In
te
r
n
ation
a
l Jou
rn
al
o
f Po
we
r
Elec
tron
ic
s an
d
D
r
ive S
y
stem
(IJ
PED
S
)
V
o
l.
10, N
o.
1, Mar
ch 20
19,
p
p.
463~
4
7
8
IS
S
N
: 2088-
86
94,
D
O
I
:
10.11
59
1
/ij
ped
s
.
v10
.
i
1.pp
4
63-
47
8
463
Jou
rn
a
l
h
o
me
pa
ge
:
ht
tp:
//i
a
e
score
.
com
/
j
o
u
r
na
l
s
/
i
n
d
e
x
.
p
hp/IJ
PED
S
NSGA-II and MOPSO based optimi
zation for sizi
n
g of hybrid
PV/win
d/b
a
ttery en
ergy storage system
Mo
h
a
ma
d
Izdi
n
Hlal
1
, V
i
g
n
a
K.
Ramach
and
a
ramu
rt
h
y
a
2
, S
an
jee
v
ik
u
m
ar Pad
man
a
b
a
n
3
,
Ham
i
d R
eza Kab
o
l
i
4
, A
ref Pou
rye
k
ta
5
,
T
u
a
n
A
b
R
a
s
h
id
b
in
T
u
a
n
A
b
du
l
l
a
h
6
1,
2
,
5
In
sti
t
ute
of P
ow
er
E
n
g
i
n
eering,
D
ep
artm
ent
of
Elect
rical P
o
w
er
Engin
eerin
g,
C
oll
e
ge
o
f
En
g
i
neerin
g
,
U
n
iv
ersi
ti
Ten
aga N
a
sio
n
al, M
a
lay
s
i
a
3
Depart
m
e
nt
o
f E
n
erg
y
T
echno
lo
gy
,
A
a
l
bo
rg
U
n
i
versity
,
D
e
nm
ark
4
U
M
P
ower Energ
y D
e
di
cated
A
dv
anced
Cent
r
e
(UMPED
AC)
,
W
i
s
m
a
R
&D
,
Uni
vers
ity
of
M
alaya
(UM
)
,
M
a
lay
s
i
a
6
Institute of Ene
r
gy P
o
l
icy and R
e
search (IEPRe), Univers
iti
Ten
aga N
a
sio
n
al, M
a
lay
s
i
a
Art
i
cl
e In
fo
ABSTRACT
A
r
tic
le hist
o
r
y
:
Re
ce
i
v
e
d
Ju
l
3
1,
201
8
Re
vise
d S
e
p 19,
201
8
A
c
c
e
pte
d
N
ov 3,
201
8
Th
is
p
ap
er
p
re
sen
t
s
a
S
t
and
-
alo
n
e
H
y
b
r
id
R
en
ewab
le
E
n
e
rgy
S
y
st
e
m
(S
HRES)
a
s
a
n
altern
ati
v
e
to
f
ossil
f
u
el
b
ased
g
en
erat
ors
.
T
h
e
Ph
otov
o
lta
ic
(P
V)
p
an
els
and
w
i
n
d
t
urbi
nes
(WT)
a
re
d
es
ig
ned
f
o
r
th
e
M
a
l
a
ys
i
an
l
o
w
wi
nd
s
peed
c
on
di
ti
ons
w
it
h
ba
t
t
e
r
y
En
ergy
S
t
o
ra
ge
(
BES
)
t
o
prov
i
d
e
electri
c
po
wer
to
t
he
l
o
a
d
.
T
h
e
a
p
p
rop
r
iat
e
s
iz
i
ng
of
each
c
om
p
o
n
e
n
t
w
as
accom
p
lis
hed
us
i
n
g
N
o
n
-
dominat
ed
S
ort
i
n
g
G
enetic
A
lgo
r
it
h
m
(
NS
G
A-II)
a
n
d
M
u
lt
i-
O
b
j
e
c
t
iv
e
P
a
r
t
i
c
le
S
w
a
rm
O
ptimi
zat
ion
(MOPSO)
t
echniq
ues.
T
h
e
op
timized
h
ybrid
s
y
s
t
e
m
w
a
s
ex
am
i
n
ed
i
n
MATLAB
usin
g
two
cas
e
s
tudies
to
f
i
nd
th
e
o
p
tim
u
m
numb
e
r
o
f
P
V
pan
e
ls,
wi
n
d
t
urb
i
nes
s
y
ste
m
a
nd
B
ES
th
at
m
inimi
z
es
t
h
e
L
oss
of
P
ower
S
up
pl
y
P
r
ob
abi
lit
y
(LPS
P
)
a
nd
Cost
o
f
En
ergy
(
CO
E
)
.
T
h
e
hyb
rid
p
o
w
e
r
sy
st
e
m
w
as
c
o
nnect
ed
t
o
th
e
A
C
b
us
t
o
in
ves
t
i
g
ate
th
e
sy
st
e
m
p
erf
o
rmance
i
n
s
up
pl
y
i
ng
a
r
u
r
al
s
ettlem
en
t.
R
eal
weat
her
dat
a
a
t
the
lo
catio
n
o
f
i
nteres
t
was
uti
l
i
z
e
d
i
n
t
h
is
p
aper.
T
h
e
resul
t
s
o
b
t
a
i
n
e
d
f
r
o
m
t
h
e
t
w
o
s
c
e
n
a
r
i
o
s
w
e
r
e
u
s
e
d
t
o
c
o
m
p
a
r
e
t
h
e
s
u
i
t
a
b
i
l
ity
o
f
the
NSG
A
-II
and
MO
PSO
m
e
thod
s
.
T
he
N
SGA
-II
m
e
t
h
od
i
s
show
n
to
b
e
mo
re
accur
a
t
e
w
h
e
rea
s
t
he
M
O
P
SO
m
etho
d
is
f
as
t
e
r
i
n
e
x
ecut
i
n
g
t
he
o
p
timi
zati
on.
Hen
ce,
b
ot
h
th
es
e
m
e
tho
d
s
can
b
e
u
s
ed
f
o
r
t
echno
-econ
o
m
i
c
op
tim
izati
o
n
of
SHRES.
K
eyw
ord
s
:
Cost of ene
r
gy
H
ybri
d
r
enew
a
b
l
e
e
ner
gy
system
Loss o
f
pow
er
s
upp
l
y
pro
b
ab
i
lit
y
MOPS
O
Mu
lt
i ob
jec
t
i
v
e
s
N
SG
A
_
II
Co
pyri
gh
t © 2
019 In
stit
u
t
e
of Advanced
En
gi
neeri
n
g
an
d
S
c
ien
ce.
All
rights
res
e
rv
ed.
Corres
pon
d
i
n
g
Au
th
or:
Mo
ham
a
d I
z
di
n H
l
al,
Inst
i
t
u
t
e
of P
o
w
er
E
ngineer
i
n
g,
D
e
p
a
r
tm
ent of
E
lec
t
ric
a
l
P
ow
er
Engine
eri
ng,
Col
l
e
g
e
of
E
n
g
i
ne
er
in
g ,
U
niv
e
rsit
i
Tena
ga
N
a
s
i
o
na
l,
Jala
n
IK
RA
M
-
U
N
I
TEN, 43
0
00 K
a
ja
ng,
M
a
l
ay
sia.
Em
ail:
moha
ma
di
z
d
i
n
h
l
a
l
@
y
aho
o
.
c
om
1.
I
N
TR
OD
U
C
TI
O
N
I
n
r
e
cent
year
s,
t
he
i
ncre
as
i
ng
c
onc
ern
on
t
he
d
e
p
let
i
on
o
f
fo
s
sil
fue
l
a
nd
g
l
oba
l
w
a
rm
ing
ha
s
ca
t
a
lyz
e
d
t
he
g
row
t
h
of
r
en
ew
a
b
l
e
e
nerg
y
so
urce
s
d
u
e
to
t
he
i
r
p
r
o
mi
si
ng
e
c
o
nomic
a
nd
e
n
v
iron
m
e
nt
a
l
bene
f
i
t
s
[
1],
[2
].
W
in
d
tu
r
b
in
e
s
a
n
d
s
olar
p
h
o
t
ovo
l
t
ai
c
ar
e
com
monl
y
use
d
i
n
t
h
e
rene
w
a
ble
ene
r
gy
sy
stem
t
o
sup
p
l
y
p
ow
e
r
t
o
c
ons
um
ers
i
n
t
he
r
em
o
t
e
re
gio
n
s
be
cau
se
t
he
re
i
s
n
o
f
u
e
l
c
o
st
i
n
vol
ve
d,
e
a
s
y
to
i
ns
t
a
ll
a
n
d
are
also
n
o
n
-po
l
lu
tin
g.
N
ev
e
r
the
l
ess,
d
e
s
ign
i
ng
a
rene
wable
e
n
e
rgy
sys
t
em
c
an
b
e
a
c
h
a
l
len
g
e
.
Thus,
k
n
o
w
l
e
dg
e
o
f
a
l
l
a
sp
ect
s
th
at
i
n
f
l
u
en
ce
s
sy
st
em
p
e
r
fo
rman
c
e
a
n
d
c
om
po
nen
t
s
iz
in
g
is
a
p
re
co
nd
i
tio
n
for
a
n
a
c
c
u
r
a
t
e
S
H
R
E
S
d
e
s
i
g
n
.
L
a
r
g
e
f
l
u
c
t
u
a
t
i
o
n
s
i
n
c
l
i
m
a
t
i
c
a
n
d
m
e
t
e
o
rolo
gic
a
l
c
on
d
i
t
i
ons
c
ause
i
n
t
erm
i
t
t
e
n
cy
o
f
rene
w
a
ble
e
n
e
r
gy
s
ourc
e
s.
M
a
l
ays
i
a
w
h
i
c
h
l
i
es
c
l
o
se
t
o
t
h
e
e
qua
to
r
h
a
s
sea
s
o
n
a
l
w
i
nd
s
p
ee
d
a
n
d
doe
s
no
t
have
a
c
ompr
ehen
si
ve
w
ind
assessme
n
t
[3].
T
his
prob
l
e
m
ca
n
pra
c
ti
call
y
b
e
o
v
erc
o
me
by
us
in
g
bat
t
er
y
ene
r
g
y
s
tora
ge
s
ys
tem
[
4
].
T
he
y
de
si
g
n
a
nd
o
p
t
i
m
iza
t
io
n
m
e
t
h
o
d
s
a
re
i
mp
or
t
a
nt
a
s
p
e
c
t
s
f
or
S
H
R
ES
t
o
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
46
3 –
47
8
46
4
g
u
a
ra
n
t
ee
s
up
pl
y
rel
i
a
bi
li
ty
a
nd
s
ecu
rit
y
,
a
nd
a
l
so
t
o
e
n
su
re
m
a
x
i
m
um
u
ti
li
za
tio
n of
P
V
p
a
nel
s
,
w
i
n
d
t
ur
bi
nes
and
ba
t
t
e
r
y e
n
e
r
gy
stora
g
e,
base
d
o
n the
l
o
a
d
profile
[5]
, [6].
The
r
e
a
r
e
seve
ral
me
th
odo
lo
gie
s
t
h
a
t
u
t
il
i
z
e
trad
it
iona
l
o
p
t
i
mi
za
tio
n
m
e
th
o
d
s
to
d
es
i
gn
a
t
e
c
h
n
o
-
ec
onom
ic
h
y
b
rid
e
n
erg
y
s
ys
te
m
based
on
th
e
Loss
o
f
P
o
w
e
r
S
uppl
y
P
robabili
ty
(
LPS
P
)
[7]-[9
]
.
A
c
o
m
mon
draw
back
o
f
t
h
ese
o
p
tim
i
z
a
t
io
n
me
th
od
s
i
s
t
he
l
ow
c
alc
u
la
t
i
o
n
e
ffi
c
i
e
n
cy,
the
r
efor
e
con
s
um
in
g
e
x
cessi
ve
c
e
n
t
ral
p
r
o
c
e
ssi
ng
unit
ti
me.
