Int
ern
at
i
onal
Journ
al of
P
ower E
le
ctr
on
i
cs a
n
d
Drive
S
ystem
s
(
IJ
PEDS
)
Vo
l.
12
,
No.
1
,
M
a
r 202
1
, p
p.
499
~
51
0
IS
S
N:
20
88
-
8694
,
DOI: 10
.11
591/
ij
peds
.
v12.i
1
.
pp
499
-
51
0
499
Journ
al h
om
e
page
:
http:
//
ij
pe
ds
.i
aescore.c
om
Hardw
are
in
the
loop sim
ula
ti
ve set
up for testi
ng the c
om
bined
heat
po
wer gen
eratin
g w
ind turbi
ne
Iho
r
Shch
ur
1
, Vsev
ol
od
S
hc
hur
2
, Ihor
Bi
ly
ak
ovskyy
3
, My
kh
ailo K
h
ai
4
1,3,4
Depa
rtment
o
f
Elec
tro
mecha
tr
onic
s
and
Comp
ute
ri
ze
d
Elec
tro
me
ch
ani
c
al Sys
te
ms,
Inst
it
ut
e
of
Pow
er
Engi
ne
ering
and
Contro
l
Sys
t
em
s, Lviv
Polytechnic Nat
ion
al
Univer
sity
,
Ban
der
a
,
Ukrai
n
e
2
Mita
-
T
ekni
k
Ltd., L
viv
,
Ukra
in
e
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
J
un
1
1
, 2
0
20
Re
vised
Jan
1
, 20
21
Accepte
d
Ja
n
22
, 2
0
2
1
Thi
s
pap
er
d
esc
r
ibe
s
th
e
design
and
i
mpl
e
me
nt
ation
of
h
ard
ware
in
th
e
loop
(HIL)
sys
tem
ba
sed
on
indu
ct
ion
mo
tor
wind
tur
bine
em
u
lator
fo
r
th
e
study
of
the
op
era
t
ion
of
a
co
mbi
n
ed
hea
t
-
power
(CHP
)
gene
rating
w
ind
ene
rgy
conve
rsion
sys
te
m
(W
ECS).
T
he
ene
rgy
g
enerat
ion
pa
rt
of
t
he
WE
CS
consists
of
two
spec
ially
design
ed
g
ene
r
at
ors
th
at
ar
e
p
lace
d
on
a
com
mon
ver
tical
axi
s,
which
is
conn
ecte
d
to
the
inductio
n
mot
or
through
a
ge
arb
ox.
The
first
g
ene
r
a
tor
is
an
e
lectr
i
c
two
-
arm
a
ture
a
xia
l
PM
SG
and
the
sec
ond
one
is
a
th
ermal
el
e
ct
rom
agne
t
ic
ret
ard
er.
The
sof
twar
e
par
t
of
th
e
HIL
setup
simul
ates
th
e
intera
c
ti
on
of
the
wind
flow
wi
th
a
v
ert
i
ca
l
axi
s
wind
turbi
n
e
(VA
WT
)
and
is
i
mpl
e
me
nt
ed
in
a
progr
am
m
able
logi
c
cont
rol
ler
base
d
on
the
mod
el
d
evel
oped
in
the
MA
TL
AB/S
im
ul
ink.
The
r
esult
s
of
e
xper
imental
studie
s
of
the
C
HP
WE
CS
wi
th
th
e
cr
ea
t
ed
HIL
si
mul
a
ti
ve
se
t
up
a
t
both
consta
nt
and
tur
bule
nt
wind
spe
eds
have
show
n
good
agr
ee
m
en
t
with
the
cor
responding
r
esult
s
of
co
mpu
te
r
simu
la
t
ion.
The
cr
eate
d
HI
L
simul
at
iv
e
setup
wil
l
b
e
used
for
th
e
d
evel
opme
nt
of
an
en
erg
y
ma
nag
eme
nt
sys
te
m
for
CHP
WE
CS
.
Ke
yw
or
d
s
:
Ele
ct
ro
ma
gnet
ic
h
eat
generato
r
Hardwa
re
in
the
loop
simulat
or
Perma
ne
nt ma
gn
et
sync
hrono
us
gen
e
rato
r
Ver
ti
cal
ax
is
wi
nd
t
urbine
Win
d
e
nerg
y
c
onve
rsion s
ys
t
em
Сom
bin
e
d heat
-
pow
er
gen
e
rati
on
This
is an
open
acc
ess arti
cl
e
un
der
the
CC
BY
-
SA
l
ic
ense
.
Corres
pond
in
g
Aut
h
or
:
Ihor S
hchu
r
In
sti
tute
of P
ower
Enginee
rin
g
a
nd Co
ntr
ol
Sy
ste
ms
Lviv
P
oly
te
ch
ni
c N
at
ion
al
Un
iversity
Ba
nd
e
ra s
tr
. 12, L
viv
,
79
013,
Ukraine
Emai
l:
ihor.z.s
hchu
r@
lp
nu.
ua
1.
INTROD
U
CTION
Re
new
a
ble
e
ne
rgy
de
velo
pme
nt
rates
ar
e
ste
adily
i
nc
reasin
g
in
ma
ny
co
untrie
s.
Win
d
powe
r
gen
e
rated
by
hi
gh
-
po
wer
wind
tu
rb
i
nes
(
W
T)
occupies
a
promi
nen
t
pla
ce
in
this
proc
ess
[
1].
At
the
same
ti
me,
low
-
po
w
er
(
up
to
10
kW)
wind
e
nergy
c
onve
rsi
on
sy
ste
ms
(WEC
S)
wh
ic
h,
as
a
ru
le
,
sta
nd
al
on
e
ar
e
now
becomi
ng
more
wi
des
pread
[
2
]
,
[
3].
S
om
e
c
onsume
r
s
us
e
th
em
f
or
powe
r
sup
ply
in
the
a
bs
e
nc
e
of
centrali
zed
power
gr
i
ds
or
i
n
the
pr
ese
nc
e
of
t
he
la
tt
er
for
ad
diti
onal
powe
r
sup
ply
in
orde
r
to
r
edu
ce
el
ect
rici
ty co
ns
umpti
on from
a
gri
d, imp
rove
en
e
rgy
e
ff
ic
ie
ncy an
d
e
nsure
uninter
rupte
d powe
r
s
upply.
Lo
w
-
ca
pacit
y
WTs
are
i
ns
ta
ll
ed
directl
y
ne
ar
co
nsumer
s
wh
e
re
wi
nd
is
mainly
c
har
act
erized
by
lo
w
aver
a
ge
s
pee
ds,
f
reque
nt
gusts,
c
ha
ng
e
of
directi
on,
and
hi
gh
t
urb
ulence.
Th
e
locat
ion
an
d
s
pecifi
c
char
act
e
risti
cs
of
t
he
wind
ac
count
for
the
sign
ific
a
nt
di
ffe
ren
ce
bet
wee
n
lo
w
-
pow
er
WT`s
str
uctu
r
es
an
d
tradit
ion
al
hi
gh
-
powe
r
ones,
in
par
ti
cula
r
at
the
us
e
of
W
Ts
with
ver
ti
cal
a
xes
of
r
otati
on
(VAWT)
[
4].
These
VAWTs
wor
k
eff
ect
ivel
y
with
gusty
winds,
const
antly
pe
rc
ei
ve
winds
of
diff
e
re
nt
direct
ion
s
,
a
nd
sta
rt
at
low
wind
sp
ee
ds
than
ks
to
the
di
rect
(
gear
le
ss
)
dri
ve
of
a
lo
w
-
s
pee
d
pe
rm
anen
t
ma
gnet
sy
nc
hro
nous
ge
ner
at
or
(PMS
G)
[
5]
.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
499
–
51
0
500
In
sta
nd
-
al
one
W
ECS,
due
to
the
st
och
as
ti
c
natu
re
of
el
ect
rici
ty
ge
ne
rati
on
a
nd
c
on
s
umpti
on,
el
ect
ro
che
mica
l
batte
ries (B)
are
mo
st
c
om
monly
use
d
to
en
s
ur
e
the
po
w
er
s
uppl
y
a
nd b
et
te
r
us
e
of
ge
ner
at
e
d
el
ect
rici
ty
[
6].
Howe
ver,
t
heir
hi
gh
c
os
t
an
d
relat
ively
l
ow
li
feti
me
si
gn
i
f
ic
antly
i
ncr
eas
e
the
total
cost
of
a
WECS
-
base
d
a
utonomo
us
po
wer
s
uppl
y
s
yst
em.
To
save
money
,
we
can
reduce
t
he
ca
pacit
y
of
t
he
i
ns
ta
ll
ed
Bs;
as
a
res
ult,
th
e
a
mou
nt
of
gen
e
rated
e
le
ct
rici
ty
that
cannot
be
dire
ct
ly
c
on
s
um
e
d
or
acc
umulat
ed
will
increase.
