Int
ern
at
i
onal
Journ
al of Ele
ctrical
an
d
Co
mput
er
En
gin
eeri
ng
(IJ
E
C
E)
Vo
l.
10
,
No.
3
,
June
2020
,
pp. 3
074
~
3085
IS
S
N: 20
88
-
8708
,
DOI: 10
.11
591/
ijece
.
v10
i
3
.
pp3074
-
30
85
3074
Journ
al h
om
e
page
:
http:
//
ij
ece.i
aesc
or
e.c
om/i
nd
ex
.ph
p/IJ
ECE
Improve
d
s
trat
egy of an
MPPT
b
ased on th
e s
lid
ing m
ode
co
nt
rol f
or a PV
syst
em
Tao
u
ni A
bder
rahim
1
,
T
ouati
A
bdelw
ahed
2
,
M
ajdoul
Ra
do
u
an
e
3
1
La
bora
tor
y
of
E
le
c
tri
c
al S
y
st
ems
and
Con
trol E
n
gine
er
ing,
fac
u
lty
of
sci
ences Aï
n
Chock,
Hass
an
II
Unive
rsit
y
,
Moro
c
co
2
,3
La
bora
tor
y
of
Struct
ura
l
En
g
in
ee
ring
,
In
te
l
li
ge
nt
S
y
st
ems
and El
e
ct
ri
ca
l
En
erg
y
ENSA
M,
Hass
an
II
Unive
rsit
y
,
Moroc
co
Art
ic
le
In
f
o
ABSTR
A
CT
Art
ic
le
history:
Re
cei
ved
Ma
y
9
, 2
019
Re
vised
Dec
12
,
20
19
Accepte
d
J
an
8
, 2020
The
en
erg
y
pro
duce
d
usi
ng
a
photovol
taic
(P
V)
is
m
ai
nl
y
d
epe
nden
t
on
wea
the
r
f
actors
such
as
te
m
per
a
t
ure
and
solar
ra
dia
ti
on
.
Given
t
he
high
cost
and
low
y
i
el
d
o
f
a
PV
s
y
stem,
it
m
ust
oper
at
e
at
m
axi
m
um
power
poin
t
(MP
P),
which
var
ie
s
a
cc
ording
to
cha
nges
in
lo
ad
and
wea
th
er
con
dit
ions
.
Thi
s
cont
ribu
ti
o
n
pre
sents
an
i
m
prove
d
m
axi
m
um
po
wer
point
tracki
ng
(MP
PT)
cont
rollers of
a
PV
s
y
ste
m
in
var
ious
cl
i
m
at
ic
condi
t
ions.
Th
e
first
i
s
a
slidi
ng
m
ode M
PP
T
tha
t
desig
ned
to
be appl
i
e
d
to
a
buck conv
ert
er in
orde
r
to
ac
h
ie
ve an
opt
imal
P
V a
rra
y
o
utput
voltage
.
T
he
sec
ond
MP
PT
is ba
sed
on
the
in
cre
m
ental
conducta
nc
e
algorithm
or
Pert
urb
-
and
-
Obs
erv
e
al
gor
it
hm
.
It
provide
s
the
output
ref
er
ence
PV
volt
age
to
the
slidi
ng
m
ode
cont
rol
ler
ac
t
ing
on
the
dut
y
c
y
cle
of
th
e
D
C
-
DC
conve
rte
r
.
Sim
ula
ti
on
is
c
a
rrie
d
out
in
Sim
Po
wer
toolbox
of
Matl
a
b/Sim
uli
nk.
Si
m
ula
ti
on
resul
t
s
conf
irm
the
eff
e
ct
iv
ene
s
s
of
the
slidi
ng
m
ode
cont
ro
l
MP
PT
under
th
e
par
amete
r
var
iation envi
ro
nm
ent
s a
nd
sho
wn t
hat t
h
e
con
tr
oll
ers
m
eet its o
bje
c
ti
ves.
Ke
yw
or
d
s
:
Buck
c
on
ver
te
r
In
c
rem
ental
c
on
duct
ance
Pertu
rb an
d
o
bse
rv
e
a
lg
or
it
hm
PV
m
odule
Sli
din
g m
od
e,
MPPT
Copyright
©
202
0
Instit
ut
e
o
f Ad
vanc
ed
Engi
n
ee
r
ing
and
S
cienc
e
.
Al
l
rights re
serv
ed
.
Corres
pond
in
g
Aut
h
or
:
Tao
un
i
A
bd
e
rrahim
,
Lab
or
at
ory
of
Ele
ct
rical
System
s an
d
Co
ntr
ol E
ng
i
neer
i
ng
,
Hassa
n
I
I U
nive
rsity
,
Moh
am
m
adia Schoo
l
of E
ng
i
neer
s
, Ra
bat
,
Mor
occo
,
Em
a
il
:
ta
ou
ni40
@
hotm
a
il
.
com
1.
INTROD
U
CTION
Photo
vo
lt
ai
c
syst
e
m
s
us
e
so
la
r
cel
ls
or
a
group
of
them
to
conver
t
sunli
ght
directl
y
into
el
ect
rici
t
y.
The
powe
r
produce
d
by
ph
otovo
lt
ai
c
syst
e
m
s
dep
e
nds
on
the
ir
ra
diati
on
a
nd
te
m
per
at
ur
e
of
t
he
so
la
r
cel
ls
[1]
.
Hab
it
ually
,
the
energy
produce
d
by
so
la
r
cel
ls
is
us
e
d
to
pro
vide
power
to
a
lo
ad
an
d
the
rem
ai
nin
g
energy
is
store
d
in
batte
ries
.