In
a
d
d
it
i
o
n
,
t
he
o
pti
m
i
z
at
io
n
met
ho
ds
c
a
n
n
o
t
fi
n
d
t
he
b
es
t
com
p
r
o
mi
se
p
o
i
nt
betw
ee
n
the
o
b
jec
tiv
e
f
u
nct
i
ons.
O
n
t
he
o
t
h
er
h
a
nd,
A
rtif
i
c
ia
l
In
te
l
lige
n
ce
(A
I)
m
e
t
ho
ds
a
re
a
b
l
e
t
o
a
c
h
ie
ve
all
th
e
co
nditio
n
s
,
su
ch
a
s
LPSP
an
d
C
O
E.
A
u
t
ho
rs
i
n
[1
0
]
-[
12
]
p
r
es
en
t
e
d
a
me
t
h
odo
lo
gy
t
o
ca
l
c
ul
a
t
e
t
h
e
op
tim
al
n
umbe
r
of
P
V
panels
,
w
i
nd
ge
ne
rat
i
on,
a
nd
ba
t
t
e
r
y
using
g
e
n
e
tic
a
lg
ori
t
hm
a
ppr
oa
ch
by
ca
l
c
u
l
a
t
i
n
g
LPSP
an
d
s
y
stem
c
o
s
t.
I
n
[
1
3
]
,
A
I
m
eth
o
d
s
su
ch
a
s
the
No
n
-
do
m
i
n
at
ed
S
o
r
t
i
ng
G
e
n
et
i
c
A
l
g
o
r
i
t
h
m
(N
SGA
-
II)
and
Mult
i-
O
b
jective
P
a
rticle
S
warm
O
ptim
ization
(
M
OP
SO
)
wer
e
use
d
t
o
pro
duc
e
a
P
a
r
e
t
o
-op
t
i
m
al
so
l
u
t
i
o
n
i
n
a
si
n
g
le
s
im
ula
t
i
on
ru
n.
S
e
v
era
l
a
u
t
h
o
r
s
h
a
v
e
stu
d
i
e
d
t
h
e
h
yb
ri
d
opt
i
m
iz
at
ion
s
y
st
e
m
u
si
ng
NSGA-
II.
I
n
20
17
,
M
o
sl
e
m
Y
o
u
s
e
f
i
e
t
a
l
.
[
14
]
used
N
SG
A-
II
a
l
g
o
r
ithm
a
n
d
HO
M
E
R
sof
t
ware
t
o
f
i
nd
t
h
e
rob
u
st
p
ro
ject
d
e
s
ig
n
o
f
S
H
R
ES
b
y
t
h
e
var
y
in
g
e
n
gine
l
oa
ds
w
i
t
h
o
pt
im
al
A
nn
ua
l
Ene
r
gy
Rec
o
ver
y
(
A
E
R)
and
t
o
ta
l
c
o
s
t
o
f
t
h
e
sys
t
em
.
Mu
ch
a
tten
t
io
n
w
a
s
pa
id
t
o
di
sc
us
s
S
H
RE
S
a
nd
Ba
tter
y
E
nerg
y
S
t
ora
g
e
(
B
ES
)
si
z
i
n
g
.
R
ef
e
r
e
n
c
e
[
1
5
]
d
i
s
c
u
sse
d
th
e
e
c
onomi
c
a
p
p
r
o
ach
o
f
mu
lt
i
-o
pt
imiz
at
ion
of
a
s
t
a
nda
l
one
h
y
b
ri
d
P
V
-
Wi
n
d
–
B
a
tter
y
a
n
d
d
i
e
se
l
ge
nera
t
o
r
sy
st
em
t
hrou
g
h
t
he
a
pp
lic
at
ion
of
m
ult
i
-o
b
j
ec
ti
ve
u
sin
g
N
S
G
A
_
II
me
tho
d
.
The
e
c
on
omic
bene
f
its inc
lude
t
he
m
inim
iz
at
i
o
n
o
f
pow
er
gener
a
tio
n cos
t
a
nd
m
aximiz
in
g th
e
usef
ul
life
o
f
t
h
e
b
a
t
ter
y
,
inc
l
ud
i
ng
th
e
l
i
fe
l
o
ss,
f
ue
l,
e
nv
iro
n
me
nt
a
l
,
a
nd
m
a
i
n
t
e
na
nce
cos
t
.
In
a
ddi
t
i
o
n
,
i
t
h
as
con
s
i
d
ere
d
t
he
l
i
f
e
t
im
e
char
acte
r
istic
s
of lea
d
-
ac
i
d
b
a
tter
i
es.
Ce
S
han
g
e
t
a
l
.
[1
6]
f
oc
use
d
o
n
the
ba
tte
ry
e
nerg
y
st
or
a
g
e
sy
st
em
s
i
z
in
g
in
s
t
a
nd-
alo
n
e
hy
br
id
pow
er
s
y
s
tem
to
g
uara
n
t
ee
r
e
lia
b
ili
t
y
a
n
d
m
in
i
m
ize
the
lev
e
lize
d
c
o
s
t
o
f
ene
r
g
y
u
sin
g
N
S
G
A
_
II
m
e
t
hod.
I
n
reference
[
17],
tec
h
no-econom
i
cal
o
p
t
imiza
t
i
o
n
for
H
R
ES
w
a
s
a
pp
l
i
e
d
u
si
ng
N
S
G
A
II
m
e
tho
d
t
o
ana
l
yz
e
the
trade
-
o
ff
be
tw
een
t
hre
e
c
on
flict
i
ng
ob
j
e
c
t
iv
es:
tot
a
l
c
o
st,
aut
o
n
o
m
y
leve
l
,
a
nd
w
a
ste
d
e
n
e
rgy
r
a
te.
H
o
w
e
ver
,
the
o
p
tim
al
s
i
z
in
g
of
t
he
s
ystem
com
p
one
nt
s
w
a
s
n
o
t
c
ons
ide
r
ed.
T
he
h
y
b
ri
d
s
o
l
a
r
/
w
i
nd
sy
stem
w
ith
t
h
e
traditiona
l
foss
il
fue
l
-fired
gene
rat
o
rs
w
as
d
escr
i
b
e
d
i
n
[1
8
]
.
W
h
i
l
e
t
wo
o
b
j
ecti
v
es
t
h
a
t
co
nt
roll
ed
t
h
e
N
SGA
-
II
pr
ocedure
w
a
s
pro
p
o
sed
t
o
m
in
imiz
e
the
cos
t
a
n
d
e
m
i
s
s
io
n.
T
o
a
c
h
i
e
v
e
t
h
e
b
e
s
t
c
o
m
p
r
o
m
i
s
e
s
o
l
u
t
i
o
n
,
t
h
e
Fu
zz
y
p
r
io
rit
y
r
an
k
i
ng
h
as
u
se
d
.
T
h
e
p
ap
er
p
re
se
n
t
ed
e
f
f
ec
t
u
ali
ty
o
f
th
e
alg
o
ri
thm
fo
r
e
v
alua
t
i
n
g
t
h
r
ou
gh
so
l
v
i
ng
c
o
s
t
a
nd
em
issi
o
n
d
i
s
pa
tch
iss
u
e
w
i
t
h
ou
t
c
ons
ide
r
in
g
t
h
e
pow
e
r
g
e
n
era
t
i
o
n,
f
or
c
ompa
rison
rea
s
on
s
and
r
e
su
lts
w
e
r
e
com
p
ar
ed
w
it
h
m
e
t
h
od
s
c
o
nta
i
ned
in
t
he
lite
rat
u
r
e.
R
eferen
c
e
[1
9
]
p
re
se
nt
e
d
s
mal
l
h
y
b
r
i
d
rene
w
a
ble
s
y
s
t
em
d
e
p
e
n
d
s
on
the
c
o
st
a
nd
e
nvir
onm
enta
l
cr
ite
ria
u
si
n
g
NS
GA_
II
t
echn
i
q
u
e
,
wi
t
h
t
w
o
in
t
e
grate
d
e
ne
rgy
st
ora
g
e
u
n
i
t
s
from
ba
tte
ry
b
an
ks
a
n
d
h
y
d
r
o
ge
n
s
tora
ge
s
ys
tem
com
b
i
n
ed.
H
e
nce
,
minim
i
z
e
d
the
CO
E
and
gre
e
nh
o
u
se
g
as
e
m
i
ssi
on
(
C
O
2
)
.
The
m
a
in
c
o
n
t
ri
but
io
n
of
t
h
i
s
w
o
rk
w
as
t
h
a
t
t
h
e
com
p
u
tin
g
of t
ota
l
gre
en
h
ous
e ga
s em
issi
o
n
s
a
cc
ordi
n
g
t
o l
i
fe
c
yc
le
a
na
l
y
sis of
e
a
c
h
sys
t
e
m
’
s c
o
mpone
nt.
Ma
ny
r
e
s
ear
ch
ers
ha
ve
i
n
v
es
ti
ga
ted
t
h
e
hy
bri
d
s
ys
tem
usi
ng
MO
P
S
O
met
h
od
.
Au
tho
r
s
i
n
[
20
]
p
r
e
s
ent
e
d
t
h
e
op
ti
mi
zati
o
n
o
f
a
n
o
f
f
-
g
r
i
d
hybri
d
m
i
c
ro
-g
ri
d
sy
s
te
m
t
o
d
e
t
er
m
i
ne
t
he
o
p
t
i
m
al
s
i
z
i
ng
in
t
w
e
lve
Swed
i
s
h
re
gi
on
s.
T
h
e
o
pt
i
m
a
l
d
e
s
i
g
n
was
se
l
e
c
t
ed
a
ft
e
r
r
u
nni
ng
t
h
e
m
u
lti-o
b
j
ec
ti
ve
o
ptim
i
z
a
t
io
n
me
tho
d
t
o
d
e
t
e
rmin
e
t
h
e
t
r
ad
e
-
of
f
b
e
t
w
ee
n
th
e
th
re
e
o
b
j
e
ct
iv
e
s
:
LP
SP,
C
OE
,
and
the
e
n
vi
r
onme
n
ta
l
i
m
pac
t
(
C
O
2).
A
hy
bri
d
r
enew
a
b
le
e
ner
g
y
s
y
st
em
w
i
t
h
m
u
lt
i-
stor
age
sy
stem
c
on
fi
g
u
r
at
io
n
for
bu
il
d
i
ngs
i
n
Ca
na
da
t
ha
t
a
d
o
p
t
s
the
MO
P
S
O
m
e
t
h
o
d
w
as
p
ro
pos
ed
i
n
[
2
1]
f
or
o
p
t
i
m
a
l
e
c
o
nomic
oper
a
t
i
o
n
i
n
o
r
d
e
r
t
o
m
i
n
i
m
i
z
e
t
h
e
t
o
t
a
l
N
e
t
P
r
esent
Cost
(
N
P
C)
a
nd
CO
2.
R
efe
r
enc
e
[
22]
p
ro
p
o
se
d
t
h
e
des
i
gn
of
a
s
ta
nd-
al
o
n
e
h
y
b
ri
d
ge
nera
tin
g
system
to de
t
e
r
mine
t
h
e
op
tim
um si
z
i
n
g
of the
n
um
b
e
r and t
ype
o
f
P
V
p
a
nels
,
w
i
n
d
t
ur
b
i
ne
s
,
ba
t
te
r
y
b
a
n
k,
as
w
e
ll a
s
die
s
el
g
e
n
er
at
o
r
s
loc
a
t
i
o
n
us
ing
MO
P
S
O
m
e
t
h
od.
T
he
s
i
z
ing
w
a
s
d
o
n
e
b
a
se
d
o
n
a
o
ne-
y
ear
d
a
t
a
to
m
inim
iz
e
the cost
a
nd emission.