T
his
dump
e
nerg
y
i
s
ei
ther
lo
st
by
re
duci
ng
t
he
VAWT
po
wer
ge
ne
rated
or
i
s
util
iz
ed
by
loadi
ng
the
ge
ner
at
or
on
t
hermoelec
tri
c
he
at
ers.
T
he
la
tt
er
are
place
d
i
n
a
hot
-
water
boil
er
us
e
d
to
heat
bu
il
di
ngs
an
d
for
hot
water
s
upply
[7].
WE
CS
that
produc
es
el
ect
rical
and
ther
mal
ene
r
gy
ca
n
be
cal
l
ed
a
co
mb
i
ned
heat
-
powe
r
(C
HP)
ge
ner
at
in
g o
ne.
In
the
know
n
CHP
WECSs
,
heat
is
obta
ined
f
rom
el
ec
tric
it
y
al
ready
ge
ner
at
e
d.
T
he
heat
fl
ow
capaci
ty
th
us
l
imi
ts
the
per
m
issi
ble
powe
r
of
the
el
ect
ric
gen
e
rato
r.
A
ne
w
al
te
rn
at
ive
so
luti
on
is
to
us
e
i
n
WECSs
,
in
a
ddit
ion
to
the
t
r
aditi
on
al
P
M
S
G,
a
heat
gen
e
rator
that
will
trans
form
the
mecha
nical
en
ergy
of
the
WT
directl
y
int
o
the
rmal
energ
y,
bypass
ing
t
he
inter
me
diate
el
ect
rical
li
nk
[
8
]
,
[
9].
The
heat
ge
nerat
or
is
an
el
ect
r
om
a
gnet
ic
de
vice
-
re
ta
rd
er
,
in
w
hic
h
the
e
dd
y
cu
r
ren
ts
ge
ner
at
e
d
in
t
he
ro
t
or
arr
a
y
by
t
he
m
agn
et
ic
flu
x
pr
oduce
d
by
an
e
xcita
ti
on
windin
g
duri
ng
r
oto
r
r
otati
on
,
e
nsure
i
ts
heati
ng.
T
he
ge
ner
at
ed
he
at
is
trans
ferred
to
the
water
that
i
s
pe
rio
dical
ly
pump
e
d
into
a
heat
c
ollec
tor.
The
sp
eci
al
he
at
ge
ner
at
or
is
mu
c
h
simpler
a
nd
ha
s
a
sig
nificantl
y
higher
s
pecif
ic
power
c
omp
ared
t
o
the
el
ect
ric
on
e
,
a
nd
he
nce
it
s
lo
w
c
ost
[9].
Su
c
h
a
so
l
utio
n
has
se
ve
ral
adv
a
ntage
s
-
e
nhanci
ng
t
he
f
unct
ion
al
it
y
of
CHP
W
ECS,
i
ncr
easi
ng
it
s
e
nerg
y
eff
ic
ie
nc
y,
a
nd
re
duci
ng
the
total
c
os
t
due
t
o
t
he
possi
bili
t
y
of
not
us
in
g
Bs
[
8].
The
e
ff
ic
ie
nt
oper
at
ion
of
CHP
W
ECS
r
equ
i
res
the
de
velo
pm
e
nt
of
a
ded
ic
at
ed
e
ne
rgy
ma
nagem
ent
sy
ste
m
t
ha
t
will
per
f
or
m
the
functi
ons
of
op
ti
mal
c
on
tr
ol
of
po
wer
extracti
ng
f
rom
the
wi
nd,
el
ect
rici
ty
and
heat
ge
ner
at
i
on
a
nd
consu
mp
ti
on,
a
s w
el
l as
monit
or
i
ng of e
nerg
y p
r
ocesses
.
In
order
to
c
ondu
ct
ex
pe
rime
ntal
rese
arc
h
i
n
t
he
a
bove
dir
ect
ion
s,
it
is
ne
cessar
y
t
o
c
re
at
e
a
s
pecial
instal
la
ti
on
.
In
the
la
tt
er,
it
is
ad
visable
to
rep
la
ce
the
mo
st
dif
ficult
aer
odynamic
pa
rt
of
WT
by
the
corres
pondin
g
simulat
or.
Mu
ch
w
ork
has
be
en
done
to
de
velo
p
su
c
h
t
oo
ls,
cal
le
d
em
ulators
or
Ha
rdw
are
-
In
-
the
-
L
oop
(HIL
)
simulat
ors
for
the
stu
dy
of
WECSs
a
nd
th
e
dev
el
opme
nt
of
c
ontrol
s
ys
t
ems
[
10
]
-
[
14].
The
y
are
disti
nguis
he
d
by
the
de
gree
of
e
mu
la
ti
on,
hardware
a
nd
softwa
re
s
olu
ti
ons,
a
nd
t
he
purpose
of
re
search
that wil
l be
conducte
d u
sin
g
t
hese i
ns
ta
ll
at
ion
s.
In
al
mo
st
al
l
s
imulat
or
s
or
e
mu
la
to
rs,
wind
flo
w
a
nd
it
s
interact
io
n
with
WT
a
re
mat
hemati
cal
l
y
simulat
ed.
Nea
mmanee
et
al
.
[10]
ha
ve
mod
el
ed
the
wind
flo
w
usi
ng
the
Va
n
der
H
ov
e
n
s
pect
r
um
,
an
d
th
e
tor
qu
e
rip
ple c
ause
d
by t
ow
e
r
eff
ect
was
ta
ke
n
int
o
acc
ount
. W
T
wo
rk
has
b
ee
n
sim
ulate
d by in
du
ct
io
n moto
r
(IM)
w
hose
t
orq
ue
has
bee
n
co
ntr
olled
by
an
in
ver
te
r
.
A
feature
of
the
Win
d
T
urb
ine
Em
ulato
r
(
WTE
)
pr
ese
nted
in [
11]
is
t
he
sim
pl
e
model
in
g
of
t
he
aer
od
yn
a
mic
chara
ct
erist
ic
s
of
the W
T
usi
ng
a
DC mo
t
or
with
separ
at
e
e
xcita
ti
on
an
d
a
seri
es
powe
r
resis
tor.
S
uc
h
an
i
mp
le
me
ntati
on
do
es
not
requ
ire
a
con
t
ro
ll
er
with
com
plex
co
ntr
ol
al
gorith
ms,
bu
t
t
he
c
har
ac
te
risti
cs
ob
ta
in
ed
in
the
ope
n
sy
ste
m
are
ve
ry
cl
os
e
to
t
he
real
char
act
e
risti
cs
of
the
WT
.
In
[
12],
the
DC
m
otor
c
on
t
ro
ll
ed
by
a
half
-
br
i
dge
DC
-
DC
c
on
ver
te
r
in
do
ub
l
e
-
lo
op
con
t
ro
l
sy
ste
m
physi
cal
ly
pro
vid
es
the
an
gula
r
velocit
y
of
WT
giv
e
n
by
t
he
c
ontrolle
r
.
Garg
a
nd
Da
hiy
a
[13
]
dev
el
op
e
d
t
he
WT
E
wh
ic
h
is
base
d
on
a
DC
mo
t
or
dr
iving
a
sel
f
-
e
xcite
d
i
nduction
ge
ne
rato
r
t
hat
is
pr
act
ic
al
ly
not
use
d
in
ei
the
r
la
r
ge
or
s
mall
W
ECSs.
I
n
[
14],
the
la
bo
ra
tor
y
-
scal
e
W
T
E
sim
ulate
s
both
the
op
e
rati
on
of
t
he
WT
with
a
w
ound
r
otor
inducti
on
gen
e
rator
a
nd
the
i
nv
e
rter
co
nnec
te
d
to
the
network.
Emphasis
is
placed
on
pro
vidi
ng
a
real
-
li
fe
wind
s
pee
d
pr
of
il
e
as
well
as
m
od
el
in
g
the
sta
rting
of
WT
an
d
it
s
con
t
ro
l
by a
pitchin
g
mec
ha
ni
sm.
In
rec
ent
yea
rs
[
15
]
-
[
19],
the
te
rm
"e
mu
la
to
r
"
is
oft
en
re
pla
ced
by
H
IL
se
t
up
or
HI
L
platf
orm.
T
hese
are
usual
ly
cl
os
e
d
-
lo
op
s
ys
te
ms
that
co
ve
r
both
hard
wa
re
an
d
s
of
tw
a
re
pa
rts
of
th
e
sy
ste
m,
w
hich
are
connecte
d
t
hro
ugh
a
po
wer
amplifie
r
with
inter
face
al
go
rithm.
In
the
pro
po
se
d
s
olu
t
ion
,
a
wind
ve
locit
y
prof
il
e,
WT
c
ha
racteri
st
ic
s
an
d
co
ntr
ol
sy
ste
ms
are
sim
ulate
d
by
the
HIL
sy
ste
m.
