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.
3, S
ep 2019,
pp.
1
5
7
5
~1
5
9
1
ISSN: 2088-
8694,
DOI
:
10.11591
/ijpeds.
v10.
i
3.pp1575-1591
1575
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
Modeling of phot
ovoltaic syste
m with maximum pow
er point
tracking control by ne
ural netw
o
r
k
s
Far
i
d
S
aad
ao
u
i
1
,
K
h
aled
. Mammar
2
,
A
b
da
l
d
j
a
ba
r
.
H
a
z
z
a
b
3
1,
3
Res
earch
Lab
o
r
at
ory
Co
nt
ro
l
A
n
al
ysis and
O
pti
m
ization of Sy
s
t
e
m
s El
ec
t
r
o-E
n
erget
i
c (CAOSE
E
),
U
n
iv
e
r
s
i
ty
o
f
T
a
hr
i
M
o
h
a
me
d
,
A
lg
e
r
i
a
2
D
e
partm
e
n
t
o
f El
ectri
cal E
ngi
n
eer
i
n
g
,
U
niv
e
rsity
of T
ahri
M
oham
ed,
A
l
geri
a
Art
i
cl
e In
fo
ABSTRACT
A
r
tic
le hist
o
r
y
:
R
e
c
e
i
v
e
d
Dec
2
7
,
2
018
Re
vise
d Mar
1,
201
9
A
c
c
e
pte
d
A
pr 3,
201
9
Th
is
p
aper
p
resen
t
ed
t
he
s
t
udy,
d
ev
elop
m
e
nt
a
n
d
i
m
p
l
e
m
e
nt
a
tio
n
of
t
h
e
ma
x
i
mu
m
po
w
e
r
po
int
o
f
a
p
ho
tov
o
ltai
c
e
nerg
y
gen
e
r
a
tor
adap
ted
by
elev
ato
r
c
onv
erter
an
d
co
ntro
ll
e
d
b
y
a
m
a
x
i
m
u
m
p
o
w
e
r
po
int
co
mm
an
d.
I
n
ord
e
r
t
o
i
m
p
rove
pho
to
vo
lt
a
i
c
s
y
s
t
e
m
p
erf
o
rm
ance
and
to
f
o
r
ce
t
he
ph
ot
ov
olt
a
ic generat
or
t
o operat
e
a
t it
s
m
a
x
i
m
u
m
po
wer
p
o
i
n
t,
t
he
i
d
e
a
of
t
he
c
o
nte
x
t
of
t
hi
s
pa
p
e
r
d
e
a
l
s
wit
h
t
he
e
x
p
loita
ti
on
o
f
th
e
te
c
h
ni
qu
e
o
f
t
he
artif
ic
i
a
l
i
n
t
e
lligence
m
e
ch
anis
m
(n
eural
net
w
o
r
k)
cert
a
in
ly
b
a
s
e
d
on
t
h
e
th
ree
part
s
o
f
t
h
e
p
hot
ov
ol
taic
s
y
s
tem
(p
ho
to
vo
lt
aic
m
odul
e
i
n
p
ut
s
(tem
p
e
rat
u
re
a
nd
s
o
l
ar
r
adi
a
tio
n
),
p
ho
to
vo
ltai
c
m
odu
le
a
nd
c
on
t
rol
(MPP
T
))
th
at
h
av
e
be
en
a
d
o
p
t
ed
w
it
h
i
n
a
simu
la
t
i
o
n
tim
e
of
2
4
ho
urs.
I
n
a
ddition,
t
o
reach
t
h
e
o
p
timal
o
p
e
ratin
g
p
o
i
n
t
reg
a
rdles
s
o
f
variatio
ns
i
n
c
li
ma
t
i
c
con
d
i
t
i
ons
,
t
h
e
us
e
of
a
n
eu
ron
net
w
o
r
k
b
a
sed
di
st
urb
a
nce
an
d
o
bs
ervati
on
a
l
go
rith
m
(
P
&
O
)
is
p
u
t
i
n
t
o
se
rv
ic
e
of
t
he
s
y
s
te
m
g
i
v
e
n
its
r
el
iabi
lit
y
,
i
t
s
sim
p
licit
y
an
d
v
i
ew
t
h
a
t
at
a
n
y
t
im
e
i
t
can
f
ol
low
t
h
e
d
e
sire
d
m
a
ximu
m
po
wer.
T
h
e
e
nt
ire
sy
st
em
i
s
im
plem
en
ted
i
n
t
h
e
M
atlab
/
S
i
m
u
li
nk
env
i
ro
nm
ent
whe
r
e
s
i
m
u
l
a
ti
on
r
esul
ts
obt
ain
e
d
are
very
p
rom
i
sin
g
an
d
hav
e
sh
ow
n
the
eff
ectiv
enes
s
an
d
sp
eed
o
f
n
e
ural
t
ec
h
n
o
l
og
y
th
a
t
s
ti
l
l
r
equire
a
learn
i
n
g
b
as
e
s
o
t
o
im
p
r
ov
e
t
h
e
perf
o
r
m
a
n
ce
of
p
h
o
t
ovo
lt
aic
sys
te
m
s
a
n
d
exp
l
o
i
t
t
h
em
i
n
energ
y
p
ro
du
ctio
n,
a
s
well
a
s
t
h
is
t
ech
niq
u
e
h
a
s
pro
v
ed
t
hat
th
ese
res
u
l
t
s
are
m
u
ch
b
et
ter in term
s
(of
i
ts
v
ery
great
p
re
ci
si
on and
sp
eed
o
f
com
p
utatio
n)
t
h
a
n
t
hos
e
o
f
t
h
e
c
on
tro
l
l
e
r
bas
e
d
o
n
t
he
c
onv
ent
i
o
n
al
M
PPT
m
e
th
od
P
& O.
K
eyw
ord
s
:
PV
GPV
MPP
T
ANN
P
&
O
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:
Fa
rid S
aada
o
u
i
,
D
e
pa
rtem
ent o
f
E
lectr
i
c
a
l E
n
gi
nee
r
in
g,
Tahri
Mohamed
U
ni
v
e
rsi
t
y,
Bp4
1
7
Bechar, A
lg
eri
a
.
Em
ail:
fsaa
d
a
o
u
i1
96
9
@
gm
ai
l
.
com
1.
I
N
TR
OD
U
C
TI
O
N
Th
e
w
o
rld
'
s
en
ergy
d
e
m
a
n
ds
a
nd
n
e
e
d
s
a
re
c
on
st
a
n
tl
y
gro
w
i
n
g
,
a
n
d
e
n
o
r
mou
s
r
es
earch
h
as
b
ee
n
con
d
u
cte
d
a
nd
e
xp
l
o
i
t
e
d
o
n
t
h
e
pr
od
uc
t
i
o
n
a
nd
c
o
ns
umpt
i
on
o
f
r
e
new
a
ble
ene
r
gy,
w
h
i
ch
h
a
s
t
he
a
d
v
a
n
ta
ge
of
b
e
i
ng
f
le
xib
l
e,
r
elia
b
l
e
a
n
d
n
on-
po
l
l
u
tin
g.
T
he
c
o
n
s
t
rai
n
t
i
mpo
s
e
d
i
s
t
o
u
se
a
n
ec
ono
mi
ca
l
a
n
d
effi
c
i
ent
source
o
f
ener
gy t
o
m
in
i
m
ize
fuel
s
up
pl
y pr
oble
m
s
t
o
p
ro
t
ect
o
u
r
e
nviro
n
m
ent.