The
no
nline
ar
cu
rr
e
nt
volt
age
(
I
-
V)
c
harac
te
risti
c
of
P
V
presents
a
m
ajo
r
chall
enge
in
their
us
age
,
w
hi
ch
le
ads
to
a
sing
le
m
axi
m
um
po
we
r
poi
nt
MPP
in
the
powe
r
volt
age
(
P
-
V)
curve
[2,
3]
. T
his
po
i
nt v
a
ries
w
it
h
te
m
per
at
ur
e
and s
olar
ra
diati
on
t
hat m
akes it
harder
to
ac
hieve.
The
m
ajo
r
p
ri
nciple
of
MP
PT
is
to
extra
ct
the
m
axi
m
u
m
avail
able
po
we
r
f
ro
m
PV
m
od
ule
by
m
aking
t
hem
op
e
rate
at
the
m
os
t
eff
ic
ie
nt
vo
lt
age
(m
axim
u
m
po
wer
point)
[4
-
6]
.
M
or
e
over,
the
m
axim
u
m
powe
r
va
ries
accor
ding
to
the
le
vel
of
sol
ar
ra
diati
on
,
the
am
bient
te
m
per
at
ur
e
an
d
the
te
m
per
at
ur
e
of
the
phot
ovolta
ic
cel
ls
[7]
.
The
MPP
T
f
unct
ions
as
an
integrate
d
sy
s
tem
,
including
a
com
bin
at
ion
of
a
hard
war
e
pa
rt
co
ns
ti
tuted
by
a
DC
-
DC
c
onve
rter,
a
nd
the
co
ntr
ol
al
gorithm
act
ing
as
a
softwa
re
par
t
of
the MPPT
. T
hi
s co
m
bin
at
io
n
determ
ines the
value o
f
the
P
V
syst
em
[8]
.
Indee
d,
t
he
r
ol
e
of
t
he
MP
P
T
is
tu
ning
th
e
operati
ng
po
int
of
the
PV
syst
e
m
,
so
as
to
e
xtract
the
m
axi
m
u
m
powe
r
pro
vid
e
d
by
the
P
V
s
yst
e
m
.
The
design
of
a
n
ef
fici
ent
MPPT
s
chem
e
m
us
t
e
ns
ure
accurate
a
nd
r
obus
t
trac
king,
overc
om
ing
the
pr
ese
nce
of
m
od
el
ing
e
rro
rs,
t
he
non
-
li
ne
arit
y
of
t
he
c
urren
t
-
vo
lt
age
c
har
ac
te
risti
c
cur
ves
of
the
phot
ovoltai
c
cel
l
a
nd
the
ra
pid
changes
in
te
m
per
at
ur
e
and
so
la
r
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Impr
oved strat
egy
of an M
PP
T base
d o
n
the
sli
din
g m
ode c
on
tr
ol for
a PV
system
(
T
aoun
i Ab
derr
ah
i
m
)
3075
inso
la
ti
on.
T
o
reach
the
MP
P
,
m
any
pap
er
s
hav
e
bee
n
pre
sented
with
di
ff
e
ren
t
c
ontrol
schem
es
wh
ic
h
ca
n
extract
m
axi
m
um
of
so
la
r
po
wer
as
m
uch
a
s
possible.
So,
seve
ral
m
e
thods
hav
e
bee
n
i
m
ple
m
ented,
nam
ely
trackin
g
syst
em
s
Perturb
an
d
Ob
se
rv
e
(
P&
O)
[9
-
11]
,
incr
e
m
ental
con
du
ct
ance
[12]
,
op
en
ci
rcu
it
volt
age,
an
d
sh
ort
ci
rcu
it
c
urren
t
[
13]
.
I
n
order
to
e
nha
nce
their
perf
or
m
ance
an
d
to
m
ini
m
iz
e
th
e
m
at
erial
,
m
o
dified
te
chn
iq
ues
h
a
ve
b
ee
n pro
po
se
d.
The
P&
O
m
et
hod
is
exte
ns
ively
us
e
d
du
e
to
it
s
si
m
ple
i
m
pl
e
m
entat
ion
.
It
con
sist
s
in
the
per
t
urbati
on
of
the
pan
el
vo
lt
age
per
i
od
i
c
al
ly
;
by
the
var
ia
ti
on
(inc
rea
sing
an
d
dec
re
asi
ng)
of
the
c
yc
li
ca
l
rati
o
of
t
he
c
onve
rter
t
o
reac
h
MPP
T.
In
in
crem
ental
condu
ct
a
nce,
t
he
volt
age
a
nd
cu
rrent
of
t
he
s
olar
pa
nel
are
m
easur
ed
to
au
gur
the
ef
f
ect
of
the
va
riat
ion
of
the
vo
lt
age
on
the
outpu
t
p
ow
e
r.
T
he
MPPT
al
gorithm
s
above
ca
n
be
i
m
ple
m
ented
with
a
c
onve
rg
e
nt
sta
bili
ty
gu
a
ran
te
e
d.
Howe
ver,
in
bo
t
h
m
et
ho
ds
rap
i
d
conve
rg
e
nce
r
equ
i
res
a
la
rge
ste
p
of
pe
rturbati
on
res
ulti
ng
m
or
e
chat
te
ring
a
rou
nd
MPP.
I
n
ad
di
ti
on,
the
ou
t
pu
t
po
wer
cha
nge
re
flect
s
not
on
l
y
the
ef
fect
of
the
va
riat
ion
o
f
the
vo
lt
ag
e
pe
rtu
rb
e
d
be
cause
the po
wer can
be
c
hange
d by
changin
g
m
et
eorolo
gical
con
diti
on
s
[
14
]
.