B
y
usi
ng
s
uc
h
H
IL
s
ys
te
m,
it
is
possi
ble
t
o
re
pro
duce
t
he
act
ual
op
e
rati
ng
c
onditi
ons
of
W
T`s
wor
king
with
diff
e
re
nt
ty
pes
of
ge
ne
rators
in
the
la
borat
ory
e
nviro
nme
nt
.
In
a
ddit
ion
,
this
ap
proac
h
al
lo
ws
to
c
onduct
othe
r
in
ves
ti
gations
direct
ed
at
the
dev
el
opme
nt
of
co
ntr
ol
s
trat
egies
of
W
ECS,
s
uc
h
as
pitch
c
ontr
ol
a
nd
M
PPT
[
16
]
,
[
18]
an
d
te
sti
ng
of
wind tu
r
bin
e
na
cel
le
s f
or ele
c
tric
al
certi
ficat
ion [
19].
This
arti
cl
e
is
about
creati
ng
a
new
C
HP
W
ECS
s
ta
nd
with
H
IL
sim
ulati
ve
set
up
of
V
AWT’s
wor
k.
It
co
ns
ist
s
of
t
he
dr
i
ve
I
M
t
ha
t
emulat
es
the
V
A
WT
t
orq
ue
un
der
the
act
ion
of
the
wi
nd
flo
w
of
giv
e
n
prof
il
e
and
t
wo
sp
ec
ia
ll
y
desig
ned
gen
e
rato
rs
-
a
n
el
ect
ric
two
-
armatu
re
P
MSG
of
a
xial
typ
e
a
nd
a
ther
ma
l
el
ec
trom
a
gn
et
i
c
retard
e
r.
T
w
o
ge
ner
at
or
s
a
re
place
d
on
a
commo
n
ve
rtic
al
axis.
Soft
war
e
par
t
of
t
he
H
IL
simulat
ive
set
up,
w
hich
si
mu
l
at
es
the
inte
rac
ti
on
of
wi
nd
fl
ow
with
V
A
W
T
an
d
ge
ner
at
e
s
the
c
orres
pondin
g
ref
e
ren
ce
for
the
f
reque
ncy
c
onve
rter
that
c
on
t
ro
ls
the
I
M
torque,
is
im
pl
emented
in
t
he
PLC,
base
d
on
t
he
model
de
velo
ped
i
n
t
he
MATL
AB/Si
mu
l
ink
s
of
tw
are.
This
m
od
el
e
ns
ures
the
e
quivale
nc
e
of
t
he
real
VAWT
’s
wor
k
with
s
pecific
par
a
mete
rs
a
nd
the
H
IL
si
mu
l
at
ive
set
up.
T
he
ap
plica
ti
on
of
the
c
reated
sta
nd
in
exp
e
rime
ntal
r
esearch
will
al
low
t
o
m
od
el
var
i
ou
s
wi
nd
conditi
ons
wit
h
the
possibil
i
ty
of
their
ide
ntica
l
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Ha
r
dw
ar
e
-
in
-
t
he
-
l
oop
si
mu
l
at
iv
e setup
f
or
t
est
ing
t
he
c
ombine
d hea
t
-
po
we
r g
e
ner
ating
…
(
I
hor
Shch
ur
)
501
rep
et
it
ion
a
nd
to
dev
el
op
e
ff
ect
ive
fle
xibl
e
al
gorith
ms
of
ma
na
geme
nt
of
powe
r
f
lows
,
pro
vid
i
ng
t
he
consu
mer wit
h el
ect
ric an
d
th
ermal e
nergies.
To
ma
ke
the
con
t
rib
ution
of
t
his
pa
per
c
le
arer,
the
a
rtic
le
will
be
st
ru
ct
ur
e
d
as
f
ol
lows
:
t
he
dev
el
opment
of
t
he
hard
war
e
pa
rt
of
the
sta
nd
is
bri
efly
hig
hlig
hted
in
S
ect
ion
2.
T
he
n,
the
de
velo
pme
nt
of
the
software
pa
rt
of
the
HIL
simulat
ive
set
up
is
intr
od
uc
ed
in
detai
l
in
Sect
ion
3.
T
he
ob
ta
ine
d
sim
ul
at
ion
resu
lt
s
ba
sed
on
t
he
MATL
AB/Si
mu
li
nk
s
of
t
war
e
a
nd
e
xp
e
rime
ntal
re
su
lt
s
obta
ined
on
t
he
sta
nd
of
C
H
P
WECS
for
t
he
modes
of
po
wer
a
nd
heat
gen
e
rati
on
are
com
pared
i
n
Sect
ion
4.
Fin
al
ly,
the
c
oncl
us
io
n
is
g
ive
n
in
Secti
on
5.
2.
HARD
WA
RE
PART O
F T
HE STA
N
D
2
.
1
.
El
ectric
al
a
n
d h
e
at ge
ne
rators
The
PMSG
tha
t
was
desi
gn
e
d
an
d
man
ufact
ur
e
d
f
or
t
he
study
of
th
e
CH
P
W
ECS
sta
nd
has
an
a
xial
const
ru
ct
io
n.
It
co
ns
ist
s
of
t
w
o
i
den
ti
cal
a
rm
at
ur
es
bet
wee
n
w
hich
th
ere
is
a
ste
el
dis
k
r
ot
or
with
perma
nen
t
mag
nets
pl
ace
d
on
bo
t
h
si
de
s
of
t
he
disk
a
s
sho
wn
in
Fi
gure
1
.
T
o
el
i
minate
a
to
oth
tor
que,
the
ge
ner
at
or
armatu
res
ha
ve
a
s
moot
h
(to
oth
le
ss)
c
on
st
ru
c
ti
on
.
T
he
wind
ing
coils
are
gl
ued
to
the
s
urf
ace
of
the
la
minate
d
disk
yo
ke.
The
same
na
me
phase
windin
gs
of
both
ar
matu
res
a
re
set
f
or
t
h
in
ph
ase
a
nd
co
nnect
ed
i
n
series.
The para
mete
r
s of the
P
M
S
G
are give
n
i
n
T
able 1.
(a)
(b)
Figure
1. Tw
o
-
armatu
re a
xial
PM
S
G man
ufa
ct
ur
e
d
f
or the
s
ta
nd
:
(a)
first a
rmatu
re a
nd
di
sk
ro
t
or w
it
h
permane
nt ma
gn
et
s
, (b
)
sec
ond ar
mature
Table
1.
Para
m
et
ers
of t
he
P
MSG
P
a
ra
m
e
t
er
V
a
l
u
e
R
a
t
e
d
p
o
we
r
[
W
]
300
R
a
t
e
d
p
h
a
se
v
o
l
ta
g
e
[V
]
18
R
a
t
e
d
s
p
e
e
d
[
r
p
m
]
250
R
a
t
e
d
t
o
r
q
u
e
[
N·m
]
1
1
.
5
N
u
m
b
er
o
f
p
ai
r
s
o
f
p
o
l
e
s
12
P
h
a
s
e
w
i
n
d
i
n
g
r
e
s
is
t
a
n
c
e
[Ω
]
0
.
5
2
8
W
i
n
d
i
n
g
i
n
d
u
c
t
a
n
ce
[H
]
0
.
0
0
1
1
F
l
u
x
l
i
n
k
a
g
e
b
y
PM
[
W
b
]
0
.
0
5
9
4
In
our
pr
opos
e
d
ro
ta
ti
ng
el
ec
trom
a
gnet
ic
tr
ansfo
rmer
of
mecha
nical
en
ergy
i
nto
heat
(ET
M
E
H)
a
s
sh
ow
n
in
Fig
ure
2
[
8],
t
he
fi
xed
in
du
ct
or
1
has
t
he
ci
rc
ul
ar
placed
ste
el
te
et
h
2,
on
w
hich
t
he
c
oils
3
a
re
arr
a
ng
e
d.
To
increase
the
c
r
os
s
-
sect
ion
al
a
rea
of
the
ma
gnet
ic
ci
rcu
it
in
the
ai
r
gap,
th
e
te
et
h
are
e
quipp
e
d
with
the
ste
el
ti
ps
4.
The
ste
el
disk
5
is
fixe
d
to
the
ve
rtic
al
ro
ta
ti
on
sh
a
ft
6
of
the
V
A
WT
a
nd
is
pla
ced
at
minimu
m
distance
f
r
om
t
he
ti
ps
.
T
he
thi
n
la
ye
r
7
of
a
non
-
fe
rroma
gn
et
ic
mate
rial
with
hi
gh
el
ect
rical
cond
uctivit
y
(c
oppe
r)
is
ap
plied o
n
the surfa
ce
of
the
dis
k
5.
