AX
Si
m
u
l
a
t
i
on
En
vi
r
onm
ent
Param
e
ters
Syste
m
Para
m
e
te
rs
Sim
u
lation tim
e
(
s
ec)
3600
Data traffic rate
10 Mbps
Basic r
a
te
1.
5 M
bps
Service Class
Na
me
Gold/UGS
Antenna Gain
15 dBi
PHY
pr
ofile
W
i
r
e
less OFDM
A
20
M
H
z
M
a
x.
Tr
ans
m
it power
0.
5
W
a
tt
Path loss
Pedestrian
BS MAC
address
Distance based
Tabl
e 3.
UM
T
S
Si
m
u
l
a
t
i
on E
nvi
ro
nm
ent
Param
e
ters
Syste
m
Para
m
e
te
rs
Sim
u
lation tim
e
(
s
ec)
3600
UMTS
M
N
cell
st
ate
CELL_
DCH
UM
T
S
RL
C pr
ocess tim
e
0.
015 sec
CPI
C
H trans
m
ission Power
1W
att
Shadow fading Sta
ndar
d
deviation
10
Pr
ocessing tim
e
0.
02 sec
Path loss
Pedestrian
UM
T
S
GM
M
T
i
m
e
r
15/30/
10
Tw
o di
f
f
e
r
ent
appl
i
cat
i
o
ns ar
e use
d
:
HT
TP
and
Dat
a
base.
The st
udi
e
d
Wi
-M
AX a
n
d
U
M
TS net
w
o
r
k
m
odels consist
s
of two Base
Stations a
n
d two cells;
each
cell has four
fixed wi
nd t
u
rbi
n
e m
achines which a
r
e
equi
ppe
d
wi
t
h
an
IESC
R
M
t
o
ser
v
e al
l
ap
pl
i
cat
i
ons t
y
pes
.
Fi
gu
re
9.
G
P
R
S
o
v
e
r
UM
TS
wi
n
d
par
k
m
o
n
ito
rin
g
OPNET
scen
ario
Fi
gu
re 1
0
. Wi
-
M
ax
wi
n
d
par
k
m
oni
t
o
ri
n
g
O
P
NET
scenari
o
The c
ove
ra
ge
of
o
n
e cel
l
i
s
app
r
oxi
m
a
t
e
ly
2km
by
2
k
m
of a
r
ea.
UM
T
S
m
odel
sho
w
n i
n
fi
g
u
re
9
com
p
ri
ses user
equi
pm
ent
s
, R
a
di
o Net
w
o
r
k C
o
nt
r
o
l
l
e
r (
R
NC
) w
h
i
c
h i
s
con
n
ect
e
d
t
o
t
h
e pac
k
et
swi
t
c
he
d
net
w
or
k
vi
a G
P
R
S
S
u
p
p
o
rt
No
de
(S
GS
N)
and t
h
r
o
ug
h
G
P
R
S
Gat
e
way
Su
pp
ort
N
ode
(G
GS
N)
w
h
i
c
h
i
n
t
u
r
n
is connected t
o
the IP
Network. Base
d on fi
gure 11,
t
h
e a
v
era
g
e re
spons
e
tim
e
in database query for
GPRS
has an
u
n
st
abl
e
beha
vi
o
r
.
It
i
s
del
a
y
e
d w
i
t
h
alm
o
st
127 tim
e
s co
m
p
ared to th
e s
a
m
e
tim
e
for
W
i
-M
ax
network.
In
ot
her si
de, it is clear
from
figure 12 that the avera
g
e tra
ffi
c received in
data
base que
r
y inc
r
ease
si
gni
fi
ca
nt
l
y
w
i
t
h
t
i
m
e
i
n
creasi
ng a
n
d i
t
t
a
ke
s a fi
xe
d m
a
xi
m
u
m
val
u
e for
bot
h
GPR
S
a
n
d
W
i
-M
ax
net
w
o
r
k
s
.
The tra
ffic
rec
e
ived is m
u
ch
m
o
re im
portan
t
fo
r the
W
i
-M
ax c
o
m
p
ared t
o
GPRS a
n
d the differe
n
ce
be
twee
n
t
h
em
i
s
about
8
0
%.
Fi
gu
re
1
1
.