Othe
rs
MPPT
con
t
ro
l
syst
em
s
based
on
f
uz
zy
log
ic
hav
e
been
int
rod
uce
d
[15
-
18]
.
The
f
uzzy
log
ic
con
t
ro
ll
ers
(
FL
Cs)
ar
e
s
ucces
sfu
ll
y
a
pp
li
ed
in
the
im
ple
m
e
ntati
on
of
MP
PT.
They
offe
r
bette
r
pe
rfor
m
ance.
Their
c
on
tr
ol
is
robust
and
does
not
re
qu
ire
exact
pr
io
r
kn
ow
le
dg
e
of
the
m
at
he
m
atical
m
od
el
of
the
syst
e
m
.
Fu
zzy
-
base
d
MPPT
al
gorith
m
s
su
rp
as
s
th
os
e
us
i
ng
P
&
O
an
d
incre
m
ental
con
duc
ta
nce
beca
us
e
of
thei
r
rob
us
tness
[
19
]
.
A
n
a
rtific
ia
l n
eur
al
netw
ork (
ANN) ha
s also
b
ee
n use
d
wit
h
the
FLC
for
t
he
MPP
T
of
t
he
PV
m
od
ule
[20,
21]
.
The
Fu
z
zy
and
A
NN
-
bas
ed
MPPT
al
gorithm
s
exceed
al
l
oth
ers
bec
ause
of
th
ei
r
r
obus
t
respo
ns
e.
H
owe
ve
r,
t
he
c
ost
of
c
al
culat
ion,
the
com
plexity
of
t
he
i
m
ple
m
entat
ion
an
d
the
te
nd
ency
to
conve
rg
e
to
wa
rd
s
the l
ocal m
axim
a in case o
f
p
a
rtia
l sha
din
g are
the
disa
dv
a
ntage
s.
Lat
el
y,
m
any
research
e
s
are
unde
r
way
in
th
e
fiel
d
of
no
n
-
li
near
co
ntr
ol.
Ma
ny
non
-
lin
e
ar
co
ntr
oller
s
are
al
so
us
e
d
to
track
PV
m
od
ule
MPPs,
includi
ng
the
sli
din
g
m
od
e
and
bac
ks
te
ppin
g
co
ntr
ol
[
22
-
2
7
]
.
A
P
hoto
vo
lt
ai
c
cel
l
is
a
non
-
li
near
.
Sim
i
la
rly
,
the
c
onve
rte
rs
a
re
al
s
o
m
od
el
ed
ass
um
in
g
th
at
they
res
pond
li
near
ly
to
the
var
ia
ble
duty
cy
c
le
.
Howe
ve
r,
they
al
s
o
descr
i
be
nonlinear
be
hav
i
or.
The
rob
us
tne
ss
an
d
sta
bili
ty
that
a
nonli
near
co
nt
ro
ll
er
ca
n
rea
li
ze
wh
e
n
t
rac
king
t
he
MPP
can
not
be
ob
t
ai
ned
by
t
he
ot
her
s
con
t
ro
ll
ers
[29
]
.
In
this
pa
pe
r,
the
sli
di
ng
m
od
e
con
tr
ol
(
SMC
)
is
pro
pose
d
t
o
r
e
gu
la
te
the
du
ty
cy
cl
e
of
DC
-
DC
buck
c
onve
rter
to
br
i
ng
t
he
P
V
m
od
ule
to
ope
ra
te
at
it
s
m
axi
m
u
m
po
we
r
point.
This
co
ntr
ol
strat
eg
y
pr
ese
nt
at
tract
ive
feat
ur
es
,
s
uc
h
go
od
sta
bili
ty
,
si
m
plicity
,
high
acc
ur
acy
,
r
obus
tne
ss
a
nd
a
finite
re
s
pons
e
tim
e
[
29
-
3
1
]
.
T
w
o
MPPTs
ve
rsions
are
de
velo
ped,
the
first
is
a
Sli
ding
Mod
e
C
on
t
ro
l
le
r
MPPT
that
br
i
ngs
the
so
la
r
pan
el
to
w
ork
m
axim
u
m
po
wer
point
Fi
gure
1.
Seco
nd
ly
,
a
n
MPPT
al
gorith
m
P&O
or
i
ncre
m
ental
cond
uctance
,
pro
vid
es
to
a
sli
ding
m
od
e
c
on
trolle
r
t
he
ou
t
put
re
fe
re
nce
P
V
vo
lt
age
.
T
hi
s
co
ntr
oller
act
s
on
the s
witc
h of
t
he bu
c
k
c
onve
r
te
r
f
or the s
olar
p
a
nel outp
ut
volt
age
rejoin
s t
he refe
ren
ce
volt
age F
i
gure
2.
Figure
1. Co
ntr
ol
s
c
hem
e w
it
h
s
li
ding
m
od
e
c
on
t
ro
ll
er M
PP
T
Figure
2. Co
ntr
ol
s
c
hem
e w
it
h
s
li
ding
m
od
e
c
on
t
ro
ll
er
an
d M
PPT
a
lg
ori
th
m
2.
SY
STE
M MO
DELING
2.1.
P
V
ch
ar
ac
teristic
A
so
la
r
cel
l
is
a
basic
com
ponen
t
of
a
P
V
s
yst
e
m
.
Gen
eral
ly
,
on
ly
a
few
watt
s
(2W)
of
powe
r
are
pro
du
ce
d
by
each
s
olar
cel
l
and
c
orres
ponds
t
o
the
open
ci
rcu
it
volt
age
is
in
bet
ween
0.5
a
nd
0.7
V
dep
e
ndin
g
up
on
the
m
at
erial
of
s
olar
cel
l.