Re
ne
w
a
b
l
e
en
er
gy
s
ourc
e
s
p
l
ay
a
n
i
m
port
a
nt
r
o
l
e
in
e
lec
t
ric
ity
ge
nera
t
i
on.
S
ol
a
r
e
ner
gy
is
one
o
f
t
h
ese
i
m
po
rt
ant
a
l
t
e
rn
ati
v
e
so
u
r
c
e
s
t
h
at
d
ep
end
s
o
n
th
e
at
mo
sp
h
ere,
t
e
m
perature
a
nd
so
lar
ra
dia
t
i
o
n
t
h
a
t
v
a
ri
es
c
on
si
d
e
r
a
b
l
y
du
ri
n
g
t
h
e
d
ay
a
nd
i
t
s
o
pt
i
m
al
u
se
r
equi
res
t
a
k
in
g
in
to
a
ccou
n
t
d
i
f
fer
e
nt
t
y
p
es
o
f
c
h
a
r
ges
suc
h
a
s
(batt
e
ries,
li
g
h
ti
ng
s
y
stem
s,
l
am
ps,
and
mo
tors).
I
n
or
der
t
o
p
e
r
m
a
nent
l
y
g
e
t
t
he
m
a
x
imum
p
ow
er
from
t
h
i
s
p
h
o
t
o
v
o
l
ta
ic
(
P
V
)
tha
t
m
ust
run
a
t
t
he
ir
m
a
x
i
m
um
pow
e
r
po
i
n
t
trac
k
i
n
g
(
M
P
P
T
)
despite
c
lim
at
e
cha
nge
.
F
o
r
t
h
i
s
r
ea
son
a
bet
t
e
r
e
stabl
i
s
h
m
e
n
t
a
nd
e
x
p
l
oi
ta
tio
n
o
f
p
hoto
v
o
l
t
a
i
c
s
y
s
t
e
ms
s
u
r
e
l
y
l
ead
t
o
a
good
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N: 2
0
8
8
-
86
94
I
n
t
J Po
w
Elec
&
Dr
i
Sy
st,
Vo
l. 1
0
,
No
. 3
,
S
e
p
2
019
:
1
5
7
5
–
1
591
1
576
im
por
tan
t
p
r
o
duc
t
i
o
n
o
f
s
o
l
a
r
ener
gy
w
h
i
c
h
ta
ke
s
i
n
t
o
a
c
c
ou
n
t
t
he
p
r
o
b
l
e
m
o
f
de
gr
a
d
at
i
on
o
f
t
he
pow
e
r
sup
p
l
i
e
d
b
y
t
he
s
olar
g
ener
at
o
r
due
t
o
t
h
e
var
i
a
t
i
on
o
f
t
he
r
ad
ia
t
i
o
n
whi
ch
can
r
e
ach
50 %
[1].
To
m
a
x
i
m
ize
t
h
e
e
l
ec
tric
ity
p
ro
duc
tio
n
o
f
p
h
o
t
o
v
o
l
t
a
i
c
p
a
nel
s
i
n
a
n
o
p
t
i
m
al
w
ay
w
e
use
a
Bo
o
s
t
c
o
n
v
er
ter
w
h
ic
h
i
s
a
b
l
e
t
o
g
ive
a
m
a
xim
u
m
pow
er
u
nd
er
o
ptim
al
c
on
d
iti
o
n
s wha
t
e
v
er are
the
v
aria
tio
ns
of th
e
r
a
dia
t
i
on
and
inde
pe
nde
n
t
l
y
o
f
t
h
e
m
e
teor
olo
g
ica
l
c
o
n
d
i
t
i
o
n
s
o
f
t
he
v
ar
iat
i
on
o
f
t
he
l
oa
d.
The
m
a
i
n
f
ac
t
o
r
s
t
ha
t
in
fl
ue
nc
e
the
ef
f
i
c
i
e
n
c
y
o
f
th
e
pr
oce
ss
o
f
c
o
ll
ect
i
ng
th
is
s
o
l
ar
r
adia
ti
o
n
a
re
t
he
e
f
ficie
n
c
y
o
f
t
h
e
sol
a
r
cel
ls,
t
h
e
i
n
te
nsi
t
y
o
f
t
h
e
sour
ce
ra
di
a
t
i
o
n
a
n
d
t
h
e
sto
r
ag
e
t
echn
i
qu
es.
T
h
e
e
f
fi
ci
en
c
y
o
f
the
s
e
solar
ce
l
l
s
i
s
l
imi
t
ed
b
y
t
h
e
ma
t
e
r
i
a
l
s
use
d
i
n
the
m
a
nuf
actu
r
e
of
cel
l
s
.
It
i
s
part
i
c
u
l
arl
y
d
i
f
fi
c
u
lt
t
o
si
gn
if
ica
n
t
l
y
i
m
p
r
ove
t
he
c
ell
'
s
per
f
or
m
a
nc
e,
w
hic
h
l
i
m
its
t
he
ef
fici
en
cy
o
f
th
e
o
v
e
ra
l
l
c
o
l
l
ect
ion
pr
o
c
ess
[
1
]
,
[
2]
.
I
n
t
h
i
s
s
t
ud
y
w
e
a
r
e
i
nte
r
es
t
e
d
in
a
pp
l
y
i
n
g
t
h
e
ne
ur
a
l
n
e
t
w
o
r
k
s
i
n
thr
e
e
inde
pe
n
d
en
t
par
t
s
o
f
t
he
ph
o
t
o
v
o
lta
i
c
(
P
V
)
c
h
ain
a
s
s
how
n
i
n
F
ig
ur
e
1,
(
incom
i
ng
r
a
dia
t
i
on
a
nd
t
e
m
p
er
atur
e,
P
V
pane
l
a
nd
m
a
ximum
pow
e
r
p
oi
n
t
t
r
a
c
k
ing
(
M
P
P
T
)
c
ontr
o
l)
,
i
n
o
r
d
er
t
o
stu
d
y
t
he
i
nf
l
u
enc
e
o
f
ar
ti
f
i
c
i
a
l
i
n
t
e
lli
ge
nc
e
on
th
is be
h
av
i
o
r.
2.
PHO
T
OVO
LTAIC GENER
AT
OR
M
ODE
L
ING
2.
1.
Ph
o
t
ovo
l
t
ai
c
c
e
l
l
m
od
el
in
g
T
h
e
pho
tov
o
l
t
a
i
c
p
a
n
el
i
s
co
mp
o
s
e
d
o
f
sev
e
ral
sili
con
cel
ls
c
o
n
ne
c
t
ed
i
n
s
e
r
i
es
a
n
d
i
n
pa
r
a
lle
l.
T
he
y
a
b
sor
b
s
ol
a
r
e
ne
r
g
y
an
d
tr
a
n
sfor
m
t
h
em
i
nto
e
l
e
c
tr
ic
a
l
c
ur
r
e
nt.
T
h
i
s
tr
an
sfor
ma
tio
n
is
d
one
b
y
c
o
n
v
e
r
t
e
r
s
i
n
or
d
e
r
t
o
r
egu
l
a
t
e
the
vo
l
t
age
a
nd
cur
r
en
t
ou
t
p
ut
o
f
this
s
olar
ce
l
l
F
ig
ure
2.
Ph
D
R
p
II
I
I
(
1
)
()
ex
p
1
S
Ds
qV
I
R
II
nK
T
(2
)
F
i
gur
e
1.
O
ver
v
i
e
w
o
f
a
m
a
x
imum
pow
e
r
poin
t
t
r
a
ck
i
ng,
M
P
P
T-
c
ont
r
o
lle
d
pho
tov
o
lta
ic,
P
V
ge
n
era
t
or
sys
t
em
with ne
u
ral
ne
twork AN
N
F
i
gur
e
2.
