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.
11, N
o.
1, Mar
ch 20
20,
p
p.
342~
3
4
9
IS
S
N
: 2088-
86
94,
D
O
I
:
10.11
59
1
/ij
ped
s
.
v11
.
i
1.pp
3
42-
34
9
342
Jou
rn
a
l
h
o
me
pa
ge
:
ht
tp:
//i
j
p
eds.i
a
esco
re
.com
Application of artificial neural
network in sizing a stand-alon
e
photovoltaic system: a review
Ahma
d Fa
t
e
h
Mo
h
a
ma
d
No
r,
Suria
n
a
Sa
l
imin,
M
o
h
d
N
o
o
r
A
bdulla
h,
M
u
h
ammad Nafis Ismail
Green
a
nd
S
u
st
a
i
n
a
bl
e En
ergy
(GS
E
n
e
rgy
)
F
ocus
G
ro
up,
F
acu
lty
o
f
El
e
ct
rical
an
d
El
ect
ro
nic
Engin
eerin
g,
Un
iv
e
r
s
i
t
i
T
u
n
H
us
s
e
in
O
n
n
M
a
l
a
ys
ia
,
M
a
l
a
ys
ia
Art
i
cl
e In
fo
ABSTRACT
A
r
tic
le hist
o
r
y
:
R
e
c
e
i
v
e
d
Au
g
1
9
,
2
019
Re
vise
d O
c
t
2
7
,
20
1
9
Ac
ce
p
t
ed
No
v
2
5
,
2
019
Artifi
cia
l
N
eural
Net
w
o
r
k
(AN
N
)
t
echni
qu
es
a
re
b
e
c
o
m
in
g
u
s
ef
u
l
i
n
the
curren
t
e
ra du
e
to
th
e
vas
t
devel
op
men
t
o
f
t
h
e
cu
rrent
c
om
p
u
ter
te
c
h
nol
og
ies
.
AN
N
h
a
s
been
u
s
e
d
i
n
v
ario
u
s
f
iel
d
s
esp
ecia
l
ly
i
n
th
e
f
i
e
l
d
o
f
scien
ce
and
tech
no
log
y
.
One
o
f
t
h
e
a
dv
ant
a
ge
t
hat
m
a
kes
A
NN
so
i
nteres
ti
ng
is
t
he
AN
N’s
ab
ilit
y
to
l
earn
th
e
in
p
u
t
and
ou
tp
ut
r
elati
onsh
i
p
ev
en
t
ho
ugh
t
h
e
relat
i
onship
i
s
n
o
n-l
i
ne
ar.
In
a
ddition
,
A
N
N
i
s
also
u
se
f
u
l
f
o
r
m
odelli
ng,
op
timizati
on,
p
red
i
ct
io
n
,
f
o
r
eca
s
t
i
n
g
,
a
nd
c
on
tro
l
li
ng
s
y
s
t
e
ms.
T
h
e
m
a
i
n
ob
ject
ive
o
f
t
his
p
a
per
i
s
t
o
pres
ent
a
r
e
v
i
ew
o
f
the
A
N
N
t
e
c
h
n
iques
f
o
r
si
zin
g
a
s
t
a
nd
-a
l
o
n
e
p
hot
ov
o
ltaic
(
P
V
)
sy
s
t
e
m
.
Th
e
revi
ew
i
n
th
is
p
ap
er
sh
ow
s
th
e
p
o
t
e
nti
a
l
o
f
A
N
N
a
s
a
desi
gn
t
ool
f
or
a
s
tan
d
-alon
e
P
V.
I
n
addition,
ANN
i
s
very
u
sef
u
l
t
o
i
mprove
t
he
s
i
z
in
g
process
of
t
he
s
ta
n
d
-a
lo
ne
P
V
s
y
s
t
e
m
.
T
h
e
s
i
z
i
n
g
p
r
o
c
e
s
s
i
s
o
f
p
a
r
a
m
o
u
n
t
i
m
p
o
r
t
a
n
c
e
t
o
a
s
t
a
nd-alone
P
V
s
yste
m
in
o
rder
t
o
m
a
k
e
s
u
r
e
the
sy
s
t
em
c
an
g
en
erat
e
am
pl
e
e
l
ectrical
energ
y
t
o
s
u
p
p
l
y
t
he l
oad
dem
a
nd.
K
eyw
ord
s
:
A
r
tificia
l ne
ural
n
etw
o
rk
S
y
stem
sizi
ng
Re
new
a
b
l
e
ene
r
gy
Solar
ele
c
tric
i
t
y
S
t
and-
a
l
o
n
e P
V
system
Th
is
is a
n
o
p
en acces
s a
r
ti
cle u
n
d
e
r t
h
e
CC
B
Y
-S
A
li
cens
e
.
Corres
pon
d
i
n
g
Au
th
or:
A
h
m
a
d F
a
teh
Moha
ma
d N
o
r
,
G
r
e
e
n a
nd S
u
st
aina
ble
Ener
g
y
(
G
S
Energy)
F
o
c
u
s
G
roup,
Fa
cult
y
o
f
E
l
e
c
t
rica
l
and
E
l
e
c
t
r
on
ic E
n
g
i
n
ee
rin
g
,
U
n
i
v
ersi
ti
T
un
H
u
ssei
n
O
nn
Ma
lays
ia,
86
4
00 Pa
rit Ra
ja,
Ba
tu P
ah
at,
Johor,
Malays
ia.
Em
ail:
afa
t
e
h
@u
thm
.
e
du.my
1.
I
N
TR
OD
U
C
TI
O
N
Th
e
amoun
t
o
f
e
l
e
ct
ri
cit
y
g
e
n
e
r
at
ed
fro
m
re
n
e
wabl
e
re
sou
r
c
e
s
su
ch
a
s
so
l
a
r
ener
g
y
h
as
i
ncre
ase
d
ann
u
a
l
l
y
f
or
t
h
e
p
a
s
t
deca
de.
It
h
a
s
b
een
e
x
p
ec
t
e
d
t
h
a
t
t
h
e
g
e
n
e
r
ati
on
of
e
l
e
c
t
ri
c
ity
f
rom
p
hot
ovo
l
t
a
i
c
(
P
V
)
tech
n
o
l
o
gie
s
c
an r
eac
h m
o
re tha
n
90
0 M
W
[1-4].
Th
is si
t
u
a
t
i
o
n
i
s
due
t
o the
e
f
fec
t
o
f g
l
oba
l
w
a
r
m
ing a
nd t
h
e
c
r
i
t
i
c
al
d
epl
e
t
i
o
n
o
f
f
o
ssil
f
u
el
a
nd
n
at
u
r
al
g
as
es
t
h
a
t
h
a
ve
b
e
e
n
t
h
e
m
a
i
n
s
o
u
rce
of
g
e
n
e
r
at
in
g
elec
t
r
ic
ity
a
ll
thi
s
t
ime
[5-7].
T
he
d
e
p
endency
on
f
o
ssi
l
fu
el
a
nd
n
a
tu
ra
l
g
a
ss
e
s
i
n
ge
ne
rat
i
n
g
e
l
e
c
t
r
i
c
ity
r
es
ult
e
d
in
gree
n
h
o
u
se
g
a
s
e
m
i
ssi
ons.
H
e
nce
,
P
V
t
e
ch
n
o
l
o
gy
i
s
a
pop
ula
r
o
p
t
i
o
n
o
f
g
e
n
era
t
i
n
g
elec
tric
ity
b
e
c
a
u
se
o
f
i
t
s
a
d
v
a
nt
ag
es
o
f
cl
e
a
n
lin
e
ss
and
s
u
ffi
ci
e
n
cy
[
8].
Ge
n
e
ra
l
l
y
,
PV
s
y
st
e
m
c
an
b
e
di
vid
e
d
in
t
o
t
wo
t
yp
es
w
hic
h
a
re
gri
d
-con
ne
c
t
e
d
P
V
sy
stem
a
nd
sta
nd-
al
one
P
V
syste
m
.
A
gri
d
-c
onn
ect
ed
P
V
sy
s
t
em
i
s
wh
e
n
t
h
e
b
u
i
l
d
i
n
g
i
s
bei
n
g
su
p
p
l
i
e
d
by
b
o
t
h
P
V
and
gri
d
.
The
c
o
nsum
er
s
c
a
n
u
s
e
th
e
P
V
t
o
p
o
w
e
r
e
l
e
c
t
r
i
c
a
l
a
p
p
l
i
a
n
c
e
s
a
s
w
e
l
l
a
s
sel
lin
g
e
x
c
e
ss
P
V
e
lectric
i
t
y
t
o
the
gr
i
d
[
9]
.
A
sta
nd-
al
o
n
e
P
V
s
y
s
t
e
m
o
n
t
h
e
o
t
h
e
r
h
a
n
d
r
e
f
e
r
s
t
o
a
n
e
l
e
c
t
r
i
c
a
l
loa
d
t
ha
t
is
b
e
i
ng
pow
e
r
ed
c
om
plete
l
y
b
y
t
h
e
P
V
system
.
A
sim
p
l
e
an
d
ba
sic
sta
n
d-
alo
n
e
P
V
s
ystem
c
a
n
be
do
ne
b
y
c
o
n
n
e
c
ti
n
g
t
h
e
P
V
pane
l
to
t
he
l
oa
d
l
i
ke
D
C
motor.
T
hi
s
s
i
m
p
l
e
P
V
s
y
s
t
e
m
i
s
u
s
u
a
l
l
y
u
s
e
f
o
r
pow
er
in
g
w
a
te
r
pumpi
n
g
s
ys
tem
in
r
em
ot
e
a
r
ea
s
[10].
A
st
a
n
d-al
o
ne
P
V
syste
m
i
s
mor
e
s
ui
ta
ble
t
o
pow
er
ho
uses
o
r
bu
i
l
d
in
gs
i
n
r
e
m
o
te
a
r
eas
w
he
re
t
he
c
o
s
t
o
f
b
r
i
ngi
n
g
t
h
e
grid
i
s
ve
ry
h
i
g
h
[1
1-
13]
.
The
m
o
st
impor
ta
nt
t
h
i
n
g
i
n
P
V
s
ys
t
e
m
is
t
he
s
iz
in
g
o
f
t
he
P
V
sys
t
em
.
S
i
zing
a
s
t
a
nd-a
l
o
n
e
P
V
s
y
s
tem
is
m
ore
c
r
uci
a
l
tha
n
s
iz
i
n
g
gr
id-
c
o
nnec
t
e
d
P
V
sys
t
em
.
This
i
s
d
u
e
to
t
he
f
ac
t
t
h
a
t
ev
ery
co
mp
on
e
n
t
in
t
h
e
s
t
a
nd
-a
lon
e
P
V
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
E
l
e
c
&
D
ri S
yst
IS
S
N
:
2088-
86
94
Ap
pl
ic
at
ion
o
f
ar
t
i
f
i
ci
a
l
neu
r
a
l
net
w
ork
i
n
siz
i
n
g
a st
a
nd-a
l
o
n
e ph
o
t
ov
o
l
tai
c
... (Ahm
ad
F
a
t
e
h Mo
h
a
m
a
d
Nor)
34
3
system must
be
s
ized ver
y
ca
refu
ll
y si
nce
th
e elec
tri
c
i
t
y is
1
0
0 pe
rce
n
t rel
y
on t
h
e P
V
system
. A
ny m
i
sta
k
e
in
si
z
i
n
g
th
e
PV
s
y
s
t
e
m
ca
n
cau
se
i
n
s
uffi
ci
ent
e
l
ect
ri
c
a
l
p
o
we
r
t
o
s
u
p
p
ly
t
he
l
oa
d
re
qu
ire
m
e
n
t
s
[
14,
1
5
]
.
The
me
tho
dol
o
gy
o
f
s
iz
i
ng
P
V
s
y
s
tem
i
n
v
o
l
ves
a
num
ber
of
c
a
l
cu
la
t
i
o
ns
a
nd
proce
dures.
Th
e
s
e
ca
l
c
ula
t
i
o
n
s
a
n
d
proce
dures
o
f
siz
i
n
g
t
he
P
V
sys
t
e
m
can
b
e
ti
m
e
c
o
n
sumi
n
g
a
nd
m
i
ght
a
lso
a
ffe
c
t
t
he
a
c
c
ura
c
y
o
f
t
he
s
iz
i
n
g
ou
tc
ome
.
I
n
or
de
r
t
o
m
a
k
e
t
h
e
P
V
s
izi
n
g
p
r
oced
ur
e
ea
sie
r
a
nd
l
e
ss
t
i
m
e
c
ons
um
in
g,
a
n
umbe
r
o
f
r
e
s
e
a
r
c
h
ers
have
c
o
m
e
o
u
t
w
it
h
a
var
i
et
y
of
P
V
si
z
i
ng
m
e
tho
d
i
n
c
l
u
d
i
n
g
t
he
a
p
p
lica
t
i
o
n
o
f
A
r
tific
ia
l
N
e
ural
N
e
t
w
o
rk
(A
NN
)
[11,
1
4
,
16-20]
.
The
ma
i
n
p
ur
p
o
se
o
f
th
is
p
a
p
er
i
s
to
p
re
sent
a
r
ev
i
e
w
o
f
t
he
a
p
p
li
cation
of
ANN
i
n
si
z
i
ng
a
st
a
n
d-alo
n
e
PV
s
y
s
tem
.
2.