I
f
the
coils
a
r
e
co
nn
ect
e
d
i
n
serie
s,
t
heir
e
nds
a
re
tur
ned
one
at
a
ti
me
a
nd
dir
ect
cu
rr
e
nt
fl
ows
t
hro
ugh
t
he
coils,
a
c
on
sta
nt
ma
gnet
ic
flu
x
of
al
te
r
na
ti
ng
po
la
rity
will
be
cl
os
e
d
th
rou
gh
the
dis
k
body.
As
t
he
disk
r
otate
s,
this
flo
w
at
eac
h
po
i
nt
in
the
dis
k
bo
dy
will
al
ready
be
var
i
able
,
wh
ic
h
will
be
acc
ompan
ie
d
by
t
he
ge
ne
rati
on
of
E
MF
in
t
he
disk
a
nd,
acc
ordin
gl
y,
e
ddy
currents
.
T
he
y
will
be
pa
rtic
ularly
sig
nific
ant
in
t
he
non
-
fe
rroma
gn
et
ic
la
ye
r
on
t
he
disk.
J
oule
he
at
from
eddy
cu
rr
e
nts
is
a
posit
ive
e
ffec
t
of
t
he
E
T
M
E
H.
T
o
us
e
it
,
the
dis
k
m
ust
ha
ve
good
th
ermal
c
onta
ct
with
t
he
coo
la
nt
li
quid
8,
w
hich
will
trans
fer
heat
to
the
sto
rag
e
water
heater.
The
ETME
H
is
pla
ced
in
the
ho
usi
ng
9,
wh
ic
h
is
well
insu
la
te
d
the
rm
al
ly.
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
499
–
51
0
502
Figure
2. The
s
ketch o
f
the
Е
TМЕ
H
A
mag
nitu
de
of
the
ge
n
e
rated
heat
flo
w
will
de
pend
on
t
he
ge
om
et
r
y
of
the
E
PMEH
desi
gn,
t
h
e
mag
netiz
ing
force
of
the
in
duct
or
,
wh
ic
h
ca
n
be
c
on
t
ro
ll
e
d
by
cha
ngin
g
the
excit
at
ion
current,
t
he
r
otati
on
sp
ee
d
of
the
di
sk
,
a
nd
it
s
str
uc
ture.
T
he
design
a
nd
opti
miza
ti
on
of
the
E
TME
H
pa
rame
te
rs
wer
e
ca
rr
i
ed
ou
t
us
in
g
t
he
ANSYS
softwa
re
for
fiel
d
m
ode
li
ng
by
the
fi
nite
el
ements
method
(in
thi
s
case
ma
gnet
ic
and
thermal
fiel
d) [20].
Accor
ding
t
o
the r
es
ults of
th
e
cal
culat
io
ns
,
the
e
xperime
nt
al
prot
otype
of
ET
M
E
H
with
the
i
nducto
r
consi
sti
ng
of
20
te
et
h
with
th
e
coils
was
des
ign
e
d
a
nd
ma
nufact
ur
e
d.
It
is
rated
for
a
the
r
mal
outp
ut
power
of
1000
W
at
a
r
ot
at
ion
al
sp
ee
d
of
250
prm,
a
n
excit
at
ion
c
urr
ent
of
2.0
A,
a
nd
a
n
e
xcita
ti
on
wi
nd
i
ng
volt
age
of
12
V.
At
the
same
ti
me,
25
W
of
el
ect
ric
al
powe
r
is
use
d
f
or
the
exc
it
at
ion
,
th
at
is,
the
ET
M
E
H
rate
d
eff
ic
ie
nc
y
is
not
lowe
r
tha
n
0.95.
The
oute
r
diamet
er
of
t
he
in
du
ct
or
yoke
is
340
m
m,
it
s
coil
heigh
t
is
42
mm. The
b
asi
c
d
esi
gn eleme
nt
s o
f
the e
xperi
mental
pr
oto
ty
pe
a
re show
n
i
n
Fi
gure
3.
(a)
(b)
(c)
Figure
3. Ba
sic
eleme
nts
of th
e d
esi
gn of the
ETME
H
e
xper
imenta
l p
ro
t
otype: (a
)
in
duct
or w
it
h 20 pole
s
, (b)
ro
t
or
with c
op
per surface
lay
er, (c)
pro
t
otyp
e assem
bled
2.2.
H
ardwar
e pa
r
t of t
he
WTE
To
imple
ment
the W
TE
hard
war
e
, th
e s
qu
ir
rel cage IM of a rate
d
po
wer
of
1.5 kW
a
nd
a rated
s
peed
of
2860
r
pm
w
as
us
ed
.
T
o
ma
ke
it
co
nfo
rm
t
o
t
he
P
M
S
G
a
nd
E
TME
H
s
pe
ed,
a
w
orm
ge
arbo
x
with
th
e
gear
rati
o
of
i
=
12
w
as
app
li
ed
.
T
he
con
tr
ol
f
unct
ion
of
the
IM
i
s
assigne
d
to
t
he
f
reque
ncy
c
onve
rter
Le
ns
e
8200
Vecto
r
with
a
powe
r
of
1.5
kW,
powe
re
d
from
a
sin
gle
-
ph
ase
net
work
of
220
V
.
T
he
c
ho
ic
e
of
this
typ
e
o
f
fr
e
qu
e
nc
y
c
onver
te
r
is
du
e
to
it
h
avi
ng the
opti
on of
vecto
r c
on
t
ro
l
of the
e
le
ct
ro
ma
gn
et
ic
torq
ue.
The
functi
onal
schem
a
of
t
he
sta
nd
f
or
th
e
CHP
WECS
study
i
n
Fi
gure
4
s
hows
t
ha
t
the
W
TE
include
s
t
he
I
M
,
the
f
re
qu
e
nc
y
c
onve
rter
F
C
an
d
the
w
orm
gearb
ox
G
t
hat
c
onnects
th
e
I
M
to
the
el
e
ct
rical
PM
S
G
a
nd
t
o
t
he
heat
ge
ner
a
tor
ET
M
E
H.
A
ll
con
tr
ol
f
un
ct
ion
s
i
nclu
ding
WTE
c
on
t
ro
l
a
re
impleme
nte
d
by
the
PLC.
F
or
this
purpose
,
it
con
ta
in
s
on
-
li
ne
w
orkin
g
s
ubr
ou
ti
ne
s
f
or
s
ha
ping
the
tu
rbul
ent
wi
nd
s
peed
vw,
corres
pondin
g
to
th
e
value
of
the
mecha
nical
W
T
t
orq
ue
T
WT,
as
well
as
the
c
orrespo
ndin
g
refe
re
nce
of
the
M
I
tor
que
T
*.
To
cal
c
ulate
th
e
la
tt
er
two,
th
e
an
gu
la
r
velo
ci
ty
of
WT
is
r
equ
i
red.
An
i
nductio
n
pulse
s
ens
or
IP
S
meas
ur
es
the IM
sp
ee
d b
y op
e
rati
on du
r
ing
eac
h
a
ppr
oa
ch of
t
he blad
es of a c
ooli
ng
fan. T
he pu
lse
s
f
r
om
IP
S
go
t
o
a
pulse
input
of
t
he
PLC
w
hich
by
i
ts
own
pro
gr
a
m
pe
rforms
dig
it
al
filt
erin
g
an
d
cal
c
ulate
s
the
angular
v
el
ocity
of P
M
S
G bas
ed on t
he gea
r rat
io i.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Ha
r
dw
ar
e
-
in
-
t
he
-
l
oop
si
mu
l
at
iv
e setup
f
or
t
est
ing
t
he
c
ombine
d hea
t
-
po
we
r g
e
ner
ating
…
(
I
hor
Shch
ur
)
503
The
sta
nd
al
s
o
inclu
des
to
ols
that
pro
vid
e
va
rio
us
possibil
it
ie
s
for
ge
nerat
ing
a
nd
stori
ng
el
ect
rical
and
the
rmal
e
ne
rgy,
as
well
a
s
ma
nag
i
ng
the
flo
ws
of
t
hese
ene
rg
ie
s
.
T
he
el
ect
rici
ty
ge
ne
rated
by
t
he
P
M
S
G
can
be
st
or
e
d
i
n
an
el
ect
r
oc
he
mica
l
batte
ry
B
that
is
conn
ect
ed
to
the
P
M
S
G
ar
matu
re
windin
gs
via
a
diode
br
i
dg
e
VD
a
nd
a
DC
-
DC
1
tra
ns
ist
or
boost
conve
rter.
T
he
DC
-
DC1
c
on
t
r
ols
the
po
wer
t
ran
s
mit
te
d
to
the
B.