E
qu
i
v
ale
n
t
cir
c
u
it
o
f
s
olar
c
e
l
l
3
1
ex
p
g
s
E
IK
T
K
T
(3
)
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
Mo
de
li
ng o
f
ph
ot
ov
ol
t
a
ic
syst
em
wi
t
h
m
a
xim
u
m
po
wer p
o
i
n
t tr
acki
n
g
c
o
n
t
ro
l by
… (
F
arid
Sa
ada
o
u
i)
1
577
I
s
:
i
s
t
he
sa
t
ura
tio
n
curre
nt o
f
t
h
e
d
i
ode
t
K
T
V
q
(4)
t
V
: is t
he
therm
ic
v
ol
ta
ge
at
t
e
mpe
r
at
ure
T
S
Rp
P
VI
R
I
R
(5)
from
equa
ti
o
n
(
1) w
e obta
i
n
the
expr
ess
i
on
of c
hara
ct
eris
tic
I-
V of the
c
h
o
se
n
model
ex
p
1
Ss
Ph
s
t
VI
R
V
I
R
II
I
nV
Rp
(6)
()
Ph
sc
i
n
G
II
K
d
T
G
(7)
G
: S
o
l
a
r r
a
diat
i
o
n
and
n
G
:
N
o
m
i
n
a
l
s
ol
ar
r
ad
i
a
ti
o
n
0
exp
1
sc
i
s
CV
t
IK
d
T
I
VK
d
T
Va
(8)
a
:
d
i
od
e
i
d
e
a
lit
y f
a
ct
o
r
()
()
ex
p
1
SS
S
P
P
S
S
S
P
P
P
P
Ph
PP
s
tP
S
S
P
P
VR
N
N
I
V
R
N
N
I
IN
I
N
I
Va
R
N
N
(9)
P
P
N
: is the
n
u
m
b
e
r
o
f pho
t
o
v
o
l
ta
ic
m
odu
l
e
s i
n
p
ar
alle
l =
1
SS
N
:
i
s the
num
be
r
of
pho
tov
o
lta
ic
m
odu
les i
n
s
e
r
ies= 3
6
The
ch
o
i
ce
o
f
a
P
V
m
odel
ge
ner
a
l
l
y
d
e
pen
d
s
on
t
h
e
use
t
h
at
w
e
w
ant
to
m
a
k
e
and
t
h
e
infor
m
a
tion
ava
ila
b
l
e
t
o
d
e
t
e
r
m
i
ne
t
he
p
a
r
am
eter
s.
E
ach
m
odel
m
u
s
t
b
e
base
d
o
n
clas
s
i
cal
p
r
i
nc
i
p
l
e
s
[3]
Ta
ble
1
.
In
o
r
d
er
to
i
mpr
ove
t
h
e
e
lec
t
r
i
c
a
l
m
ana
g
em
en
t
o
f
P
V
system
,
it
is
e
ssen
t
i
a
l
to
m
o
d
e
l
and
a
n
al
y
s
e
t
h
e
s
e
s
p
eci
fic
in
t
r
ins
i
c
char
acte
r
istic
s
[4].
T
a
b
l
e 1.
The
ele
ctrica
l
char
act
e
r
ist
i
cs
o
f p
h
o
t
o
v
o
l
ta
ic
m
od
ul
e
Par
a
met
e
r
s
V
al
u
e
s
G
r
:
Sola
r ra
di
a
t
i
o
n re
f
e
r
e
n
c
e
1000
(
W
/
m
²
)
R
s
: E
quiva
le
nt
s
e
r
ie
s
r
e
si
st
a
n
c
e
of
t
h
e
re
s
istiv
e
c
ont
a
c
t
s
R
s
h:
S
hunt
r
e
s
istanc
e
r
e
pre
s
e
n
ting
the
lea
k
ing
c
u
rr
e
n
ts
q:
E
l
e
c
t
ron
c
h
a
r
ge
K
:
B
ol
t
z
m
a
nn c
o
n
s
ta
nt
T
: is
th
e
e
f
f
ec
tiv
e
tem
p
e
r
a
t
ur
e
of the
ce
l
l
T
i
nit: Initi
a
l
t
e
m
p
e
r
a
tur
e
of
t
h
e PV Ce
l
l
I
s
c:
Sh
o
r
t
-
cir
c
u
i
t
cu
r
r
en
t
Voc
: Initia
l
ope
n-c
i
r
c
uit volta
g
e
Ki : T
emp
e
r
a
tu
r
e
v
ersu
s
cu
rren
t
c
o
e
ff
ici
e
n
t
E
g
: Ga
p
e
n
e
rgy
si
lic
iu
m
0.
5
O
h
m
s
500
O
h
m
s
1.
602*
10^-1
9C
1.
38*1
0^
-
23
J
/
°
K
°K
298
°K
3.
33
A
21.
3
V
0.
0025A
/°C
1,
12
e
V
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N: 2
0
8
8
-
86
94
I
n
t
J Po
w
Elec
&
Dr
i
Sy
st,
Vo
l. 1
0
,
No
. 3
,
S
e
p
2
019
:
1
5
7
5
–
1
591
1
578
2.
2.
A
r
t
i
f
i
cia
l
n
e
u
r
a
l
n
e
t
w
or
k
m
o
d
e
l
The
a
r
ti
f
i
cia
l
n
eur
a
l
ne
tw
or
k
(
A
N
N
)
i
s
a
m
athem
a
tica
l
c
om
puta
t
i
o
n
co
n
cept
th
a
t
b
e
l
ong
s
to
t
h
e
fam
i
l
y
o
f
m
e
t
h
o
d
s
of
a
r
tific
ia
l
i
n
te
l
lige
n
ce
,
whose
sche
ma
t
i
c
d
e
s
ign
r
e
sem
b
l
e
s
t
h
e
neur
ons
o
f
the
hum
an
br
ai
n
ba
se
d on
l
o
g
i
ca
l
r
e
as
o
n
i
n
g.
I
t
s
s
t
r
uc
tur
e
is
c
o
m
pose
d
o
f
t
h
r
e
e l
a
yer
s
a
s sh
ow
n
in de
t
a
ils be
l
ow
c
ho
ic
e i
n
Fi
g
u
re
3
.
I
n
p
u
t
l
a
yer
:
c
o
m
pose
d
o
f
thr
ee
neur
o
n
s
tha
t
t
r
a
nsmi
t
the
va
lu
e
of
t
he
i
n
p
u
t
s
i
g
nal
tha
t
c
or
r
e
spon
ds
t
o
th
e
var
i
a
b
le
o
u
t
p
u
t
vol
t
a
ge
o
f
t
h
e
ph
o
t
o
v
o
l
t
a
i
c
pa
nel
t
o
t
he
n
e
x
t
l
ay
e
r
.
the
h
i
dde
n
la
ye
r
:
w
hich
i
s
co
mpose
d
o
f
te
n
ne
ur
ons
w
i
t
h
d
e
f
i
ne
d
a
c
tiva
t
i
on
f
u
nct
i
ons.
the
o
u
tp
u
t
l
a
y
e
r
:
c
o
mpose
d
o
f
one
n
eur
on,
t
h
e
out
p
u
t
o
f
n
e
u
r
on
whic
h
is
a
0
or
1 lo
gic
v
a
r
i
ab
le.
F
i
gur
e 3.