Since
this
powe
r
is
insu
ff
ic
ie
nt
to
supp
ly
m
os
t
dev
ic
es,
the
volt
age
m
us
t
be
increa
sed
a
nd
the
refor
e
incr
easi
ng
the
powe
r.
P
V
a
rr
ay
m
odule
co
ns
ist
s
of
num
ber
of
se
rie
s
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
3
,
J
une
2020
:
3
0
7
4
-
3085
3076
par
al
le
l
co
nnec
te
d
cel
ls
as
pe
r
th
e
re
qu
i
red
r
at
ing
of
P
V
a
rray
[33
]
.
S
o,
to
increase
t
he
vo
lt
age
these
cel
ls
are
cl
us
te
red
in
se
r
ie
s an
d
in p
ar
al
le
l t
o
increase t
he
curre
nt.
So
l
ar cell
s p
resen
t
n
at
ur
al
ly
n
onli
near
cha
racteri
sti
cs
I
-
V
an
d
P
-
V,
wh
ic
h
de
pe
nd
on
t
he
s
olar
rad
ia
ti
o
n
an
d
the
te
m
per
at
ure
of
t
he
cel
l.
A
sim
ple
el
ec
tric
al
equ
i
valent
m
od
el
with
sin
gle
diode
is
a
dopt
ed.
As
s
how
n
i
n
Fi
g
ure
3,
t
he
eq
ui
valent
ci
r
cuit
of
a
P
V
ce
ll
has
a
cur
re
nt
sourc
e
(
),
a
diod
e
connecte
d
in
par
al
le
l,
a
series
resist
or
(
)
and
a
pa
r
al
le
l
resist
or
(
)
[34
].
The
e
quat
ion
wh
ic
h reflect
s
the r
el
at
io
n between
the
volt
age a
nd curre
nt
in the PV
m
od
ule can
b
e
writ
te
n
as:
0
.
e
x
p
1
p
v
p
v
S
p
v
p
v
S
p
v
p
h
Tp
V
I
R
V
I
R
I
I
I
a
V
R
(1)
w
he
re
I
ph
is
t
he
c
urre
nt
ge
ner
at
e
d
by
the
incide
nce
of
li
gh
t.
Io
is
t
he
re
ver
se
sat
ur
at
ion
cu
rr
e
nt,
V
T
is
the
j
unct
io
n
th
erm
al
vo
lt
age
(
V
T
=N
s
.
kT/q
)
of
N
S
series
-
c
onnected
s
olar
cel
ls
pe
r
m
odule,
q
is
the
el
ect
ron
charge
(
1.6
0217646×
10
-
19
C)
,
k
is
the
Bolt
z
m
ann
co
ns
ta
nt
(1
.
380650
3×
10
-
23
J/K
),
T
i
s
the
tem
per
at
ur
e
of
the
p
-
n
jun
ct
io
n
i
n
Kelv
in
a
nd
a
is
the
di
ode
ideal
it
y
facto
r.
The
li
gh
t
ge
ner
at
e
d
c
urre
nt
I
ph
an
d
the
sat
ur
at
i
on
current
de
pend
on th
e
tem
per
at
ur
e a
nd s
olar radiat
io
n
that c
an be e
xpresse
d by the
f
ollowi
ng
e
quat
io
n:
,
p
h
p
h
n
i
n
n
G
I
I
k
T
T
G
(2)
w
he
re
I
ph,n
is
the
li
gh
t
ge
ne
rated
c
urren
t
at
the
no
m
inal
conditi
on
,
G
n
is
the
irra
diance
at
ST
C
,
G
is
the s
urface ir
ra
diance
of the
c
el
l and
k
i
is
the
shor
t ci
rcu
it
c
urren
t c
oe
ff
ic
ie
nt.
3
0
0
,
11
e
x
p
g
n
n
n
qE
T
II
T
a
k
T
T
(3)
w
he
re
E
g
is
th
e
band
gap
e
ne
rg
y
of
the
se
m
ic
on
duct
or
(
1.12
e
V
for
th
e
sil
ic
on
),
I
o,n
and
V
oc,n
re
presents
resp
ect
ively
t
he
no
m
inal
sat
ur
at
io
n
c
urr
ent
an
d
t
he
open
ci
rcu
it
volt
age
at
STC,
al
so
T
n
a
nd
T
are
,
resp
ect
ively
,
th
e nom
inal and
the act
ual tem
per
at
ur
e
s in
K
el
vin
.
Figure
3. Eq
ui
valent
c
irc
uit o
f
s
olar
c
el
l
The
P
V
ar
ray
consi
der
e
d
in
this
w
ork
is
th
e
m
on
oc
rista
li
n
10
0W
Sun
powe
r
Fle
xib
le
so
la
r
pan
el
.
The
c
orrespo
ndin
g
el
ect
rical
char
act
e
risti
cs
are
li
ste
d
in
Ta
ble
1.
In
order
t
o
use
the
so
la
r
pa
nel
in
the
f
ollo
wing
si
m
ulati
on
s;
th
e
ide
ntific
at
ion
of
sin
gle
d
i
od
e
m
od
el
pa
r
a
m
et
ers
is
pe
r
form
ed.
The
m
et
ho
d
adopted
to
determ
ine
I
ph
,
I
o
,
has
R
S
and
R
P
is
the
detai
le
d
in
[
35
]
.
I
t
is
a
si
m
ple
and
acc
ur
at
e
m
et
ho
d
that
al
lows
us
i
ng
data
s
uppli
ed
by
the
m
anu
fact
ur
e
r
to
ca
lc
ulate
these
pa
ram
et
ers.