In
a
ddit
ion
,
po
wer
ca
n
be
ext
racted
f
r
om
th
e
PMSG
a
nd
c
onve
rted
to
he
at
wh
e
n
the
t
he
rm
oelect
r
ic
he
at
ers
TEH
a
re
c
onne
ct
ed
to
t
he
a
rmatu
re
winding
an
d
placed
in
a
sto
ra
ge
water
heater.
The
heat
ge
ne
rati
on
capaci
ty
ca
n
be
re
gu
la
te
d,
f
or
e
xam
ple,
by
per
i
od
ic
al
ly
co
nn
ect
in
g
the
T
EH
t
o
t
he
tria
c
switc
he
s
T
VR
[
21].
The
re
gu
la
ti
on
of
the
the
rmal
fl
ow
gen
e
rate
d
by
t
he
ET
MEH
is
acc
ompl
ished
by
varyi
ng
the
c
urre
nt
in
it
s
excit
at
ion
windin
g by mea
ns
of the
DC
-
DC
2
c
onve
rter.
Figure
5
s
hows
pi
ct
ur
es
of the c
reated sta
nd.
A numbe
r of st
ud
ie
s
w
e
re c
ondu
ct
e
d
to
d
et
e
r
mine the
basi
c
par
a
mete
rs of t
he
c
reated sta
nd.
The
t
orq
ue
of
dry
f
rict
ion
on
t
he
IM
s
haf
t
Tс.s
(IM)
is
de
te
rmin
e
d
by
means
of
mea
su
ri
ng
by
a
dynam
om
et
e
r
the
f
or
ce
F
ap
pl
ie
d
to
a
threa
d
w
ound
on
t
he
cylindri
cal
su
r
face
of
the
c
ouplin
g
with
a
rad
i
us
rm
=
0.05 m.
A
s
a r
es
ult, Tс
.s(I
M
)
=
F rm
=
1.03 N·m.
The
mec
ha
nic
al
char
act
e
risti
cs
of
t
he
sta
nd
durin
g
it
s
i
dle
str
oke
we
re
ob
ta
ine
d
th
anks
to
the
po
s
sibil
it
y
of
measu
rin
g
t
he
r
otati
on
al
s
pe
ed
of
the
I
M
and
the
el
ect
r
om
a
gn
et
ic
t
orqu
e
giv
e
n
to
i
t
by
the
fr
e
qu
e
nc
y
c
onver
te
r
.
A
s
a
r
esult,
the
value
of
the
c
oe
ff
ic
ie
nt
of
vis
cous
f
rict
ion
on
t
he
I
M
s
ha
ft
of
bs
(
I
M
)
=
0.001
3 Nm
·s was
det
ermine
d.
The
m
om
e
nt
of
i
ner
ti
a
of
t
he
sta
nd
relat
ive
to
t
he
I
M
sh
a
f
t
Js(
I
M
)
was
de
te
rmin
e
d
by
the
At
woo
d
machine
meth
od
-
m
otio
n
unde
r
t
he
act
ion
of
a
lo
ad
.
To
e
xclu
de
f
rom
t
he
cal
culat
io
n
the
f
rict
ion
f
orce,
it
is
necessa
ry
t
o
c
arry
ou
t
tw
o
e
xp
e
rime
nts
wit
h
tw
o
diff
e
re
nt
loads
[
22].
Ea
ch
of
t
hem
wa
s
hung
i
n
tu
rn
on
a
cord
thr
own
over
the
blo
c
k
and
wou
nd
on
the
same
s
urface
of
the
c
ouplin
g
with
ra
diu
s
r
m.
U
nde
r
t
he
influ
e
nce
of
it
s
weig
ht,
the
load
was
dro
pped
e
ve
nly
from
th
e
hei
gh
t
h
=
2.4
4
m,
w
hile
the
ti
me
of
it
s
moveme
nt
was
meas
ured
.
A
fter
a
series
of
e
xp
e
rime
nts
an
d
ave
rag
i
ng
t
he
resu
lt
s,
we
obta
ined
t
he
f
ollow
i
ng
:
for
a
loa
d
of
m
1
=
4
kg,
the
l
oweri
ng
ti
me
w
as
t1
=
1.3
85
s,
for
a
loa
d
of
m
2
=
6
k
g,
t2
=
0.9
87
s
.
T
he
deter
mined
values
of
acce
l
erati
on
s
we
re
2
11
2
a
h
t
=
=
2.5
44
m/s2
a
nd
2
22
2
a
h
t
=
=
5.0
1
m/s2
respec
ti
vely.
The
m
om
e
nt
of
inerti
a of t
he
st
and w
a
s e
qu
al
to 0.02
47 kg
·m
2
as
a r
e
su
lt
of
the cal
culat
io
n by the e
xp
res
sion [
22]
(
IM
)
=
2
2
(
2
ℎ
−
1
1
2
)
−
1
(
2
ℎ
−
1
2
2
)
2
−
2
−
1
−
2
(1)
Figure
4. F
un
ct
ion
al
sc
hema
of the
stan
d for t
est
ing
of CH
P
WECS
P
L
C
φ
P
M
S
G
E
T
M
E
H
i
a
i
f
*
CS
s
IM
VS
А
G
PC
QF
FC
I
Р
S
T
V
R
T
E
H
VD
v
a
B
А
А
PA
1
PA
2
PA
3
DC
-
DC
2
220
V
12
V
W
T
E
T
*
DC
-
DC
1
i
b
*
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
499
–
51
0
504
Figure
5. Stan
d f
or
t
he
te
sti
ng
of CHP
WEC
S
2.3.
PLC
for
HIL sim
ula
tiv
e setup
The
PLC
WP1
30
MK
I
I
pro
duced
by
M
it
a
-
Tek
nik
Ltd
a
s
sh
ow
n
in
Fi
gur
e
5
[
22]
,
[23
],
con
t
ro
ls
the
exp
e
rime
ntal
s
ta
nd
.
T
his
P
L
C
is
s
pecial
ly
desig
ne
d
to
c
ontr
ol
W
Ts
with
a
rate
d
pow
er
of
up
to
1
MW.
It
include
s
a
set
of
dif
fer
e
nt
I/
O
cha
nnel
s
f
or
bo
t
h
di
gital
a
nd
a
nalo
g
si
gnal
s.
This
ma
ke
s
it
po
ssi
ble
to
us
e
the
con
t
ro
ll
er
as a
sta
nd
-
al
one
de
vice (
with
ou
t a
dd
it
io
nal m
odul
es)
to c
ontr
ol less co
mp
le
x
s
yst
ems.
With
t
he
S
of
t
war
e
Dev
el
opment
Kit
c
reat
ed
by
M
it
a
-
Te
kn
i
k
f
or
the
M
A
TLAB/Si
m
ulink,
it
was
po
s
sible
to
c
ompi
le
the
Sim
ulink
model
source
c
od
e
f
or
t
he
W
P13
0
MK
I
I
co
ntr
oller
platfo
rm.
T
o
do
t
his,
on
e
m
us
t
pre
-
set
blo
c
ks
of
input
an
d
outpu
t
va
riables
in
the
model
that
will
li
nk
M
A
TLAB/Si
m
ulink
al
gorithms
to
the
co
ntr
oller
pro
gr
a
m
va
riables
a
nd
th
e
in
pu
ts/
ou
t
puts
o
f
t
he
i
nte
rf
ace
cha
nnel
s
.
T
he
PEPT
OOL
pro
gr
a
mmin
g
e
nv
iro
nm
e
nt
the
n
al
lows
t
he
c
ompil
ed
co
de
t
o
be
buil
t
into
the
a
ppli
cat
ion
for
t
he
WP130
M
K
II
con
t
ro
ll
er.
T
hu
s,
by
c
reati
ng
a
vi
rtual
obje
ct
of
the
c
on
t
ro
l
sy
ste
m
in
M
A
TLAB/Si
m
ulin
k,
we
can
a
pp
l
y
a
de
velo
p
ed
an
d
de
bugged
al
gori
thm
f
or
a
real
ob
je
ct
c
on
tr
ol
sy
ste
m
base
d
on
t
he
M
it
a
-
T
ekn
i
k
con
t
ro
ll
er.
To
ob
ta
in
t
he
r
esults
of
the
c
ontr
oller
W
P13
0
M
K
II,
as
we
ll
as
to
de
bug
t
he
pr
ogram
,
th
e
con
t
ro
ll
er
op
e
rati
ng
s
ys
te
m
pro
vid
es
the
recordi
ng
of
values
of
an
y
var
ia
bles
f
rom
the
data
base
t
o
a
file
in
real
ti
me.
The reco
r
ded va
riable
data ca
n be
represe
nted
as
w
a
vefo
rm
s or tab
ular
v
al
ues.
3.