Char
a
cter
i
s
t
i
c
of t
h
e
P
V
ce
l
l
The
n
u
m
b
e
r
o
f
no
des
in
e
ac
h
laye
r
v
a
r
i
es
a
nd
d
e
p
e
n
d
s
o
n
t
h
e
u
s
e
r
.
T
h
e
A
N
N
i
n
p
u
t
v
a
r
i
a
b
l
e
s
o
r
si
gna
ls
a
r
e
the
P
V
gener
a
tor
inp
u
t
par
a
m
e
t
e
r
s
suc
h
a
s tem
p
er
a
t
u
r
e
a
n
d
s
o
l
ar
r
adiat
i
on,
w
hile t
he out
pu
t
si
gna
l
is
t
he
m
a
x
imu
m
i
dent
i
f
ie
d
p
o
w
e
r
.
Th
e
choice
o
f
ANN
is
m
ad
e
acco
r
ding
t
o:
t
h
e
t
y
p
e
o
f
n
e
uro
n
s
,
th
e
num
b
e
r
of
n
e
u
r
on
l
a
yer
s
a
nd
t
h
e
st
ud
y
m
e
t
h
ods
,
the
i
r
ad
va
n
t
a
g
es
i
n
br
oa
d
b
a
n
d
use
d
i
n
in
d
u
s
t
r
y
i
s
t
h
e
st
or
a
g
e
o
f
a
n
en
or
m
ous
m
ass
o
f
info
rmation
and
th
e
direct
s
to
ra
ge
i
n
r
eal
t
im
e
o
f
t
h
i
s
i
n
for
m
a
t
i
on,
a
l
l
ow
in
g
to
m
a
k
e
l
o
gica
l
dec
i
s
i
o
n
s,
ef
ficie
n
c
y
t
o
so
lv
e
p
r
ob
lems
o
f
non
-lin
earities.
T
h
e
ANN
i
s
n
ot
e
d
f
or
s
ens
i
tiv
it
y
to
noise.
Ge
n
e
ra
ll
y
t
h
e
ne
ur
on
o
u
t
p
ut
w
il
l
be
o
ne
o
r
m
o
r
e
r
e
f
er
ence
s
i
gna
l(
s)
[
5]
.
The
ANN
p
e
rf
o
r
ms
t
h
r
ee
c
alcu
lation
o
p
e
ration
s
on
it
s in
p
u
t
s
[
6]:
The
po
nde
r
a
t
i
on
(
w
ei
g
h
t
)
:
th
e
m
u
l
t
i
p
li
c
a
t
ion
o
f
a
ny
i
npu
t
by
a
s
p
e
c
i
fic
fact
or
(
whic
h
i
s
t
h
e
c
o
n
n
ec
ti
on
wei
g
h
t
).
The
sum:
acc
o
r
di
ng t
o
al
l
in
p
u
t
s
weig
h
t.
The
ac
tiva
t
io
n func
tio
n
( f).
The
ar
chitectur
a
l
s
y
nopti
c
o
f
the
pr
opose
d
n
e
u
ral
network
(
3
-
1
0
-
2
)
co
ns
ist
s
o
f
a
n
i
n
p
u
t
la
ye
r
wi
t
h
3
i
n
p
u
t
s
(
s
o
l
a
r
r
a
d
i
a
t
i
o
n
G
,
t
e
m
p
e
r
a
t
u
r
e
T
a
n
d
v
o
l
t
a
g
e
V
)
,
a
h
i
d
d
en
l
a
y
er
o
f
1
0
n
eu
ro
ns
a
n
d
an
o
u
t
put
l
ay
er
wi
th
2
o
ut
put
(
cu
rren
t
I
a
nd
v
olt
a
g
e
V
)
F
i
gu
r
e
2
2
.
T
h
e
m
o
d
e
l
s
u
se
d
in
M
A
TLA
B
f
o
r
bu
i
l
d
i
ng
n
e
u
r
o
ns
a
r
e
di
ff
er
e
n
t
,
u
si
n
g
t
he
h
yper
b
o
l
i
c
t
an
ge
n
t
s
igm
o
i
d
t
r
a
ns
fer
f
u
nct
i
on
i
n
t
he
h
i
dde
n
lay
e
r
“
t
ans
i
g”
a
n
d
t
he
t
r
a
nsfe
r
f
u
n
c
tio
n
“
pur
e
l
i
n
”
f
o
r
t
h
e
o
u
t
pu
t
layer
.
T
he
h
id
de
n
la
yer
tr
a
n
s
fe
r
fu
nc
t
i
o
n
i
s
a
s
i
gm
oi
d
f
unc
tio
n
de
f
i
ned
a
s
f
o
l
l
o
w
:
)
.
(
1
1
=
)
(
u
d
e
u
f
(
10)
Whe
r
e
d : is t
h
e
s
l
o
pe
par
am
eter
.
The
en
tr
y
of
t
h
e
h
i
d
de
n
lay
e
r
is
e
x
p
r
e
ssed
b
y
t
he
f
o
llow
i
n
g
r
el
at
io
nshi
p
:
)
(
=
1
=
i
i
ij
n
j
b
x
w
u
(
11)
Whe
r
e
W
ij
i
s
th
e
weig
ht
o
f
th
e
jth
i
np
ut
X
j
t
o
t
h
e
i
th
n
eur
o
n
i
n
t
he
h
id
de
n
laye
r
an
d
bi
i
s
the
b
i
as.
I
f
the
f
u
nct
i
o
n
i
n
t
h
e
o
u
t
pu
t
laye
r
i
s
l
i
n
ea
r
t
h
e
n
t
he
n
e
u
r
a
l
ne
tw
or
k
s
e
t
i
s
e
x
p
re
sse
d
b
y
the
fo
llowi
n
g
re
l
a
ti
on
sh
ip:
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
Mo
de
li
ng o
f
ph
ot
ov
ol
t
a
ic
syst
em
wi
t
h
m
a
xim
u
m
po
wer p
o
i
n
t tr
acki
n
g
c
o
n
t
ro
l by
… (
F
arid
Sa
ada
o
u
i)
1
579
))
(
(
=
=
)
(
=
1
=
1
=
1
=
i
i
ij
N
j
o
ij
N
j
i
i
o
ij
N
j
k
b
x
w
f
w
u
b
u
w
u
y
(12)
k
y
: is the
o
u
t
put
s
i
gna
l
of t
he
k
th
o
u
t
pu
t ne
uro
n
s a
nd
o
ki
w
is the
w
e
ig
ht o
f
t
h
e it
h ou
t
p
u
t
i
u
to t
h
e
kiem
e
neur
ons
i
n
the
ou
t
p
u
t
l
a
y
er
.
The
bi
as
a
nd
w
e
i
g
ht
i
n
g
va
l
u
es
o
f
t
h
e
ne
uron
a
re
d
e
t
er
mined
as
a
f
un
c
t
io
n
of
t
h
e
u
s
a
ge
a
lg
o
r
i
t
h
m
.
W
ij
i
s
t
h
e
w
e
i
g
h
t
o
f
t
h
e
c
o
nne
ct
i
on
be
tw
e
e
n
neuro
n
s
i
a
n
d
j
an
d
e
a
c
h
n
eu
r
o
n
i
s
c
o
n
n
e
ct
ed
t
o
t
h
e
s
e
t
o
f
neur
ons
o
f
t
h
e
ne
x
t
l
a
y
e
r
by
t
h
ese
w
e
ig
hts
w
h
ic
h
a
r
e
re
al
numbe
rs.
The
con
s
t
u
tu
t
i
o
n
a
nd
type
o
f
al
g
o
ri
thms
o
f
t
h
e
ANN
a
u
to
mati
cal
l
y
af
fect
t
h
e
M
P
P
T.