SI
Z
I
NG
A
S
T
A
ND
-ALONE
P
V
SYSTEM
A
sta
nd-a
l
one
P
V
sys
t
e
m
c
an
b
e
w
e
ll-
de
fine
d
as
a
n
e
l
ec
t
r
i
c
a
l
s
y
s
t
e
m
t
ha
t
o
n
l
y
c
ons
ume
s
P
V
mo
d
u
l
e
s
a
s
t
he
s
ol
e
e
l
ect
ri
cal
p
o
w
e
r
s
o
u
r
ce.
T
h
e
m
a
i
n
ob
j
ecti
v
e
o
f
th
e
PV
s
y
s
t
e
m’s
si
zi
ng
p
ro
c
e
dure
i
s
t
o
ma
ke
s
ure
t
h
a
t
t
he
e
le
c
t
r
i
c
a
l
pow
er
p
r
o
duc
ed
b
y
t
h
e
P
V
s
yste
m
i
s
ade
q
u
a
te
t
o
su
pp
l
y
t
he
e
le
c
t
rica
l
l
o
a
d
requ
ire
m
e
n
t
[1
4]
.
The
siz
i
n
g
p
roce
dure
co
ns
ist
of
a
n
umbe
r
of
s
te
ps.
Befo
re
s
tar
t
i
ng t
h
e
siz
i
n
g
p
roce
d
u
r
e,
t
he
ac
cura
t
e
p
ea
k
sun
s
h
i
ne
d
a
t
a
need
t
o
be
k
n
o
wn.
Th
is
d
a
t
a
ca
n
be
o
b
t
ai
ned
from
t
h
e
loc
a
l
m
e
te
oro
l
o
g
ica
l
depa
r
t
me
n
t
w
e
b
s
ite
e
t
c
.
The
peak
s
u
n
sh
i
n
e
data
i
s
very
i
mporta
n
t
beca
use
be
s
i
de
s
d
i
ffe
r
ent
lo
ca
t
i
on
s
hav
e
di
ffe
re
nt
p
ea
k
sun
s
h
i
ne
d
a
t
a,
t
he
p
ea
k
s
u
n
s
h
i
ne
h
o
u
r
s
p
er
d
a
y
a
s
well
a
s
t
h
e
sol
a
r
i
r
ra
d
i
a
t
ion
wi
ll
a
ff
e
c
t
th
e
o
u
t
p
ut
o
f ev
e
r
y P
V
p
an
e
l
[21
].
Th
e
f
i
rst
st
ep
i
s
a
d
e
t
a
i
l
e
d
e
l
ect
ri
cal
l
o
a
d
ana
l
ysi
s
.
Lo
a
d
a
na
ly
si
s
co
ns
is
t
of
t
he
a
c
c
u
mul
a
t
i
o
n
o
f
t
h
e
to
t
a
l
e
l
e
c
t
rica
l
ene
r
gy
use
d
i
n
a
spec
ific
p
e
r
io
d
of
t
ime
.
U
sua
l
l
y,
t
he
l
o
a
ds
f
or
s
ta
nd-
alo
n
e
P
V
s
yste
m
ar
e
alm
o
st
c
o
n
st
an
t
an
d
o
n
ly
h
a
v
e
sm
all
cha
n
g
e
s.
I
f
the
loa
d
i
s
var
i
a
ble
th
rou
g
h
o
u
t
t
h
e
da
y,
t
he
n
t
h
e
h
i
ghe
st
val
u
e
o
f
t
ha
t
l
o
ad
s
hou
ld
b
e
ta
ke
n
i
n
t
o
a
c
c
o
u
n
t
.
A
n
o
t
he
r
very
i
m
por
ta
nt
d
a
t
a
tha
t
i
s
nee
d
e
d
i
s
the
l
o
ad’s
opera
tin
g
t
i
m
e
.
This
i
s
due
t
o
the
e
l
e
c
t
rica
l
l
o
ad
s
rar
e
ly
o
p
e
r
a
t
e
a
t
t
h
e
s
a
m
e
t
i
m
e
.
F
o
r
e
x
a
m
p
l
e
,
t
h
e
l
i
g
h
t
s
a
r
e
on
ly
u
se
d
dur
i
ng
n
i
g
h
tt
ime
,
w
hile
t
he
k
et
t
l
e
is
u
sua
l
ly
u
se
d
ea
rl
y
i
n
t
he
m
orni
ng
or
e
ven
i
ng.
T
he
o
pera
t
i
n
g
ti
m
e
o
f loa
d
i
s t
h
e
to
ta
l
num
b
e
r
of
h
o
u
rs
p
er
d
ay
t
ha
t
the loa
d
i
s
op
e
r
at
in
g.
N
o
t
a
l
l
l
o
a
ds
a
re
b
e
i
ng
u
s
e
d
ev
e
r
y
day.
F
or
e
xa
mple,
a
w
a
s
h
in
g
ma
chi
n
e
ope
rate
s
for
2
h
our
s
eve
r
y
w
e
ek
h
a
s
a
n
e
q
uiva
le
nt
o
pera
ti
ng
t
ime
o
f
0.29
h
our
per
day
[
1
4
,
22-2
4
].
The
fo
llow
i
n
g
s
te
ps
a
fte
r
l
o
a
d
ana
l
y
s
is
a
re
b
asica
lly
r
e
g
a
r
d
i
n
g
t
he
s
iz
in
g
of
e
ve
ry
c
om
pone
n
t
s
i
n
t
h
e
st
a
nd-al
o
n
e
P
V
s
ys
t
e
m.
A
s
tan
d
-a
lo
ne
P
V
sys
t
em
c
on
si
s
t
s
of
P
V
p
a
n
el
o
r
PV
a
rray,
i
nverter,
s
olar
c
harge
con
t
ro
l
l
er
a
nd
ba
tte
ries
[
1
5
,
25,
2
6].
The
c
onf
igur
at
ion
o
f
a
s
tanda
r
d
s
ta
nd-
a
l
o
n
e
P
V
s
ys
tem
is
d
ep
ic
ted
in
F
i
gure
1.
P
V
a
r
r
a
y
w
i
ll
co
n
v
e
r
t
sun
lig
h
t
t
o
D
C
e
le
c
t
ric
i
t
y
[
2
7
]
.
The
requi
re
d
c
u
rre
n
t
a
n
d
vo
l
t
a
g
e
tha
t
m
ust
be
pro
duce
d
b
y
t
h
e
P
V
a
rra
y
n
ee
d
t
o
b
e
determ
ine
d
d
e
p
e
nds
b
y
t
h
e
l
o
a
d
dem
a
nd
(o
b
t
ai
ne
d
from
lo
a
d
a
nal
y
sis)
.
Th
is
e
le
ctric
i
t
y
w
i
l
l
c
h
a
r
ge
t
he
b
a
tter
y
b
a
nk.
T
he
s
o
l
ar
c
har
g
e
c
on
tro
l
ler
a
c
ts
t
o
pre
v
en
t
t
h
e
ba
t
t
e
r
y
fr
om
bei
n
g
o
v
erc
h
ar
ge
d.
F
i
n
a
lly,
inve
r
t
er
i
s
use
d
t
o
c
o
n
v
e
r
t
D
C
p
r
o
d
uc
e
d
b
y
t
h
e
P
V
p
ane
l/ar
r
ay
from
t
h
e
b
a
tter
y
ban
k
t
o
A
C
e
le
c
t
r
i
c
ity
[
9].
F
i
gure
1.
C
o
n
fi
gura
t
i
o
n
o
f
a
st
a
nd-a
l
o
n
e
P
V
s
ys
t
e
m
I
n
t
e
r
m
s
o
f
s
i
z
i
n
g
t
h
e
c
h
a
r
g
e
c
o
n
t
r
o
l
l
e
r
,
o
n
e
h
a
s
t
o
m
a
k
e
s
u
r
e
t
h
a
t
t
he
c
ha
rge
c
o
ntr
o
lle
r
sho
u
l
d
be
rate
d
a
t
l
ea
st
1
2
5
%
of
t
he
m
a
x
i
m
um
c
urre
nt
from
t
h
e
P
V
a
rra
y
[1
4
,
2
3
]
.
Th
i
s
i
s
to
t
ak
e
i
n
to
a
cc
oun
t
the
cha
nge
s
t
h
a
t
m
igh
t
o
cc
urr
e
d
in
c
ur
rent
v
alue
f
r
o
m
t
h
e
P
V
a
rr
ay.
A
nother
c
r
ucia
l
pa
rt
i
s
siz
i
n
g
t
he
c
ha
rg
e
con
t
ro
l
l
er
i
s
t
o
m
a
k
e
sur
e
t
ha
t
t
h
e
c
h
arge
c
on
tro
l
ler
is
a
bl
e
t
o
o
p
era
t
e
w
i
t
h
t
he
b
a
t
tery
s
ystem
volta
ge
.
F
o
r
in
sta
n
c
e
,
if
t
he
b
at
t
e
ry
s
ys
tem
volta
ge
i
s
12V
,
the
n
a
c
h
a
rge
c
on
tro
ller
w
i
t
h
a
v
o
l
tage
r
a
t
i
ng
of
1
2V
s
h
o
u
ld
b
e
selec
t
e
d
.A
b
at
t
e
ry
b
an
k
is
a
n
um
ber
of
b
a
tte
r
i
e
s
a
re
b
ein
g
c
on
n
ec
ted
t
o
ge
t
h
e
r
w
he
ther
i
n
pa
ralle
l,
i
n
series
o
r
bo
th.
A
t
t
hi
s
st
e
p
,
t
h
e
re
qu
ired
a
m
p
ere
h
our
a
nd
sys
t
em
v
ol
ta
ge
o
f
the
b
a
tter
y
b
an
k
is
d
eter
mi
ned.
U
sua
l
l
y
,
the
s
y
ste
m
v
o
l
ta
ge
o
f
t
h
e
ba
tte
ry
b
a
n
k
is
d
e
t
e
r
m
i
ne
d
a
c
c
o
r
d
in
g
to
t
he
s
iz
e
o
f
t
he
s
y
s
tem
’
s
pow
er.
U
s
ua
lly
1
2
V
or
2
4
V
bat
t
ery
ba
n
k
v
olt
a
ge
i
s
used.
B
a
tter
y
b
a
nk
i
s
s
ize
d
t
o
st
ore
eno
u
gh
e
n
e
r
g
y
to
s
up
p
l
y
t
h
e
loa
d
s
requ
ire
m
e
n
t
[2
3]
.
F
i
nal
l
y
,
the
i
n
ve
rter
n
ee
d
t
o
b
e
si
z
e
d
s
o
t
h
a
t
i
t
i
s
ab
le
t
o
sup
p
l
y
a
t
lea
s
t
the
sam
e
a
m
oun
t
o
f
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 V
ol.
11,
N
o.
1
, Ma
r
202
0
:
342
–
34
9
34
4
pow
er
a
s
the
t
o
ta
l
A
C
p
ow
e
r
d
em
and.
N
o
n
e
the
l
ess
it
i
s
h
i
g
h
l
y
su
gg
est
e
d
t
h
a
t
t
h
e
s
i
z
e
of
t
h
e
i
nv
e
r
t
e
r
ha
v
e
t
o
be
a
b
it
h
i
gher
tha
n
t
he
t
o
t
a
l
A
C
p
o
w
e
r
de
ma
nd
by
1
0
t
o
4
0
%
.
A
n
o
t
h
e
r
vi
ta
l
pa
rt
i
n
s
i
z
i
n
g
t
he
i
nve
rte
r
i
s
th
a
t
the
i
n
verte
r’s
o
u
t
pu
t
vol
ta
ge
m
ust
be
s
am
e
w
i
t
h
t
he
A
C
sys
t
em
v
o
lta
g
e
.
For
exam
pl
e
i
f
t
he
A
C
s
y
s
t
em
vo
lta
ge
i
s
2
4
0
V
A
C
t
he
n
the
in
ver
t
e
r
o
utp
u
t
vo
lta
ge
m
us
t
als
o
b
e
2
40
V
A
C.
O
n
the
ot
he
r
ha
n
d
,
the
D
C
i
n
p
u
t
vo
lta
ge o
f t
h
e
inve
r
t
er
m
ust e
qua
l w
i
t
h
t
he
b
att
e
r
y
sys
tem
vol
t
ag
e
[
2
3
]
.
3.
ARTIFI
C
IAL N
E
URAL
N
E
T
WO
R
K
A
r
t
i
f
i
c
i
a
l
N
e
u
r
a
l
N
e
tw
ork
or
A
N
N
i
s
a
m
e
t
h
o
d
t
hat
ada
p
t
s
t
he
n
e
ur
on
s
y
ste
m
i
n
t
h
e
h
u
m
a
n
brai
n
in
to
s
o
l
vin
g
e
ng
ine
e
ri
ng
or
t
ec
h
n
o
lo
gy
prob
le
m
suc
h
a
s
predic
t
i
on
val
u
es,
class
i
f
i
ca
t
i
o
n
,
o
b
serva
tio
n
e
t
c
.
ANN
i
s
v
ery
u
s
eful
f
o
r
m
o
d
e
l
l
i
n
g
o
r
p
redict
ing
ev
en
t
hou
gh
t
h
e
i
n
pu
t
ou
t
p
u
t
d
a
t
a
rel
a
ti
o
n
shi
p
i
s
un
kn
ow
n
[
2
8
-
32
].
A
s
i
m
p
l
e
con
f
i
g
u
r
at
i
o
n
o
f
ANN
i
s
b
u
i
lt
u
p
o
f
t
h
r
ee
l
ay
er
s
.