The
N
S
G
A
-II
a
nd
MO
P
S
O
are
the
mode
rn
r
andom
o
p
t
i
m
iza
t
i
o
n
m
e
tho
d
s
t
h
at
a
re
a
b
l
e
to
f
i
n
d
Pa
reto.
H
e
nce,
b
o
t
h
the
s
e
m
e
th
o
d
s
are
a
p
p
lie
d
to
d
es
ig
n
the
S
H
R
ES
a
nd
t
o
mi
ni
mi
z
e
t
h
e
m
ul
ti
-obj
ect
iv
es
suc
h
a
s
LP
S
P
a
nd
CO
E.
T
h
i
s
pa
per
presen
ts
a
c
om
par
a
t
i
ve
e
va
lua
t
ion
of
t
he
p
erform
ance
o
f
NSG
A
-II
a
nd
MO
P
S
O
t
o
d
et
e
r
mine
t
he
o
pt
im
al
s
izi
n
g
o
f
S
H
R
ES
u
sin
g
P
are
t
o
o
p
t
i
m
iza
tio
n.
I
n
orde
r
to
d
e
t
e
r
m
i
n
e
t
he
b
e
s
t
com
b
i
n
a
t
i
o
n
o
f
e
ne
rg
y
so
urce
s
and
t
o
e
ns
u
r
e
their
sea
m
l
e
ss
i
n
t
e
gr
at
i
o
n
i
n
t
o
t
he
d
is
tri
but
i
on
s
y
s
t
e
m
t
o
be
a
t
the
o
p
tima
l
s
i
z
e,
t
he
num
bers
o
f
P
V
p
a
n
e
l
s,
W
T
sys
t
em
,
and
ba
t
t
er
i
e
s
ar
e
use
d
a
s
the
de
ci
sio
n
v
ar
ia
ble
s
t
o
minim
i
z
e
t
he
L
P
S
P
a
nd
CO
E.
T
he
r
e
s
t
of
t
he
p
a
p
er
i
s
orga
niz
e
d
a
s
fol
l
o
w
s
.
A
brie
f
o
v
e
r
v
i
ew
o
f
hy
bri
d
WT
/P
V
mode
ls
a
n
d
B
ES
i
s
prese
n
t
e
d
i
n
S
ec
tio
n
2.
T
he
o
p
t
ima
l
c
on
fig
u
ra
tio
n
of
t
he
s
t
a
nd-a
l
o
n
e
h
ybr
id
syste
m
i
s
e
x
p
l
aine
d
in
S
ec
ti
on
3.
T
he
c
o
m
pa
rati
ve
a
n
a
l
y
se
s
of
t
h
e
s
i
m
ul
a
t
ion
ar
e
d
i
sc
usse
d
in
S
ecti
o
n
4,
fo
l
l
ow
e
d
by
th
e
c
o
n
c
l
u
si
on i
n
S
ec
t
i
on
5.
2.
PO
WER CIR
C
U
I
T OF
T
HE (S
H
RES
)
I
n
o
rde
r
t
o
pre
d
i
c
t
t
h
e
S
H
RE
S
pe
rform
a
nc
e
,
t
he
e
nerg
y
so
urc
e
s
n
e
e
d
t
o
be
p
ra
ct
ical
l
y
d
es
ig
ne
d
to
me
et
t
he
l
oa
d
de
ma
n
d
.
A
t
t
he
s
am
e
t
i
m
e
,
the
p
o
w
e
r
o
b
ta
i
n
ab
le
f
r
o
m
a
hyb
ri
d
rene
wa
b
l
e
s
y
st
e
m
h
as
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
NSG
A
-II
and
MO
PSO
base
d
opt
im
i
z
a
t
io
n f
o
r siz
i
n
g
o
f
hy
br
id PV
/wi
n
d
/
b
a
ttery
...
(
M
oha
m
ad Izdi
n H
l
a
l
)
46
5
sign
ifica
n
t
fluc
tua
t
ions
d
ue
t
o
w
e
a
t
he
r
c
o
n
d
i
ti
ons
a
nd
he
nc
e
,
t
he
c
ons
ta
n
t
l
o
a
d
dem
a
n
d
m
ay
n
ot
b
e
me
t.
T
o
miti
ga
te
t
h
i
s
is
sue
,
a
b
at
ter
y
b
a
n
k
c
a
n
b
e
inte
gra
t
ed
t
o
the
hy
bri
d
s
yste
m.
H
ow
ever
,
the
hig
h
c
os
t
of
b
a
tter
i
es
is
a
n
issue
in
r
ene
w
able
e
ne
rgy
s
y
s
t
em
s.
T
h
u
s,
o
pt
imiz
ing
t
h
e
s
i
z
e
o
f
t
h
e
P
V
-
W
T
-
B
E
S
s
y
s
t
e
m
b
e
c
o
m
e
s
essen
tia
l.
T
he
s
e
c
o
n
tri
b
ut
ion
s
r
educ
e
the
ca
pi
ta
l
c
o
st
a
n
d
i
ncr
e
a
se
t
he
c
hanc
e
s
o
f
i
n
ve
stm
e
n
t
i
n
rene
w
a
ble
ene
r
g
y
p
l
a
n
t
i
n
s
t
a
ll
a
tio
n.
T
he
r
e
fore,
the
op
t
i
ma
l
c
o
mb
ina
t
i
o
n
o
f
rene
w
a
bl
e
p
o
w
e
r
resour
c
e
s
w
ith
a
p
p
r
o
pria
te
stor
age
siz
i
n
g
,
as
p
ro
po
sed
i
n
t
h
i
s
w
o
rk,
w
ill
gi
ve
a
v
it
a
l
c
on
tri
b
uti
on
fo
r
the
fu
ture
e
c
o
n
o
m
i
c
feas
ib
i
l
i
t
y
o
f
suc
h
pla
n
t
s,
t
h
u
s
ma
king
t
h
e de
sig
n
m
ore
a
t
tra
c
t
i
ve f
or in
v
e
st
o
rs.
2.1.
Win
d
tu
r
b
in
e
mod
e
lin
g
Th
e
wind
i
s
cha
r
ac
t
e
ri
ze
d
by
i
ts
s
p
e
e
d
a
nd
d
i
r
e
c
t
i
o
n
and
i
s
a
f
f
ect
e
d
b
y
fa
ct
o
r
s,
s
u
c
h
as
g
eo
g
r
ap
hi
c
pos
it
io
n,
m
eteoro
lo
g
i
ca
l
fac
t
ors
and
he
ig
ht
a
bo
ve
g
r
oun
d
l
e
ve
l.
W
in
d
t
u
r
b
ine
rea
c
ts
t
o
t
h
e
w
i
n
d
,
capt
u
rin
g
a
part of i
t
s
ki
ne
tic
e
ne
r
gy
a
n
d
s
w
itc
hi
n
g
i
t
in
t
o
u
sab
l
e
e
n
e
r
g
y
.
T
he
o
u
t
pu
t po
w
e
r
of
w
in
d
tu
rbi
n
e
i
s
de
t
er
mine
d
as
a
f
unc
t
i
on
of
t
he
r
ate
d
w
i
n
d
sp
e
e
d
(
V
r
)
,
t
he
c
ut-
i
n
w
i
nd
s
p
ee
d
(
V
ci
)
and
t
h
e
c
u
t-
out
w
i
n
d
spee
d
(
V
co
)
ac
cord
in
g t
o
th
e
f
ol
l
o
w
i
ng
(1):
P
WT
=
0
(1
)
Where,
P
WT
i
s
t
h
e
outp
u
t
p
o
w
er
by
w
i
nd
t
urbi
ne,
i
s
t
h
e
ra
ted
w
i
n
d
pow
e
r
,
V
i
s
t
h
e
w
i
n
d
speed,
,
an
d
repr
esen
t
the
cut-
in
w
in
d
spe
e
d,
nomina
l
w
in
d
speed,
and
cu
t-o
u
t
w
i
n
d
s
p
e
ed
respe
c
t
i
ve
l
y
.
T
h
e
tur
b
i
n
e
c
u
t-in
s
pe
ed
i
s
sma
l
l,
w
h
i
c
h
e
n
h
a
n
c
es
t
h
e
e
f
f
e
c
t
i
v
e
op
era
t
i
on
of
t
h
e
s
y
s
t
e
m
e
v
en
un
der
low
w
i
n
d
spee
d
[2
0
]
-
[2
3].
2.2.
S
olar
P
V
arr
a
y mod
e
l
i
n
g
S
o
lar
pane
l
s
a
r
e
d
ef
i
n
e
d
a
s
a
gro
up
of
c
e
l
l
s
c
on
nec
t
e
d
i
n
p
a
ra
l
l
e
l
a
n
d
se
ri
es
t
o
ge
ne
rate
t
he
r
eq
u
i
red
elec
tr
ical
p
owe
r
b
ased
o
n
m
e
t
e
oro
l
o
g
i
cal
f
a
c
t
ors
s
u
c
h
a
s
so
lar
r
a
dia
t
io
n
a
n
d
tem
p
era
t
ure
.
T
he
c
urre
n
t
m
ode
l
use
d
to
pred
ic
t the
o
u
t
pu
t
po
w
e
r
of a
P
V m
o
d
u
le c
an be
e
xpr
esse
d
thro
ug
h
the
fo
l
l
ow
in
g (2)
and
(3)
[24]:
P
PV
= P
V
ST
C
1
.
(
2)
= T
a
+ (0.
0256
* G).
(
3)
Whe
r
e,
P
V
ST
C
i
s
t
h
e
n
o
m
i
na
l
pow
e
r
i
n
(
k
W),
G
i
s
t
he
g
l
o
bal
s
o
lar
ra
di
a
tio
n
(kW/m2)
,
i
s
t
h
e
so
l
a
r r
a
diat
i
on un
der
S
T
C (10
0
0
/
m
2
),
T
C
is the
tem
p
era
t
u
r
e of P
V
c
e
ll,
=
2
5
C
°
,
is t
h
e PV tem
pera
tu
r
e
c
o
eff
i
ci
ent
,
3
.7
*
10
(1/
Ċ
) a
nd T
a
is the
surrou
n
d
i
n
g tem
p
era
t
ur
e.
2.3.
Batt
ery
s
tora
ge
mod
elin
g
T
h
e
ty
p
i
c
a
l
b
a
tt
e
r
i
e
s
th
a
t
a
r
e
u
s
e
d
f
o
r
h
y
b
r
id
e
n
e
r
g
y
s
y
s
t
e
m
i
n
a
r
e
a
s
o
f
l
o
w
w
i
n
d
s
p
e
e
d
a
n
d
in
t
e
rm
i
t
te
nt
s
o
l
ar
r
adia
t
i
o
n
c
o
n
d
iti
on
s
ar
e
t
h
e
l
e
ad-a
c
i
d
a
n
d
l
i
t
h
i
um
-ba
s
e
d
b
at
terie
s
[
25]
.
Com
m
onl
y,
bot
h
the
s
e
ba
t
t
e
r
ies
ar
e
em
ploye
d
i
n
m
ost
la
rge-
sc
ale
e
n
erg
y
s
t
o
ra
ge
p
ro
j
e
c
t
s
b
e
c
a
us
e
of
t
h
e
i
r
l
o
w
c
ost
,
l
on
g
li
fe
spa
n
,
an
d dura
b
i
l
i
t
y
,
in
a
d
d
i
t
i
o
n
t
o
the
i
r
c
omm
e
rc
ial
a
v
a
i
la
bi
l
it
y
[
26].
Me
m
b
rane
b
ase
d
l
ea
d
a
c
i
d
bat
t
er
ie
s
a
r
e
als
o
a
vai
l
a
b
le
i
n
the
m
a
rket
p
re
sent
l
y
.