A
v
e
r
age
res
p
onse
t
i
m
e
i
n
Dat
a
bas
e
Q
u
ery
Fi
gure 12. Ave
r
age
traffic
re
ceiv
ed
i
n
Database Qu
ery
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN
:
208
8-8
7
0
8
Wind Turbi
n
e
Fault Detection
System
in Re
al Time
Re
mot
e
Monitoring
(
S
aad
C
h
a
k
ko
r)
89
0
Wh
ereas, referring
to
th
e sim
u
la
tio
n
resu
l
t
s illu
strated
in
fi
g
u
re
1
3
,
W
i
-Max is mu
ch
faster t
o
provide we
b page in HTTP s
e
rvice and it e
x
ceeds GPRS
w
ith about 187 times in aver
age. This pa
ge respons
e
tim
e
has an unstable evol
ution. Con
cerning
the avera
g
e tra
ffic receive
d
in HTT
P
service for
GPRS a
n
d
W
i
-
Max
sho
w
n
in
figu
re 14
, it can
b
e
ob
serv
ed
th
at th
e v
a
lu
e
of th
is p
a
ram
e
te
r is g
r
eater fo
r
W
i
-Max
th
an
GPRS
with
abo
u
t
94%. Ind
e
ed
, as illu
strated
in
figu
re15
, av
er
ag
e n
e
twork
activ
atio
n
d
e
lay for W
i
-Max
is m
u
l
tip
lied
13
7 t
i
m
e
s t
h
an t
h
at
of GPR
S
.
The obt
ai
ned
resul
t
s
can
be ju
st
i
f
i
e
d by
t
h
e fact
t
h
at
GPR
S
net
w
o
r
k co
vers a
wid
e
area bu
t
it d
a
ta tran
smissio
n
is o
n
l
y ab
le to
ach
i
e
v
e
a d
a
ta rat
e
b
e
tween
35Kb
it/s and
87Kb
it/s in
practice. This is
m
u
ch less tha
n
W
i
-Ma
x
whi
c
h can
r
each
75Mbit/s.
W
i
-M
ax is able to bring m
o
re ba
ndwidth
an
d pro
v
i
d
e
s
m
o
re co
nn
ectio
n qu
ality services to
b
e
n
e
fit
fro
m
th
e wind
p
a
rk
m
o
n
ito
ri
ng
.
Figure 13. Ave
r
age
Pa
ge re
sp
ons
e t
i
m
e i
n
HTTP
service
Figure 14. Ave
r
age
traffic rec
e
ived i
n
HTTP
service
Furt
herm
ore,
GPR
S
has
a
h
uge
fl
uct
u
at
i
o
n i
n
s
p
ee
d
wh
ereas
W
i
-M
ax
gi
ves a
co
nst
a
nt
ban
d
w
i
d
t
h
w
h
i
c
h
results i
n
a c
onstant and a
fast
est Inte
rnet se
rvice.
W
i
-Max re
m
a
ins the m
o
st approp
riate technology c
h
oice to
satisfy the
QoS re
quirem
e
nts
and real tim
e data traf
fic
fo
r
wind
turb
in
es
co
n
t
ro
l and
m
o
n
ito
ri
n
g
app
licatio
n
.
Figure 15.
A
v
e
r
age
net
w
o
r
k
a
c
t
i
v
at
i
on
del
a
y
8.
CO
NCL
USI
O
NS
W
i
nd
turbine
machines nee
d
pe
ri
odical
proactive
m
a
inte
nance
t
o
prolong thei
r electrom
ecanical
com
pone
nts lifetim
e.
The m
a
in contri
bution of this
pa
pe
r is to
design, by
study
and a
n
a
l
ysis, an intelligent
wireless
rem
o
te m
onitoring a
n
d control syst
e
m
according
t
o
wind turbi
n
e
s
features
a
n
d requirem
ents. Both
har
d
ware m
o
d
u
l
e
an
d
ope
rat
i
n
g
m
ode ha
ve
been
desc
ri
be
d in
d
e
tail.
Th
is
syste
m
b
a
sed
on
IP
co
mm
u
n
i
c
a
tio
n
com
b
ines
W
e
b and data
base
client/ser
ver technology to
copy data m
easur
e
m
ents receive
d from
the dif
f
erent
s
sen
s
o
r
s in
stalled
in
t
h
e
wind tu
rb
in
e m
ach
in
es an
d th
erefo
r
e
send
s it to th
e base statio
n th
rou
g
h
a
wireless
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
IJECE Vol. 4, No. 6, D
ecem
ber 2014
:
882 – 892
8
91
co
nn
ection
.
The sim
u
latio
n
co
m
p
ariso
n
,
with
Pro
t
eu
s
ISIS
and
OPNET
si
m
u
lato
rs b
e
tween
GPRS
an
d
W
i
-
Max
im
p
l
e
m
e
n
tatio
n with the stud
ied
m
o
du
le,
favorites
W
i
-M
ax
net
w
or
k
d
u
e t
o
t
h
e
i
r be
ne
fi
t
s
o
v
e
r
GPR
S
n
e
two
r
k
.