T
h
e
c
o
n
t
r
i
b
u
tio
n
o
f
t
he
n
e
u
ral
ne
t
w
or
k
is
t
he
refor
e
u
sed
t
o
appr
ox
im
ate
a
nd
e
v
a
l
ua
te
t
h
e
m
aximum
pow
er
o
f
the
P
V
m
odu
le
a
n
d
t
o
dri
v
e
the
Boos
t
(D
C-D
C
)
pow
er
con
v
er
t
e
r
t
o
opera
t
e
c
lo
se
t
o
the
MPPT,
desp
i
t
e
t
h
e
d
i
ffe
r
en
t
o
pe
r
a
ting
c
o
n
d
i
t
i
ons
t
ha
t
c
h
a
n
ge
t
he
cha
r
ac
t
e
ris
tics
of
a
p
h
o
to
vo
l
t
a
i
c
gener
a
tor
ov
er
t
i
m
e
an
d
t
h
a
t
im
pose
t
h
e
fo
r
m
a
t
io
n
of
t
he
n
e
u
ra
l
ne
tw
ork
i
n
a
peri
od
ic
m
a
n
n
e
r
t
o
r
esu
lt in a pref
e
r
r
ed
M
P
P
T.
The
b
oos
t
c
o
n
v
erter
is
a
l
s
o
know
n
as
a
s
te
p-up
c
o
n
v
erter.
I
t
i
s
g
e
n
era
l
ly
u
se
d
in
c
onv
e
r
t
i
ng
a
l
ow
inpu
t
vol
tage
t
o
a
hi
g
h
o
utpu
t
vo
ltage
[7].
T
h
e
M
PP
T
co
nt
ro
l
al
l
o
w
s
y
ou
to
b
e
a
t
a
n
y
tim
e
fr
om
t
he
s
u
n
t
o
t
h
e
ma
ximum
po
w
e
r
of
our
P
V
system
.
To
s
im
ulate
t
h
e
m
a
ximum
p
o
w
e
r,
w
e
f
i
r
s
t
c
o
n
n
e
c
t
e
d
o
u
r
P
V
s
y
s
t
e
m
direc
t
l
y
t
o t
h
e
loa
d
,
an
d
the
n
w
e
con
n
e
c
te
d the
b
o
o
st
c
o
nve
rter
c
ontro
lle
d by
t
h
e
MP
P
T
b
e
t
w
e
en t
he
l
oa
d
an
d
the P
V
system
[7].
3.
CHARACTE
R
IS
T
I
CS AN
A
L
Y
S
I
S
OF A PHOT
OVOLTAIC GENE
R
A
T
OR
3.1.
Charac
t
er
ist
i
c I-V
&
P-V
The
c
u
r
r
ent
I
ph
pr
od
uce
d
b
y
the
c
e
l
l
is
d
i
r
ectl
y
p
ro
port
i
o
n
al
t
o
t
h
e
so
la
r
ra
d
i
a
tio
n
G
,
w
it
h
a
sma
ll
vo
lta
ge
v
a
r
ia
ti
on
a
nd
t
he
o
p
e
n
c
i
r
c
u
it
vo
l
t
a
g
e
w
i
ll
decr
e
a
se
o
n
ly
s
li
ght
ly
w
it
h
rad
i
a
t
i
on.
T
h
u
s,
t
he
o
p
t
i
m
al
pow
er
o
f
t
h
e
ce
l
l
i
s
pr
act
ica
l
l
y
p
r
o
p
o
r
t
i
ona
l
to
t
he
r
a
d
i
a
tio
n
a
n
d
the
m
a
x
i
m
u
m
pow
er
poin
t
s
ar
e
appr
ox
im
ate
l
y
a
t
t
he sa
m
e
vo
l
t
a
g
e
[8].
3.1.
1. In
f
lu
e
n
ce of
th
e
t
emp
é
r
a
t
u
r
e
o
n
t
h
e
I
-
V
&
P
-
V
c
h
a
ra
c
t
eri
stic
F
i
gure
4
a
n
d
F
i
g
u
re
5
s
h
o
w
s
t
he
c
har
acte
r
i
s
t
i
cs
o
f
the
I-V
a
nd
P
-
V
of
t
he
p
h
o
t
o
vol
ta
ic
m
odu
l
e
w
i
t
h
a
m
a
ximum
se
lec
t
e
d
p
ow
e
r
o
f
ap
pr
oxim
a
t
e
ly
5
5
W
,
i
n
a
t
em
pera
tur
e
v
aria
tio
n
from
(
10
°
C
t
o
6
0
°C)
.
T
hi
s
tem
p
era
t
ur
e
i
n
c
r
ea
se
h
as
a
d
irec
t
i
n
fl
uenc
e
o
n
t
he
b
e
h
a
v
i
our
o
f
t
he
P
V
,
i
t
ca
u
s
e
s
a
r
educ
t
i
o
n
i
n
t
h
e
o
p
en
circ
u
i
t
v
o
l
t
a
g
e
of
a
b
out
2
m
V
per
de
gre
e
w
it
h
a
ne
g
l
i
g
i
b
le
i
nfl
uenc
e
o
n
t
h
e
p
h
o
t
o-c
u
rr
ent
co
ns
ide
r
ed.
In
t
he
sam
e
w
ay,
the
p
o
w
e
r
pr
oduc
ed
d
e
c
re
ases
s
lig
h
tly
f
ol
low
i
n
g
t
he
v
ol
tag
e
d
ec
li
n
e
,
whi
c
h
di
re
ctl
y
r
e
f
l
ect
s
th
e
dec
r
ea
se
i
n
th
e
m
a
ximum
a
v
a
ila
b
l
e
p
o
w
e
r,
t
here
fore
,
o
n
t
he
e
ffi
ci
ency
o
f
th
e
cel
l.
C
on
seq
u
e
nt
ly
,
th
e
un
des
i
rab
l
e
e
f
fe
ct
o
f
t
e
mpe
r
ature
on
a
ph
ot
o
v
o
l
t
a
ic
p
an
el
c
a
n
d
am
ag
e
t
h
e
sys
t
e
m
a
nd
s
hor
te
n
i
t
s
li
fe
,
espec
i
al
ly
i
n
t
h
e hot
tes
t
a
rea
s
.
3.1.
2. In
f
lu
e
n
ce of
th
e
r
ad
iation
on
th
e I-
V &
P-V
carac
t
ér
ist
i
c
Th
e
in
fl
u
e
n
ce
o
f
s
o
l
ar
r
a
d
ia
t
i
on
o
n
c
h
aract
eri
s
t
i
c
I-V
i
s
c
on
si
dera
bl
e
,
e
spe
c
ia
l
l
y
o
n
t
he
p
h
o
t
o
-
c
u
rre
n
t
and
we
ak
o
n
t
h
e vo
lt
a
g
e
:
ph
c
II
A
G
(13)
α:
i
s
the l
i
g
h
t
a
bsor
pt
i
on coe
f
ficie
n
t
an
d A
:
i
s
t
h
e
a
c
t
i
ve
sur
f
ace
of
the c
e
l
l
.