Ba
t
t
e
r
y
stor
age
is
s
iz
e
d
t
o
m
ee
t
t
h
e
l
o
ad
d
e
m
a
nd
du
ri
ng
a
s
h
o
rt
ag
e
or
in
t
e
rrup
t
ion
of
r
ene
w
able
e
n
e
rgy
so
urc
e
,
usua
ll
y
r
e
fe
rre
d
to
a
s
A
u
t
onomo
u
s
D
a
y
s
(AD).
R
e
g
u
l
a
rly
,
AD
i
s
take
n t
o
be o
n
e
t
o
t
hree
d
a
y
s a
c
c
ord
i
n
g
t
o (4).
T
ypica
lly ba
t
te
ry
c
apa
c
i
t
y de
si
gn
de
pe
nd
s
o
n
t
he
l
oad an
d A
D
.
Th
us, ba
t
t
e
r
y
c
apac
i
t
y
can
b
e
calc
u
la
t
e
d usi
ng
the f
o
l
l
owi
n
g
eq
ua
t
i
on
:
C
B
=
(
E
L
*
AD) /
(DOD*
μ
bat
* μ
inv
)
4)
Where,
C
B
i
s
th
e
b
a
t
t
e
ry
c
ap
ac
i
t
y
,
E
L
i
s
the
l
o
a
d
d
em
an
d,
D
O
D
i
s
the
de
pt
h
of
d
isc
h
ar
g
e
,
μ
ba
t
is
t
h
e
e
ffi
c
i
en
cy
o
f
ba
t
t
e
r
y
a
nd
μ
inv
is
t
he
i
nver
t
e
r
e
ff
i
c
ie
nc
y.
I
t
is
n
o
t
e
w
or
th
y
th
a
t
,
t
h
e
pr
ece
d
i
n
g
e
x
p
r
essi
on
i
s
o
n
l
y
use
d
w
he
n
t
h
e
hy
bri
d
P
V
/
W
T
s
yste
m
is
u
n
a
ble
to
s
u
p
p
l
y
the
r
e
q
u
i
re
d
e
n
e
r
gy
[
20
].
T
he
h
yb
rid
PV
/
W
T
a
n
d
bat
t
ery
sys
t
em
i
s
f
o
rm
u
l
a
t
e
d
a
s
a
m
u
lti-
ob
je
ct
ive
o
p
t
i
m
i
z
a
t
i
o
n
pr
ob
le
m
to
i
m
p
ro
ve
t
he
t
ec
h
n
o
-ec
onom
i
c
perform
ance
si
m
ulta
ne
ou
sl
y.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
46
3 –
47
8
46
6
2.4.
Hyb
r
id
e
n
ergy
man
a
ge
men
t
syste
m
Th
e
u
n
cert
a
inty
o
f
re
newa
ble
e
n
e
r
gy
s
up
pl
i
e
s
(E
RE
)
is
t
he
m
a
i
n
li
m
i
ta
tio
n
of
h
ybr
i
d
r
ene
w
ab
le
ene
r
g
y
p
l
a
n
t
s.
T
her
e
fore
,
a
n
e
ner
g
y
ma
nag
e
m
e
nt
s
trate
g
y
is
r
e
q
u
i
r
ed
t
o
com
p
lem
e
nt
t
h
e
e
xc
ha
nge
o
f
p
o
w
e
r
f
r
o
m
t
h
e
g
en
era
t
in
g
sou
r
ce
s
a
n
d
t
h
e
l
o
a
d
u
nd
er
v
a
r
i
a
ble
wea
t
h
e
r
c
on
dit
i
o
n
s.
I
t
i
s
c
a
l
cu
la
te
d
us
ing
t
h
e
fo
l
l
ow
i
n
g
(5)
a
nd (
6
):
∆
Enet
(
t
)
=
E
RE
(t) –
E
L
(t)
(
5)
(t) = N
WT
*
E
WT
(t) + N
PV
*
E
PV
(
t
)
(
6
)
Whe
r
e,
∆
En
e
t
i
s
th
e
n
e
t
en
ergy
o
f
S
HR
ES,
N
WT
i
s
t
h
e
nu
mbe
r
o
f
wind
t
u
r
bi
n
e
s,
N
PV
i
s
t
h
e
n
u
m
b
er
o
f
PV
p
an
el,
E
WT
i
s
t
h
e
e
n
er
gy
gene
ra
te
d
b
y
t
he
w
ind
t
u
rb
in
es,
E
PV
i
s
t
h
e
en
ergy
g
e
n
e
r
at
ed
f
ro
m
th
e
PV
p
an
el
s,
and
E
L
(t)
i
s
the
load
d
em
a
nd
a
t
ho
u
r
(
t
) w
h
ere (t)
equals o
n
e
hour.
T
h
e
f
o
l
l
o
w
i
n
g
c
a
s
e
s
a
r
e
t
a
k
e
n
i
n
t
o
a
c
c
o
u
n
t
i
n
t
h
i
s
a
r
t
i
c
l
e
,
t
o
sim
u
la
t
e
a
n
ener
g
y
m
ana
g
em
en
t
stra
te
gy,
a
s
dep
i
c
t
ed
i
n
F
i
g
u
re
. 1:
a.
Whe
n
t
he
g
e
n
e
r
ated
p
ow
er
i
s
hi
g
h
er
t
ha
n
t
h
e
l
o
a
d
d
e
m
a
n
d,
t
he
s
u
rp
lu
s
pow
er
i
s
e
m
ploy
ed
t
o
c
h
arge
t
h
e
bat
t
ery
ba
n
k
.
b.
Whe
n
t
he
g
ene
r
ated
p
ow
er
i
s
hi
ghe
r
t
h
a
n
t
he
l
oad
de
ma
nd
a
n
d
t
h
e
s
ta
te
o
f
ch
ar
ge
o
f
t
h
e
ba
tt
e
r
y
ban
k
i
s
fu
l
l
, the
surp
l
u
s
e
ne
rgy is c
o
n
s
um
ed
i
n
a
du
mp
l
oa
d.
c.
Whe
n
t
he
g
ene
r
ated
e
ner
g
y
is
l
ow
e
r
t
ha
n
t
h
e
l
o
ad
d
e
m
a
n
d,
t
h
e
b
a
tte
ry
b
a
n
k
is
d
i
s
cha
r
ge
d
t
o
s
uffi
c
i
e
n
tl
y
sup
p
l
y
t
he
l
oa
d
d
e
ma
nd.
F
i
gur
e 1.
F
lo
w
c
har
t
o
f t
h
e
h
y
b
ri
d e
n
e
r
gy s
y
s
t
em
3.
OPTIMAL C
O
NFIGURATIO
N
OF TH
E
S
T
A
ND-
ALO
N
E
H
Y
BRID SY
S
T
E
M
O
n
c
e
t
he
h
y
b
r
i
d
c
o
mpo
n
e
n
t
spec
i
f
i
c
a
t
i
o
ns
h
a
v
e
be
en
d
e
t
er
mined,
tw
o
case
s
w
i
l
l
be
i
nve
st
i
g
ate
d
base
d
o
n
m
u
l
t
i
-o
bjec
tive
o
p
t
imiza
t
i
o
n
u
s
i
ng
N
S
G
A
-
I
I
a
n
d
M
O
P
S
O
m
e
t
h
o
d
s.
I
n
t
h
e
first
ca
se,
the
hy
bri
d
syste
m
w
ill
c
o
ns
i
s
t
o
f
P
V
pa
nel
s
,
WT,
an
d
bat
t
ery
ba
nk,
w
hi
l
e
t
he
s
ec
ond
ca
se
c
o
n
sis
t
s
of
P
V
pane
ls
a
n
d
bat
t
ery
ba
nk
on
ly.
The
fo
llow
i
ng
tw
o
s
u
b-sec
tio
ns
i
llu
stra
t
e
t
h
e
de
fini
t
i
o
n
s
o
f
t
he
o
b
j
ec
t
i
ve
f
u
n
c
ti
on
s
i
n
d
et
ail
.
3.1
.
R
e
liabi
lit
y
anal
ys
is
Th
ere
are
t
w
o
ap
p
r
o
a
c
h
es
t
o
d
e
t
e
rmi
n
e
t
h
e
l
o
ng
-t
erm
p
e
rforman
c
e
o
f
LP
S
P
i
n
a
s
t
a
nd-a
l
one
h
y
b
ri
d
syste
m
,
nam
e
ly,
c
h
ro
no
l
ogic
a
l
me
t
hod
a
nd
pro
b
ab
i
list
i
c
te
c
h
n
i
qu
e
s
[
2
7
]
.
Th
e
c
h
ro
nol
o
g
i
c
al
m
et
hod
i
s
mo
re
com
m
on
an
d
a
ccur
a
te,
espec
i
a
lly
t
o
de
t
e
rm
ine
t
h
e
e
n
erg
y
p
r
o
duce
d
from
the
ba
t
t
er
y
an
d
t
h
e
com
p
ut
a
tio
na
l
ti
m
e
is ty
p
i
c
a
l
l
y
large
r t
h
a
n
t
h
a
t o
f
pro
ba
bi
l
i
stic mo
d
e
l
s.
It
i
s
com
mon i
n
t
he chro
n
o
l
o
gi
c
a
l m
ode
ls t
o pe
rform
a
on
e
-
y
e
a
r
s
i
m
u
l
a
t
i
on
wit
h
a
o
n
e
-h
ou
r
t
i
m
e
s
t
ep
.
Co
mpu
t
ati
on
ti
me
i
s
e
s
pe
ci
a
l
l
y
n
ec
e
ssary
b
ec
a
u
s
e
t
hi
s
ki
nd
o
f
m
od
e
l
i
s
gen
e
ral
l
y
u
s
e
d
fo
r
co
mp
on
e
n
t
si
z
e
op
ti
mi
z
a
t
i
on
t
h
a
t
r
e
q
ui
re
s
se
vera
l
i
t
e
r
ati
ons.
H
e
nc
e
,
t
he
chro
no
l
o
g
i
ca
l
me
thod
is
u
til
iz
ed
i
n
th
is
p
a
p
er.
The
LPS
P
i
s
de
f
in
e
d
a
s
t
h
e
p
r
o
b
a
bil
ity
o
f
un
me
t
lo
ad
o
v
e
r
t
h
e
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
NSG
A
-II
and
MO
PSO
base
d
opt
im
i
z
a
t
io
n f
o
r siz
i
n
g
o
f
hy
br
id PV
/wi
n
d
/
b
a
ttery
...
(
M
oha
m
ad Izdi
n H
l
a
l
)
46
7
t
o
t
a
l
en
ergy
p
ro
du
c
e
d
[2
6
]
,
a
s
m
en
t
i
on
e
d
i
n
t
h
e
fi
rs
t
obj
e
c
ti
v
e
.
The
unm
et
l
oa
d
c
a
n
be
c
a
l
c
u
la
te
d
by
ut
i
liz
in
g
the de
fici
t p
o
w
e
r
betw
e
e
n the
loa
d
a
n
d
so
u
rce
s
in S
H
RES thro
ug
h
t
he f
ollowing (7)
[
3
]
:
LP
SP
=
∑
∗
∑
(
7)
3.2.
Ec
on
omi
c
a
n
a
ly
si
s
COE
i
s
d
efi
n
e
d
a
s the
ave
r
ag
e
cost pe
r
ki
l
o
watt-h
o
u
r (
$
/
k
Wh) of
ele
c
t
ric
a
l
ener
g
y
prod
u
ce
d b
y
t
he
hy
bri
d
ene
rg
y sys
t
em
[24
]
,
w
hi
c
h
c
an be
a
c
h
ie
ve
d
vi
a
the
fo
l
l
ow
in
g e
q
ua
tio
ns
:
COE
=
(
CR
F
*
TAC
)
/ E
L
(
8)
CR
F =
(
9)
Where
,
CRF
i
s
the
ca
pita
l
re
cover
y
f
a
c
t
or,
whic
h
ca
lcu
l
a
t
es
t
he
p
r
e
sen
t
v
a
l
ue
o
f
s
y
s
t
em
c
ompo
ne
nts
by
c
onsi
d
er
in
g
t
h
e
in
tere
st
r
ates
(
i
=
ap
pr
ox
im
ated
a
s
6%
)
a
n
d
pro
j
ec
t
li
fe
s
pa
n
(T=
2
0
y
e
a
rs
i
n
our
c
ase).
