Indonesian Journal of
Electrical
Engineer
ing and
Computer Science
V
o
l.11
,
No
.1
, Ju
ly 20
18
, pp
. 16
1
~
16
8
ISSN: 2502-4752,
DOI: 10.
11591/ij
eecs.v11
.i1.pp161-168
1
61
Jo
urn
a
l
h
o
me
pa
ge
: http://iaescore.c
om/jo
urnals/index.php/ijeecs
Sustainable Environmental Econ
omic Dispatch Optimization
with Hybrid Metaheuristic Modification
M.
R.
M. Ridz
uan
,
E.E.
H
a
s
s
an,
A
.
R
.
Abd
u
llah, A.
F.
A. Kadir
Faculty
of Electr
ical
Eng
i
neering
,
Universiti Tekn
ikal Mala
y
s
ia M
e
lak
a
, Hang
Tuah Jay
a
, Durian
Tunggal, 76100,
Mala
y
s
ia
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
Ja
n 19, 2018
Rev
i
sed
Mar
23
, 20
18
Accepted Apr 21, 2018
Today
’
s Econo
mic Dispatch
(ED) soluti
ons ar
e featured with environmental
obligat
ions. He
nce,
the signi
fi
cant ob
jec
tive
f
unctions con
t
rib
u
te to
cost
m
i
nim
i
zation
,
lo
wer em
ission and less total s
y
s
t
e
m
losses. As an alt
e
rnat
ive
,
New Meta Heu
r
istic Evo
l
ution
a
r
y
Progr
amming (NMEP)
technique was
proposed to op
timize
the
individual
ED prob
lem categorized
as Single
Object
ive
Envir
onm
ental E
c
ono
m
i
c Dispatch (S
OEELD), d
e
velo
ped from
an
integr
ation of or
iginal Me
ta Heu
r
istic Evo
l
ution
a
r
y
Program
m
i
ng (Meta-EP)
with Artifici
a
l I
m
m
une
Sy
st
em
(AIS) with new
arrangem
e
nt in t
h
e m
u
tation
and clon
ing processes. The
comparativ
e analy
s
is was
conducted
b
e
tween th
e
original Meta-EP and classical met
hod of
Hadi Saadat to
verif
y
th
e
performance of
NMEP
method. Each pa
rticular
objective fun
c
tio
n identif
ied
the best possible outcomes through the
NMEP method. Th
e simulations wer
e
conducted using
MATLAB programming wh
ich tested both stan
dard IEEE
26 and 57
bus s
y
stems.
K
eyw
ords
:
Artificial In
tellig
en
ce
Econom
ic Dispatch
Mu
lti Obj
ective Fun
c
tion
Copyright ©
201
8Institute of
Ad
v
anced
Engineeri
ng and Scien
c
e.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
M.R.M. Rid
z
uan
,
Facu
lty of Electri
cal Engineering,
Un
i
v
ersiti Tekn
ik
al Malaysia Melak
a
,
H
a
ng
Tu
ah
Jay
a
, Dur
i
an
Tun
g
g
a
l,
7
610
0, Malaysia.
Em
a
il: m
o
h
a
mad
r
ad
zi199
2@g
m
ai
l.co
m
1.
INTRODUCTION
Po
wer sy
st
em
opt
i
m
i
zati
on i
s
a vi
t
a
l
st
udy
for a
n
o
p
t
i
m
al
po
we
r o
p
erat
i
o
n t
o
p
r
ovi
de s
m
oot
h an
d
sust
ai
na
bl
e l
o
a
d
dem
a
nd [
1
]
.
The ri
se
of e
n
e
r
gy
dem
a
nd a
n
d i
n
s
u
f
f
i
c
i
e
nt
o
f
ene
r
gy
res
o
ur
ces are re
qui
re
d f
o
r
q
u
a
lity an
d
secu
red
d
i
sp
atch
[2
]. A
well-coord
i
n
a
ted
a
n
d
op
ti
m
i
zed
p
o
wer syste
m
o
p
e
rat
i
o
n
h
e
lp
in
sati
sfyin
g
Eco
nom
i
c
Di
spat
ch
(E
D)
am
on
g
use
r
s
of
p
o
we
r
net
w
or
ks
. He
nce,
st
u
d
i
e
s nee
d
t
o
be c
o
n
d
u
ct
ed i
n
or
der
t
o
anal
y
ze an
d
de
vel
o
p
new
t
o
ol
s so
t
h
at
t
h
e
o
p
t
im
i
zat
i
on i
ssu
es i
n
E
D
c
o
ul
d
be
o
v
erc
o
m
e
.