SOFTW
AR
E
PAR
T
OF
TH
E HIL SI
M
U
LATIVE
SET
UP
To
f
orm
t
he
I
M
t
orq
ue
c
ontr
ol
la
w
that
ade
qu
at
el
y
sim
ula
te
s
the
V
A
WT
oper
at
ion
on
a
gi
ven
wind
prof
il
e,
we
c
on
sider
t
wo
el
ect
romecha
nical
s
ys
te
ms
as
s
hown
in
Fig
ur
e
6
:
(a)
t
he
VAW
T
directl
y
dri
ve
s
the
PM
S
G;
(b)
t
he
WT
E
with
t
he
co
ntr
olled
I
M
dri
ves
the
P
MSG
t
hro
ugh
t
he
gea
rbo
x
with
the
gea
r
rati
o
і.
The
tor
qu
e
balanc
e
of
the
ge
ne
rator’s
s
haf
t,
un
der
the
c
onditi
on
of
t
he
a
bsolute
ri
gid
it
y
of
mec
han
ic
al
transmissi
on, f
or both
of the
s
e sy
ste
m
s is
de
scribe
d by the
fo
ll
owin
g
(2)
:
-
f
or
WECS
wi
th VA
WT
W
T
e
m
c
d
ω
ω
d
T
T
T
b
J
t
=
+
+
+
(2)
wh
e
re
T
em
is
th
e
el
ect
romag
ne
ti
c
torque
creat
ed
by
the
P
M
S
G,
T
c
is
the
t
orqu
e
of
dry
f
ric
ti
on
i
n
the
VAWT
’s
s
ha
ft, and
J
Σ
is t
he
total
mome
nt of
iner
ti
a
of
t
he VA
WT
w
it
h t
he
rotors
of
t
he gene
rato
r;
-
f
or the
stan
d wit
h WTE
ne
gl
ect
ing
the
b
ac
klash
in mec
ha
nical
tra
nsmi
ss
ion
IM
е
m
с
.s
(
G
)
s
(
G
)
s
(
G
)
d
ω
ω
d
і
T
T
T
b
J
t
=
+
+
+
(3)
wh
e
re t
he
s
ubs
cript s
(G) m
ea
ns
belo
ngin
g
t
o
the
stan
d rela
ti
ve
to th
e
gen
e
rator s
haf
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Ha
r
dw
ar
e
-
in
-
t
he
-
l
oop
si
mu
l
at
iv
e setup
f
or
t
est
ing
t
he
c
ombine
d hea
t
-
po
we
r g
e
ner
ating
…
(
I
hor
Shch
ur
)
505
(a)
(b)
Figure
6. Sc
he
mes
of
elec
tr
ome
cha
nical
sys
te
ms: (a) W
ECS, (b
)
sta
nd w
it
h WTE
Con
si
der
i
ng th
at
с
.
s
(
G
)
с
.
s(
I
M
)
T
i
T
=
,
2
s
(
G
)
s
(
I
M
)
J
i
J
=
,
2
s
(
G
)
s
(
I
M
)
b
іb
=
(4)
and
re
movin
g
from
(
2)
a
nd
(
3)
T
em
,
we
ob
t
ai
n
the
t
orq
ue
con
t
ro
l
la
w
on
the
IM’s
s
ha
ft
w
hich
will
pr
ov
i
de
identic
al
work
of the
WECS
a
nd the
sta
nd
with WTE si
m
ula
ti
ng
the
V
A
W
T’s o
per
at
io
n:
(
)
(
)
(
)
22
I
M
W
T
c
с
.s
(
I
M
)
s
(
I
M
)
s
(
I
M
)
1d
ω
ω
d
T
T
T
i
T
b
і
b
J
і
J
it
=
−
−
−
−
−
−
(5)
Fo
r
ver
i
ficat
ion
t
he
acc
uracy
of
t
he
obta
ine
d
i
n
(
5)
,
c
omp
uter
simulat
io
ns
of
the
el
ect
romecha
nical
sy
ste
ms
of
t
he
WEC
S
a
nd
th
e
sta
nd
with
WTE
sho
wn
i
n
Fig
ur
e
6
were
pe
rforme
d
in
MATL
AB/Si
mu
li
nk.
Also
,
the
simu
la
ti
on
stu
dies
sh
owe
d
th
at
,
f
or
t
he
phys
ic
al
modeli
ng
of
the
op
e
rati
on
of
the
el
ect
ro
-
th
ermal
W
ECS
with
t
w
o
-
a
rmatu
re
P
MSG
an
d
the
ET
M
E
H
with
the
above
-
me
ntioned
pa
rameters
,
it
is
fit
the
V
AWT
with
rated
el
ec
tric
po
wer
of
P
el
=
0.5
kW,
w
hi
ch
is
reache
d
a
t
a
nominal
wind
s
pee
d
of
V
w.n
=
10
m/s.
We
chose
a
th
ree
-
blade
H
-
ro
t
or
V
A
W
T
with
t
he
ra
diu
s
r
.
Its
ae
r
odynamic
c
ha
racteri
sti
c
С
Р
(λ)
-
de
pende
nce
of
t
he
po
wer
coeffic
ie
nt
on the
wind ti
p spe
ed rat
io (TSR
)
λ
=
ω
r
/
V
w
was c
hosen
as the
foll
ow
i
ng for t
hi
s r
esearc
h:
6.0
λ
P
9.47
(
λ
)
1.14
1
λ
Ce
−
=−
(6)
The
de
pende
nc
e
(
6)
pr
ov
i
de
s
the
ma
ximum
value
of
P
m
a
x
C
=
0.3
514
at
th
e
opti
mum
va
lue
of
λ
opt
=
3.
67
5.
Th
e
V
AWT’s
pa
r
amet
ers
for
t
he
ex
per
ime
ntal WECS we
r
e
ca
lc
ulate
d
by
the
ex
pr
es
sio
n
as
sh
ow
n
in Ta
ble 2.
е
l
.
n
W
T
.
n
Σ
η
P
P
=
,
W
T
.n
3
a
P
m
a
x
w
.n
0.5
P
А
СV
=
,
2
A
r
=
,
o
p
t
w.
n
n
ω
V
r
=
,
W
T
.
n
W
T
.
n
n
ω
Р
T
=
(7)
W
he
re
η
=
0,8
5
is
the
total
e
ff
ic
ie
nc
y
of
t
he
P
M
S
G
a
nd
mecha
nical
tra
ns
missi
on,
A
i
s
the
s
wep
t
are
a
of
the
VAWT
.
These
par
a
mete
rs
we
re
inc
orporated
i
nto
th
e
com
pu
te
r
m
odel
to
cal
culat
e
the
curre
nt
T
WT
values
at
diff
e
re
nt w
i
nd
sp
ee
ds
a
nd d
i
fferent a
ngula
r v
el
ociti
es o
f
the
VAWT
.
Table
2.
Para
m
et
ers
of t
he VA
WT fo
r
t
he
sta
nd of C
HP
WE
CS
W
ECS
VAWT
P
е
l
.n
[kW
]
P
WT
.
n
[kW
]
А
[m
2
]
r
[m]
ω
n
[
rad/
s]
T
W
T
.
n
[N
·
m]
J
[kg
·
m
2
]
0
.5
0
.58
8
2
.72
4
1
.41
2
6
.0
2
2
.6
14
.0
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
499
–
51
0
506
Figure
7
s
hows
a
c
ommo
n
c
omp
uter
model
of
the
ge
ne
ral
c
on
t
ro
l
sy
ste
m
i
mp
le
me
nted
i
n
M
A
TLAB/Si
m
ulink
with
the
connecte
d
in
put/
ou
t
pu
t
ports
of
the
PLC.
T
his
c
on
tr
ol
s
yst
em
impleme
nt
s
the
wind
sp
ee
d
si
mu
la
to
r
(
Win
d
Simulat
ion),
the
V
AW
T
m
od
el
(
Wi
nd
T
urbine
S
ubs
ys
te
m),
as
well
as
the
su
bsyste
ms
of
opti
mal
co
ntr
ol
of
t
he
PMS
G
loa
ding
by
the
TE
H
(Puls
Co
ntro
l
S
ub
s
ys
te
m)
a
nd
op
ti
mal
con
t
ro
l
of
t
he
ETME
H
e
xcita
ti
on
(
If
ET
MEH)
.
T
he
m
od
el
has
s
om
e
I/O
po
rts
in
dicat
e
d
by
dif
fer
e
nt
colo
rs:
on
e
discrete
i
nput
po
rt
(
blu
e
)
DI_Rot
orFre
q
-
the
pu
lse
rate
of
t
he
I
M
s
pee
d,
tw
o
a
nalo
g
ou
tpu
t
ports
(
yellow
)
-
AO_T
orq
ueSP_V
is
the
el
ect
romag
netic
tor
qu
e
ref
e
re
nce
of
the
I
M
an
d
AO_If
_E
P
M
E
T_
V
is
the
exc
it
at
ion
vo
lt
age
of
the
ETM
EH
,
an
d
one
disc
rete
ou
t
pu
t
port
(yel
low)
-
D
O
_L
oa
dSwit
chCo
ntr
ol
are
t
he
pu
l
ses
of
con
t
ro
l
of the
PM
S
G
l
oad
i
ng by th
e t
hermo
el
ect
ric h
eat
ers
. Th
e
main
d
at
a (av
e
ra
ge
a
nd
tur
bu
le
nt w
i
nd
sp
ee
d,
init
ia
l
torque
for
fast
VAW
T
sta
rt)
e
nter
the
gr
ee
n
port
s
f
rom
t
he
P
L
C
pro
gr
a
m.