T
he
f
i
r
st
l
ayer
i
s
al
so
known
a
s
input
l
a
y
e
r
.
T
h
i
s
i
s
w
h
e
r
e
t
h
e
d
a
t
a
t
h
a
t
h
a
s
b
e
e
n
s
e
l
e
c
t
e
d
a
s
i
n
p
u
t
i
s
i
n
sert
ed
i
nt
o
t
h
e
ANN
co
nfi
g
u
r
atio
n
.
T
h
e
se
c
o
n
d
l
a
y
e
r
i
s
k
no
wn
a
s
th
e
h
i
dd
en
l
ay
er.
Th
e
s
econ
d
l
a
ye
r
co
n
sis
t
o
f
neurons,
w
ei
ghts
as
t
he
s
ynapses
that
con
n
ec
t
a
ll
t
h
e
neur
ons
t
oge
t
h
er
.
The
t
h
ird
laye
r
is
t
he
f
i
n
al
an
d
o
u
t
p
ut
l
a
y
er
t
hat
g
i
ve
a
w
a
y
the
proc
essed
d
a
t
a
o
r
th
e
re
su
l
t
f
r
o
m
t
h
e
ANN
co
n
f
i
g
u
r
at
i
o
n
[
3
3
]
.
ANN
can
b
e
c
o
nfi
gured
i
n
t
o
a
n
u
m
b
er
o
f
d
i
ffe
r
en
t
con
f
ig
ura
tio
ns.
N
e
vert
hele
ss,
t
he
m
os
t
c
o
mm
on
an
d
broa
dl
y
use
d
A
N
N
c
o
n
f
ig
ur
at
i
o
n
is
t
he
m
u
l
t
ila
ye
r
perc
ep
tron
w
i
t
h
ba
ck
p
ropa
g
a
t
i
o
n
[
1
7
].
I
t
i
s
c
a
lle
d
ba
ck
p
r
opa
g
a
t
i
o
n
beca
use
t
h
e
err
o
r
b
e
twee
n
the
pr
edic
te
d
o
u
t
p
ut
s
f
r
o
m
ANN
wi
t
h
t
he
t
ar
g
e
t
wil
l
b
e
s
ent
b
ack
t
o
t
h
e
hi
d
d
e
n
l
a
y
e
r
fo
r
we
i
ght
a
dju
s
t
m
en
t
[3
4
]
.
F
i
gure
2
[
3
5]
s
how
s
t
h
e
c
o
n
f
i
g
ura
t
i
o
n
of
t
he
m
u
l
t
ilaye
r
pe
rce
p
t
r
o
n
w
i
t
h
b
ac
k
pro
p
a
g
a
t
i
o
n
ne
tw
ork.
D
ue
t
o
t
h
e
c
u
rrent
c
omp
u
t
er
t
ec
hnol
og
y
,
A
N
N
h
a
s
b
e
e
n
wid
e
ly
a
p
p
l
i
e
d
in
v
a
r
io
us
e
n
g
i
n
e
erin
g
fie
l
ds
i
nc
l
u
d
i
ng
the P
V
fi
e
ld [
36-39]
.
F
i
gure
2.
C
o
n
f
i
gur
a
tio
n
of t
he
m
ult
i
l
a
yer
perc
eptr
on w
ith ba
c
k
p
r
op
a
g
ati
on o
f
ANN
4.
APPLI
C
AT
ION OF A
N
N
I
N SI
Z
I
N
G
A
S
T
A
N
D
-
A
LO
N
E
P
V
S
Y
STEM
I
n
t
his sec
t
i
o
n, t
he
l
i
t
e
r
at
ure
re
view
on t
h
e a
p
plic
a
t
i
o
n
of A
N
N
in s
iz
in
g a
sta
nd-a
l
one
P
V
sys
t
em
a
re
prese
n
t
e
d.
T
he
liter
a
ture
a
r
e
s
e
l
ect
ed
b
ase
d
o
n
the
a
p
plica
t
i
o
n
o
f
ANN
i
n
v
ar
i
o
u
s
a
sp
ect
o
f
PV
s
yst
e
m
si
zing
.
The
as
pec
t
s i
n
c
l
u
d
e
t
h
e
s
i
z
i
n
g
o
f
t
h
e
PV
s
y
s
te
m’s
com
p
on
ent
s
s
u
c
h
a
s
PV
p
a
n
el
s,
b
a
t
t
e
ri
es
e
t
c
,
t
h
e
p
r
ed
i
c
ti
on
of
l
oa
d
po
w
e
r,
t
he
p
re
d
i
ct
ion
of
t
he
e
l
e
c
t
ric
a
l
p
ow
er
t
ha
t
is
a
ble
t
o
b
e
ge
ner
a
te
d
b
y
t
he
P
V
system
,
find
ing
sui
t
abl
e
t
i
l
t
ang
l
e and
so
on
.
On
e
o
f
t
h
e
a
dvan
t
ag
e
of
ANN
t
h
at
m
ak
es
A
NN
s
u
i
t
a
bl
e
f
o
r
si
zi
ng
a
PV
s
y
s
t
e
m
i
s
t
h
a
t
ANN
h
a
s
t
h
e
abi
l
i
t
y
t
o
lea
r
n
the
c
o
n
n
ec
tio
n
be
tw
ee
n
th
e
in
p
u
t
a
n
d
o
u
t
p
ut
p
a
ramet
e
r
s
.
ANN
can
b
e
u
sed
to
a
c
h
i
e
v
e
t
h
a
t
by
l
e
a
r
ni
ng
t
h
e
p
re
v
i
ou
s
re
co
rd
ed
d
a
t
a
[17
]
.
Th
is
i
s
al
so
known
a
s
ANN
t
r
a
i
n
in
g
d
a
t
a
.
ANN
act
s
l
i
k
e
a
“
b
l
a
ck
bo
x”
w
here
i
n
this
“
b
l
ac
k
b
o
x
”
t
he
l
e
a
rn
i
ng
and
t
r
ai
n
i
n
g
p
roce
s
s
take
p
lac
e
.
P
r
evio
us
r
es
ea
rche
r
w
o
r
k
i
n
[17]
h
a
s
ap
pli
e
d
ANN
t
o
p
red
i
ct
t
h
e
s
i
z
in
g
co
e
f
f
i
ci
e
n
t
of
a
s
t
a
n
d
-
al
o
n
e
P
V
s
ys
tem
.
I
n
th
i
s
p
ape
r
,
a
n
u
m
b
er
o
f
d
i
f
f
e
ren
t
ANN
con
f
ig
u
r
at
ion
s
a
re
t
est
e
d
c
o
mp
ared
.
Th
e
r
e
sul
t
s
h
o
w
ed
t
h
a
t
ANN
con
f
ig
u
r
at
ion
t
h
at
c
o
n
s
i
s
t
s
of
o
ne
h
i
dde
n
laye
r
w
i
t
h
8
n
e
u
ro
ns
g
a
v
e
the
m
o
st
acc
urate
r
e
sul
t
.
I
n
add
i
t
i
o
n
,
t
h
e
ANN
mo
d
e
l
d
e
v
e
l
oped
i
n
th
i
s
p
a
p
er is
ve
r
y
su
ita
b
l
e
a
nd
fi
ts N
orth
A
fri
c
a
n
c
ou
n
t
ries su
c
h
a
s Al
geria
.
F
rom
ther
e, an
am
ou
n
t
o
f r
e
s
e
a
r
ch
are
a
c
an
b
e
d
o
n
e
i
n
o
r
d
er
f
or
t
h
i
s
sys
t
em
t
o
wor
k
i
n
othe
r
cou
n
tries
es
pe
cia
l
l
y
M
ala
y
s
i
a.
T
his
is
d
u
e
t
o
t
h
e
f
a
c
t
t
h
a
t
t
h
e
cli
m
a
t
e
o
f
M
a
l
a
y
si
a
is
v
ery
d
i
f
f
e
r
en
t
t
h
at
t
h
e
c
l
ima
t
e
of
N
o
r
th
A
fric
a
n
c
oun
tries.
T
he
c
lim
ate
espec
i
al
ly
t
he
a
moun
t
o
f
s
o
l
ar
r
adia
ti
o
n
h
a
v
e
a
grea
t
i
m
pac
t
o
f
t
he
a
m
oun
t
o
f
e
lec
t
r
i
c
ity
t
ha
t
t
h
e
ca
n
be
gene
ra
ted
b
y
t
he
P
V
syste
m
.
The
fin
d
i
n
g
s
in
[
4
0
]
ha
ve
a
lso
sh
ow
n
th
at
ANN
co
n
f
i
g
u
r
at
i
o
n
o
f
o
n
e
h
id
d
e
n
layer
w
i
t
h
8
n
e
u
ro
ns
g
ave
t
h
e
m
o
st
a
c
c
ura
t
e
o
u
t
p
ut
v
a
l
ue
s.
H
ow
e
ve
r,
t
h
i
s
doe
s
no
t
m
e
an
t
ha
t
8
n
u
m
b
er
o
f
n
e
u
r
o
n
i
s
t
h
e
b
est
nu
mb
er
f
o
r
e
v
e
ry
ANN
co
nf
igu
r
ati
o
n
.
T
h
i
s
i
s
bec
a
u
s
e
t
h
a
t
t
he
n
u
m
ber
of
n
e
u
ro
n
varie
s
f
r
o
m
o
n
e
ANN co
n
f
i
g
u
r
ati
o
n
t
o
a
not
h
e
r
d
e
pen
d
in
g
on
t
h
e
ty
p
e
o
f
da
t
a
,
s
i
ze of data, initi
a
l
w
eight va
lue etc.
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
E
l
e
c
&
D
ri S
yst
IS
S
N
:
2088-
86
94
Ap
pl
ic
at
ion
o
f
ar
t
i
f
i
ci
a
l
neu
r
a
l
net
w
ork
i
n
siz
i
n
g
a st
a
nd-a
l
o
n
e ph
o
t
ov
o
l
tai
c
... (Ahm
ad
F
a
t
e
h Mo
h
a
m
a
d
Nor)
34
5
The
rese
arc
h
ers
in
[
2
0
]
ha
s
st
ate
d
t
ha
t
A
N
N
has
bee
n
w
ide
l
y
use
d
i
n
t
he
f
i
e
l
d
o
f
rene
w
a
ble
e
n
erg
y
suc
h
a
s
t
h
e
s
i
z
i
n
g
o
f
P
V
s
ys
t
e
m
in
r
em
ote
a
r
ea
s.
T
hey
ha
ve
a
l
s
o
s
u
c
cessf
ull
y
a
pp
li
ed
ANN
t
o
p
redi
ct
t
h
e
s
i
z
i
n
g
c
o
e
f
f
i
c
i
e
n
t
o
f
a
P
V
s
y
s
t
e
m
.
A
l
s
o
i
n
t
h
i
s
p
a
p
e
r
,
t
h
e
y
h
a
v
e
i
m
p
r
ov
e
t
h
e
t
r
ain
i
ng
s
y
s
t
e
m
o
f
t
h
e
ANN
b
y
com
b
i
n
in
g
i
t
w
it
h
gene
t
i
c
a
l
gor
i
t
hm
(
G
A
).
T
he
r
e
s
ults
h
a
v
e
prov
e
d
t
ha
t
the
impr
o
v
e
d
t
he
A
N
N
s
t
r
uc
ture
i
s
able
t
o
pre
d
i
c
t
the
siz
i
n
g
c
oe
fficie
n
t
o
f
a
P
V
system
s
iz
in
g
c
u
rve
w
i
th
l
ess
in
i
tia
l
in
fo
rm
atio
n
.
T
he
r
esu
l
t
s
f
r
o
m
t
hi
s
p
a
p
e
r
h
a
v
e
p
r
o
v
e
d
t
h
e
a
d
v
a
nt
ag
e
o
f
ANN
w
h
i
c
h
i
t
c
an
l
earn
o
r
di
s
c
over
the
relations
h
ip
b
e
t
ween
t
h
e
i
npu
t
and
t
h
e
t
a
rg
e
t
v
a
l
u
e
s
e
v
en
t
h
ough
t
h
e
r
el
ati
o
n
s
h
i
p
i
s
not
c
l
e
ar
o
r
le
ss
i
n
f
o
rma
tive
.
O
t
h
er
t
ha
n
th
at,
the
c
o
mbi
n
a
t
i
on
of
A
N
N
an
d
G
A
ha
s
o
p
e
ned
a
h
u
g
e
rese
arc
h
a
r
ea.
Th
is
c
a
n
b
e
a
p
p
l
i
e
d
in
v
ar
ious
P
V
app
l
ica
t
i
o
n esp
e
cial
l
y
i
n the
fi
eld o
f
s
i
z
ing
a
st
a
nd-al
o
n
e
PV
sy
stem
.
A
N
N
in
[
4
1
]
h
a
s
be
en
a
pp
l
i
e
d
t
o
fore
ca
st
s
olar
i
r
r
ad
iat
i
on
a
nd
loa
d
p
ow
er
c
onsump
t
i
o
n
for
the
P
V
syste
m
.
They
h
a
v
e
use
d
a
nd
c
ompa
red
a
n
u
mbe
r
o
f
d
i
ffe
rent
c
on
fi
gur
at
ions
s
uc
h
as
d
i
f
fer
e
n
t
n
umbe
r
of
layer
s
,
d
i
ffere
n
t
se
t
o
f
i
np
u
t
s
an
d
di
ffere
n
t
n
u
mbe
r
o
f
hidde
n
ne
ur
ons.