The
c
h
ara
c
teri
s
t
i
c
s
hows t
h
at
t
he
effici
enc
y
d
epe
n
d
s
o
n t
h
e
rad
i
a
ti
on:
ma
x
m
a
x
ma
x
.
i
PV
I
PG
A
(14)
Thi
s
f
a
c
to
r
i
s
v
ari
a
b
l
e
th
roug
hou
t
th
e
d
a
y
a
n
d
c
a
nno
t
b
e
i
n
c
rea
se
d,
t
here
fore
on
t
h
e
p
o
w
e
r-vol
ta
ge
cha
r
ac
t
e
ris
tic,
depe
n
d
i
n
g
on
t
h
e
re
l
a
tio
nsh
i
p
(
13)
a
n
d
f
or
b
ett
er
P
V
pe
rfor
m
a
n
c
e
re
sol
u
ti
on,
t
h
e
s
o
l
u
t
ion
i
s
t
o
cho
o
se
a
l
a
r
ge
a
re
a
o
f
hig
h r
a
dia
t
ion
t
o
e
x
p
o
s
e
th
e
so
lar
pane
l
s
[
9
].
F
i
gures
4
a
n
d
5
s
how
t
he
v
a
r
iat
i
o
n
o
f
t
h
e
I
-
V
and
P
-
V
cha
r
ac
t
e
r
istic
s
of
t
he
P
V
modu
le
a
t
d
i
ffere
n
t
tem
p
era
t
ur
es
w
hic
h
r
a
n
ge
f
r
o
m
0
°
C
t
o
6
0
°
C
w
h
en
t
h
e
r
adia
t
i
o
n
i
s
k
e
p
t
c
o
nsta
n
t
a
t
1
000
W
/
m2.
A
s
t
he
tem
p
era
t
ur
e
in
c
r
ea
ses,
t
he
s
hor
t
c
i
r
c
u
i
t
cur
r
ent
inc
r
ea
ses,
b
u
t
t
he
o
pe
n-
circu
it
vol
ta
ge
d
e
c
re
ases.
Th
u
s
,
the
ou
tpu
t
pow
er
o
f the
ce
ll
i
s
a
ls
o
dec
r
ea
se
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
Int J
P
o
w
El
e
c
&
D
ri S
yst
,
V
ol.
10,
N
o.
3
, S
e
p
2
0
1
9
:
1575
– 1
591
1
580
F
i
gures 6
a
n
d
7
s
h
o
w
t
h
e
va
r
i
at
i
on of
t
h
e
I
-V
a
nd
P
-V
c
ha
rac
t
e
r
i
s
t
i
c
s
of
t
h
e
P
V
mo
dul
e
a
t
a
c
on
st
a
n
t
a
m
b
i
e
n
t
t
e
mp
e
r
a
t
u
r
e
T
=
25
°
C
,
by
m
odi
fy
in
g
t
h
e
ra
di
at
ion
v
a
l
u
e
s
from
25
0
W
/
m2
t
o
100
0
W
/
m2.
If
t
he
radia
t
i
o
n incr
e
a
se
s the m
a
xi
m
u
m
pow
er
is hi
g
h
er
a
nd a
l
so
t
he
c
u
rre
n
t
a
ls
o incre
a
ses.
F
i
n
a
l
l
y
w
e
c
a
n
e
a
s
i
l
y
s
a
y
f
r
o
m
t
h
ese
figures
that
t
he
operating
p
oi
nt
o
f
th
e
maxi
mu
m
p
o
w
er
gene
ra
ted
is
due
t
o
t
h
e
i
n
cr
e
a
se
o
f
the
so
la
r
r
a
diat
i
on
w
h
ic
h
c
ause
s
an
i
ncr
ease
o
f
t
he
c
urre
nt
d
ue
t
o
t
h
e
circ
u
i
t
of c
ut
i
s
of c
ourse
t
he
vo
l
t
a
ge
i
n
the
ope
n
circ
u
it
rem
ains a
l
m
os
t cons
ta
n
t
.
F
i
gure
4. V
olta
ge
cur
rent
c
har
a
cter
i
s
t
i
c,
I
-V
of
a
p
hot
ovo
lt
a
i
c mo
dul
e, P
V f
o
r d
i
ff
e
r
en
t
t
e
mp
erat
u
r
e
s
a
nd co
ns
ta
nt r
adia
t
i
o
n
F
i
gure
5.
V
olta
ge
pow
er
c
har
acte
r
isti
c,
P
-V of a
p
hot
ovo
l
t
a
i
c
mo
dul
e, P
V f
o
r d
i
ff
e
r
en
t
t
e
mp
erat
u
r
e
s
a
nd co
ns
ta
nt r
adia
t
i
o
n
F
i
gure
6.
V
olt
a
ge c
urre
nt
c
ha
r
acte
r
ist
i
c
I-V
o
f
a
p
hot
ovo
l
t
a
i
c
mo
dul
e P
V
f
o
r
d
i
f
f
e
re
nt
ra
d
i
a
t
i
o
n
s
an
d
c
o
nst
a
nt
t
e
m
p
e
rat
u
re
F
i
gur
e 7.
V
olta
ge
pow
er
cha
rac
t
e
r
is
t
i
c
P
-
V
of a
ph
o
t
o
v
o
lta
ic
m
od
ule
P
V
for
diffe
ren
t
r
adia
ti
ons
a
nd
c
o
nst
a
nt
t
e
m
p
e
rat
u
re
4.
Photov
olt
a
ic
s
y
s
t
e
m
cont
rol b
y
neur
o
n
array
4.1.
PV system b
a
se
d
on
n
eu
ron
-
n
e
twor
k
The
fo
ll
ow
in
g
fi
gures
s
how
t
he
c
l
i
m
a
tic
v
aria
ti
o
n
s
me
asure
d
on
the
06
/
0
1
/
201
1
by
t
he
w
e
a
t
her
service
s
i
n
s
o
m
e
Saha
r
an cit
ies of
A
lger
ia
f
or
a
d
ura
tio
n of
2
4
ho
u
rs
F
i
g
ur
e 8
.
B
ec
har c
i
t
y
ha
s
bee
n
ch
ose
n
as
a
sam
p
le.
S
i
m
u
la
t
i
o
n
r
es
ults
o
b
t
a
i
n
e
d
in
2
D
and
3D
u
n
d
e
r
M
at
la
b
/S
im
ul
i
nk
sh
ow
t
h
e
p
er
form
ance
o
f
P
V
syste
m
b
ehav
i
o
r
con
t
rol.
F
i
gur
e 8.
Rad
i
a
ti
o
n
var
i
a
tio
n
sc
hem
a
f
or
A
l
g
e
r
ia
n sou
t
her
n
m
eas
ure
d
by
t
h
e
w
e
at
her
service
s
on
0
6
Jan
u
a
ry
2011
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
Mo
del
i
ng o
f
p
h
o
t
o
v
o
l
t
ai
c
syst
em
w
i
t
h
m
a
x
i
m
u
m
po
w
e
r p
o
i
n
t
trac
ki
n
g
c
ont
ro
l by
… (
F
a
r
i
d
Sa
a
d
a
o
u
i)
1
581
4.
1.
1.
A
p
p
lica
t
ion
of
A
N
N
o
n
r
a
d
i
at
ion
The
q
u
a
d
r
a
t
i
c
so
lar
ra
di
a
tio
n
error
is
v
e
r
y
ne
g
lig
i
b
l
e
;
it
i
s
o
n
the
or
de
r
of
1
0
-4
w
h
i
c
h
e
xp
la
in
s
t
h
e
im
por
tance
o
f
ANN
a
s
inp
u
t
o
f
t
h
e
P
V
p
a
n
e
l
.
Da
ta
b
a
s
e
radia
tio
n
v
ar
ia
t
i
on
o
f
B
e
cha
r
c
it
y
m
easur
e
d
b
y
th
e
w
e
ather
ser
v
i
c
es
on Ja
nuar
y
06,
2
01
1
o
v
e
r
a
p
e
r
io
d
o
f
2
4 h
o
ur
s
w
i
t
h
out
n
eur
a
l
ne
tw
or
k
as show
n
i
n
F
igur
e 9.