The
res
ults
obta
in
ed
are
sh
ow
n
in
Ta
bl
e
2.
Fi
gure
4
a
nd
Fig
ur
e
5
s
how
t
he
I
-
V
c
urves
an
d
P
-
V
re
la
te
d
to
so
la
r
pa
nel
us
e
d
f
or
di
ff
ere
nt
Tem
per
at
ur
e
L
evels
a
nd
di
ff
e
ren
t
ir
ra
diance
le
vels.
T
he
P
-
V
c
urves
pass
t
hro
ugh
m
axi
m
um
s
in
w
hich
t
he
P
V
gen
e
rates
m
or
e
powe
r
tha
n
ot
her
point
s.
T
he
re
are
m
axi
m
um
po
we
r
po
i
nts
for
eac
h
cu
r
ve
.
The
refore
,
duri
ng
on
e
d
ay
t
he
m
axim
u
m
p
ow
e
r po
i
nt of a
P
V v
aries acco
r
ding
to
the
irr
a
dia
nc
e an
d
te
m
per
at
ur
e.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Impr
oved strat
egy
of an M
PP
T base
d o
n
the
sli
din
g m
ode c
on
tr
ol for
a PV
system
(
T
aoun
i Ab
derr
ah
i
m
)
3077
Table1
. Solar
a
rr
ay
d
at
asheet
p
aram
et
ers
I
m
p
5
.7A
V
m
p
1
7
.7V
P
m
p
1
0
0
.89
W
Isc
6
.1A
Vo
c
2
1
.7V
Ns
32
Ki
0
.01
5
%/°C
Kv
-
0
.34
%/°C
Table
2.
Param
et
ers
o
btaine
d of
t
he
s
olar
p
a
nel
So
lar
p
an
el para
m
eter
Valu
e
a
1
.3
R
S
(
)
0
.16
2
7
2
6
2
9
R
P
(
)
2
9
9
.42
I
O
(
A)
9
.34
2
9
e
-
9
Iph
(
A)
6
.10
3
3
1
(a)
(b)
Figure
4. (a
)
C
urren
t
–
v
oltage
and (
b) P
ow
e
r
–
v
oltage
c
urve
s of PV
a
rr
ay
a
t 25°C
(a)
(b)
Figure
5. (a
)
C
urren
t
–
volt
age
and
(
b) P
ow
e
r
–
v
oltage
c
urve
s of PV
a
rr
ay
a
t 100
0W
/m
2
2.2.
Buck
con
vert
er
model
ing
The
ave
rag
e
d
m
od
el
of
the
bu
ck
c
onver
te
r
is
ob
ta
ine
d
by
app
ly
in
g
the
f
undam
ental
law
s
gove
r
ning
the
f
unct
ion
i
ng
of
the
syst
em
.
The
co
nv
er
te
r
sc
hem
e
is
sh
ow
n
in
Fig
ure
6.
The
dyna
m
ic
equ
at
ion
s
of
this
conve
rter ca
n be e
xpresse
d
a
s:
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
3
,
J
une
2020
:
3
0
7
4
-
3085
3078
{
1
=
2
−
2
=
1
1
2
−
1
1
(4)
w
he
re
=
[
1
2
]
=
[
]
,
=
as
a
n
e
qu
i
valent
loa
d
connecte
d
to
P
V
pan
el
a
nd
is
duty
cy
cl
e
.
The
bu
c
k
c
onve
rter
is
powe
re
d
by
the
s
olar
pan
el
.
Th
e
in
put
volt
age
var
i
es
co
ntinuo
u
sl
y
with
the
wea
ther.
Ther
e
f
or
e,
the
du
ty
cyc
le
s
hould
c
ha
ng
e
to
foll
ow the m
axim
u
m
p
ow
er
po
int o
f
the
ph
otovo
lt
ai
c ar
ray.
Figure
6.
S
c
he
m
e o
f
t
he DC
-
DC
b
uc
k
c
onve
rter
3.
SLIDI
NG M
ODE
CONTR
OL
DESI
GN:
Eq
uation (
4) c
an be
wr
it
te
n
a
s foll
ow
:
̇
=
(
)
+
(
)
+
(5)
w
he
re
(
x
)
=
(
0
x
2
R
pv
C
1
)
,
g
(
x
)
=
(
x
2
L
−
x
1
C
1
)
a
nd
=
(
−
0
)
Accor
ding
to
the
P
V
c
urves
sh
ow
n
for
var
i
ou
s
ra
diati
on
le
vels
in
Fig
ure
4
(
b
)
,
the
m
axim
u
m
po
we
r
point
MPP ex
pressi
on is c
har
act
e
rized
by:
∂
P
pv
∂
V
pv
=
∂
(
V
pv
I
pv
)
∂
V
pv
=
∂
I
pv
∂
V
pv
V
pv
+
I
pv
=
0
(6)
This e
xpressi
on is c
hosen
as t
he
sli
di
ng surfa
ce f
or
t
he
SM
C M
PPT
S
(
x
)
=
∂
P
pv
∂
V
pv
(7)
An
al
yz
in
g
t
he
relat
io
ns
hi
p
be
tween
t
he
sta
te
of
S
(
x
)
an
d
the
posi
ti
on
of
th
e
oper
at
ing
po
int
on
Figure
(
7
)
,
it
is n
ecessa
ry to
m
ake th
e c
hoic
e of the
foll
ow
i
ng contr
ol law:
=
{
0
(
)
≥
0
1
(
)
<
0
(8)
The
eq
uatio
n
of
the
ph
oto
v
oltai
c
curre
nt
is
giv
en
i
n
(
1)
.