Th
is
r
e
s
e
a
rc
h
has
fo
u
n
d
o
u
t
t
ha
t
t
h
e
d
i
f
f
e
ren
t
c
onf
ig
u
r
at
ion
s
w
il
l
g
i
v
e
e
f
f
e
ct
t
o
th
e
ANN
p
r
ed
ict
i
on
’s
acc
urac
y.
D
i
f
fer
e
n
t
c
onf
igur
at
ion
s
c
a
n
b
e
use
d
t
o
im
prove
a
ccur
acy
v
al
ues.
H
ow
ever
,
the
me
t
h
o
d
o
f
de
te
rmi
n
i
ng
t
h
e
b
es
t
o
r
s
ui
t
a
bl
e
ANN
con
f
ig
ura
tio
n
i
s
y
e
t
t
o
be
d
isc
ove
re
d.
T
hi
s
is
b
e
cause
,
as
m
e
n
t
i
o
n
e
d
p
r
e
v
i
o
u
s
ly
,
d
i
f
f
e
rent
ANN
co
n
f
i
gurat
i
on
varie
s
f
rom
on
e
a
p
p
l
ica
tio
n t
o
a
no
t
h
er.
He
n
ce,
a wide
researc
h
a
r
e
a
o
n
thi
s
to
p
i
c
i
s av
ail
a
b
l
e to
b
e e
xplo
r
ed
.
C
e
y
l
an
e
t
a
l
.
in
[
4
2
]
h
av
e
u
s
ed
ANN
t
o
p
r
e
d
i
ct
t
h
e
e
f
f
i
c
i
e
n
c
y
o
f
P
V
m
od
u
l
e
for
a
ll
r
e
gi
o
n
s
in
T
urk
e
y.
T
he
ANN
co
n
f
ig
u
r
at
io
n
con
s
i
s
t
s
o
f
t
h
e
measu
r
ed
s
ol
ar
r
ad
i
a
ti
o
n
and
m
e
a
s
u
r
e
d
P
V
m
o
d
u
l
e
a
m
b
i
e
n
t
a
i
r
t
e
mp
erat
u
r
es
a
s
t
h
e
inp
u
t
s.
W
hil
e
t
h
e
ANN
o
u
t
pu
t
i
s
t
he
P
V
mo
d
u
le
b
ack
s
ide
te
mpe
r
at
ure
s
.
PV
m
odu
les
tem
p
era
t
ur
e
w
e
re
c
hosen
a
s
t
h
e
o
u
t
p
ut
b
ec
a
u
se
P
V
m
odule
s
t
em
per
at
ure
has
grea
t
im
p
act
i
n
t
h
e
effic
i
ency
o
f
a P
V
syste
m.
The au
t
hors a
l
so
have
sta
t
e
d
i
n
t
h
is r
esea
rch w
o
rk
t
h
a
t
ANN
h
a
s b
e
en
w
i
d
ely
u
s
ed
to
p
r
e
d
i
c
t
th
e
efficie
n
c
y
a
n
d
m
axim
um
power
of
t
h
e
P
V
system
.
Ba
pt
is
t
a
e
t
a
l
.
in
[
3
7
]
ha
ve
a
p
p
lie
d
A
N
N
t
o
pred
ict
the
e
l
e
c
tri
c
a
l
e
ner
gy
pro
duc
t
i
o
n
by
a
P
V
s
ystem
.
The
pred
ic
t
i
o
n
o
f
e
l
e
c
tr
ica
l
e
n
e
rgy
pr
o
duc
t
i
o
n
b
y
a
P
V
s
ystem
i
s
very
i
mp
orta
nt in
the
s
i
z
i
n
g
p
r
o
ce
ss
fo
r
c
o
st
and
system
e
ff
i
c
ie
nc
y.
T
he
A
N
N
used
c
o
n
s
i
sts
o
f
t
hre
e
l
a
y
ers
w
h
i
c
h
are
th
e
i
n
put
l
a
y
e
r
,
h
i
dd
en
l
a
y
e
r
a
nd
ou
tpu
t
l
a
y
er.
The
au
t
hors
ha
v
e
r
aised
one
o
f
t
h
e
pr
oble
m
o
f
ap
p
lyi
n
g
ANN
i
s
t
h
e
d
et
er
mi
n
a
tio
n
of
numb
e
r
of
hi
dde
n
ne
ur
on
s.
T
his
i
s
b
ec
a
u
se
t
here
a
re
no
s
p
e
c
i
f
ic
m
e
t
ho
ds
t
o
i
d
ent
i
fy
t
he
c
orre
ct
num
ber
of
h
id
de
n
neur
ons
f
or
e
ach
A
N
N
c
onfi
gura
t
i
o
ns.
H
e
nc
e,
t
his
p
a
per
i
nve
st
ig
a
t
es
t
w
o
d
iffere
nt
m
etho
ds
t
o
de
ter
m
ine
t
h
e
numbe
r
o
f
h
i
d
den
n
e
ur
ons.
Th
ose
me
t
hod
s
a
r
e
K
o
lm
og
oro
v
’s
T
h
e
ore
m
and
rule
o
f
thumb.
T
his
r
e
search
w
o
rk
h
a
s
c
onc
l
u
de
d
tha
t
t
he
K
olmo
gor
ov’s
The
o
re
m
ga
v
e
b
e
t
t
e
r
re
s
u
l
t
s
com
p
are
d
t
o
t
h
e
ru
l
e
o
f
t
h
um
b.
I
n
add
i
tio
n,
t
h
i
s
p
a
pe
r
has
suc
c
e
ssfu
l
l
y
a
pp
l
i
e
d
A
N
N
t
o
the
fi
e
l
d
t
h
a
t
i
s
m
o
s
t
r
e
l
a
t
e
d
t
o
P
V
s
i
z
i
n
g
w
h
i
c
h
i
s
t
h
e
ene
r
g
y
t
ha
t
c
a
n
b
e
pro
d
u
ce
d
by
t
h
e
P
V
s
ystem
.
T
hi
s
te
c
h
ni
que
c
a
n
be
f
urt
h
er
e
xpa
nde
d
i
n
t
o
p
red
i
c
t
i
ng
th
e
ou
tpu
t
pow
er
f
rom
other
P
V
c
om
po
ne
nt
s suc
h
a
s ba
tte
ry ba
nks
e
tc
.
A
r
un
in
[
4
3
]
ha
s
de
vel
o
ped
a
tw
o-sta
g
e
A
N
N
confi
gur
a
t
i
o
n
for
pr
e
d
i
c
t
t
h
e
s
i
z
i
n
g
c
urve
f
or
t
h
e
st
a
nd-al
o
n
e
P
V
s
yst
e
m
i
n
i
sola
te
d
are
a
s
in
s
ou
t
h
ern
I
ndia
.
T
he
fi
r
s
t
st
age
o
f
t
h
e
ANN
co
nfi
g
u
r
at
i
o
n
co
n
s
i
s
t
s
of
t
he
g
e
ogra
p
hica
l
c
oor
dina
te
s
of
t
he
l
oca
t
i
on
as
t
he
i
n
p
u
t
s.
The
o
u
tp
uts
for
th
is
f
irst
s
t
a
ge
a
re
t
he
m
inim
um
regu
la
t
e
d
P
V
a
rra
y
ratin
g
a
n
d
the
cor
r
esp
o
n
d
i
n
g
reg
u
la
te
d
ba
tte
r
y
c
a
p
a
c
i
t
y
.
I
n
t
h
e
s
e
c
o
n
d
s
t
a
g
e
o
f
t
h
e
A
N
N
con
f
ig
ura
tio
n,
t
he
g
e
o
grap
h
i
c
a
l
co
ord
i
na
tes
a
nd
the
sta
nda
rdize
d
P
V
a
rr
a
y
r
ati
ngs
a
re
b
ein
g
s
e
l
e
c
te
d
a
s
t
he
in
put
s,
w
hil
e
t
he
c
orre
spon
d
i
ng
sta
nda
r
d
ize
d
s
t
o
rage
c
a
p
a
c
i
t
i
es
a
re
s
e
l
ect
e
d
a
s
t
h
e
outp
u
t
s
.
Thi
s
t
e
c
h
n
i
qu
e
ca
n
be
f
ur
ther
e
xpa
n
d
ed
i
nto
m
u
lti
pl
e
sta
g
e
of
A
N
N
con
f
i
gura
t
i
o
n
.
O
t
h
er
t
h
a
n
th
a
t
,
t
h
i
s
r
e
s
ea
rch
al
s
o
c
a
n
b
e
exp
l
ored
m
or
e usin
g Ma
l
a
y
s
ia
’s clim
ate.
C
h
at
terje
e
a
n
d
K
eyha
ni
i
n
[
4
4]
h
a
v
e
implem
ente
d
A
N
N
t
o
pred
ict
t
h
e
i
d
e
a
l
ti
lt
a
n
g
l
e
o
f
P
V
a
r
ray
at
a
sp
ecif
i
ed
l
o
c
a
t
ion
.
T
hi
s
i
s
t
o
p
r
e
d
i
c
t
t
h
e
amo
u
n
t
of
e
l
ectri
c
al
e
nerg
y
ge
nera
t
e
d
from
t
he
P
V
arr
a
y.
T
he
t
ilt
ang
l
e
of
t
he
P
V
arr
a
y
i
s
v
er
y
im
p
o
r
t
a
n
t
i
n
o
r
d
er
t
o
m
a
ke
s
ure
t
h
e
P
V
array
i
s
a
bl
e
to
g
en
era
t
e
max
i
mu
m
elec
tr
ici
t
y
t
h
a
t
t
he
P
V
a
rra
y
is
a
ble
t
o
.
N
o
t
o
n
l
y
t
ha
t,
t
he
t
i
lt
a
ngl
e
of
t
h
e
P
V
array
ca
n
a
l
l
o
w
th
e
rai
n
w
at
er
t
o
f
l
o
w
w
h
e
n
rain
ing
.
T
h
i
s
resear
ch
w
o
r
k
h
a
s
co
n
c
lu
d
e
d
th
at
ANN
h
a
s
the
a
b
il
it
y
to
l
e
a
rn
t
he
n
o
n
li
ne
ar
rela
tio
ns
hip
be
tw
een
t
he
P
V
a
rra
y
tilt
a
n
g
l
e
s
,
l
a
t
i
t
u
de,
gro
u
n
d
r
e
f
lect
i
v
ity
w
i
t
h
t
h
e
s
o
la
r
ir
radiat
i
on
re
c
e
ive
d
w
itho
u
t
a
n
y
c
o
m
plex
a
nal
y
tic
al
m
e
t
hod.
N
e
v
er
the
l
es
s,
t
he
i
de
n
t
i
fi
ca
ti
o
n
o
f
t
h
e
t
i
l
t
an
gl
e
o
f
t
h
e
P
V
mo
du
l
e
i
s
very cr
u
c
i
al i
n a
r
ea
s
w
h
ich is loca
t
e
d
c
l
o
ser
t
o
t
he
e
a
r
t
h
’s
p
o
l
e
s
su
ch
a
s
t
h
e
Eu
rop
e
an
c
ou
nt
ri
e
s
a
nd
A
us
t
r
ali
a
.
For
c
o
u
n
t
r
i
e
s
like
Ma
la
ys
ia
w
hic
h
i
s
loc
a
t
e
d
n
ea
r
to
t
he
e
qua
t
o
r,
t
he
t
ilt
ang
l
e
is
n
o
t
m
uch
a
n
i
ss
ue
s
inc
e
t
he
loca
t
i
o
n
o
f
the
sun
is m
ore
less the
sam
e
t
hr
o
ugh
ou
t t
h
e
yea
r
.
5.
SUMMAR
Y
OF THE AN
N IMPLE
M
ENTATIO
N
I
N S
I
ZI
NG A
S
TAND-
AL
ONE PV
S
Y
ST
E
M
I
t
c
an
b
e
co
mp
r
e
h
e
nd
ed
fro
m
t
h
e
p
r
evi
o
us
s
ect
i
on
t
h
a
t
ANN
h
a
s
b
ee
n
w
i
del
y
u
se
d
i
n
t
h
e
s
iz
i
n
g
of
a
P
V
s
yste
m.
T
he
a
bi
l
ity
o
f
A
N
N
t
o
lea
r
n
th
e
no
n-l
i
ne
ar
r
e
l
a
t
i
o
n
s
hi
p
b
e
tween
in
put
a
nd
o
ut
put
m
ak
es
ANN
very
s
ui
ta
ble
for
th
is
pur
pose
w
h
e
t
he
r
to
p
r
e
dic
t
t
he
s
ize
of
P
V
p
a
n
e
l
s
o
r
P
V
a
r
r
a
y
s
,
s
i
z
e
o
f
b
a
t
t
e
r
i
e
s
,
s
o
l
a
r
irra
d
i
a
tio
n
e
t
c.
I
n
a
d
d
i
ti
on,
t
he
p
re
vio
u
s
se
cti
o
n
a
l
s
o
c
on
vey
s
t
ha
t
t
h
ere
i
s
a
w
ide
ar
e
a
o
f
the
a
p
p
lic
at
io
n
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 V
ol.
11,
N
o.
1
, Ma
r
202
0
:
342
–
34
9
34
6
o
f
ANN
i
n
s
izi
n
g
PV
s
y
s
t
e
m
t
h
at
can
b
e
ex
pl
o
r
e
ev
en
m
o
r
e.