I
n
F
igur
e
1
0
is
a
r
c
hi
t
e
c
t
ur
a
l
s
yn
o
p
tic
o
f
r
a
d
i
a
tio
n
pr
o
pose
d
b
y
n
eur
a
l
net
w
or
k
(
1
-
1
0-
1)
.
I
n
F
igur
e
1
1
i
s
A
N
N
b
l
o
c
S
imulink
o
f
r
ad
ia
t
i
on
.
I
n
F
ig
u
r
e
1
2
is
ANN
train
i
n
g
o
f
r
ad
ia
ti
o
n
w
ith
M
a
tla
b
/
Sim
u
li
nk.
I
n
Fig
u
re
13
is
th
e
sq
u
a
red
er
ro
r
o
f
th
e
ANN t
r
ai
n
i
n
g
o
f
r
ad
iation
with
M
a
tlab/
Sim
u
lin
k.
F
i
gur
e 9.
Da
t
a ba
se r
adia
t
i
o
n
var
i
a
t
i
o
n
o
f Be
c
h
ar
c
it
y
m
easure
d
b
y
t
h
e
w
e
ather
ser
v
ice
s
on
Ja
nuar
y
06,
2
0
1
1
over
a
per
i
o
d
o
f
24
h
o
u
r
s
w
ith
ou
t
ne
ur
a
l
n
e
t
w
o
r
k
F
i
gur
e
1
0
.
A
r
c
h
i
t
ec
tur
a
l
sy
no
ptic
o
f
r
a
dia
t
i
o
n
pr
o
pose
d
b
y
ne
u
r
al
n
et
wo
rk (1
-
1
0
-1
)
F
i
gur
e
1
1
.
A
N
N
bl
oc
S
i
m
ul
i
n
k
of
r
adia
t
i
o
n
F
i
gur
e
1
2
.
A
N
N
Tr
ainin
g
o
f
ra
di
a
tio
n
w
ith
M
a
t
la
b/S
i
m
u
l
i
n
k
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
Int J
P
o
w
El
e
c
&
D
ri S
yst
,
V
ol.
10,
N
o.
3
, S
e
p
2
0
1
9
:
1575
– 1
591
1
582
F
i
g
u
r
e 13
.
Th
e
sq
u
a
red
er
ro
r o
f
th
e
ANN
t
r
a
i
nin
g
o
f
rad
i
atio
n
w
it
h
M
a
tl
ab/
S
i
m
u
l
in
k
4.1.
2. A
p
p
l
i
ca
t
i
on
of
AN
N
on
tem
p
e
r
a
tu
r
e
Tem
p
era
t
ure
v
a
riat
ion
sc
hem
a
f
or
A
l
g
eria
n
so
ut
hern
m
ea
su
r
e
d
by
t
h
e
we
ather
se
rv
ic
es
o
n
0
6
Jan
u
ar
y 20
1
1
a
s
sh
ow
n i
n
F
ig
ure 1
4
. D
a
ta ba
s
e tem
p
era
t
ur
e varia
t
i
on
o
f
B
e
c
h
ar
c
i
t
y
m
ea
su
red
b
y
th
e
weat
h
e
r
service
s
o
n
Ja
nuar
y
06,
201
1
o
v
er
a
p
erio
d
of
24
ho
ur
s
w
i
t
hou
t
ne
ura
l
n
etw
o
rk
a
s
sh
ow
n
in
F
ig
ure
15.
I
n
F
i
gure
1
6
i
s
ar
c
h
itec
t
ur
al
s
y
n
o
p
tic
o
f
t
h
e
pr
op
ose
d
n
e
u
ral
ne
t
w
or
k
of
T
e
m
pe
rature
(
1-1
0
-1)
.
I
n
F
i
g
u
re
17
i
s
A
N
N
b
l
o
c
s
i
mul
i
nk o
f
t
em
pe
ra
t
u
r
e
a
s show
n.
F
i
gur
e 1
4
.
Temper
ature
var
i
a
t
i
o
n sc
hem
a
for
A
l
ge
ria
n
s
o
u
t
h
er
n
m
easure
d
by the
w
e
at
her
service
s
on 0
6
Ja
n
uary
201
1
F
i
gure
1
5
.
D
a
ta
b
ase
t
e
mpe
r
ature
va
ria
t
i
o
n
o
f
Bec
har
c
ity
m
e
a
sure
d by t
h
e
w
e
a
t
he
r
se
rvices
o
n Jan
u
a
r
y 0
6
,
2
011
o
v
e
r a
p
e
ri
od
o
f 24
h
ou
r
s
wi
t
hou
t
n
e
u
r
a
l
ne
t
w
o
r
k
F
i
gure
1
6
.
Ar
c
h
i
t
ec
tura
l
sy
no
p
t
ic
o
f the
pr
op
ose
d
n
eura
l
ne
t
w
or
k of tem
pe
r
a
ture
(
1-10-1)
0
20
40
60
80
100
12
0
14
0
160
10
-4
10
-3
10
-2
10
-1
10
0
10
1
B
e
s
t
T
r
a
i
n
i
n
g
P
e
r
f
or
m
a
n
c
e
i
s
0
.
0
0
0170
06
at
e
po
ch
1
71
M
e
a
n
S
qua
r
e
d
E
r
r
o
r
(
m
se
)
171
E
p
o
c
hs
Tr
a
i
n
Be
s
t
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
Mo
del
i
ng o
f
p
h
o
t
o
v
o
l
t
ai
c
syst
em
w
i
t
h
m
a
x
i
m
u
m
po
w
e
r p
o
i
n
t
trac
ki
n
g
c
ont
ro
l by
… (
F
a
r
i
d
Sa
a
d
a
o
u
i)
1
583
F
i
gur
e
1
7
.
A
N
N
bloc
s
imu
l
in
k
of
t
e
m
pe
r
a
tu
r
e
W
e
w
i
l
l
u
s
e
t
w
o
n
e
t
w
o
r
k
s
o
f
n
e
u
r
o
n
s
a
t
a
t
i
m
e
;
t
h
e
f
i
r
s
t
n
e
t
w
o
r
k
wh
ose
ro
le
i
s
to
e
st
ima
t
e
the
r
a
dia
t
i
on
a
n
d
t
h
e
sec
o
nd
i
s
t
o
estima
t
e
t
h
e
t
e
m
p
er
a
t
ur
e
w
h
ich
g
i
v
e
t
h
e
corr
e
s
p
o
ndi
ng
c
urren
t
a
nd
v
o
l
t
a
g
e
o
f
th
e
PV
p
an
el.
I
n
F
igu
r
e
1
8
is
ANN
t
r
ain
i
n
g
o
f
t
emp
e
ratu
r
e
w
i
t
h
Ma
t
l
a
b
/S
i
m
uli
n
k.
I
n
F
i
gur
e
19
is
t
he
s
q
u
ar
ed
e
r
r
o
r
of
t
he
A
N
N
t
r
ain
i
ng
o
f
te
m
p
e
r
atur
e
w
i
t
h
M
at
la
b
/
S
i
muli
n
k
.
I
n
F
ig
u
r
e
2
0
i
s
sc
he
ma
o
f
r
a
dia
t
i
o
n
a
n
d
tem
p
er
at
ur
e
var
i
a
tio
n
Bec
har
cit
y
a
s
Januar
y
06,
201
1
i
n
d
u
r
a
t
i
on
of
2
4
h
w
i
t
h
ne
ur
al
n
e
t
w
o
r
k
F
i
gur
e
1
9
.