W
he
n
we
n
eglect
the
eff
ect
of
the r
esi
sta
nces
RS an
d
RP
, th
e
curre
nt Ip
v be
com
es:
=
ℎ
−
(
exp
(
)
−
1
)
(9)
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Impr
oved strat
egy
of an M
PP
T base
d o
n
the
sli
din
g m
ode c
on
tr
ol for
a PV
system
(
T
aoun
i Ab
derr
ah
i
m
)
3079
So
,
S(x
)
ca
n be
w
ritt
en
as:
(
)
=
+
=
ℎ
−
(
1
+
)
exp
(
)
+
(10)
The
e
quivale
nt
contr
ol in
pu
t
u
eq
is o
btained
from
the cond
it
io
n:
̇
(
)
=
0
̇
(
)
=
(
)
=
−
(
+
2
)
exp
(
)
̇
=
0
⇒
̇
=
0
is t
he
n deduct
ed fr
om
(
4)
:
1
=
2
1
=
ℎ
≠
0
(11)
The
e
quivale
nt
com
m
and
is
eff
ect
iv
e
on
c
e
the
tra
j
ect
or
y
of
the
syst
e
m
state
reache
s
the
sli
di
ng
su
r
face
(
sli
ding
m
od
e).
A
rea
chin
g
la
w
m
us
t
be
ex
pr
es
sed
to
bri
ng
the
tra
j
ect
ory
to
the
s
li
din
g
s
urface
duri
ng
the con
verge
nc
e m
od
e.
The
sim
plest f
or
m
that can ta
ke
is as
foll
ow
s:
=
−
1
.
(
(
)
)
w
he
re
k
1
re
pr
e
s
ent
a
po
sit
ive
c
on
sta
nt
of
t
he
synthesis
disco
ntinuo
us
la
w.
The
sli
di
ng
m
od
e
c
on
t
ro
l
la
w
can
be writt
en
as
bel
ow
:
1
=
1
+
(12)
To veri
fy co
nv
erg
e
nce i
n
fi
nite t
i
m
e, a candidate Ly
ap
unov
fun
ct
io
n
is
de
f
ined by:
(
)
=
1
2
2
(
)
Its d
y
nam
ic
, r
el
at
ive to tim
e, i
s g
i
ven b
y:
̇
(
)
=
(
)
̇
(
)
<
0
To
e
nsure
con
verge
nce in a
f
init
e tim
e, w
e
can c
ho
ise
:
̇
(
)
=
−
(
+
2
)
exp
(
)
In case:
S
(
x
)
>
0
Wh
e
nce
S
(
x
)
S
̇
(
x
)
<
0
In case:
S
(
x
)
<
0
and
<
0
⇒
̇
(
)
>
0
Wh
e
nce
S
(
x
)
S
̇
(
x
)
<
0
It is d
em
on
stra
te
d
that t
he
sys
tem
is asym
pto
ti
cal
ly
stable.
4.
SLIDI
NG M
ODE
CONTR
OL
LE
R
WIT
H MPPT
ALGORI
THM
In
this
sect
io
n,
the
desig
n
of
a
sli
din
g
m
od
e
con
tr
oller
(
S
MC
)
us
in
g
a
re
fer
e
nce
volt
ag
e
prov
i
ded
by
an
MPP
T alg
or
it
h
m
w
il
l be
presented
.
F
or th
is co
ntro
l,
the
s
li
din
g su
rf
ace
beco
m
es:
(
)
=
2
−
(13)
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
3
,
J
une
2020
:
3
0
7
4
-
3085
3080
w
he
re
V
pv
ref
is
vo
lt
age
pr
ov
i
ded
by
an
MP
PT
a
lgorit
hm
(P
erturb
&
O
bs
er
ve
or
I
ncr
em
ental
cond
uctance
).
The
c
ontrol la
w
f
or this
ca
se
is desc
ribe
d by:
{
0
(
)
<
0
1
(
)
>
0
(14)
The
e
quivale
nt
contr
ol in
pu
t
u
e
q2
is o
btained
from
the cond
it
io
n:
̇
(
)
=
0
̇
(
)
=
̇
2
−
̇
=
0
̇
(
)
=
1
1
2
−
2
1
1
−
̇
=
0
⇒
2
=
2
1
−
1
1
̇
2
=
−
1
̇
ℎ
≠
0
a
reac
hing law
is t
aken
:
=
2
.
(
(
)
)
w
he
re
k
2
is a
pos
it
ive co
nst
ant.
The
sli
di
ng m
od
e c
on
t
ro
l l
a
w ca
n be
wr
it
te
n as bel
ow
:
2
=
2
+
(15)
4.1.
M
axim
u
m po
w
er
refer
ence v
olt
age
The
r
ol
e
of
the
MPPT
al
go
rit
hm
is
to
cal
cul
at
e
the
op
ti
m
a
l
ou
tp
ut
volt
age
of
the
ph
otov
oltai
c
pan
el
cal
le
d
Vpv
ref
wh
ic
h
al
lows
the
PV
syst
e
m
to
fu
ncti
on
at
it
s
MPP.
Fo
r
t
hat
two
well
-
kn
own
al
gorithm
s
per
t
urb
-
a
nd
-
O
bs
er
ve
al
gorith
m
w
it
h
adap
ti
ve
step a
nd Inc
r
e
m
enta
l C
onduct
ance alg
or
it
hm
are
te
ste
d.
4.1.1
Adap
tiv
e P&O
a
lg
orit
hm
The
al
gorithm
ad
opte
d f
or ad
aptive Pe
rtu
r
b & Obse
rv
e
is
de
scribe
d
as
foll
ow
s:
{
(
)
=
(
−
ℎ
)
+
&
.