Th
is
in
tellig
en
t d
e
vice h
a
s lo
w co
st and
it can b
e
u
s
ed
in
o
t
h
e
r ap
p
lication areas. Main
ten
a
n
c
e
issu
es are
no
w ab
le t
o
b
e
reso
lv
ed
with
t
h
is ef
ficien
t co
mm
u
n
i
cato
r
too
l
.
Th
e
fu
t
u
re
wo
rk
will b
e
fo
cu
sed
o
n
th
e test and
app
licatio
n
o
f
th
e wh
o
l
e m
o
n
itorin
g
syste
m
in
th
e
p
r
actice.
REFERE
NC
ES
[1]
M. L. Sin
,
W
.
L. Soong
and
N. Er
tugrul, “
On-Line Conditio
n Monitoring
a
nd Fault Diagn
osis - a Survey
”,
Australian
Univ
ersities Power Engin
eer
ing Conf
erence, New Zealand 2003
.
[2]
Ouadie B
e
nnou
na et
al, “
Con
d
ition
Monitori
ng
&
Fault Diagnosis System
for Offshor
e
W
i
nd T
u
rbin
es
”,
https://zet10
.ip
e
e.pwr
.
wro
c
.pl/reco
rd/425/f
iles/in
vited_p
aper_3
.p
df
[3]
Elie
Al-Ahm
ar et al,
“
W
i
nd
En
er
gy Conversion
Systems Fa
u
lt
Diagnosis Using W
a
velet
Analysis
”, Internation
a
l
Review of
Electr
ical Eng
i
ne
ering
3, 4 2008
, pag
e
s
:
646-652.
[4]
M.L. Sin
et al, “
Induction
Machine On-Line Condition
Monitoring and
Fault D
i
agno
sis –
A
Survey
”,
http:/
/www
.acad
em
ia.edu
/41644
1/Induction_M
a
c
hine
_on_
Line_
C
ondition_Moni
toring_a
nd_Faul
t
_Diagnosis_A_Su
rvey
[5]
K.
K. Pandey
et al, “
Review
on
Fault Diagnosis
in
Thr
ee-Phase Induction
Moto
r
”, MEDHA
– 2
012, Proceed
ing
s
published b
y
International Journ
a
l of
Computer
Applications (IJ
CA).
[6]
E.
Al Ahmar
et
al,
“
Advan
ced
Signal Pr
ocessing
T
echniqu
es for F
ault Detection
and Diagnosis
in a
W
i
nd
T
u
rbin
e
Induction
Generator Drive T
r
a
i
n:
A
Comparative Stud
y
”, I
E
EE En
er
g
y
Conv
ersion Congr
ess and
Exposition
(ECCE),
Atlanta: États-Unis 201
0.
[7]
Hamid
A.
T
o
liyat et
al., “
Ele
c
tr
ic
Machin
es Mo
deling
,
Cond
itio
n Moni
toring,
a
nd Fau
l
t
Diagn
osis
”, CRC
P
r
es
s
T
a
y
l
or & Francis Group NW
2013, ISBN-13: 978
-1-4200-0628-5.
[8]
R.
W
.
H
y
ers
et
al
.,
“
Condition
monitoring
and
pr
ognosis of ut
ilit
y sca
l
e wind
turbines
”, Institute of Mate
ria
l
s,
Minerals
and Mining and
W
.
S.
Maney
& Son Ltd Ener
g
y
M
a
ter
i
als 2006,
V
o
l.1
No.3, p
a
ges 187
-203.
[9]
Julia Ni
lsson e
t
al.
,
“
Maintenan
ce Manag
ement
of W
i
nd
Power
Syst
ems Using
Condition
Monit
o
ring Syst
ems-Lif
e
Cycle Cost
Analysis for T
w
o Cas
e
Stud
ies
”, IEEE
T
r
ansactions
On Ener
g
y
C
onvers
ion,
V
o
l. 22
, No
. 1, March
2007
[10]
M
ohanraj.M
et al.
,
“
A
CAN Bus
based system fo
r monitoring and fault d
i
agnosis in W
i
nd T
u
rbine
”, Intern
ation
a
l
Conference on
Emer
ging
T
r
end
s
in
VLSI, Embedded S
y
s
t
em,
Nano Electron
i
cs and
T
e
lecommunication S
y
s
t
em
(ICEVENT), T
i
r
uvannamalai,
7-
9 Jan 2013
, p
a
ges 1-3.
[11]
Chun-Liang Hs
u, “
Constructin
g transmitting
interf
ace of
running parameters of small-scaled wind-pow
er
electricity gen
e
rator with WSN
modules
”,
Expert S
y
stems with
Applications J
ournal 37
, Elsevier 2010
, pag
e
s
3893–3909.