T
he
s
ys
te
m
pr
ovides
recordi
ng in
t
he
PLC me
mor
y o
f
the
b
asi
c
va
riables
(
V
w,
ω
,
λ
,
С
Р,
T
IM
) b
y
t
he ora
nge
ports.
4.
RESU
LT
S
AND DI
SCUS
S
ION
The
a
dequac
y
of
t
he
WTE
w
ork
was
ver
ifi
ed
by
c
ompa
ring
the
resu
lt
s
ob
ta
ine
d
e
xper
imenta
ll
y
o
n
the
HI
L
sim
ulati
ve
set
up
with
the
a
nal
og
ic
al
resu
lt
s
of
c
ompu
te
r
sim
ulati
on
i
n
the
M
A
TL
AB/Si
mu
li
nk
of
the
op
e
rati
on
of
t
he
stu
died V
A
WT wit
h
t
he direct
-
dri
vi
ng elec
tric
al
an
d t
he
r
mal ge
ner
at
or
s
.
4.1.
P
ow
er
g
e
nera
tion
Fo
r
powe
r
ge
ne
rati
on
re
searc
h,
the
sim
plest
co
nfi
gurati
on
of
the
sta
nd
w
as
ta
ke
n
in
the
f
orm
of
a
non
-
re
gu
la
te
d
PM
S
G
loa
ding
thr
ough
a
diode
br
i
dg
e
to
the
B
with
a
volt
age
of
13.
2
V.
I
n
t
his
cas
e,
the
corres
pondin
g win
dings
of th
e two ge
n
e
rato
r
ar
matu
res we
re
placed i
n phase
and c
onnec
te
d
in
series.
Stea
dy
sta
te
mo
de
s
of
W
EC
S’
s
work
at
th
e
const
ant
wi
nd
s
peed
s
fro
m
3
to
8
m/s
w
ere
init
ia
ll
y
inv
est
igate
d.
C
omparis
on
of
the
res
ults
obta
ined
from
t
he
c
ompu
te
r
simul
at
ion
of
the
e
xperime
ntal
W
E
CS
in
M
A
TLAB/Si
m
ulink
an
d
the
r
esults
of
a
simi
la
r
stu
dy
c
onduct
ed
i
n
e
xperi
ment
as
the
wa
veforms
of
the
basic
var
ia
bles
rec
orded
by
t
he
P
LC
show
s
tha
t
the
ma
ximum
de
viati
on
of
the
e
xperim
ental
res
ults
from
t
he
theo
reti
cal
o
ne
s does
not exce
ed 10%
.
To
te
s
t
the
a
de
qu
ac
y
of
WT
E
op
e
rati
on
in
dyna
mic
m
od
es
,
a
WECS
’s
w
ork
was
perfor
med
i
n
the
mode
of ele
ct
rici
ty g
e
ner
at
in
g at
turb
ulent
wind spee
d. T
he
t
est
w
in
d
s
pee
d was
modele
d b
y
the
exp
ressio
n
(
)
(
)
w
w
.
a
v
e
r
w
w
2
.
2
s
i
n
ω
0
.
8
0
.
7
c
o
s
8
ω
1
.
1
V
V
t
t
=
+
−
−
+
(8)
W
it
h
an
a
ver
a
ge
wind
s
peed
of
V
w
.aver
=
5.0
m/s
a
nd
a
lo
w
er
ci
rcu
la
r
fr
e
qu
e
nc
y
of
wind
sp
ee
d
ch
an
ge
o
f
w
ω
0
.
2
4
2
π
6
0
=
s
-
1
.
Figure
7. Ge
ne
ral co
ntr
ol s
ys
te
m for ex
pe
rimental
r
es
earc
h
im
pleme
nted
in MATL
AB/Si
mu
li
nk s
of
t
w
are a
nd
relat
ed
to
WP1
30 MK
II c
on
tr
oller
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
P
ow Elec
& Dri S
ys
t
IS
S
N: 20
88
-
8
694
Ha
r
dw
ar
e
-
in
-
t
he
-
l
oop
si
mu
l
at
iv
e setup
f
or
t
est
ing
t
he
c
ombine
d hea
t
-
po
we
r g
e
ner
ating
…
(
I
hor
Shch
ur
)
507
Figure
8
s
how
s
the
res
ults
of
simulat
io
n
a
nd
e
xp
e
rime
nt
a
t
the
wi
nd
te
st
prof
il
e
(8).
Co
mp
a
rison
of
these re
su
lt
s s
hows
their
h
i
gh
consi
ste
ncy (
m
aximum e
rror
do
e
s
no
t e
xcee
d 15%).
(a)
(b)
Figure
8. Ba
sic
v
a
riables
ob
ta
i
ned by t
he
sim
ulati
on
(a) a
nd
for
the
ex
pe
rimental
WECS’
s work
with
HIL
simulat
ive set
up
(b) fo
r
el
ect
ri
ci
ty g
e
ner
at
in
g an
d direct
c
ha
rg
i
ng of the
b
at
te
ry
at t
he
te
st
tur
bu
le
nt w
i
nd
sp
ee
d
(
8)
4.2.
He
at g
e
n
erat
i
on
Ex
per
ime
ntal
stud
ie
s
of
the
E
TME
H
ope
rati
on
with
the
HIL
simulat
ive
s
et
up
wer
e
perf
ormed
in
a
cl
os
ed
-
lo
op
s
yst
em
with
opti
mal
co
ntr
ol
of
the
ex
ci
ta
ti
on
to
obta
in
ma
xi
mu
m
he
at
at
var
ia
ble
wind
sp
ee
ds
.
The
st
ru
ct
ur
e
of s
uch a c
on
t
rol
sy
ste
m is
sho
wn in Fi
gure
9.
As
a
res
ult
of
t
he
pr
e
vious
e
xperime
ntal
stu
dies
[
18]
,
the
de
pende
ncies
of
the
ou
t
pu
t
t
he
rmal
power
of
ET
M
E
H
(h
e
at
flu
x)
P
h
a
nd
the
c
orres
pondin
g
mecha
nic
al
tor
qu
e
T
HG
on
t
he
s
haf
t
of
t
he
t
her
mal
ge
ne
rator
from
the
an
gula
r
velocit
y
ω
a
nd
the
e
xcita
tio
n
c
urren
t
І
f
w
ere
ob
ta
i
ned.
T
hese
de
pe
nd
e
nc
es
in
the
sim
ul
at
ion
model are
pre
s
ented
i
n
th
e L
ook
-
up
-
ta
ble
i
n Fi
gure
9
t
he dot
te
d
sec
ti
on si
mu
la
te
s the
ET
M
E
H’
s
work
.
Figure
9. Bl
oc
k diag
ram of
th
e ET
M
EH
cont
ro
l s
ys
te
m
The
ET
M
E
H
c
on
t
ro
l
syst
em
has
tw
o
st
ru
ct
ur
es
that
ca
n
be
use
d
in
diff
e
ren
t
CHP
WE
CS’s
m
od
es
(sw
it
che
d
by
t
he
ke
y
S
):
1)
f
eedb
ac
k
c
on
t
r
ol
subsyst
em
with
a
powe
r
con
t
ro
ll
er
PC
f
or
sta
bili
zi
ng
the
heat
flu
x
acc
ordin
g
to
the
it
s
ref
e
re
nce
*
h
P
;
2)
fee
dfo
rw
a
rd
co
ntr
ol
s
ub
s
ys
te
m
w
hic
h
f
orms
a
n
opti
mal
ref
e
re
nce
of
the
e
xitat
ion
c
urren
t
*
f
I
with
re
gard
t
o
t
he
V
A
WT
a
ng
ular
ve
locit
y.
The
op
t
imum
de
pe
nd
e
nce
(
)
*
f
ω
I
wh
ic
h
is
wr
it
te
n
i
n
the
c
on
tr
ol
bloc
k
OCB
is
obta
ined
by
im
po
si
ng
the
c
urve
of
the
opti
mal
VAWT
powe
r
3
W
T
.
o
p
t
a
Р
.
m
a
x
w
0
.
5
ρ
P
А
r
С
V
=
on
the
ch
arac
te
risti
cs
(
)
hf
ω,
PI
.