(
Δ
Δ
)
Δ
=
(
)
−
(
−
ℎ
)
Δ
=
(
)
−
(
−
ℎ
)
&
=
3
.
|
Δ
|
w
he
re
k
P&
O
an
a
dap
t
ive
ste
p
is
de
pe
nd
e
nt
on
t
he
cha
nge
of
the
PV
po
wer
a
nd
h
is
the
ste
p
ti
m
e
.
T
he
posit
ive
const
ant
gain
k
3
is ad
justed t
o o
btain th
e
best
fun
ct
io
ning
of t
he
MPP
T.
4.1.2.
Increm
ent
al c
onduc
t
an
ce
algorit
h
m
The
al
gorithm
ad
opte
d f
or Inc
rem
ental
Con
duct
ance is
d
es
c
ribe
d
as
f
ollows:
{
(
)
=
(
−
ℎ
)
+
.
(
I
V
+
Δ
I
Δ
V
)
Δ
=
(
)
−
(
−
ℎ
)
Δ
=
(
)
−
(
−
ℎ
)
=
4
.
|
Δ
|
w
he
re
k
I
n
c
an
a
dap
t
ive
ste
p
is
de
pe
nd
e
nt
on
the
change
of
the
PV
powe
r
a
nd
h
is
the
ste
p
ti
m
e.
The
posit
iv
e
const
ant
gain
k
4
is ad
justed t
o o
btain the
best
s
uitable
per
for
m
ances of th
e
MPPT.
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Impr
oved strat
egy
of an M
PP
T base
d o
n
the
sli
din
g m
ode c
on
tr
ol for
a PV
system
(
T
aoun
i Ab
derr
ah
i
m
)
3081
5.
SIMULATI
O
N
RESU
LT
S
To
ve
rify
the
validit
y
and
ef
fecti
ven
es
s
of
the
pro
po
s
ed
s
li
din
g
m
od
e
con
t
ro
ll
ers
,
sim
ulati
on
s
a
re
dr
a
w
n
to
il
lust
rate
their
perform
ance.
The
c
on
t
ro
ll
ed
syst
em
is
a
bu
c
k
dc
-
dc
c
onve
rter
with
the
c
om
po
ne
nt
s
ind
ic
at
ed
in
T
able
3.
T
he
pa
ram
et
ers
of
the
desig
ned
c
ontr
ollers
are
c
ho
s
en
as
re
por
te
d
in
Table
4.
In
this
stud
y,
the
m
on
o
c
rysta
ll
ine
100W
Sun
powe
r
Fle
xib
le
s
ola
r
pan
el
is
sel
e
ct
ed
as
the
source
of
phot
ovoltai
c
energy.
T
he
num
ber
of
s
olar
pa
nels
ch
os
e
n
to
pro
vi
de
50
0W
m
axi
m
u
m
powe
r
is
5
m
odules
c
onnect
ed
in
series.
T
he
sim
ula
ti
on
tim
e
is
cho
se
n
s
hort
to
sho
w
the
respo
ns
e
tim
e
of
the
stud
ie
d
c
ontrolle
rs
.
Irrad
ia
nce
pro
f
il
e
is
sh
own
in
F
ig
ur
e
7
(
f
)
and
the
F
ig
ures
7
(
a
-
c
)
repr
esent
resp
e
ct
iv
el
y
the
evo
luti
on
of
vo
lt
age
,
c
urre
nt
an
d
powe
r
of
the
P
V
ar
ray
in
res
ponse
to
the
ir
rad
i
ance
var
ia
ti
on
us
i
ng
sli
ding
m
od
e
con
t
ro
ll
er
SM
C
M
PPT.
The
volt
age,
c
urre
nt
an
d
powe
r
of
t
he
P
V
a
rr
a
y
us
in
g
the
S
MC
with
Pe
rturb
a
nd
Ob
se
r
ve
MPP
T
al
gorithm
a
re
il
lustrate
d
i
n
F
ig
ures
8
(
a
-
c)
a
nd
t
hose
corres
pondin
g
to
the
SMC
with
increm
ental
co
nductance
are
in
Fi
gures 9
(a
-
c
)
.
The
loa
d
volt
ages
co
rr
es
po
nd
ing
to
the
dif
fe
ren
t
co
ntr
ollers
are
sh
ow
n
in
Figure
7
(
d
)
,
F
igure
8
(
d)
and
Fig
ure
9
(
d
)
.
T
he
var
ia
ti
on
s
of
du
ty
cy
cl
e
are
dep
ic
te
d
in
Fi
gure
7
(
e
)
,
Fig
ure
8
(
e
)
a
nd
Fig
ure.
9
(
e
)
.
Fr
om
si
m
ulatio
ns,
the r
ob
us
tness o
f
dev
el
oped
c
ontr
ollers
is
ob
se
rv
e
d.
T
he
m
axi
m
u
m
p
ow
e
r
is
achie
ve
d
with
excell
ent
accur
acy
seen
the
va
riat
ion
of
irra
diance.
F
r
om
th
e
va
r
i
ou
s
s
im
ul
a
t
i
on
r
e
s
ul
t
s
,
w
e
c
a
n
c
on
c
l
u
de
t
ha
t
t
he
s
l
i
di
ng
-
m
ode
a
nd
t
he
i
nc
r
em
e
nt
a
l
-
c
on
du
c
t
a
nc
e
c
on
t
r
o
l
m
od
e
s
ha
ve
f
a
s
t
e
r
c
on
ve
r
g
e
nc
e
t
im
e
s
t
o
r
e
a
c
h
t
he
sm
a
ll
e
r
M
PP
po
w
e
r
po
i
nt
c
om
pa
r
e
d
t
o
t
he
pe
r
t
ur
be
d
a
n
d
o
bs
e
r
ve
d
a
l
g
or
i
t
hm
.