[12]
Ammar Zah
e
r
et al., “
Database
Management
for High R
e
solution
Condition
Moni
toring of
W
i
nd
T
u
rbines
”,
P
r
oceedings
of
the 44th Int
e
rna
tional Univ
ers
i
t
i
e
s
P
o
wer Engin
eering Conf
eren
ce (UP
E
C), Gl
a
s
gow 1-4 S
e
pt.
2009.
[13]
M
.
C. M
a
l
lika
r
ju
ne Gowda
et
al
,
“
Impr
ovement of th
e Performance o
f
W
i
nd T
u
r
b
ine Gen
e
rator
Using Condition
Monitoring T
ech
niques
”, Proc
ee
dings of 7th Int
e
rnat
ional Conf
e
r
ence on In
te
llig
ent S
y
st
em
s and Control (ISCO
2013), IEEE 201
2.
[14]
Y
a
s
s
i
ne
Am
irat
et
a
l
,
“
W
i
nd T
u
rbines Condition Monitoring
and Fault Diag
nosis Using Generator Curr
ent
Amplitude Demo
dulation
”, I
E
EE
International
En
er
g
y
Confer
ence and
Exhibition
(
E
ner
g
y
C
on), 20
10 .
[15]
Luci
an M
i
h
e
t
P
opa
et
a
l
.
,
“
Con
d
ition
Moni
toring of
W
i
nd G
e
ne
rators
”, IEEE In
dustr
y
Applic
ations Society
38
th
Annual Meeting
,
IAS'
03
, Salt Lake C
i
ty
, Utah
U
S
A, Oct
ober
200
3,
V
o
l. 3
IEEE S
i
gnal
Processing
Society
,
p.1839
-
1846.
[16]
Christopher J.
Crabtree, “
Survey of Commer
c
ia
lly
A
v
ai
lable Co
ndition Mon
itoring Systems for
W
i
nd T
u
rbines
”,
Durham University
School of
En
gineer
ing and
C
o
m
puting Scien
c
es, 2nd
November 2010, Rev
i
sio
n
: 05.
[17]
Shuangwen Sheng and Paul
V
e
ers, “
W
i
nd T
u
rbine Drive train Co
nd
ition Monitoring – an Overview
”, M
ech
ani
cal
Failures Prev
ention Group:
Applied S
y
st
ems Health
Manag
e
ment Confer
ence
V
i
r
g
inia Beach,
V
i
r
g
inia May
10
–
12, 201
1.
[18]
Xiang Gong
,
“
On-line Non-I
n
trusive Condition Mon
itoring
and
Fau
l
t D
e
tection
for W
i
nd T
u
rbin
es
”,
DigitalCommons@University
of
Nebraska –
Lin
c
oln.
[19]
M
a
rtin Klug
e
a
nd M
i
cha
e
l
Dan
its
chek
, “
Condition Monitoring
Systems (
C
MS)
in wind
turbines
”, ifm
el
ec
tronic
gmbh, update version: 22.07
.201
0.
[20]
Juha T
uominen
,
“
Hydraulic Bo
om
Monitoring W
ith
IEEE
802
.1
1 Based
W
i
r
e
less Sensors Network
”
,
Ma
ste
r
of
Science
Th
esis,
T
a
mper
e Univer
sity
of
T
e
chnolo
g
y
, Octob
e
r 200
9.
[21]
I.
Ahm
e
d et
al
.
,
“
A
Baselin
e S
t
udy for On-Lin
e Condition
Mo
nitoring of Ind
u
ction Ma
chine
s
”
,
Austra
la
sia
n
Univers
iti
es
P
o
wer Engin
eer
ing
Conferen
ce
(AUP
EC), 26-29 S
e
ptember 200
4,
Brisbane,
Australia.
[22]
C. J. Cr
abtree et al., “
Detecting
Incipient W
i
nd
T
u
rbine Gearbox Fail
ur
e:
A
Sig
nal
Analysis M
e
thod for On-lin
e
Condition Monitoring
”,
European
W
i
nd Ener
g
y
Conference (
E
WEC),
W
a
rsaw 19
April 2010.
[23]
W
u
Chunming and Cheng
Liang
,
“
The
Monitori
ng Syst
em
for th
e
W
i
nd
Pow
e
r G
e
neration
Based
on
th
e
W
i
r
e
les
s
Sensor Network
”, In
tern
ation
a
l
Conferenc
e
on
Future
Ele
c
tr
ica
l
Power and En
er
g
y
S
y
stems 201
2, En
er
g
y
Procedia
17, p
.
1020
– 10
27.
Evaluation Warning : The document was created with Spire.PDF for Python.