As
a
res
ult,
the
f
ollow
i
ng
de
pe
nd
e
nce
was
ob
ta
ine
d
(
)
*2
f
ω
1
.
4
4
2
1
0
.
0
7
7
8
2
ω
0
.
0
0
5
1
9
1
ω
I
=
−
+
(9)
The
a
uto
m
at
ic
con
t
ro
l
of
t
he
ETME
H
e
xcita
ti
on
c
urren
t
is
pro
vid
e
d
in
a
c
losed
-
lo
op
co
nt
ro
l
s
ys
te
m
with
a
n
e
xcita
ti
on
c
urre
nt
c
ontr
oller
CC
.
D
ue
to
the
small
value
of
t
he
e
le
ct
ro
ma
gn
et
ic
ti
me
co
ns
ta
nt
of
the
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8
694
In
t J
P
ow
Ele
c
&
D
ri
S
ys
t,
V
ol
.
12
, N
o.
1
,
Ma
rch
20
21
:
499
–
51
0
508
excit
at
ion
windin
g
e
f
f
T
L
R
=
=
0.3
28
H
/
3.97
Ω
=
0.0
82
6
s,
the
CC
has
a
pro
porsi
on
al
struct
ur
e
with
the
gai
n
of 50.
Іn
t
he
PLC
as
sh
ow
n
in
Fig
ure
7,
the
ET
MEH
e
xcita
ti
on
con
t
ro
l
s
ubsy
st
em
is
impleme
nted
in
t
he
blo
c
k
If
EP
MET
based
on
the
str
uctu
re
s
how
n
in
Fi
gure
9.
The
e
xc
it
at
ion
cu
rr
e
nt
refe
re
nce
si
gnal
is
transmitt
ed
throug
h
the
ADC
port
AO_If
_EPMET
_V.
Figure
10(a
)
sh
ows
the
obta
ined
res
ults
of
co
mputer
s
imulat
ion
of
t
he
e
xp
e
rime
ntal
WECS
’s
op
e
rati
on
i
n
t
he
m
od
e
of
the
r
mal
ene
r
gy
ge
ner
at
io
n
us
i
ng
the
ET
M
E
H
w
it
h
the
gi
ven
w
ind
spe
ed
te
st
prof
il
e
(8).
As
can
be
seen
from
the
ob
ta
ine
d
ti
me
dep
e
ndence
s,
t
he
op
ti
mal
c
on
trol
of
the
exci
ta
ti
on
c
urren
t
of
the
ETME
H
le
a
ds
to
su
c
h
mech
anical
loadi
ng
on
t
he
s
haf
t
t
ha
t
prov
i
des
the
op
ti
m
um
V
A
WT
a
ngular
ve
locit
y
(from
20
s
to
180
s
a
nd
f
rom
210
s
to
3
40
s)
.
T
hat
in
tur
n
e
ns
ures
th
e
C
P
value
cl
ose
to
it
s
maxi
mu
m
,
a
n
d
hen
ce
t
he
ma
ximu
m
t
hermal
powe
r
ge
ne
rati
on
in
the
E
TME
H
r
otor.
I
n
a
ddit
ion
,
a
minim
um
value
of
V
AWT
angular
velocit
y
of
10
ra
d/s
was
set
in
t
he
con
t
ro
l
s
ys
te
m.
T
her
e
fore,
at
low
wind
s
peeds
,
the
exc
it
at
ion
vo
lt
age
ref
e
re
nce
be
gins
to
be
adj
us
te
d
by
t
wo
-
posit
io
ns
-
t
urns
on
a
nd
t
urns
off
mainta
inin
g
th
e
sta
ble
minimu
m
VAWT a
ngular
ve
locit
y
(
from
18
0
s t
o 2
10 s).
Figure
10(b)
s
hows
i
n
the
s
ame
arra
ngem
ent
as
Fig
ur
e
10(a)
t
he
simi
la
r
wa
veforms
ob
ta
ine
d
i
n
exp
e
rime
ntal
stud
ie
s
of
t
he
C
HP
WECS
’s
w
ork
with
the
H
IL
sim
ulati
ve
s
et
up
i
n
the
rma
l
energy
ge
nerat
ion
mode. Com
pari
so
n of
t
he
ti
me d
epe
ndencies
o
f
the s
ame va
riables,
ob
ta
ine
d
in the p
hy
sic
al
an
d
mat
hem
at
ic
al
exp
e
rime
nts,
s
hows
thei
r
s
uf
fici
ent
c
onve
r
gen
ce
-
the
e
rro
r
i
n
t
he
w
orki
ng
range
of
w
ind
s
peeds
do
es
no
t
exceed
20%
.
(a)
(b)
Figure
10. Basi
c v
a
riables
ob
t
ai
ned
by sim
ul
at
ion
(a)
a
nd
for
the
experi
me
ntal WECS
’s work
w
it
h HI
L
simulat
ive set
up
(b)
in
the
mode
of h
eat
ge
ne
rati
ng by t
he
E
TME
H
at
t
he
t
est
w
in
d
s
pee
d (8)
5.
CONCL
US
I
O
N
The
c
reati
on
of
CH
P
ge
ne
rati
ng
WECSs
is
a
pro
misi
ng
a
rea
f
or
the
de
ve
lop
me
nt
of
lo
w
-
ca
pacit
y
wind
power
th
at
has
no
t
yet
gaine
d
e
nough
popula
rity
.
T
hi
s
pap
e
r
pro
poses
the
HI
L
si
mu
la
ti
on
set
up
for
the
study
of
C
HP
ge
ne
rati
ng
W
ECS,
w
hich
c
on
sist
s
of
the
W
TE
co
ntr
olled
by
IM
t
ha
t
dri
ves
tw
o
s
pecial
gen
e
rato
rs
-
el
ec
tric
al
PM
S
G
a
nd
the
rmal
ET
M
E
H.
Su
c
h
a
set
up
ma
kes
it
possible
t
o
phys
ic
al
ly
sim
ula
te
the
work
of
C
HP
gen
e
rati
ng
WE
CS
in
dif
fer
e
nt
wind
c
onditi
ons,
pro
mp
tl
y
a
nd
with
the
re
quire
d
re
pro
duc
ibil
it
y.
This
al
lo
ws
c
onduct
in
g
a
wide
range
of
re
se
arch
f
or
t
he
de
velo
pm
e
nt
of
c
on
t
ro
l
s
ys
te
ms
of
opti
mal
V
AWT
loading
at
eac
h
wind
s
peed
by
eac
h
of
the
ge
ne
rato
rs,
as
well
as
syst
ems
of
e
nerg
y
flo
w
c
ontr
o
l
f
or
bo
t
h
gen
e
rato
rs
with
a
gi
ven
pri
or
it
y
or
t
he
re
quired
powe
r
ge
ne
rated
[
24]
.
A
nothe
r
pro
misi
ng
area
of
resea
rch
for
the
de
velo
ped
set
up
is
the
operati
on
at
wi
nd
s
pee
ds
that
exceed
t
he
nominal
VAW
T
rati
ng.
I
n
th
is
case,
diff
e
re
nt
act
iv
e
sta
ll
c
on
tr
ol
m
et
hods
[25]
are
re
qu
ire
d
in
or
der
to
li
mit
the
powe
r
that
the
VAWT
ha
rv
est
s
from
the
wind.
T
he
us
e
of
C
HP
ge
ne
rati
ng
pr
ese
nts
ne
w
oppo
rtu
niti
es
to
increase
the
e
f
fici
ency
of
W
ECS
at
high
wind
s
pe
eds
beca
us
e
of
the
possibil
it
y
to
c
onsume
t
wo
ty
pes
of
ge
ner
a
te
d
e
nerg
y.
H
oweve
r,
the
re
are
new
c
halle
nge
s
to
mai
ntaini
ng
th
e
mecha
nical
stre
ngth
of
V
A
WT,
as
well
as
de
velo
ping
ap
propriat
e
al
gorithms
an
d co
ntr
ol laws.
REFERE
NCE
S
[1]
R.
L
acal
-
Arán
tegui
,
“Globalizat
i
on
in
the
wind
ene
rgy
industry:
cont
r
ibut
ion
an
d
e
conom
i
c
im
p
ac
t
of
European
com
pan
ie
s”,
R
en
ewabl
e
Ene
rg
y
,
vol.
134
,
Apri
l
2
019,
pp
.
612
-
62
8
,
2019
.
[2]
Z.
Si
mi
c
,
J
.
Ha
vel
ka
and
M.
Vrhovca
k,
“Sm
al
l
wind
tu
rbines
-
a
unique
seg
me
nt
of
the
wi
nd
power
ma
rk
et
”
,
Re
newab
le
Ene
r
gy
,
no
.
50
,
Feb
ru
ary
2013,
pp
.
10
27
-
1036,
2014
.
Evaluation Warning : The document was created with Spire.PDF for Python.