T
h
e
se
a
r
e
a
s
a
r
e
e
x
p
l
a
i
n
e
d
m
o
r
e
i
n
t
h
e
fo
l
l
ow
i
n
g
sub-
sect
io
ns.
O
n
e
of
t
he
m
os
t
c
o
mm
on
m
a
t
te
rs
t
ha
t
th
e
pas
t
r
esea
rche
rs
i
n
try
in
g
to
i
nve
sti
g
a
t
e
is
t
he
s
u
i
tab
l
e
numbe
r
o
f
n
e
u
r
on
espec
i
a
lly
t
he
w
or
ks
t
hat
have
b
e
e
n
do
n
e
i
n
[3
7,
40]
.
Th
is
i
s
due
t
o
the
fa
ct
t
ha
t
t
h
er
e
ar
e
no
s
t
a
ndard
o
r
suita
ble
me
th
od
to
d
e
t
e
r
m
i
n
e
t
he
n
umbe
r
of
n
euro
n
s
i
n
th
e
ANN.
T
h
e
r
e
i
s
no
g
u
a
rant
ee
t
h
a
t
t
h
e
in
c
r
ea
se
numb
e
r
o
f
n
eu
ron
s
c
a
n
i
n
c
re
ase
th
e
ac
cu
ra
c
y
o
f
t
h
e
ANN
or
v
i
c
e
v
e
rsa.
S
o
m
e
research
ers
su
ch
as
i
n
[4
5]
u
sed
10
ne
ur
ons
i
n
the
h
i
dde
n
l
a
y
e
r.
T
his
is
b
ec
ause
1
0
n
u
m
b
e
r
o
f
n
e
u
r
o
n
s
a
r
e
t
h
e
d
e
f
a
u
l
t
n
e
u
r
o
n
n
u
m
b
e
r
in
t
h
e
M
ATLA
B
ANN
t
o
olb
o
x
.
M
an
y
research
ers
h
a
v
e
c
o
m
e
o
u
t
w
ith
a
n
um
be
r
of
m
etho
d
i
n
o
r
d
er
to
d
e
t
e
r
m
i
ne
t
h
e
s
u
ita
ble
n
u
m
b
er
o
f
neur
ons
[
4
6
-4
9].
The
c
o
mbi
n
a
t
i
o
n
of
G
A
in
A
N
N
is
a
goo
d
i
d
ea
i
n
o
r
d
e
r
to he
l
p de
te
rm
inin
g
t
h
e
s
u
i
t
a
b
l
e
num
ber
of h
i
dde
n
ne
uro
n
.
A
n
o
t
her
r
e
sear
ch
a
re
a
i
n
t
his
t
o
p
i
c
is
t
he
d
eterm
i
nat
i
on
of
t
he
i
np
u
t
a
nd
o
u
t
p
ut
o
f
number
o
n
neur
ons
i
n
A
N
N
.
S
imilar
to
t
he
d
e
t
e
r
m
i
na
ti
on
o
f
num
ber
of
n
e
u
r
o
n
s
,
no
s
p
e
ci
fi
ed
m
e
t
ho
d
th
at
i
s
av
a
i
lab
l
e
on
how
t
o
se
lect
w
hic
h
d
a
t
a
or
i
n
f
orm
a
tio
n
as
t
he
i
n
p
u
t
.
Bes
i
de
s
t
hat,
t
he
num
ber
of
i
n
p
u
t
s
or
s
e
t
o
f
in
put
s
ar
e
als
o
c
a
n
b
e
i
n
ves
t
i
g
a
t
ed
i
n
fut
u
re
r
esea
rch.
A
s
m
e
nti
o
ned
in
[
4
1],
di
ffe
r
e
nt
s
e
t
s
o
f
i
np
ut
s
w
ill
g
i
ve
d
iffere
nt
leve
l
of
a
c
c
ura
c
y
o
f
t
he
A
N
N
s
ys
tem
.
S
o
for
t
h
e
P
V
s
yste
m
sizi
n
g,
a
n
u
m
ber
of
s
e
t
s
in
pu
ts
c
a
n
b
e
pre
p
are
d
.
Th
en
c
o
m
p
a
r
e
d
t
h
es
e
s
e
t
s
w
it
h
t
h
e
accu
r
a
cy
o
f
t
h
e
ANN.
A
s
far
a
s
the
d
e
ter
m
ina
t
i
o
n
o
f
t
he
A
N
N
’s
o
ut
p
u
t
i
s
conc
er
ned,
S
ecti
o
n
4
ha
s
s
h
ow
n
tha
t
t
he
o
u
t
pu
t
ca
n
be
w
het
h
er
a
n
y
c
o
m
po
ne
nt
s
o
f
t
he
P
V
sys
t
e
m
,
the
ene
r
g
y
p
r
oduc
ed
b
y
the
s
y
st
e
m
,
the
PV
p
a
n
e
l
t
ilt
a
n
g
l
e
a
n
d
the
tem
p
er
at
ure
o
f
t
he
P
V
pane
l.
T
hi
s
a
r
e
a
c
a
n
be
w
i
de
n
eve
n
m
ore
b
y
i
nc
l
u
d
i
ng
the
s
i
z
e
o
f
ot
her
P
V
s
ys
t
e
ms’
c
o
mp
o
n
e
nts
suc
h
a
s
t
h
e
s
i
ze
o
f
c
h
ar
ge
con
t
ro
l
l
er
a
nd i
nver
t
e
r
.
Ano
t
her
rese
arc
h
a
re
a
in
t
hi
s
to
p
i
c
t
h
at
c
a
n
b
e
ex
pl
ore
in
f
ut
u
re
r
esea
rch
is
t
he
i
m
p
ro
ve
m
e
nt
o
f
the
ANN
co
n
f
ig
u
r
at
io
n
.
T
hi
s
can
b
e
do
n
e
b
y
comb
i
n
in
g
ANN
wi
th
o
th
er
a
rt
i
f
icia
l
i
n
te
ll
ige
n
t
co
nfi
gura
t
i
o
n.
F
or
i
n
s
t
an
ce,
as
d
i
s
co
v
e
r
e
d
i
n
[
20],
i
t
i
s
p
o
s
si
bl
e
t
o
c
o
m
bi
n
e
ANN
w
i
t
h
GA.
T
h
e
c
o
m
bi
n
a
ti
o
n
o
f
GA
i
n
ANN
h
a
s
show
n
gre
a
t
p
o
t
en
t
i
a
l
i
n i
m
pr
ov
i
ng t
h
e per
f
o
r
m
a
nce
of
t
he A
N
N
.
Ot
h
e
r
t
h
an
GA,
A
NN
c
a
n
al
so
b
e
co
mb
in
ed
with
f
uzz
y
l
o
g
i
c
t
o
c
r
eate
a
con
f
ig
urat
io
n
th
at
c
an
b
e
ca
lle
d
a
s
A
d
a
p
tive
N
e
uro-
F
u
zz
y
Inference
Sys
t
em
o
r
A
N
F
IS
.
AN
FIS
c
a
n
be
d
e
f
in
ed
i
n
de
ta
i
l
ed
a
s
a
s
t
ruc
t
ura
l
d
es
ig
n
th
a
t
c
on
si
st
s
a
c
o
mbin
a
t
io
n
of
A
N
N
w
i
t
h
Su
g
e
no
t
y
p
e
d
fu
zzy
lo
gi
c.
I
n
ANF
I
S
s
y
s
t
e
m,
t
h
e
ANN
met
hod
s
u
c
h
as
b
a
c
k
pr
opa
ga
t
i
on
i
s
u
s
e
d
t
o
i
m
prov
e
the
m
e
m
b
e
r
shi
p
f
u
n
c
t
ions
a
nd
r
u
l
e
s
of
t
he
f
uzz
y
s
ys
tem
[5
0].
Ta
b
l
e
1
su
mma
ri
ze
s
t
h
e
re
sea
r
ch
a
re
a
t
h
at
c
an
be
done
i
n siz
i
ng a
sta
nd-a
l
on
e
P
V
system
wit
h
t
he
a
pp
l
i
c
a
t
ion
of
ANN
.
Tabl
e 1
.
Ap
p
l
i
c
at
ion
o
f
ANN i
n
S
i
z
i
n
g
P
V
s
y
s
t
e
m
P
V
s
iz
ing m
e
thod
A
NN
a
ppl
i
c
a
tion
re
s
e
a
r
c
h
a
r
e
a
Loa
d
a
n
a
l
y
sis
A
pply
A
N
N
t
o
f
o
r
e
c
a
s
t
th
e
t
o
t
a
l
num
b
e
r
of
e
l
e
c
t
r
i
c
a
l
l
oa
d
tha
t
is
u
s
e
d
by
the
building/house
.
T
h
e
input
o
f
t
h
e
AN
N
ca
n
be
v
a
r
ie
d.
F
or
e
x
a
m
p
l
e
,
t
h
e
usua
l
ty
p
e
s
a
n
d
ra
ti
ng
of
t
he
e
l
e
c
t
r
i
c
i
t
y
a
ppl
ia
n
c
e
s
e
tc
.
Ot
he
r
t
h
a
n
t
ha
t,
t
h
e
w
eat
h
e
r
c
ond
i
tion
or
l
oc
a
t
i
on
ca
n
b
e
u
s
e
d a
s
a
p
a
r
t
of
t
he
i
nputs.
De
t
e
r
m
in
a
tion
num
b
e
r
of
P
V
pa
n
e
l
s
.
A
N
N
c
a
n
b
e
a
pplie
d
to
p
r
e
di
c
t
t
h
e
r
equi
r
e
d
num
b
e
r
of
P
V
p
a
ne
ls
.
T
h
e
input
o
f
the
A
N
N
c
a
n
a
lso
be
v
a
r
ie
d.
I
n
put
c
a
n
b
e
t
a
k
e
n
fr
o
m
t
h
e
e
q
u
a
t
i
o
n
u
s
e
d
t
o
c
a
l
c
u
l
at
e t
h
e n
u
m
b
er
o
f
P
V
p
an
el
s
.
D
i
ff
e
r
e
n
t
A
N
N
c
o
n
f
ig
ura
t
ion
ca
n
b
e
use
d a
nd
t
e
ste
d
.
Siz
i
ng
th
e
size
o
f
ba
tt
e
r
y
b
a
nk.
ANN
can
b
e
ap
p
l
i
e
d
t
o
p
r
e
d
i
ct
t
h
e
s
i
z
e.
T
h
e
input
o
f
the
A
N
N
c
a
n
a
lso
be
v
a
r
ie
d.
I
n
put
c
a
n
b
e
t
a
k
e
n
fr
o
m
t
h
e
e
q
u
a
tion
use
d
t
o ca
l
c
ula
t
e
th
e
size
o
f
ba
tt
e
r
y
b
a
nk.
D
i
ff
e
r
e
n
t
A
N
N
c
o
nf
ig
ura
t
ion
ca
n
b
e
use
d a
nd
t
e
ste
d
.
Siz
i
ng
oth
e
r
c
o
m
pon
e
n
t
s
s
uc
h
as
t
h
e
s
i
ze o
f
inve
rte
r
a
n
d
s
ola
r
c
h
a
r
ge
c
ontrolle
r
.
ANN
can
b
e u
s
ed
t
o
p
r
ed
i
c
t
t
h
e s
i
z
e
o
f
in
v
e
r
t
er
a
n
d
s
o
l
a
r
ch
ar
g
e
c
ontrolle
r
.
T
h
e
i
n
p
u
t
of
the
ANN
c
a
n
a
l
s
o be
v
a
r
ie
d.
For e
x
a
m
p
l
e
, the
input
fr
o
m
lo
ad
a
n
al
ys
i
s
can
b
e
u
s
e
d
s
o
th
at
t
h
e
A
N
N
can
l
ear
n
t
h
e
r
e
l
a
t
i
o
n
s
h
i
p
b
etw
een
l
o
ad
a
n
a
l
y
s
i
s
an
d
t
h
e
s
i
z
e
o
f
i
n
v
e
r
t
er
a
n
d
s
o
l
a
r
ch
ar
g
e
c
o
n
t
r
o
l
ler
.
D
i
ff
e
r
e
n
t
A
N
N
c
o
nf
ig
ura
t
ion
ca
n
b
e
use
d a
nd
t
e
ste
d
.
6.
CONCL
U
S
ION
Thi
s
p
ap
er
h
as
p
resent
ed
r
evi
e
w
of
t
h
e
ANN
met
h
od
s
fo
r
si
zi
ng
a
s
t
a
nd-a
l
on
e
pho
to
vo
lt
ai
c
(P
V)
syste
m
.
The
adva
n
t
age
s
o
f
AN
N
espe
cial
l
y
t
he
a
b
ili
ty
t
o
lear
n
t
h
e
i
n
p
u
t
o
u
t
pu
t
r
e
l
a
ti
ons
hi
p
w
i
th
a
m
p
le
t
r
ain
i
ng
m
ak
es
ANN
s
u
i
t
a
bl
e f
o
r
th
e
PV
s
y
s
tem
s
i
zi
n
g
p
r
o
ced
u
r
es
. Th
is pa
p
e
r
a
lso pr
ov
i
d
e i
n
for
m
a
t
i
on base
d
on
t
h
e
r
e
v
i
ew
r
e
g
ardi
ng
t
h
e
po
te
nt
ia
l
o
f
t
he
a
pp
li
c
a
t
i
on
of
A
N
N
i
n
siz
i
n
g
a
s
ta
n
d
-
a
l
one
P
V
syste
m
t
o
be
exp
l
ored
m
ore
in
f
ut
ure
rese
arc
h
.