Ta
bl
e3.
Par
ameters
of
t
he
b
uck
c
o
nver
te
r
C
1
660µF
C
2
100µF
L
120µH
R
L
1
.8Ω
Table
4.
c
ontr
ol
le
rs
p
a
ram
et
er
s
k
1
10
k
2
1
k
3
0
.00
1
k
4
4
Figure
7
.
(
a
-
c)
Re
sp
ect
ively
volt
age,
c
urre
nt
and p
ower
of P
V
a
rr
ay
, (
d)
l
oa
d vo
lt
age
Evaluation Warning : The document was created with Spire.PDF for Python.
IS
S
N
:
2088
-
8708
In
t J
Elec
&
C
om
p
En
g,
V
ol.
10
, No
.
3
,
J
une
2020
:
3
0
7
4
-
3085
3082
Figure
7
.
(e
) D
uty
cy
cl
e o
f
bu
ck
c
on
t
ro
l a
nd
(f) ir
rad
ia
nce
pro
file
u
si
ng
S
MC
MPPT (
c
onti
nu
e
)
Figure
8
.
(
a
-
c)
Re
sp
ect
ively
volt
age,
c
urre
nt
and
p
owe
r
of
PV
a
rr
ay
,
(d)
load
volt
age
(e
)
duty
cyc
le
o
f
bu
c
k
con
t
ro
l a
nd (f)
irrad
ia
nce
pro
f
il
u
sin
g
SMC
P
&O
MP
PT a
l
gorithm
Evaluation Warning : The document was created with Spire.PDF for Python.
In
t J
Elec
&
C
om
p
En
g
IS
S
N: 20
88
-
8708
Impr
oved strat
egy
of an M
PP
T base
d o
n
the
sli
din
g m
ode c
on
tr
ol for
a PV
system
(
T
aoun
i Ab
derr
ah
i
m
)
3083
Figure
9
.
(a
-
c)
R
espe
ct
ive
l
y
volta
ge,
cur
ren
t and
power
of
PV
a
r
r
a
y
,
(d
)
L
o
ad
vo
ltage
(
e)
du
t
y
c
y
cle
of
bu
ck cont
ro
l
and
(f)
irr
adi
an
ce pr
ofil
using
SM
C
i
ncr
emental cond
uct
an
ce
MP
PT
a
lgori
t
hm
6.
CONCL
US
I
O
N
In
this
pa
per
,
an
im
pr
ov
e
d
m
axi
m
u
m
po
w
er
point
trac
kin
g
(MPPT
)
co
ntr
olle
rs
of
a
PV
syst
em
in
var
i
ou
s
cl
im
a
tic
co
nd
it
io
ns
is
pr
ese
nted
.
T
he
co
ntr
ol
ba
sed
upon
the
sli
ding
m
od
e
a
ppr
oa
ch
for
P
V
syst
e
m
is
synthesiz
ed
.
Using
this
fa
st
and
r
obus
t
con
tr
ol
syst
e
m
,
su
it
able
MPPT
pe
rform
ances
are
obta
ined
.
Indee
d,
the
des
ign
of
MPPT
c
o
nt
ro
l
with
sli
ding
m
od
e
fo
r
photov
oltai
c
syst
e
m
based
a
sta
ti
c
bu
ck
co
nverte
r
represe
nt
one
of
the
m
os
t
eff
e
ct
ive
ways
to
optim
iz
e
PV
syst
e
m
s
with
high
perform
ances.
The
key
no
velty
of
the
con
t
ro
l
al
gorithm
pr
esent
ed
li
es
in
the
way
dynam
ic
of
di
ff
e
ren
t
m
e
thods
to
desi
gn
sli
ding
m
od
e
con
tr
ol
with
MP
PT.
A
c
om
par
ison
betwee
n
a
s
li
din
g
m
od
e
con
t
ro
ll
er
MP
PT,
a
co
ntr
ol
le
r
with
inc
re
m
ental
cond
uctance
al
gorithm
and
a
one
with
pe
rturb
a
nd
obser
ve
al
gorithm
i
s
en
de
d.
The
three
co
ntr
oller
s
ar
e
eff
ic
ie
nt
a
nd
r
obus
t.
Ra
pid
it
y
to
reach
t
he
m
axi
m
u
m
power
point
is
the
qual
it
y
of
the
two
ci
te
d
above
com
par
ed
t
o
th
e Pert
urb
a
nd
Ob
se
r
ve
al
go
rithm
.
REFERE
NCE
S
[1]
N.
Ham
rouni,
M
.
Jraidi,
and
A.
Chéri
f,
"S
ola
r
ra
dia
ti
on
and
ambient
t
empera
tur
e
eff
ects
on
th
e
pe
rform
anc
es
of
PV
pu
m
ping
S
y
stem,"
Journal
of Re
newabl
e
Ene
rgi
e
s
,
vol. 11, no. 1,
pp.
95
–
106
,
200
8.
[2]
S.
Prem
rude
epr
e
ec
ha
cha
rn
and
N.
Pata
n
api
rom
,
"S
ola
r
-
arr
a
y
m
odel
li
ng
and
m
axi
m
um
power
point
tr
ac
king
usin
g
neur
al ne
tworks,
" i
n
2003
IEEE Bologna Power T
ec
h
Conf
ere
nc
e
Proceedi
ngs,
B
ologna
,
It
aly
,
vo
l.
2
,
p
p
.
5
,
2003
.
Evaluation Warning : The document was created with Spire.PDF for Python.