TA
C is th
e
tot
al an
n
u
al
ize
d
c
os
t i
n
$.
The
to
tal a
n
n
u
a
liz
ed c
os
t
i
s
t
h
e
s
u
m
o
f
th
e
an
nu
al
iz
e
d
c
api
t
a
l
co
st
(
CC
),
opera
tio
n
and
m
a
inte
na
nc
e
cos
t
(
C
O&
M
)
an
d
r
e
plac
em
ent c
o
st
(
C
R
).
TAC
=
C
C
+
C
O&
M
+
C
R
(
10)
C
C
=
(
N
PV
*P
PV
*C
PV
) + (
(
N
WT
*P
WT
*C
WT
)
+
(C
WT
*N
WT
*2
0
/
10
0))
+ (N
B
*C
b
*C
B
)
+
(N
INV
*C
IN
V
)
+
(C
REG_PV
+
C
RE
G
_
WT
).
(
11
)
Where,
C
PV
,
C
WT
,
a
nd
C
b
,
a
r
e
PV
p
an
el
u
nit
-
p
r
i
c
e
,
W
T
u
n
i
t-
p
r
i
ce
,
a
nd
ba
tt
e
r
y
un
it-pr
i
ce
re
sp
ec
ti
v
e
ly
.
A
l
so
,
C
IN
V
,
C
REG
_
PV
,
a
n
d
C
REG
_
WT
a
re
i
nvert
e
r
u
n
it
price,
r
egula
t
or
o
f
P
V
p
rice
,
and
regu
la
tor
of
WT price
re
s
p
e
cti
v
e
l
y. In
a
d
dit
i
o
n
, N
INV
i
s
t
h
e
num
ber of i
n
v
e
r
ters
an
d
eq
u
a
l
2
as
w
el
l
as
t
h
e
c
o
s
t
of
th
e
w
i
n
d
tow
e
r es
tima
t
e
d
as
2
0
% of t
h
e
sys
tem
c
a
p
i
ta
l c
o
st.
T
h
e
C
O&
M
ta
ke
n a
s
1
% of t
he
t
o
t
a
l
c
os
t as ear
l
i
e
r
s
up
por
te
d
[2
8].
F
o
r the
Replace
me
nt
C
o
s
t,
e
xpre
ssed i
n
(
16) a
nd (17).
C
R
=
C
RE
P
*
SFF (i, P
R
_
LF
)
(
12)
C
REP
=
i
r
*
(
(
N
B
*
C
RB
)
+
(
N
IN
V
* C
R.
I
N
V
)
+
(
N
RE
G
_
P
V
*
C
R
_
RE
G
_
PV
)
+ (
N
REG_
W
T
*
C
R
_
REG
_
W
T
)
)
(
1
3
)
Where
S
F
F
e
xpressed
a
s
t
he
s
ink
i
ng
fu
n
d
f
actor,
P
R_
L
F
i
s
l
ifes
pa
n
o
f
c
om
po
ne
nt
s
(bat
ter
y
,
in
verter
,
a
n
d
re
gu
l
a
to
rs).
C
RB
,
C
RIN
V
,
C
R_REG
_
PV
,
a
nd
C
R
_
REG-W
T
a
r
e
t
h
e
r
e
p
l
a
c
e
m
e
n
t
c
o
s
t
o
f
t
h
e
b
a
t
t
e
r
y
,
i
n
v
e
r
t
e
r
,
P
V
regu
la
tor
and
WT r
egu
l
a
t
or r
espec
t
i
v
e
l
y.
Wh
ile
N
REG_P
V
is
t
he
num
be
r
of
v
o
lta
ge
r
egu
l
a
t
or
a
n
d
e
qua
l
1,
N
REG
_
W
T
is
t
he
n
um
ber
of
W
T
regu
la
tor
and
e
qua
l 1.
Where
a
s
the
s
ink
i
ng f
und
fac
t
o
r
is c
l
a
r
i
fie
d
b
y
the f
o
l
l
ow
i
n
g e
qua
t
i
on
:
The
refore,
the
sin
k
i
n
g fu
nd
fac
t
or
is c
l
a
r
i
f
ie
d by
t
h
e f
o
ll
ow
i
ng
eq
u
a
t
i
on
:
SF
F
(
i
,
P
R_LF
) =
_
(
14)
3.3.
S
olu
t
ion
me
th
od
olo
gy
MOP
S
O
and
N
S
G
A
-II
algor
ithms
wor
k
b
y
creat
i
ng
n
ew
r
eli
a
bl
e
s
o
lu
t
i
o
ns.
H
o
w
e
ve
r,
t
heir
w
or
k
i
n
g
me
cha
n
ism
diffe
r
s.
T
able1
sh
o
w
s
the d
i
ffere
nces
b
e
t
wee
n
t
he
t
wo
a
l
gor
it
hm
s.
Tab
l
e
1.
D
i
f
fer
e
nce
be
t
w
e
e
n
N
S
G
A
-II
a
nd MO
P
S
O
algori
t
hms
NS
GA
_II
M
O
P
S
O
S
e
l
e
ct
i
o
n
,
c
r
o
s
s
o
v
e
r
,
a
n
d
m
u
t
a
ti
o
n
a
r
e
u
s
e
d
d
u
r
in
g
each
ge
n
e
r
a
tion.
T
hos
e
individ
u
a
l
s
or
c
hrom
oso
m
e
s
a
r
e
c
om
bi
n
e
d
t
o
c
r
ea
t
e
c
hildr
e
n
m
o
de
l.
Pa
r
tic
l
e
posit
ions
a
r
e
a
f
f
ec
t
e
d
b
y
t
h
e
ir
s
e
l
f-
da
t
a
a
nd
info
rm
a
t
ion
sha
r
ing
am
ong
sw
a
r
m
m
e
m
b
e
r
.
It
u
se
s
tw
o
e
quations:
v
e
lo
c
ity
a
nd
positi
on.
G
r
e
a
t
a
t
finding
th
e
globa
l
opti
m
u
m
solution.
C
a
p
a
b
l
e
of
findin
g
the
l
o
ca
l
opti
m
u
m
M
o
r
e
c
o
m
ple
x
due
t
o
m
u
ta
t
i
on
a
nd
c
r
ossove
r;
t
a
k
e
s
e
xtra
t
i
m
e
co
m
p
a
r
ed
t
o
MOP
S
O
.
F
a
s
t
a
n
d
e
a
s
y
t
o
i
m
p
l
emen
t
as
o
n
l
y
a
f
e
w
p
a
r
a
met
e
r
s
n
eed
ad
j
u
s
t
men
t
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
46
3 –
47
8
46
8
In
r
ecent
year
s,
t
he
i
mplem
e
ntat
ion
of
N
S
G
A
-II
and
MOP
S
O
al
gor
i
t
h
m
s
b
a
s
e
d
o
n
G
A
a
n
d
P
S
O
a
r
e
t
h
e
fore
mos
t
t
ec
h
n
iq
ues
a
d
op
t
e
d
for
gl
oba
l
o
p
t
i
miza
t
i
o
n
i
n
va
rious
fie
l
ds,
s
u
ch
a
s
busi
n
ess,
e
ng
i
n
eer
ing,
a
nd
i
n
stoc
has
t
i
c
n
a
t
u
r
e
of
r
enew
a
b
l
e
e
ne
rgy
ap
p
lic
ati
ons
[
1]
.
In
t
h
i
s
s
tu
d
y
,
the
N
S
GA
-I
I
and
MO
P
S
O
me
thod
s
a
r
e
ut
iliz
e
d
t
o
s
i
z
e
the
s
t
a
nd-al
o
n
e
h
ybr
id
P
V
/
w
i
n
d
s
ys
tem
s
a
n
d
t
he
se
m
et
ho
ds
a
re
s
how
n
t
o
b
e
bet
t
er
t
ha
n
the
sing
le-
o
bj
e
c
t
i
v
e
m
e
thod
s suc
h
a
s hy
brid
g
e
n
etic
a
lg
ori
t
hm
(
G
A
)
o
r
p
art
i
cl
e swar
m o
p
t
i
m
i
z
at
io
n
(PSO
)
.
3.3.
1. Op
t
imal
con
figu
r
a
tion
b
ased
on
NS
GA
_II a
l
g
o
r
i
th
m
I
n
20
0
2
, D
eb p
rop
o
sed t
h
e us
e of N
S
G
A
-
I
I
a
lgor
it
hm
[2
9],
in w
hic
h
t
h
e
po
pul
a
t
io
n
i
s
d
ist
r
i
b
ut
ed
i
nt
o
sever
a
l
n
on-d
o
m
ina
t
i
o
n
le
ve
l
s
a
nd
eac
h
s
o
lut
i
o
n
i
s
a
ssig
n
e
d
a
fi
t
n
ess
e
q
ual
to
its
n
o
n
-dom
ina
t
i
o
n
lev
e
l
.
T
h
e
alg
o
ri
t
h
m
can be
sum
m
a
rize
d
a
s
f
oll
o
ws [1
9
]
:
a.
In the
first
ste
p
,
the req
ue
s
t
e
d
in
p
u
t da
ta are
pro
vi
de
d.
Th
i
s
data
i
n
v
o
l
ve
s t
h
e spe
c
i
f
ica
t
i
o
ns of
t
h
e hy
bri
d
syste
m
(
hourl
y
r
a
d
i
a
tio
n,
t
e
m
pera
t
u
r
e
a
n
d
w
ind
s
p
ee
d)
a
nd
a
l
so
loa
d
d
em
and,
d
a
t
a
to
c
om
pu
t
e
t
h
e
tech
n
i
ca
l
and
e
c
on
om
ic f
unc
t
i
ons
a
nd da
t
a
t
o
a
s
se
ss
t
he c
on
stra
in
t
si
tu
at
ion
s
.
b.
U
pper
and
low
e
r bou
nd
o
f
the
numbe
r of P
V-WT
a
nd
BES
c.
The
e
n
e
r
g
y
out
pu
t
of
P
V
and
w
i
n
d
t
ur
bi
ne
a
r
e
c
a
l
c
u
l
a
ted
thr
oug
h
t
he
P
V
a
n
d
w
i
n
d
m
o
d
e
l
s
b
y
usi
n
g
(1
-
3).
The
m
ode
l
of
e
nerg
y
st
ora
g
e
ba
t
t
ery
by
us
in
g
(4)
w
ith
t
he
t
ota
l
c
a
p
ac
i
t
y
(C
B)
i
s
a
l
l
o
w
e
d
to
c
ha
rg
e
a
n
d
di
sc
h
a
rg
e
u
p
t
o
a
l
i
m
it
d
ef
in
ed
b
y
th
e
max
i
mu
m
d
e
p
t
h
of
d
i
s
c
h
arg
e
(
D
O
D),
by
u
s
i
n
g
(5,
6)
a
nd
F
i
gure
1.
d.
A
random
p
are
n
t
po
p
u
la
t
i
on
(P
i)
i
s
cr
eate
d
w
it
h
si
z
e
N
.
The
n
,
t
he
p
o
p
u
l
ati
o
n
of
c
h
ildr
e
n
(Q
i)
i
ncl
u
d
i
ng
N
solu
t
i
o
n
s
i
s
p
r
o
duce
d
t
hr
ou
g
h
g
e
n
e
tic
m
a
n
ipu
l
a
t
i
on (
c
ros
s
ov
er
a
nd m
u
ta
ti
o
n
).
e.
Ca
lcu
l
a
t
e
t
h
e
o
b
jec
t
iv
e
fu
nct
i
ons
f
or
e
a
c
h in
di
v
i
d
u
a
l
o
f
P
i
po
p
u
l
atio
n
(LPSP an
d
C
O
E) u
sin
g
(7
-
1
4
)
.
f.