The
config
ura
t
i
on,
i
np
u
t
a
n
d
out
p
u
t
of
A
N
N
d
efini
t
e
l
y
ca
n
be
i
n
v
e
s
t
i
ga
te
mo
re
i
n
t
h
e
fut
u
re
i
n
o
r
d
e
r
to
i
n
c
rea
s
e
th
e
eff
i
ci
e
n
cy
o
f
t
h
e
A
NN
.
Fro
m
t
h
e
l
it
er
atu
r
e
r
e
v
i
ew,
ANN
h
a
s
succe
ssfu
lly
b
e
a
par
t
o
f
t
h
e
siz
i
ng
p
r
o
ce
d
u
r
e
b
y
pred
ict
i
ng
t
h
e
si
ze
s
of
P
V
p
a
n
e
l
s
,
effi
c
i
en
c
y
o
f
t
h
e
PV
pane
ls,
ene
r
gy
pro
duc
e
d
b
y
the
P
V
system
and t
h
e t
i
l
t
a
n
g
l
e
of
the
PV
p
ane
l
.
Evaluation Warning : The document was created with Spire.PDF for Python.
Int J
P
o
w
E
l
e
c
&
D
ri S
yst
IS
S
N
:
2088-
86
94
Ap
pl
ic
at
ion
o
f
ar
t
i
f
i
ci
a
l
neu
r
a
l
net
w
ork
i
n
siz
i
n
g
a st
a
nd-a
l
o
n
e ph
o
t
ov
o
l
tai
c
... (Ahm
ad
F
a
t
e
h Mo
h
a
m
a
d
Nor)
34
7
ACKNOW
LEDG
E
MEN
T
S
Th
e
a
u
t
h
o
r
s
wo
ul
d
l
i
k
e
to
F
a
c
u
lty
o
f
El
e
c
t
r
i
cal
a
n
d
E
l
ect
ron
i
c
En
g
i
ne
er
i
n
g
a
nd
U
n
ive
r
sit
i
T
u
n
H
u
ssei
n
O
nn
M
a
l
a
ys
i
a
REFE
RENCES
[1]
S.
E
ft
ekharnejad,
V.
V
ittal,
G
.
T
.
H
eydt,
B.
K
e
e
l
and
J.
L
oehr,
"
Im
pact
o
f
in
crea
s
e
d
pen
e
t
r
ati
o
n
of
p
h
o
to
volt
a
ic
generat
i
o
n
o
n
p
o
w
e
r
sy
st
e
m
s,
"
IEE
E
Tr
an
sactio
ns on
Po
wer S
y
st
ems
,
v
o
l
.
28,
no.
2
,
p
p.
8
93
-90
1
,
201
3.
[2]
Y.
S
un,
S
.
L
i
,
B.
L
i
n
,
X.
F
u
,
M
.
Ram
ezani
an
d
I.
J
ait
h
w
a
,
"Art
i
fi
cial
n
eu
ral
net
w
o
r
k
f
o
r
con
t
rol
an
d
gri
d
int
e
gration
of
r
esidential
so
lar
p
hot
ov
oltai
c
s
ystem
s
,
"
IEEE T
r
a
n
s
a
ct
ions o
n
Sus
tain
abl
e En
erg
y
,
v
o
l.
8
,
no
4
,
pp.
1
4
8
4
-
149
5,
20
1
7
.
[3]
M.
K
ari
m
i,
H
.
M
o
k
h
lis,
K.
N
aidu
,
S.
U
ddin
a
nd
A.
H
.
A.
B
akar,
"
Ph
otov
ol
ta
ic
p
e
n
e
t
ra
tion
i
ss
ue
s
a
n
d
imp
a
c
t
s
in
distri
buti
on
network
-
A
revi
ew,
"
Re
ne
wa
ble
an
d S
u
sta
i
na
ble
En
e
r
g
y
Re
v
i
e
w
s
, v
o
l
.
5
3
, p
p.
59
4
-
60
5, 20
1
6
.
[4]
S
.
R
.
M
a
d
e
t
i
a
n
d
S
.
N
.
S
i
n
g
h
,
"
A
c
o
m
p
r
e
h
e
n
s
i
v
e
s
t
u
d
y
o
n
d
i
f
f
e
r
e
nt
t
y
p
es
o
f
f
a
ults
a
nd
d
et
ect
i
on
tech
niq
u
es
f
or
s
olar
pho
to
vo
lt
aic s
y
stem
,
"
S
o
l
ar
E
n
e
rgy
,
vol.
1
5
8
,
p
p
.
16
1-1
85,
2
0
1
7
.
[5]
S
.
S
ob
ri,
S.
K
ooh
i-Kam
a
li
an
d
N
.
A
.
Rahim,
"
So
lar
p
h
o
t
o
voltai
c
gen
e
ratio
n
fo
recast
i
ng
m
e
t
hod
s :
A
revi
ew
,"
Ener
gy Co
nver
sio
n
and
M
a
n
a
g
em
ent
,
v
o
l
.
156,
pp.
4
5
9
-49
7
,
20
1
8
.
[6]
C.
G
henai
,
A
.
M
e
rabet,
T
.
S
a
lameh
an
d
E.
C
.
Pi
gem
,
"
Grid
-tied
a
nd
s
t
a
nd
-alo
ne
h
yb
rid
s
o
lar
po
wer
sy
stem
f
or
desal
i
n
a
ti
on
p
l
an
t,
"
D
e
sali
na
ti
o
n
, v
ol
.
4
3
5
, no
.
A
u
g
u
s
t
20
17
,
p
p
.
1
7
2
-
18
0, 2
01
8.
[7]
E
.
M
.
H
.
A
r
i
f
,
J
.
H
o
s
s
e
n
,
G
.
R
.
M
u
r
t
h
y
,
M
.
Z
.
H
.
J
e
s
m
e
e
n
a
n
d
J
.
E
.
Raja,
"An
ef
fi
cien
t
microco
n
t
r
oller
bas
e
d
su
n
tracker
c
o
n
tro
l
f
o
r
s
ol
ar
c
e
l
l
s
y
stem
s
,
"
Int
e
rn
atio
nal
Jou
r
n
a
l of E
l
ectrica
l an
d
Co
mpu
t
er
Engin
eeri
n
g
(
I
JE
CE)
,
vol
.
9
,
n
o.
4
,
p
p
.
27
43
-2
75
0,
2019.
[8]
L.
D
u,
L
.
Zh
ang
and
Xi.
T
i
an
,
"Deep
pow
er
f
orecas
ti
ng
m
ode
l
f
o
r
b
u
i
l
d
in
g
atta
ch
ed
pho
to
vo
lta
i
c
s
yst
e
m
,
"
IEE
E
Access
, vo
l
.
6
, p
p. 52
6
3
9
-
52
65
1
, 2
01
8.
[9]
M
.
A
.
M.
R
a
m
li
,
A
.
H
ie
nd
ro,
K.
S
e
d
r
a
ou
i
a
n
d
S.
T
wa
h
a
,
"
O
pt
ima
l
siz
i
ng
o
f
gri
d
-con
nect
ed
p
ho
tov
o
ltai
c
e
nerg
y
sys
t
e
m
i
n Saudi
Arabia,"
Re
n
e
w
a
bl
e
En
e
r
g
y
,
vo
l.
7
5
,
p
p
. 4
89
-49
5
, 2
01
5.
[10]
M
.
Y
a
i
c
h
i
,
M
.
-
K
.
F
e
l
l
a
h
a
n
d
A
.
T
a
y
e
b
i
,
"
A
f
a
s
t
a
n
d
s
i
m
p
l
i
f
i
e
d
m
eth
o
d
usi
ng
no
n-l
i
n
e
ar
t
ran
s
latio
n
o
f
operatin
g
poi
nts
f
o
r
P
V
m
o
d
u
l
es
e
n
e
rgy
o
u
t
p
u
t
a
nd
d
ail
y
p
u
m
ped
w
a
ter
to
p
redi
ct
t
he
p
erf
o
rm
an
ce
o
f
a
stand
-
alo
n
e
pho
to
vo
lt
aic p
u
m
p
i
ng syst
em at d
i
ff
eren
t h
eads,
"
R
e
newa
ble Ener
gy
,
vol.
1
3
3
,
p
p
.
248
-26
0
,
201
9.
[11]
R.
L
una-Ru
b
i
o
,
M
.
T
rej
o
-P
erea,
D
.
Vargas
-Va´
z
q
uez,
a
nd
G.
J
.
Rı
os-Mo
r
e
n
o
,
“
Opt
i
ma
l
siz
i
ng
o
f
re
ne
wa
ble
hy
brids
energy
sy
s
t
e
ms : A
rev
i
e
w of
m
eth
o
d
o
l
ogi
es
,
”
S
o
l
ar
E
n
e
rg
y,
vol.
86,
p
p
.
1
077
–1
088,
2
01
2.
[12]
W.
Z
han
g
,
A.
M
al
eki,
M
.
A.
R
os
en
a
nd
J
.
L
i
u
,
"
S
i
zi
ng
a
s
t
a
nd-al
on
e
s
o
lar-w
in
d-hy
dro
g
en
e
nerg
y
sy
st
e
m
u
s
i
n
g
weath
e
r
f
o
recas
tin
g
and
a
hy
bri
d
s
earch
o
pt
imizatio
n
algo
rit
h
m,
"
En
e
r
gy
Con
v
e
r
sion
an
d Man
ag
e
m
e
n
t
,
v
o
l.
1
80
,
pp.
6
0
9
-6
21,
2019
.
[13]
G.
J
im
én
ez-Ca
s
t
i
l
l
o
,
F
.
J
.
M
u
ñ
o
z-Ro
dríg
uez,
C
.
Ru
s-Casas
,
J
.
C.
H
ernán
d
ez
a
n
d
G
.
M
.
Ti
na,
"
M
on
it
ori
n
g
PWM
sign
a
l
s
in
sta
nd
-
a
lo
ne
p
ho
tov
o
l
ta
ic
sy
s
te
m
s
,
"
M
e
as
ur
ement
,
v
o
l
. 1
34,
p
p
.
4
12-4
2
5
,
2
019
.
[14]
M.
S
ulaim
a
n
,
A
.
F
.
M
.
Nor
and
R.
O
m
a
r,
"
A
GUI
b
ased
t
eachin
g
an
d
le
a
r
ning
s
o
f
twa
r
e
for
sy
ste
m
s
iz
in
g
of
a
s
ta
nd
alon
e
hy
bri
d
s
olar
e
lect
ricit
y
system
,
"
MAGNT
R
e
se
arc
h
Re
po
rt
,
v
o
l
.
3
,
no
.
6
,
p
p.
72-8
5
,
20
15.
[15]
T
.
Kh
at
ib
,
I.
A
.
Ib
rahim
and
A.
M
o
h
ame
d
,
"A
r
eview
o
n
s
i
z
i
n
g
m
eth
odo
lo
g
i
es
o
f
ph
oto
volt
a
ic
a
rray
and
sto
r
ag
e
batt
ery i
n
a
s
t
a
n
d
al
on
e
ph
ot
ov
o
ltaic
s
ystem
,
"
En
e
r
gy
Co
nv
e
r
sio
n
an
d Ma
na
ge
me
n
t
,
v
o
l.
1
20,
pp.
430
-448
,
2
01
6.
[16]
S
.
S
em
ao
ui,
A
.
H
.
Arab
,
S
.
B
ach
a
a
n
d
B.
A
zo
ui
,
"Op
tim
al
s
i
z
ing
of
a
s
tan
d
-al
one
p
hot
ov
ol
taic
s
yst
e
m
w
i
t
h
e
nerg
y
man
a
gem
e
nt
i
n
iso
l
ated
areas,
"
Ener
gy P
r
o
c
ed
ia
, v
ol
.
36
,
p
p
.
358
-36
8
,
201
3.
[17]
A.
M
ellit
and
M.
B
eng
h
anem
,
"S
iz
i
n
g
o
f
s
tand
-alo
ne
p
hot
ov
oltaic
s
ys
te
m
s
u
si
ng
neu
r
al
n
et
wo
rk
a
dap
t
i
v
e
m
o
d
e
l,
"
Desali
nati
on
,
vol.
20
9,
p
p.
6
4
-
72
,
20
07
.
[18]
A.
M
elli
t,
S
.
A
.
K
alogirou,
L
.
Hont
or
i
a
a
nd
S
.
Shaari,
"
Ar
tific
i
a
l
in
t
e
ll
ig
ence
techn
i
q
u
es
f
or
s
izin
g
ph
oto
volt
a
ic
s
y
ste
m
s
:
A r
e
v
i
e
w
,"
Ren
e
wab
l
e a
nd
Su
sta
i
n
able E
n
er
gy
R
e
v
i
ews
, v
ol
. 13
,
p
p
. 40
6
-41
9
, 20
0
9
.
[19]
A.
M
ellit,
"
ANN-ba
s
e
d
G
A
f
or
g
ener
at
ing
t
h
e
s
i
zing
c
urve
o
f
st
a
nd-al
on
e
ph
oto
v
o
l
t
a
ic
s
y
s
t
e
m
s
,
"
A
d
van
ces i
n
En
gine
e
r
in
g
S
o
f
t
w
are
,
vol.
4
1
,
pp.
6
8
7
-6
93
,
2
010
.
[20]
A
.
M
e
l
l
i
t
,
S
.
A
.
K
a
l
o
g
i
r
o
u
a
n
d
M
.