T
h
e
s
quar
e
d
er
r
o
r
of
t
he
A
N
N
t
r
a
i
n
i
ng
o
f
tem
p
er
at
ure
with
M
at
la
b/Sim
u
l
i
n
k
F
i
gur
e
1
8
.
A
N
N
Tr
ainin
g
o
f
te
mpe
r
atur
e
w
i
t
h
M
a
t
l
a
b/
Si
mu
lin
k
Th
e
s
q
u
a
red
erro
r
o
f
t
h
e
t
emperatu
re
i
s
v
e
r
y
n
egligib
le,
it
is
on
t
h
e
or
de
r
o
f
1
0
-4
w
h
i
ch
e
xpla
i
n
s
t
he
im
por
tance
o
f
A
N
N
a
nd
the
f
o
l
l
o
w
i
ng
sc
he
m
a
s
how
s
the
equ
i
vale
nc
e
w
i
t
h
F
i
gur
e
9
an
d
F
i
gur
e
1
5
.
F
i
gur
e
2
0
.
S
c
h
e
m
a
o
f
r
a
diat
io
n
and
t
e
m
p
e
r
a
t
ur
e
va
r
i
at
ion
Be
cha
r
ci
t
y
as Januar
y
06,
2
011 in
d
ur
ation
of
24h
w
i
t
h
n
e
u
r
a
l
net
w
or
k
0
5
10
15
20
25
30
35
10
-6
10
-4
10
-2
10
0
10
2
B
e
s
t
T
r
a
in
in
g
P
e
r
f
o
r
m
a
n
c
e
is
6
.8
1
8
1
e
-
0
6
a
t
e
p
o
c
h
3
5
M
ean
S
q
u
a
r
e
d
E
r
r
o
r
(
m
se)
35
E
p
o
ch
s
Tr
a
i
n
Be
s
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SSN: 2088-
8694
Int J
P
o
w
El
e
c
&
D
ri S
yst
,
V
ol.
10,
N
o.
3
, S
e
p
2
0
1
9
:
1575
– 1
591
1
584
The
fo
l
l
ow
in
g ste
p
s
of va
r
ia
t
i
o
ns
a
re
from
00h
0
0
to
07
h
0
0
:
t
he
r
adi
a
t
i
o
n
(
G
=
0)
and
t
h
e
tem
p
er
at
ure
(T
=2°
C
).
from
07h
0
0
to
12
h
0
0
:
t
he
r
adi
a
t
i
o
n
a
nd t
h
e
t
e
mpe
r
ature
incr
e
a
se
s
rapid
l
y.
f
r
o
m
1
2
h00
t
o
1
3h3
0:
t
h
e
r
ad
i
a
t
i
on
i
s
max
i
mu
m
(G=max
)
a
n
d
th
e
t
emperature
i
s
al
s
o
m
axim
um
(T=max
)
.
from
13h
3
0
to
18
h
0
0
:
t
he
r
adi
a
t
i
o
n
(
G
)
a
nd the
tem
p
era
t
ure
(
T
)
be
gin
to
d
ec
rea
s
e
r
a
pidly
.
from
18h
0
0
to
00
h
0
0
:
a
t n
i
g
h
t
,
no r
a
diat
io
n (G
=0)
an
d the
t
e
m
p
e
ra
t
u
r
e
te
nds
t
ow
a
r
ds (
T=3.
5°C).
4.1.
3. A
p
p
l
i
ca
t
i
on
of
AN
N
on
th
e
PV
gen
e
rat
o
r
To
m
ak
e
th
e
co
nv
ersio
n
o
f
s
o
l
ar
e
n
e
rgy
in
to
e
l
ect
ri
c
a
l
p
o
we
r
ef
fe
ct
ive
a
t
t
he
o
u
t
pu
t
o
f
t
he
P
V
pa
nel,
i
t
i
s
p
r
e
f
e
r
a
b
l
e
a
n
d
i
n
d
i
s
p
e
n
s
a
b
l
e
t
o
o
p
e
r
a
t
e
i
t
c
l
o
s
e
t
o
a
p
o
i
nt
o
f
ma
ximu
m
p
o
w
er,
a
s
t
hi
s
p
o
w
e
r
v
a
r
i
e
s
con
s
ta
n
tly
a
c
c
o
rd
in
g
t
o
t
he
c
lima
t
o
l
og
ica
l
c
on
d
iti
on
s
(radia
t
io
n
le
vel
s
o
lar,
t
emperature
o
f
s
u
rf
ace
and
the
vo
lta
ge
o
f
m
o
du
le.)
,
or
t
he
ope
ra
ti
n
g
p
ow
e
r
t
ha
t
is
d
e
t
e
r
m
i
ne
d
fro
m
the
s
e
th
re
e
i
npu
ts
d
i
r
ect
ly
a
ff
ect
s
t
h
e
des
i
g
n
an
d
p
erform
ance
of
th
e
system
.
Cur
r
ent
a
n
d
vol
ta
ge
v
aria
ti
o
n
s
up
plie
d
b
y
t
h
e
PV
p
a
n
e
l
w
i
t
ho
u
t
neu
r
al
n
e
t
w
o
rk
a
s
show
n
in
F
i
gure
2
1
.
I
n
F
i
gur
e
2
2
i
s
a
r
c
h
itec
t
ural
s
y
n
o
p
tic
o
f
the
pr
o
p
os
e
d
n
e
u
r
a
l
netw
ork
(3-
10-
2).
In
F
igu
r
e
23
is
ANN
t
r
ai
ni
ng
o
f
rad
i
at
io
n
and
t
e
mp
eratu
r
e
wi
th
M
at
l
a
b/
Simu
l
i
nk
.
I
n
F
i
g
u
re
2
4
i
s
t
he
s
quar
e
d
err
o
r
o
f
t
h
e
ANN
t
r
ai
n
i
n
g
o
f
radi
ati
o
n
an
d
t
e
mp
er
at
u
r
e
wi
t
h
M
a
t
l
a
b
/
S
i
mul
i
n
k
.
I
n
F
i
gure
2
5
i
s
vo
lta
ge
c
urren
t
cha
r
ac
t
e
ris
tic,
I-V
o
f
a
p
h
o
to
vo
l
t
aic
m
odu
le
f
or
t
he
d
iffer
e
nt
te
mpe
r
a
t
ure
s
a
n
d
c
on
sta
n
t
ra
di
a
tio
n
w
i
t
h
n
e
u
ra
l
netw
ork.
I
n
F
i
g
u
re
26
is
v
ol
t
a
ge
p
ow
er
c
har
acter
i
s
t
i
c,
P
-V
o
f
a
phot
ovo
lt
ai
c
modu
l
e
f
o
r
t
h
e
d
if
f
e
re
nt
tem
p
era
t
ur
es a
nd co
ns
ta
nt r
ad
iat
i
on
w
i
t
h
n
eu
ra
l
netw
ork.
Fi
g
u
r
e 21
. Cu
rre
n
t
a
nd
v
olt
a
ge
v
a
r
i
a
t
i
o
n
supp
li
ed
b
y
t
h
e P
V
p
a
n
e
l
w
it
ho
ut
n
e
u
ral
ne
tw
ork
F
i
gure
2
2
. A
rc
hi
tec
t
ura
l
sy
n
o
p
tic
o
f
t
h
e
pr
op
ose
d
n
eura
l
ne
t
w
or
k
(3-10-2)
Evaluation Warning : The document was created with Spire.PDF for Python.