The
t
w
o p
o
p
u
l
a
tio
ns a
re
c
ombi
ne
d to f
orm
the
(Ri)
p
o
p
u
l
a
tio
n
w
i
th
s
i
z
e
2N.
g.
Cla
s
s
i
fica
ti
o
n
o
f
the
Ri
p
opu
la
ti
on
i
s
m
a
de
i
n
a
c
c
o
rda
n
ce
w
it
h
th
e
P
a
re
to
f
ro
n
t
on
t
h
e
ba
se
s
of
f
i
t
nes
s
(non-dom
in
a
t
ed sorting is per
f
or
m
e
d
t
o
d
eter
mine
t
h
e
r
ank
(fron
t) of ea
ch
p
o
pula
t
io
n m
e
m
b
er
).
h.
The
nex
t
p
o
p
u
l
a
tio
n
of
o
ne
o
f
the
fr
ont
s
is
b
u
i
l
t
a
cc
o
r
d
i
ng
t
o
pr
ior
itie
s
b
y
p
erfor
m
ing
a
ge
ne
ral
com
p
aris
on o
f
the
m
e
m
be
rs of
the R
i
p
op
u
l
a
tio
n.
i.
S
i
nce
t
h
e
size
o
f
R
i
i
s
e
q
u
a
l
t
o
2
N
,
t
he
r
em
aini
ng
s
o
l
u
t
i
on
s
ca
n
si
mply
b
e
ign
o
r
e
d
be
ca
u
s
e
i
t
i
s
impo
ssi
ble
to p
lac
e
a
l
l
m
e
m
bers in
t
h
e
new
p
o
p
u
l
a
t
ion
(P
i+
1).
j.
The
e
nd,
i
f
th
e
ma
ximum
nu
mbe
r
o
f
i
t
e
r
at
i
ons
r
eferre
d
i
n
S
tep
3
i
s
re
ac
hed,
t
he
n
o
n
-d
om
ina
t
e
d
s
ort
i
n
g
resol
u
t
i
on
a
t
t
h
e
l
ast
iter
a
t
i
on
w
a
s
a
c
hie
v
e
d
a
s
the
o
p
t
i
ma
l
s
i
z
i
ng
a
nd
d
esign
fo
r
t
h
e
SHRES.
O
t
h
erwi
se
go ba
c
k
t
o nu
m
b
er
2
.
F
i
gure
2
d
i
s
p
l
a
ys
t
he
N
S
G
A
-
I
I
a
lgorithm
a
p
p
lica
t
io
n
ap
proa
ch.
T
h
i
s
a
ppro
a
c
h
i
s
ai
med
a
t
f
in
di
n
g
t
h
e
nu
mb
ers
of
v
ari
a
bl
e
s
t
h
a
t
d
e
p
e
n
d
on
t
he
s
t
a
t
e
o
f
N
PV
,
N
WT
,
and
N
BES
i
n
orde
r
to
m
ee
t
the
m
i
n
i
m
i
z
e
d
system
L
PS
P
and
COE.
T
his
is
e
xec
u
t
e
d
by
u
s
i
ng
N
S
G
A-II
optimiza
ti
on
t
o
o
lb
ox
i
n
M
A
TLA
B,
w
i
t
h
a
selec
t
e
d
pop
ul
ati
o
n
si
z
e
o
f
2
00,
t
he
c
rosso
v
e
r
va
l
u
e
of
0
.8
,
an
d
t
h
e
ma
xim
u
m
s
i
mu
l
a
tion
g
en
e
r
at
i
o
n
nu
mb
e
r
set a
t
5
00.
(a)
(b)
F
i
gure
2.
(
a) an
d
(
b)
N
SG
A
-
II
p
roce
dure
[2
7]
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
NSG
A
-II
and
MO
PSO
base
d
opt
im
i
z
a
t
io
n f
o
r siz
i
n
g
o
f
hy
br
id PV
/wi
n
d
/
b
a
ttery
...
(
M
oha
m
ad Izdi
n H
l
a
l
)
46
9
3.3.
2. Op
t
imal
con
figu
r
a
tion
b
ased
on
MOPS
O alg
o
r
i
th
m
I
n
1
9
9
5
,
K
e
n
n
e
y
a
n
d
E
b
e
r
h
a
r
t
s
h
o
w
e
d
t
h
a
t
t
h
e
P
a
r
t
i
c
l
e
S
w
a
r
m
O
p
t
i
m
i
z
a
t
i
on
(P
S
O
)
has
t
w
o
separ
a
te
conc
e
p
ts:
a)
s
ocia
l
in
te
rac
tio
n
w
h
i
c
h
is
e
x
h
ib
ite
d
b
y
s
w
a
rm
i
ng
a
n
d
b)
f
iel
d
o
f
e
vol
u
t
i
onar
y
c
a
l
c
u
la
ti
on.
I
n
P
S
O
,
the tw
o
b
e
st va
l
ues
w
i
l
l
de
t
erm
i
n
e
the
pos
it
io
n of
e
a
c
h
pa
rti
c
le.
P
S
O
is
a
n
A
I
a
ppr
oac
h
w
hic
h
i
s
base
d
on
th
e
sw
a
r
m
socia
l
i
nt
e
r
a
c
tio
n
w
i
thi
n
a
f
iel
d
o
f
e
v
o
l
ut
i
onar
y
cal
c
ul
a
tion,
a
s
proposed
[
30]-[3
2
]
.
The
algor
ithm
d
e
t
erm
i
nes
th
e
tw
o
be
st
p
o
s
i
tio
ns
f
or
e
a
c
h
p
a
r
t
ic
l
e
.
In
it
i
a
l
l
y
,
th
e
bes
t
v
a
l
u
e
i
s
ob
tai
n
e
d
,
c
a
l
l
e
d
th
e
ind
i
vi
du
al
b
e
st
(
p
be
s
t
),
a
nd
is
r
eta
i
ne
d
b
y
t
he
p
a
r
tic
le,
w
h
ile
t
he
nex
t
v
a
l
ue
i
s
deter
m
i
n
ed
b
y
the
P
S
O
opt
i
m
iz
at
i
o
n
a
l
gor
i
t
hm
w
ith
i
n
a
g
l
o
bal
be
st
p
o
p
u
l
a
tio
ns.
Indi
vid
u
a
l
part
icle
pos
i
tio
n
defi
nes
t
h
e
p
a
rtic
les
varia
b
l
e
t
ar
get
v
a
l
u
es
a
nd
ve
l
o
c
ity
t
h
a
t
is
a
p
p
l
ied
t
o
m
onitor
t
h
e
o
v
era
l
l
gl
oba
l
be
st
v
a
l
ue
(
g
be
s
t
).
T
h
e
f
i
t
n
e
ss
e
qu
at
ion
of
t
hi
s
a
l
g
o
r
i
t
h
m
i
s
to
s
ea
rch
th
e
b
e
st
s
o
l
uti
o
n
f
r
o
m
a
mong
st
a
l
l
pos
si
b
l
e
a
v
a
i
l
a
bl
e
o
p
ti
on
s,
w
ith
a
d
d
i
t
i
o
n
a
l
c
onstra
i
nts
a
dde
d.
The
a
l
g
o
r
i
t
h
m
is
b
a
s
e
d
o
n
eac
h
par
tic
le
f
itne
s
s
appra
i
sa
l, in
d
i
vid
u
a
l
a
n
d
g
l
o
bal
be
st fit
ness
upda
te
, a
lon
g
s
i
de
w
ith
p
ar
tic
l
e
posit
i
on an
d
vel
o
c
ity.
D
u
r
i
n
g
t
he
o
pe
ra
t
i
on
o
f
t
he
a
lgor
i
t
hm
,
ea
ch
p
art
i
cle
kee
p
s
t
h
e
be
st
f
itne
s
s
val
u
e
t
h
at
it
ha
s
ach
ieve
d.
The
par
t
ic
l
e
w
i
t
h
the
bes
t
f
it
n
e
ss
va
lue
i
s
c
a
l
cu
la
te
d
an
d
up
da
ted
d
u
ri
ng
i
t
e
ra
ti
o
n
s.
I
n
t
h
is
c
ase,
eac
h
part
icle
repr
esents
a
p
ote
n
ti
a
l
c
o
n
fi
g
u
ra
tio
n
of
t
he
P
V-w
i
nd
t
u
rb
i
n
e
a
n
d
b
a
t
t
e
r
y
h
y
b
r
i
d
s
y
s
t
e
m
:
N
PV
,
N
WT
and
N
BES
,
a
n
d
t
h
e
s
e
a
r
c
h
s
p
a
c
e
d
i
m
e
n
s
i
o
n
a
r
e
t
h
r
e
e
.
T
h
e
n
,
t
h
e
o
b
j
e
c
t
i
v
e
f
u
n
c
t
i
o
n
of
e
ac
h
pa
rt
i
c
le
i
s
co
m
put
e
d
,
co
rr
esp
ond
ing
to
e
ach
scen
ar
io
c
o
n
figu
ration
(
LPSP
an
d
C
O
E).
Th
e
fo
l
l
ow
i
n
g
s
t
eps
il
lu
stra
te
u
ti
l
i
ze
d
t
h
i
s
m
e
t
h
od
f
o
r
H
R
E
S
as
t
h
e
f
oll
o
w
i
ng
:
-
a.
I
n
t
h
e
f
i
r
s
t
s
t
e
p
:
I
n
i
t
i
a
l
i
z
a
t
i
o
n
,
t
h
e
r
e
q
u
e
s
t
e
d
i
n
p
u
t
d
a
t
a
a
r
e
p
ro
vi
ded.
T
h
i
s
da
t
a
i
nv
o
l
ve
s
th
e
spec
ific
a
tio
ns
o
f
t
h
e
h
y
b
rid
syste
m
(
hour
l
y
r
a
d
iat
i
on,
t
e
m
pe
rat
ure
a
n
d
w
i
n
d
s
pe
e
d
)
a
nd
a
l
so
l
oa
d
dem
a
nd,
d
ata
to
c
om
p
u
t
e
th
e
tec
h
n
i
ca
l
a
n
d
ec
on
om
ic
f
unc
tio
ns
a
n
d
data
t
o
a
s
se
ss
t
he
c
o
n
stra
in
t
situa
t
io
ns.
b.
U
pper
and
low
e
r bou
nd
o
f
the
numbe
r of P
V-WT
a
nd
BES
c.
Th
e
en
ergy
out
put
o
f
PV
a
nd
w
in
d
tu
rbin
e
a
r
e
ca
l
c
ul
at
e
d
t
h
r
o
ugh
t
he
P
V
and
w
i
n
d
m
ode
ls
b
y
usin
g
(1-3)
.
T
h
e
m
odel of e
nerg
y st
ora
g
e ba
t
t
e
r
y by us
ing (4) w
ith th
e
t
o
ta
l c
a
p
a
c
ity (C
B
) is al
l
o
w
e
d to cha
r
g
e
and
disc
harge
up
t
o
a
l
imit
d
efine
d
b
y
t
h
e
m
a
xim
u
m
dep
t
h
of
d
is
ch
a
r
ge
(
D
O
D
)
,
by
usi
n
g
(5,
6)
and
F
i
g
u
re
1
.
d.
Cons
ta
nt
s :
-
P
e
rsonal a
n
d g
l
o
b
a
l
coe
fficie
n
ts,
C1 = C
2 =
2
.
-
In
er
t
i
a wei
ght
,
w =
0
.
9
.
e.
The
pos
it
ion
a
nd
vel
o
c
i
t
y
o
f
part
icles
a
r
e
rand
om
ly
s
elec
te
d
in
o
rd
e
r
t
o
ge
n
e
ra
t
e
t
h
e
i
ni
ti
a
l
p
opu
l
a
ti
on
and
the
n
app
l
i
e
d
to
the
o
b
j
e
c
t
i
v
e
fu
nc
ti
o
n
s
to
fin
d t
h
e
op
t
i
m
u
m
fit
ness va
lu
e
, by
using
(7-14).
f.