D
r
i
f
,
"
A
p
p
l
i
c
a
t
i
o
n
o
f
n
e
u
r
a
l
networks
a
nd
g
enet
ic
a
l
gorithms
f
or
s
izi
n
g
o
f
pho
to
vo
lt
aic s
y
stem
s,"
R
e
newab
l
e Ener
gy
,
vo
l
.
35,
n
o
.
1
2
,
p
p.
2881-2
8
9
3
,
2
01
0.
[21]
M.
G
a
l
ad
a
nd
P
.
S
p
ani
k
,
"Des
ign
o
f
p
hot
ov
oltaic
s
o
l
ar
cell
m
o
d
e
l
fo
r
s
t
an
d
al
on
e
ren
e
wabl
e
s
y
st
em
,"
i
n
EL
E
K
T
R
O
,
pp.
2
8
5
-2
88,
2014
.
[22]
A.
E
l-S
h
a
h
at,
R.
J
.
Had
d
ad,
B.
G
u
h
a
an
d
Y
.
K
alaani
,
"
S
i
zin
g
r
es
id
ent
i
a
l
photo
v
o
l
t
a
ic
s
yst
e
ms
i
n
t
h
e
s
t
at
e
o
f
Georgia,"
IEEE Int
e
r
n
a
t
i
o
n
a
l
Co
n
f
eren
ce on
Sm
a
r
t G
r
i
d
Commu
n
i
catio
ns (
S
m
a
r
t
Gr
id
Co
mm)
,
p
p
.
629
-63
4
,
2
01
5
.
[23]
J.
P
.
D
u
n
l
op,
"
P
hot
ov
oltaic
S
ystem
s
,
"
2
n
d
ed.
Illi
no
is
:
Am
eri
c
an T
e
c
h
ni
cal
Pu
blishers
,
In
c.
,
2
010.
[24]
M
.
A
li,
A
.
Yousaf
a
nd
F
.
G.
S
eh
aran,
"
F
easibi
lity
e
valuat
ion
o
f
s
ta
nd
-a
lo
n
e
p
ho
to
vo
l
t
a
i
c
sys
t
e
m
s
for
re
sid
e
nt
ia
l
loads,"
T
h
e 9
t
h In
ter
n
a
t
io
nal Renewa
ble
Ener
gy
Co
ng
res
s
(
I
R
E
C 2
018
)
, 20
1
8
, pp
.
1
-4.
[25]
H.
A
.
K
a
zem
,
T.
K
hatib
a
nd
K.
S
o
p
i
a
n,
"
Sizi
ng
o
f
a
stan
dal
o
n
e
p
hotovol
t
aic/ba
ttery
s
ystem
at
m
inimu
m
c
os
t
f
o
r
remo
te
h
ou
si
ng
electrif
i
catio
n i
n
S
o
h
ar,
O
m
a
n,
"
En
e
r
gy
an
d
Bu
il
ding
s
,
vol.
6
1
,
pp.
1
08
-11
5
,
2
013.
[26]
X.
H
an
,
X.
A
i
an
d
Y.
S
u
n
,
"Research
o
n
l
a
rge-s
cal
e
d
i
s
p
at
chabl
e
g
rid
-
c
o
nn
ec
te
d
PV
s
yste
ms,"
J
o
urn
a
l of
Mod
e
r
n
Power
System
s and
Clea
n
E
n
erg
y
, vo
l
.
2
, n
o.
1,
pp
. 69
-
76
,
2
0
1
4
.
[27]
S
.
S
.
Mo
hamm
e
d
,
D.
D
ev
araj
a
n
d
T.
P
.
I.
A
h
a
med,
"
M
a
xi
mu
m
pow
er
po
int
Tracki
n
g
system
f
or
s
tan
d
a
lo
ne
s
o
l
ar
P
V
p
o
w
er
s
yst
e
m
using
adap
ti
v
e
n
eu
ro-f
uzzy
i
n
f
e
r
en
ce
sy
stem
,"
B
i
ennia
l
Internat
io
nal Confer
ence on
Po
wer
an
d
En
e
r
g
y
Sy
s
t
e
m
s:
To
w
a
rd
s S
u
s
t
a
inable Energy
(
P
ESTS
E)
, p
p.
1-4
,
20
16
.
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 V
ol.
11,
N
o.
1
, Ma
r
202
0
:
342
–
34
9
34
8
[28]
V.
S
harm
a,
S
.
Rai
an
d
A.
D
ev
,
"
A
c
o
m
p
r
eh
en
siv
e
s
tu
dy
o
f
art
i
fic
i
a
l
neu
r
al
n
etw
o
rk
s,"
Inter
n
a
t
i
onal Jou
r
n
a
l
o
f
Adva
nced
Res
e
arch
in Com
puter Sci
e
nce a
n
d
So
ft
wa
re Eng
i
n
eer
ing
,
vol.
2,
n
o.
10,
p
p.
2
78
-28
4
,
201
2.
[29]
R.
P
agari
y
a
an
d
M
.
B
art
e
re,
"Review
pap
e
r
o
n
artifici
a
l
n
e
ural
netw
ork
s
,
"
In
tern
atio
nal Jo
urn
a
l
o
f
A
d
vanced
Research in Co
mputer
Sc
ience
, vo
l
.
4
, n
o.
6,
p
p
.
4
9
-
54
, 20
1
3
.
[30]
S
.
A
s
s
a
h
o
u
t
,
H
.
E
l
a
i
s
s
a
o
u
i
,
A
.
E
l
O
u
g
l
i
,
B
.
T
i
d
h
a
f
a
n
d
H
.
Z
r
o
u
r
i,
"
A
n
e
ural
n
etw
o
rk
a
n
d
f
uzzy
l
og
ic
b
as
ed
M
P
P
T
algo
rit
h
m
f
o
r
p
h
o
t
o
v
o
l
t
a
ic
pum
pi
ng
s
y
s
tem,
"
In
ter
natio
nal
Jo
ur
nal
of Po
wer E
l
ectr
o
n
i
cs
an
d D
r
ive S
y
st
em
(IJPED
S
)
,
v
o
l
.
9
,
n
o.
4
,
p
p
.
18
23
-1
83
3,
201
8.
[31]
S
ugi
arti
,
Y
u
h
a
ndri
,
J
.
N
a
’am
,
D
.
In
dra
an
d
J.
S
anto
ny
,
"An
art
i
fi
c
i
al
n
eu
ra
l
n
e
tw
ork
app
r
oach
f
or
d
e
t
ect
ing
s
k
in
canc
e
r,"
T
E
L
K
OM
NIKA
(
T
e
l
ecom
m
u
n
icati
on
Comp
utin
g El
ectron
i
cs
a
n
d
Co
n
t
ro
l)
,
vo
l.
1
7
,
n
o.
2
,
pp
.
78
8-7
9
3
,
201
9.
[32]
D.
A
hm
ed
,
B.
M
ok
ht
a
r
a
nd
B
.
Aek,
"
DTC
hybrid
b
y
d
i
ff
eren
t
techn
iq
ues
o
f
o
b
s
ervat
i
o
n
w
it
h
artif
i
c
i
al
n
eu
ronal
netw
ork
(AN
N
)
fo
r
i
n
d
u
cti
o
n
m
achi
n
e
driv
es,"
In
ter
n
a
tio
nal Jour
na
l of Power
E
l
ect
ro
n
i
cs
a
n
d
Dr
ive Syst
em
(IJPED
S
)
,
vo
l
.
1
0
,
n
o
. 2
, p
p.
69
7
-70
8
, 20
1
9
.
[33]
Z.-J.
L
i
m,
M
.
W.
M
us
t
a
f
a
a
nd
J.
J
.
Jami
an,
"
V
oltage
s
tabi
lit
y
p
redi
ctio
n
on
p
o
w
er
s
ys
te
m
n
e
two
r
k
via
enhan
ced
hybrid
part
ic
l
e
s
w
a
rm
a
rt
if
ici
a
l
neural
n
et
wor
k
,"
Jou
r
n
a
l
of
Electrica
l E
ngin
e
erin
g
&
T
echn
o
lo
gy
,
vo
l.
1
0
(
3),
pp.
8
7
7
-8
87,
2015
.
[34]
H
.
H
.
G
o
h
,
Q
.
S
.
C
h
u
a
,
S
.
W
.
L
e
e
,
B
.
C
.
K
o
k
,
K
.
C
.
G
o
h
a
n
d
K
.
T
.
K
.
T
eo,
"Ev
a
l
u
at
io
n
fo
r
v
o
l
t
age
st
abi
l
i
t
y
in
di
ces
in
p
o
w
er s
ys
t
e
m us
in
g
art
i
fi
cia
l
n
eu
ra
l
n
e
two
r
k,"
P
r
o
c
ed
ia
En
g
i
neer
ing
,
vo
l. 1
18
,
p
p
. 11
2
7
-1
13
6,
2
01
5.
[35]
D.
Q
.
Zh
ou,
U
.
D.
A
nn
a
k
kag
e
a
nd
A
.
D
.
R
ajap
aks
e
,
"
O
nl
in
e
m
o
n
i
to
ring
o
f
v
o
l
t
ag
e
stab
ilit
y
m
a
rgi
n
u
s
i
ng
a
n
artif
i
c
i
al
neu
ral n
e
two
r
k,
"
IEE
E
Tran
sa
c
t
io
ns
on
Po
we
r S
y
ste
m
s
, v
ol
.
2
5
,
n
o
.
3
,
pp.
1
5
6
6
-
15
74,
2010
.
[36]
L
.
H
o
n
t
o
r
i
a
,
J
.
Ag
u
i
l
e
r
a
a
n
d
P
.
Z
u
f
i
r
i
a
,
"
A
n
e
w
ap
p
r
o
a
c
h
f
o
r
si
zing
s
t
an
d
a
l
on
e
phot
ov
oltaic
s
y
s
t
e
ms
b
ased
i
n
neu
r
al
netw
ork
s
,
"
So
la
r
En
e
r
g
y
,
v
o
l
. 78
, p
p. 3
13
-31
9
, 20
0
5.
[37]
D.
B
apt
i
st
a,
S
.
A
b
reu
,
C
.
T
r
av
ies
o
-g
onzál
ez
a
n
d
F
.
M
o
rgado
-
dias
,
"Hardw
are
i
m
ple
m
en
tation
o
f
a
n
art
i
fi
cia
l
n
eu
ral
netw
ork
m
o
de
l
t
o
p
red
i
ct
t
he
e
nerg
y
pro
d
u
c
tion
o
f
a
p
h
o
t
ovo
lt
ai
c
sy
s
t
em,"
Mic
r
op
ro
c
e
s
so
rs an
d
Mic
r
osy
s
te
ms
,
vol
.
4
9
,
pp
.
77
-8
6
,
2
017
.
[38]
L.
M
.
El
ob
aid,
A
.
K
.
A
bd
e
l
s
a
la
m
an
d E.
E.
Z
a
kzou
k,
"
A
r
tificial
n
e
ural
n
etw
o
rk
-based
p
hot
ov
oltaic
m
a
x
im
um
power
poi
nt t
rack
in
g
t
e
chn
i
qu
es:
A
su
rv
ey,
"
IE
T
Ren
e
wabl
e Po
wer G
e
ner
a
ti
on
,
v
o
l.
9
,
no.
8
,
pp
.
1
0
4
3
-1063
,
2
01
5.
[39]
I.
A
.
Be
lo
va
a
nd
M
.
V.
M
artin
ovich
,
"
N
eural
n
e
two
r
k
con
t
ro
l
alg
ori
t
hm
f
o
r
s
tan
d
-alo
ne
s
o
l
ar
cell
electri
cal
e
nerg
y
conv
ersi
on
s
ystem
,
"
1
6
th Int
e
rnat
ion
a
l
Confer
e
n
ce of
You
n
g
Sp
ecial
is
ts
on
Mi
cro/
Nan
o
t
echno
logi
es a
nd Elect
ron
Devices
,
p
p.
3
87-39
0,
2
01
5.
[40]
A.
M
el
lit,
M
.
Benghanem,
A
.
H.
A
rab
and
A.
G
uessou
m
,
"An
ad
apti
ve
a
rtif
ic
ial
neural
n
etwork
m
o
d
el
for
siz
i
ng
s
ta
n
d
-a
lon
e
p
ho
to
vo
lta
i
c
sy
ste
m
s:
A
pp
lic
a
t
io
n
fo
r
is
ol
a
t
ed
s
i
t
es
i
n
Alg
e
ria,"
Ren
e
wa
bl
e Ener
gy
,
vo
l.
3
0,
pp.
1
5
0
1
-
152
4,
20
0
5
.
[41]
M.
B
renn
a,
F
.
Foi
a
del
l
i
,
M
.
Lo
n
g
o
an
d
D.
Z
ani
n
el
li
,
"S
o
l
ar
r
ad
i
ati
on
an
d
l
o
ad
p
o
w
er
c
on
sum
p
ti
o
n
f
orecas
ti
ng
usin
g
neural
n
et
wo
rk,
"
6
t
h
Inter
nati
o
n
a
l
Conferen
ce on
Clea
n Elect
r
i
c
a
l
Power
(
I
CCEP)
,
p
p
.
7
26
-731
,
201
7.
[42]
I.
C
eyl
a
n,
E
.
Ged
i
k
,
O
.
E
r
k
a
ym
a
z
a
nd
A.
E
.
Gu
re
l,
"
T
h
e
artif
i
c
i
al
n
eu
ral
n
e
tw
ork
m
o
d
e
l
t
o
e
stim
ate
the
ph
ot
ovo
lt
aic
mo
du
l effi
cien
cy f
or
a
ll
reg
i
ons
of th
e Tu
rkey,
"
Ener
gy a
n
d
Buil
d
i
ngs
, vo
l
. 84
, pp
.
2
58
-2
67
,
2
0
1
4
.
[43]
P
.
A
run,
"
S
i
zi
ng
cu
rve
f
o
r
isolat
ed
p
h
o
t
ovo
lt
aic-b
a
t
t
ery
s
y
stem
s
u
s
in
g
artifici
al
n
eural
n
e
tw
orks
,"
2nd Internation
a
l
Conf
eren
ce on
Gr
een E
n
erg
y
a
n
d
Appli
c
a
t
ions (
I
CGEA
)
, p
p
. 14
7
-1
51
, 20
1
8
.
[44]
A.
C
hatt
erjee
a
n
d
A.
K
ey
hani
,
"Neu
ral
netw
ork
es
tim
a
tio
n
o
f
m
ic
rogri
d
m
axim
um
s
olar
p
o
w
er,
"
IEEE Tran
sac
t
io
ns
on Sma
r
t
Grid
,
vol
.
3
,
n
o
.
4
,
p
p
.
186
0-1
8
6
6
,
2
01
2.
[45]
A.
F
.
M.
N
or
a
nd
M
.
Sul
a
iman,
"V
olt
a
ge
i
nstability
a
nalys
i
s
bas
e
d
o
n
mo
dal
an
a
l
ys
is
t
echni
qu
e
a
n
d
art
i
f
i
cia
l
n
eu
ral
netw
ork
,
"
Ind
o
n
e
s
i
a
n
Jo
ur
nal of
E
l
ect
rica
l Eng
i
neeri
ng
and
Compu
t
er
Sci
e
nce
, v
o
l
.
13
,
n
o
. 3
, p
p.
1
2
7
4
-1
27
9, 2
01
9.
[46]
D.
H
unt
e
r,
H
.
Y
u
,
M.
S
.
Puk
i
sh,
J.
K
olbus
z
a
nd
B
.
M.
W
i
l
a
m
owski
,
"S
electi
o
n
of
p
rop
e
r
neural
n
etw
o
rk
s
izes
a
n
d
a
r
c
h
ite
c
t
ure
s
- A c
o
m
pa
ra
tive
s
tu
dy
,"
IE
EE T
r
ansa
c
ti
on
s o
n
In
dust
r
i
a
l
In
fo
rm
a
tics
,
vol.
8
(2
),
p
p
.
228
-240
,
2
01
2.
[47]
K.
G
.
S
h
eela
and
S
.
N
.
Deep
a,
"
Revi
ew
o
n
m
e
th
ods
t
o
fix
n
u
m
b
er
of
h
id
d
e
n
n
e
uron
s
in
n
eural
net
w
ork
s
,
"
M
a
t
h
em
atical Prob
lem
s
i
n
E
ngineerin
g
,
vol.
2013,
p
p
.
1
-1
1,
2
0
1
3
.
[48]
F
.
A
n
i
f
o
wose,
J.
L
abadi
n
a
nd
A
.
A
b
d
u
l
r
a
h
eem
,
"E
nsem
ble
m
o
del
o
f
a
rt
ifi
c
ia
l
neu
r
al
n
et
work
s
with
r
and
o
mi
zed
num
b
e
r
o
f
h
id
den
neu
r
on
s,
"
8t
h
Inter
n
a
t
i
o
n
a
l
Co
nf
eren
ce on
Inf
o
r
m
atio
n
T
ech
no
lo
gy i
n
A
s
i
a
(
C
ITA)
, p
p.
1-5
,
20
13
.
[49]
S
.
K
arsoliy
a,
"
A
p
p
r
oxim
a
ti
ng
n
um
ber
o
f
h
i
dden
layer
neu
r
on
s
in
mult
i
p
l
e
h
idden
layer
BP
NN
a
rchi
t
ecture,
"
Inter
n
a
t
i
onal Jo
ur
na
l
o
f
Engin
eeri
n
g
T
r
end
s
and
T
echn
o
l
o
g
y
, v
ol
. 3
,
n
o
.
6
,
p
p
. 7
14
-71
7
, 2
01
2.
[50]
A
.
F
.
M
.
N
o
r
,
M
.
S
u
l
a
i
m
a
n
,
A
.
F
.
A
.
K
a
d
i
r
a
n
d
R
.
O
m
a
r
,
"
V
o
l
t
a
g
e
s
tab
ilit
y
a
naly
s
i
s
o
f
l
o
a
d
bus
es
i
n
el
ectric
po
wer
sys
t
e
m
u
si
ng
a
dapt
iv
e
neu
r
o-f
u
zzy
i
n
f
erence
s
y
st
em
(
ANFIS
)
an
d
p
r
ob
abil
isti
c
neu
r
al
n
e
t
wo
rk
(
P
N
N),"
ARPN
Jour
na
l of Engineeri
ng
and
A
pplied
Sci
e
nces
, v
ol
. 1
2, n
o
. 5
, pp
.
1
4
0
6
-
14
10
, 20
1
7
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t
J
P
o
w
Elec
&
D
r
i
S
y
st
I
S
S
N
:
2088-
86
94
Ap
plic
at
i
on o
f
ar
ti
fic
ia
l ne
u
r
a
l
net
w
ork
i
n
sizin
g
a st
and-
a
l
o
n
e
ph
o
t
ov
o
l
t
a
i
c
..
.
(
A
hm
ad F
a
te
h
Mo
h
a
m
a
d
Nor)
34
9
BIOGRAPHI
E
S
OF
AUT
HORS
Ahm
a
d
Fa
te
h
M
o
ha
ma
d
No
r
re
ce
iv
e
d
B
a
c
he
lor
o
f
E
le
c
t
r
i
c
a
l
E
n
g
i
ne
e
rin
g
,
M
a
ster
o
f
El
ectri
cal
En
gin
eerin
g
an
d
P
h
.
D
.
i
n
E
lect
rical
E
ng
in
ee
ri
ng
(
P
o
w
e
r)
i
n
2
0
1
1
,
2
0
1
3
a
n
d
2
017,
r
esp
ecti
v
el
y
f
r
o
m
U
ni
versit
i
Tekn
ik
al
M
a
l
ays
i
a
Mel
a
ka
(
UTeM
).
C
urren
t
l
y
h
e
i
s
a
l
ecturer
a
t
th
e
D
e
p
a
rtm
e
n
t
of
E
l
ectri
cal
P
o
w
er
E
ng
ineeri
ng,
F
acult
y
of
E
lect
rical
a
n
d
E
l
e
c
tronic
E
ng
i
n
eering,
U
ni
versi
t
i
Tu
n
Hu
sse
i
n
O
n
n
M
alays
i
a.
H
is
a
rea
o
f
r
es
ear
ch
i
nt
erests
i
n
c
lud
e
s
e
lect
rical
p
o
w
er
e
ng
in
ee
ri
ng
,
vo
lt
a
g
e
i
n
stabili
t
y
an
alys
is,
s
o
l
a
r
elect
ricity,
artifici
al
n
eu
ral
n
e
tw
o
r
k
(ANN
)
an
d
ad
apt
i
v
e
n
eu
ro-
f
u
zzy
i
n
f
erence
s
y
s
t
em
(
ANF
IS
).
H
e
is
r
eg
istered
a
s
G
raduat
e
E
n
gi
neer
w
it
h
the
Board
of
E
n
g
i
n
e
e
r
s
M
a
l
a
y
s
i
a
(
B
E
M
)
a
s
w
e
l
l
a
s
G
r
a
d
u
a
t
e
T
e
c
h
n
o
l
o
g
i
s
t
s
w
i
t
h
t
he
M
a
l
ays
i
a
Bo
ard
o
f
Tech
nol
og
ists
(
M
B
OT).
S
u
ri
a
n
a
S
a
limi
n
i
s
a
l
ect
urer
i
n
F
acult
y
of
E
lect
rical
E
ngin
eer
i
n
g
(
F
KE
E
)
,
U
T
HM
.
A
f
t
e
r
grad
uated
fro
m
her
f
i
rst
degree
in
E
lect
rical
E
ngin
eeri
n
g
(U
TM
-
20
06
),
s
he
w
o
r
k
e
d
a
s
a
t
uto
r
i
n
UT
HM
f
o
r
6
m
on
th
s
an
d
continued
f
o
r
M
S
c
in
P
o
w
er
D
is
tri
b
u
t
i
on
E
n
g
i
n
eerin
g
in
N
ew
castle
Un
iversit
y
(
2007).
S
h
e
com
p
l
e
ted
h
e
r
P
h
D
in
E
lectrical
E
ngin
eer
in
g
i
n
2014
a
ls
o
f
r
o
m
New
c
a
s
tl
e
U
n
i
v
ers
ity,
UK
.
S
h
e
h
a
s
pu
blis
hed
a
f
e
w
papers
i
n
Jo
u
rn
als
an
d
pro
c
e
e
d
i
n
g
s
i
n
ce
20
13
.
Her
resear
ch
i
nt
erests
a
re
on
p
o
w
e
r
q
u
al
ity
i
m
p
rov
e
men
t
,
d
i
stri
buted
g
ener
ati
o
n
sy
stems
,
renew
a
b
l
e
en
erg
y
and
harm
on
ics
miti
gatio
n.
S
h
e
i
s
a
m
e
m
b
er
o
f
t
he
I
n
s
tit
u
t
e
o
f
E
l
ectri
cal
a
n
d
El
ectronics
En
g
ineers
(IEEE
).
M
o
h
d
N
o
o
r
A
b
d
u
l
l
ah
r
e
cei
ved
hi
s
B.E
ng.
(
Hons
)
in
E
l
ectrical
E
n
g
i
n
eerin
g
and
M
.
E
ng.
i
n
El
ectrical
E
ng
ineering
(P
o
w
er
S
ys
tem
)
fro
m
U
n
ivers
i
ti
T
ekn
o
l
ogi
M
a
l
ays
i
a
(UTM
)
i
n
2
0
08
and
20
10
r
es
pect
ively.
H
e
also
r
ecei
ved
a
P
h
.D
d
eg
ree
in
E
lectri
cal
E
n
g
i
n
eeri
ng
f
r
om
U
n
i
versity
o
f
M
a
laya
(
U
M
)
in
201
4.
H
e
has
b
een
w
it
h
U
n
i
v
ersiti
T
u
n
H
uss
e
in
O
n
n
Malaysi
a
(
UTHM)
f
r
o
m
2
0
0
8
t
o
2
0
1
4
a
s
a
t
u
t
o
r
.
H
e
i
s
c
u
r
r
e
n
t
l
y
w
o
r
k
s
a
s
a
l
e
c
t
u
r
e
r
i
n
D
ep
artm
ent
of
E
lectri
cal
P
o
w
er
En
gin
eerin
g
,
F
a
c
u
l
t
y
o
f
El
ectric
a
l
and
El
ectron
i
c
E
n
g
i
n
e
e
r
i
n
g
(
FKEE),
U
n
i
v
ersiti
T
un
H
u
s
sei
n
On
n
M
a
l
a
ysi
a
(
UT
H
M
).
H
e
als
o
a
ppo
inted
as
a
h
ead
o
f
G
r
een
a
n
d
S
ustai
n
able
E
nergy
(
G
S
E
n
e
r
g
y
)
F
o
c
u
s
G
r
o
u
p
i
n
F
K
E
E
,
U
T
H
M
.
H
e
i
s
r
e
g
i
s
t
e
r
e
d
a
s
a
m
e
m
ber
o
f
B
oa
rd
o
f
En
g
i
neer
M
a
l
a
y
s
i
a
(
B
E
M
)
a
n
d
I
n
s
t
i
t
u
t
e
o
f
E
l
e
c
t
r
i
c
a
l
a
n
d
E
l
e
c
t
r
o
n
i
c
s
E
n
g
i
neers
(IEEE)
.
His
res
earch
in
terests
are
elec
t
r
ic
p
ow
er
d
i
s
p
a
tch
,
d
istrib
u
t
ed
g
enerat
io
n,
renewab
l
e
energy
a
nd
m
et
a
-
h
e
uri
s
t
i
c
op
timizati
on
tech
ni
ques
.
M
u
h
a
mm
ad
N
afis
I
sm
ai
l
receiv
e
d
h
i
s
B.E
n
g
.
i
n
El
ectrical
E
n
g
i
n
ee
ring
f
rom
U
n
i
v
ersit
i
Tek
nol
ogi
M
alay
sia
(UTM
)
i
n
2
0
0
5
.
H
e
is
a
n
i
n
struct
or
i
n
Fa
cu
lt
y
o
f
E
l
ectri
ca
l
Eng
i
n
eering
(
F
K
E
E
)
,
U
n
i
v
e
r
s
i
t
i
T
u
n
H
u
s
s
e
i
n
O
n
n
M
a
l
a
y
s
i
a
(
U
T
H
M
)
.
H
i
s
a
r
e
a
o
f
r
esear
ch
i
n
t
erests
i
s
elect
rical
p
ower
g
enerati
o
n
.
H
e
is
r
egi
s
t
e
red
as
G
rad
u
at
e
E
n
g
i
n
e
e
r
w
i
t
h
t
h
e
B
o
a
r
d
o
f
E
n
g
i
n
e
e
r
s
Malaysia (B
E
M)
.
Evaluation Warning : The document was created with Spire.PDF for Python.