Eva
l
ua
te
t
he fit
ness
va
lue,
w
ith
m
inim
um
L
PS
P a
nd CO
E
g.
Ca
lcu
l
a
t
e
and
up
da
te
(P
b
est a
n
d
g
b
es
t)
h.
Ca
lcu
l
a
t
e
and
up
da
te
vel
oc
i
t
y
a
nd po
sit
i
o
n
o
f
ea
ch pa
r
t
i
c
l
e
i.
A
p
p
l
y t
h
e u
p
d
a
t
e
d va
lue
t
o
fi
nd
op
tim
um
v
a
l
ue o
f LP
SP
a
n
d
C
O
E
j.
The
e
nd,
i
f
t
h
e
ma
xi
m
u
m
nu
mber
o
f
i
t
e
r
at
i
ons
r
efe
rre
d
i
n
S
t
e
p
3
i
s
re
ac
h
e
d,
t
he
n
o
n
-d
om
ina
t
e
d
s
ort
i
n
g
resol
u
t
i
on
a
t
t
h
e
l
ast
iter
a
t
i
on
w
a
s
a
c
hie
v
e
d
a
s
the
o
p
t
i
ma
l
s
i
z
i
ng
a
nd
d
esign
fo
r
t
h
e
SHRES.
O
t
h
erwi
se
go ba
c
k
t
o nu
m
b
er
2.
Th
e
si
zi
ng
m
od
el
o
f
th
e
h
ybri
d
P
V/
wind
e
n
e
rgy
sy
st
ems
i
s
m
o
r
e
c
omp
l
ex
t
ha
n
the
si
ng
le-so
u
rc
e
gene
ra
ti
o
n
s
ys
tem
s
.
Thi
s
i
s
bec
a
u
s
e
the
va
riab
les
mus
t
b
e
c
ons
i
der
e
d
for
sys
t
em
opt
i
m
i
z
at
ion.
M
oreove
r,
syste
m
p
erfor
m
a
n
ce
o
v
er
a
l
on
g-term
,
ec
onom
ic
p
a
r
a
m
ete
r
s,
a
nd
r
e
l
ia
bi
li
ty
o
b
j
e
c
tive
s
s
h
o
u
ld
b
e
w
e
ll
th
oug
h
t
-o
ut
i
n
orde
r
t
o
a
c
h
i
e
ve
a
s
u
ita
ble
c
o
m
p
rom
i
se
b
e
t
w
e
e
n
C
O
E
a
n
d
L
P
S
P
.
N
S
G
A
-
I
I
a
n
d
M
O
P
S
O
a
r
e
the
a
ppr
opr
iat
e
m
etho
ds
w
i
t
h
r
ega
r
ds
t
o
g
l
o
b
a
l
o
p
t
i
m
iza
t
i
o
n
a
n
d
t
he
r
a
ndom
n
a
t
ur
e
of
r
ene
w
a
b
le
p
ow
er
source
s.
The
se m
etho
ds ha
v
e be
en u
se
d i
n
m
any h
y
b
r
i
d
app
l
i
c
a
t
i
o
ns
i
n re
cent
y
e
ars
[4],
[
33]
,
[34]
I
n
t
his
paper,
t
he
N
SGA-II
and
MO
PS
O
algorithm
s
w
ere
applied
to
s
i
z
e
t
h
e
st
a
n
d
-
a
l
on
e
hyb
rid
P
V
/WT
a
nd
b
a
tter
y
s
ys
t
e
m.
I
n
pr
inc
i
p
l
e
,
t
he
se
a
l
gori
t
hm
s
a
i
m
to
f
i
n
d
t
h
e
o
p
ti
mu
m
nu
mb
er
o
f
PV
p
an
el
s,
w
i
n
d
g
e
n
era
tio
n,
a
nd
ba
t
t
erie
s
to
m
in
imize
the
LP
SP
a
nd
C
O
E
.
Th
e
op
t
i
m
i
za
tio
n
me
t
h
od
w
a
s
d
one
u
sing
MA
TLA
B
so
ft
w
a
re
by
exe
c
u
ti
ng
the
sam
e
num
ber
of
i
tera
tio
ns
a
nd
po
pu
l
a
t
i
o
n
in
bot
h
a
l
go
rith
ms.
Fi
n
a
lly
,
the r
e
su
lts
w
ere
c
o
mpar
ed
a
nd the
al
g
o
ri
thm
wi
th
t
he
bes
t resu
l
t
w
a
s
i
d
e
n
tifie
d.
Th
e
op
ti
mi
z
a
t
i
o
n
m
et
hod
s
t
a
rt
s
wit
h
t
h
e
f
ol
lo
wing
i
npu
t
v
a
lu
e
s
:
h
ourl
y
out
p
u
t
p
o
w
e
r
for
P
V
,
WT
a
n
d
l
o
ad
p
ro
file
.
Th
en
,
t
h
e
l
o
o
p
i
t
erat
es
t
h
e
N
SG
A-
II
o
r
M
OPSO
a
l
gor
ith
m
s
.
The
optim
um
r
esults
g
e
n
er
ated
by
t
he
t
w
o
a
lg
ori
t
hms
ar
e
c
o
nve
r
t
e
d
t
o
t
h
e
nea
r
est
i
n
te
ge
r
val
u
es.
T
h
i
s
g
ive
s
t
he
s
iz
i
n
g
of
t
he
g
e
n
era
tin
g
s
o
u
r
c
e
.
T
h
e
o
p
t
i
m
u
m
s
i
z
i
n
g
a
l
o
n
g
w
i
t
h
C
O
E
a
n
d
L
P
S
P
v
a
l
u
e
s
a
r
e
r
ecor
d
e
d
f
or
eac
h
r
u
n
a
nd
s
t
or
ed
i
n
a
n
arr
a
y
for
e
a
c
h
iter
a
tio
n.
F
ina
l
ly,
a
l
l
a
ppr
opri
a
te
s
o
l
u
t
i
o
ns
a
r
e
o
b
t
a
i
ne
d
t
h
r
oug
h
n
on-
dom
ina
t
e
d
o
p
t
im
al
call
e
d
P
a
reto
f
ro
nt
[
35].
The
num
be
r
of
e
ac
h
g
e
ner
a
t
i
n
g
u
n
its
a
re
p
rov
ided.
F
l
o
w
c
h
a
rts
of
t
he
N
SGA
-II
and
MO
P
S
O a
l
gor
ithm
s
ar
e
show
n
i
n
F
ig
ure
. 3.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
46
3 –
47
8
47
0
St
a
r
t
M
e
t
h
od
ol
ogi
c
a
l
d
a
t
a
-
one
ye
ar
(
h
our
l
y
r
a
d
i
at
i
o
n
,
t
e
m
p
e
r
a
t
u
r
e
a
nd
w
i
n
d
s
p
e
e
d
)
and
al
s
o
l
oad de
m
an
d
R
a
n
d
om
i
n
i
t
i
al
p
op
ul
a
t
i
o
ns
(
nu
m
b
e
r
of
P
V
-
WT-B
E
S
)
P
V
-W
T
_B
E
S
m
o
d
e
l
s
,
u
s
i
n
g
E
q
.
(1
-
6
)
a
n
d
(
F
i
g
.
1
)
R
u
n
N
S
G
A
_I
I
O
R
M
O
P
S
O
t
o
ac
h
i
e
v
e
op
t
i
m
a
l
val
u
e
o
f
e
a
c
h
c
om
p
o
ne
n
t
I
f
cri
t
eri
a
f
u
l
f
i
l
l
C
rea
t
e
n
ew
g
en
era
t
i
on
S
e
l
e
ct
i
o
n
o
p
erat
i
o
n
t
o
m
i
n
i
m
i
ze
o
b
j
ect
i
v
e f
u
n
c
t
i
o
n
N
e
w
g
e
n
era
t
i
on
o
f
c
o
n
f
i
g
u
r
at
i
o
n
S
t
o
re
r
es
u
l
t
i
n
a
n
arra
y
S
e
l
e
c
t
t
h
e
o
p
t
i
m
al
c
onf
i
gu
r
a
t
i
on
o
f
t
h
e
hyb
r
i
d
P
V
,
W
T
a
nd
B
E
S
s
y
s
t
e
m
b
a
s
e
d
o
n
t
he
g
l
o
b
a
l
be
s
t
m
i
ni
m
u
m
L
P
S
P
a
nd
C
O
E
Se
t
of
p
os
s
i
b
l
e
s
o
l
u
t
i
on
s
u
s
i
n
g par
e
t
o F
r
on
t
EN
D
YE
S
NO
M
OP
S
O
I
n
i
t
i
a
l
i
z
e
p
a
rt
i
c
l
e
s
w
i
t
h
r
a
n
d
o
m
p
o
s
i
t
i
o
n
a
n
d
v
e
l
o
c
i
ty
v
e
c
to
r
E
v
a
l
u
a
t
e
t
he
f
i
t
ne
s
s
v
a
l
ue
,
u
si
g
E
q
.
(
7
-
1
4
)
Ca
lc
u
l
at
e
an
d
u
p
d
a
t
e
(
P
b
e
s
t
an
d
gb
e
s
t)
Ca
lc
u
l
at
e
an
d
u
p
d
a
t
e
v
e
l
oc
i
t
y
an
d
p
o
s
i
t
i
o
n
of
e
ac
h
p
a
r
t
ic
l
e
N
SGA
_
I
I
G
e
n
e
r
a
te
i
n
i
t
i
al
p
op
u
l
at
i
o
n
Fi
t
n
e
ss
e
v
a
l
u
a
t
i
o
n
,
u
s
i
n
g
E
q
.
(
7
-
1
4
)
Se
l
e
c
t
i
o
n
Cr
o
s
s
o
ve
r
Mu
t
a
t
i
o
n
F
i
gur
e 3.
N
S
G
A
II or
M
O
P
S
O
opt
imiza
t
i
o
n
fl
ow
cha
rt for
sta
nd-a
l
o
ne
hy
b
ri
d e
n
ergy
sys
t
em
4.
RE
S
U
L
T
AND DI
S
C
US
S
I
ON
The
m
e
th
o
d
o
l
ogy
w
a
s
a
p
p
l
i
e
d
t
o
f
in
d
t
h
e
op
t
i
ma
l
size
f
or
S
H
R
ES.
T
he
l
oa
d
pro
f
ile
f
or
a
t
y
p
i
cal
ru
ral
vi
ll
ag
e
i
n
M
ala
y
si
a
co
nsi
s
t
i
ng
o
f
20
h
ou
se
h
o
ld
s
i
s
s
ho
wn
in
F
i
g
ure
4
an
d
the
t
o
ta
l
energ
y
c
o
n
sum
p
t
i
o
n
per
d
a
y
is
1
38.
4
kWh
[36].
T
h
e
m
a
x
i
m
u
m
s
o
lar
rad
i
a
t
io
n
w
a
s
ap
pr
o
x
i
m
a
t
e
l
y
10
50
W/m
2
a
n
d
th
e
ma
x
i
mu
m
win
d
s
peed
w
a
s
r
ecor
d
ed
a
t
5
m
/sec
.
The
a
n
nua
l
m
e
te
oro
l
o
g
ica
l
c
o
nd
iti
o
n
s
i
n
Ma
la
ys
ia,
and
so
lar
ra
di
a
tio
n
and
w
i
n
d
s
pee
d
a
re
i
l
l
us
t
r
ate
d
i
n
F
i
g
u
re
5
a
a
nd
5b,
r
e
s
p
e
c
t
i
v
e
l
y.
T
he
d
at
a
w
a
s
obt
a
i
ne
d
from
t
he
M
a
l
ays
i
a
n
Me
teoro
l
og
ica
l
D
epar
t
m
ent.
Table
2 i
n
d
i
cate
s
t
he pa
r
am
eter
s of
P
V
p
anel
s,
w
ind,
b
at
t
e
ry a
nd i
n
verte
r
s.
F
i
gure
4.
H
ourly l
oa
d
p
r
ofi
l
e
for r
u
ral area i
n m
a
laysia
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
El
e
c
&
D
ri S
yst
I
S
S
N
:
2088-
86
94
NSG
A
-II
and
MO
PSO
base
d
opt
im
i
z
a
t
io
n f
o
r siz
i
n
g
o
f
hy
br
id PV
/wi
n
d
/
b
a
ttery
...
(
M
oha
m
ad Izdi
n H
l
a
l
)
47
1
(a)
(
b
)
F
i
gure
5. Annu
a
l
m
eteoro
lo
g
i
cal c
on
d
iti
o
n
s for
rura
l ar
ea
i
n
m
a
lays
ia,
(a) S
o
l
a
r
r
a
diation
and
(b)
wi
nd
speed
Tabl
e2.
Par
a
meter
s
o
f
PV, wi
nd
tu
rbin
e
,
b
at
te
ry
a
n
d
in
v
e
rter
PV
m
odul
e
s
S
p
eci
f
i
cat
i
o
n
s
Wind turbine
spec
ifi
c
a
tions
Ba
tte
ry
spec
i
f
ica
tions
Inve
rte
r
s
pe
c
i
fi
c
a
tions
Pow
e
r m
a
x =
320W
Ra
te
d
volt
a
g
e
=
V
m
p
p
54.
7v
Ra
te
d c
u
r
r
e
n
t =
I
m
pp
5.49A
I
n
i
tia
l
c
o
st =
$
2
90
[37,
38]
PV
r
e
gul
a
t
or
c
ost
=
$
750
[39]
L
i
f
e
t
i
m
e
=
20 ye
ars
Ra
te
d
output
pow
er
=
3
k
W
Ge
n
e
r
a
t
o
r
volt
a
g
e
=
230
V-a
c
C
u
t-in
w
ind
spee
d =
2 m
/
s
Ra
te
d w
i
nd
spee
d =
12 m
/
s
I
n
i
tia
l
c
o
st =
$
2
800
[40]
Wind
re
gul
a
t
or
c
os
t =
$
750
[
41]
L
i
f
e
t
i
m
e
=
20 ye
ars
Ra
t
e
d ca
p
a
c
ity =
1
000A
h
Ra
t
e
d
volt
a
g
e
=
2V
E
f
f
i
ci
en
cy
=
8
5
%
D
OD
=
7
0
%
I
n
it
ia
l c
o
st =
$
230
[42]
Li
f
e
Tim
e =
10
Y
e
a
r
s
R
a
t
e
d
output
powe
r
650
0
W
Inp
u
t Volta
g
e
12V
DC
/
24V
DC
F
r
e
que
n
c
y
50
H
Z
Ef
f
i
ci
e
n
cy
=
9
0
%
Initia
l c
o
st =
$ 2528 [4
3]
L
i
fe
tim
e =
12 yea
r
s
N
o
t
e
:
$1.
0 = R
M
3
.
90
I
n
t
h
i
s
pape
r,
t
w
o
s
c
e
nar
i
os
a
r
e
i
nves
t
iga
t
e
d
t
o
determ
i
n
e
t
h
e
o
p
t
i
ma
l
siz
i
ng
of
S
H
R
ES
.
In
t
he
f
i
r
st
ca
se
,
the
hy
br
i
d
s
yst
e
m
c
o
n
s
ists
o
f
P
V
p
anel
s,
w
in
d
t
u
rbi
n
es,
a
n
d
b
a
t
t
e
r
y
b
a
n
k
,
w
h
i
l
e
t
h
e
s
e
c
o
n
d
c
a
s
e
com
p
rise
d
of
P
V
pa
nels
a
n
d
b
at
tery
b
a
n
k
on
ly.
The
o
u
t
p
ut
pow
er
c
u
r
v
e
gener
a
te
d
by
t
he
f
irst
a
n
d
s
eco
n
d
ca
se
o
f
S
H
RE
S
and
the
s
t
a
t
e
of
c
harge
a
n
d
disc
ha
rge
of
t
he
b
a
t
t
er
y
o
n
a
n
hour
ly
b
as
is
u
nder
the
be
st
con
f
ig
ura
tio
n
a
r
e
re
presente
d
i
n
F
igur
e
6a
a
nd
6b,
r
e
s
pect
i
v
el
y
.
Eve
n
t
ho
ug
h
t
h
e
t
o
t
a
l
e
n
er
gy
ge
nera
te
d
w
a
s
sufficie
n
t
t
o
c
o
v
er
t
he
p
ea
k
l
o
a
d
i
n
t
h
e
eve
n
i
ng,
t
he
s
urp
l
us
p
ow
e
r
w
as
e
m
p
l
o
yed
to
c
harge
t
h
e
ba
t
t
er
y
ba
n
k
.
These
fi
gure
s
c
l
e
arl
y
i
ndic
a
t
ed
t
ha
t
the
bat
t
ery,
i
n
t
h
e
peri
od
f
r
o
m
7
P
M
t
o
7
A
M
,
w
a
s
sta
t
e
of
d
i
s
cha
r
ge
a
n
d
it
s
at
isfi
e
d
th
e
l
oa
d dem
a
n
d
,
w
h
i
l
e the
pe
ri
od from
7 A
M t
o
7 P
M
wa
s
st
at
e
of
ch
a
rg
e
. In c
o
nson
an
c
e
w
i
t
h
t
h
e
t
w
o
case
st
udi
es,
t
h
e
c
ont
rib
u
t
i
o
n
s
o
f
o
u
tp
ut
P
V,
w
i
n
d
tu
rbin
e
an
d
b
a
t
t
e
ry
b
ank
th
rou
g
h
a
on
e-y
e
ar
p
eri
o
d
are
show
n
in
F
i
g
ure
7a
a
nd 7
b
.
Base
d
on
F
i
g
u
re
6
,
i
t
c
a
n
b
e
se
e
n
t
h
a
t
the
a
re
a
ha
d
l
o
w
w
i
n
d
spee
d
a
n
d
hi
g
h
so
lar
radia
t
i
o
n.
F
urther,
it
w
a
s
e
v
i
den
t
t
ha
t
the
use
o
f
P
V
pa
ne
ls
h
a
s
a
grea
t
adva
n
t
age
be
ca
u
s
e
the
rene
w
a
bl
e
ene
r
g
y
for
th
i
s
loca
t
i
o
n
w
i
l
l
e
n
a
b
le
t
he
c
om
muni
t
i
e
s
t
o ac
ce
ss
ener
gy for
their
da
ily
liv
i
n
g.
F
i
gure
6
(a). O
u
t
p
u
t o
f
P
V
-
W
T
a
nd c
h
arge
-d
isc
h
arge
c
u
r
ve
o
f t
h
e
b
a
tter
i
es
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
I
nt
J
P
ow
Elec
& Dr
i
S
y
st, Vol. 10,
N
o.
1, Mar
c
h 2
0
1
9
:
46
3 –
47
8
47
2
F
i
gure
6
(b). Out
p
u
t
o
f
P
V
an
d c
h
arge
-d
isc
h
arge
cur
ve of t
h
e
ba
tt
e
r
i
e
s
(a)
(b)
F
i
gure
7.
(
a
)
Con
tri
but
i
on o
f
e
ne
rg
y usi
n
g fi
rst c
a
se
s
t
u
d
y
(
P
V
, w
i
nd
an
d ba
tte
ry d
uri
n
g one
ye
a
r)
and
F
i
gure
. 7(
b
) Con
t
ri
but
i
on o
f
e
ne
rg
y usi
n
g sec
o
n
d
c
a
s
e stu
dy (P
V
a
nd ba
tter
y
d
urin
g o
n
e
yea
r
)
F
o
r
t
h
e
f
i
rst
ca
se
s
t
u
d
y
,
t
h
e
se
t
o
f
s
ol
u
tio
ns
obta
i
ne
d
from
t
h
e
NSGA-I
I
a
n
d
M
OPS
O
a
p
p
r
o
ach
es
f
o
r
one
year
ar
e
sh
o
w
n
i
n F
i
gure
7a a
nd
F
ig
ure
7b, respec
tive
l
y.
Ea
ch
solu
t
ion
rep
r
esen
ts
t
he LPSP
an
d
COE
that
dem
o
n
s
t
r
ate
t
h
e
min
i
mum
v
a
l
u
e
o
f
t
he
m
ult
i
-o
bjec
t
i
ve
o
p
t
imiz
a
t
i
o
n
s
e
t
o
f
s
o
lu
t
i
ons
k
now
n
as
a
P
a
r
eto-
op
tim
al
s
e
t
o
r
P
a
re
to
fro
nt
.
T
h
is
t
ec
h
n
i
q
ue
s
elec
ts
o
ne
o
f
the
di
ffere
n
t
s
o
l
u
ti
on
s
an
d
ma
kes
de
cis
i
on
o
n
L
P
S
P
aga
i
ns
t
CO
E
b
a
se
d
on
t
he
n
um
ber
of
P
V
(N
P
V
),
w
ind
t
u
r
b
ine
(N
WT)
a
nd
b
a
t
ter
y
e
ne
rgy
s
t
orage
(N
BES
)
.
A
ny
of
t
he
s
o
l
ut
ion
s
c
a
n
b
e
c
ons
ide
r
ed
o
pt
im
um
,
w
h
ich
m
eans
t
h
a
t
no
i
mp
ro
v
e
me
nt
c
a
n
b
e
ach
i
e
v
e
d
on
o
n
e
of
t
he
o
bjec
t
i
v
e
func
ti
ons
w
i
t
ho
u
t
a
g
g
ra
vat
i
ng
t
h
e
o
t
h
e
r
obje
c
t
i
v
e
f
u
n
c
t
i
o
n
.
In
o
r
d
er
t
o
m
a
ke
t
he
b
e
s
t
dec
i
si
o
n
,
a
nu
m
b
er
o
f
po
i
n
t
s
on
t
h
e
Par
e
to
fron
t
w
as
s
elec
t
e
d,
an
d
t
h
en
t
he
o
p
tima
l
s
ol
uti
o
ns
w
e
r
e
cho
s
e
n
base
d
on t
h
e
t
r
adeo
ff be
t
w
een
c
ost an
d
relia
b
i
l
ity.
F
i
gure
8c
s
h
o
w
s
t
h
e
c
o
m
p
a
r
iso
n
o
f
spa
c
e
of
o
per
a
t
i
n
g
poi
nts
(n
o
n-dom
i
n
a
t
ed)
;
t
h
e
m
atc
h
in
g
betw
ee
n
tw
o
m
e
tho
d
s
o
f
o
p
t
im
iz
at
i
on
re
sul
t
s
w
a
s
qu
ite
c
lose.
Th
e
N
S
G
A
-II
algori
t
h
m
had
g
o
od
gl
o
b
al
sear
ch a
bi
lit
y b
u
t a
sl
ow
er
c
onver
g
e
n
ce
spee
d
than
MO
P
S
O
.
F
i
gure 9a a
n
d
F
igure 9
b
s
ho
w
the
P
a
re
t
o
o
ptim
um fi
n
a
l
v
alue for
t
h
e
s
e
c
o
n
d
ca
se
s
t
udy.
O
n
c
e
ag
a
i
n,
it
p
rove
d
tha
t
,
com
p
ared
t
o
MOPS
O
,
t
he
N
S
G
A-II
algorithm
had
a
ro
bus
t
se
arc
h
c
a
p
ab
i
l
i
t
y
b
u
t
a
s
low
e
r
con
v
er
ge
nce
spee
d.
Evaluation Warning : The document was created with Spire.PDF for Python.