Basically,
th
e p
r
i
n
cip
a
l ob
j
e
ctiv
e o
f
lo
ad
d
i
sp
atch
is to
m
i
n
i
m
i
ze
th
e to
tal
fu
el co
st wh
ile satisfyin
g
the requirem
ents of som
e
important
o
p
erat
i
onal
pa
ram
e
t
e
r
s
. In t
o
day
’
s e
nvi
ronm
ent, efficient load dis
p
atch
r
e
qu
ir
es
no
t on
ly to
sch
e
du
l
e
th
e pow
er
gen
e
r
a
tion
at
the least cost but also to
c
o
n
s
i
d
er ot
he
r per
f
o
r
m
a
nce
fact
or
s
t
o
be o
p
t
i
m
i
zed
i
n
p
o
w
er fl
o
w
ov
er t
h
e
net
w
o
r
ks. The obl
i
g
at
i
o
n
of
soci
al
at
t
e
n
t
i
ons has
i
n
fl
u
e
nc
e
d
t
h
e red
u
ct
i
o
n o
f
ener
gy
co
nse
r
vat
i
o
n an
d p
o
l
l
u
t
i
on em
i
ssi
on pr
o
duce
d
by
po
we
r pl
ant
s
[
3
]
.
Hence, t
h
e
t
o
t
a
l
co
st fu
n
c
tion
alo
n
e
is no
longer
su
itab
l
e as t
h
e m
a
in
f
o
cu
s
in
op
ti
m
i
zin
g
th
e ED
pr
ob
lem
s
. I
n
o
r
d
e
r
to r
e
du
ce
pol
l
u
t
i
o
n as a
resul
t
o
f
el
ect
ri
cal
po
wer
gene
rat
i
o
n, m
i
ni
m
i
zat
i
on on
em
i
ssi
on sh
o
u
l
d
be ad
de
d
t
o
t
h
e
ob
ject
i
v
e f
u
nct
i
on o
f
ED w
h
i
c
h i
s
gener
a
t
i
o
n cost
m
i
nim
i
zat
i
on [4]
.
H
o
w
e
ver
,
ED p
r
o
b
l
e
m
s
are al
so sub
j
ect
to
th
e op
eration
a
l con
s
train
t
s an
d
security c
r
iteria o
f
a
power system
so that the
secure
d and ec
onom
ic loads
are dispatched equally.
Po
wer sy
st
em
ope
rat
i
o
n i
s
ge
t
t
i
ng m
o
re cha
l
l
e
ngi
n
g
d
u
e t
o
t
h
e l
a
r
g
e n
u
m
ber of va
ri
ab
l
e
s wo
rki
n
g
to
g
e
th
er with
u
n
c
ertain
p
a
rameters so
, th
e math
e
m
atic
a
l
s
o
lu
tion
s
fo
r it is b
eco
m
i
n
g
mo
re co
m
p
licate
d
[5
].
So
lu
tion
s
to
po
wer system
p
r
ob
lem
s
o
f
ten
in
vo
lv
e
so
l
v
ing
o
p
tim
izat
io
n p
r
ob
lem
s
in
wh
ich
o
b
j
ective and
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
250
2-4
7
5
2
I
ndo
n
e
sian
J Elec Eng
& Com
p
Sci, V
o
l. 11
,
No
.
1
,
Ju
ly
20
18
:
1
61–
168
16
2
con
s
t
r
ai
nt
fo
r
m
ul
at
i
ons
a
r
e no
n
-
di
f
f
ere
n
t
i
a
bl
e
a
n
d
res
u
l
t
e
d
i
n
no
nl
i
n
ea
r sol
u
t
i
o
ns. T
h
u
s
,
m
a
ny
st
udi
es have
been
co
n
duct
e
d t
o
o
v
e
r
com
e
com
p
l
i
cat
ed op
t
i
m
i
zat
i
on p
r
o
b
l
em
i
n
po
we
r s
y
st
em
operat
i
o
n.
Ove
r
t
h
e l
a
st
20 y
ears
,
m
o
st
opt
i
m
i
zat
i
on t
echni
q
u
es ha
ve bee
n
cat
eg
ori
z
e
d
i
n
t
o
t
h
r
ee di
ffe
rent
categories nam
e
ly conve
ntional
m
e
thods,
int
e
lligent searches and fuzzy set a
pplication [6]. As re
porte
d
in a
st
udy
[
7
]
,
t
h
e Gra
d
i
e
nt
base
d co
nve
nt
i
o
nal
appr
oac
h
es s
u
ch as Ne
wt
o
n
M
e
t
h
o
d
s, l
i
n
ear pr
o
g
ram
m
ing a
n
d
qua
d
r
at
i
c
pr
o
g
r
am
m
i
ng m
a
y
resul
t
i
n
p
o
o
r
s
o
l
u
t
i
o
ns s
o
l
v
i
n
g p
r
obl
em
s wh
i
c
h are
n
o
n
-
co
nve
x,
n
o
n
-c
ont
i
n
u
o
u
s
an
d
h
a
v
e
h
i
gh
ly n
o
n
-lin
ear so
lu
tion
s
. Altern
ativ
ely, th
e meta-heuristic approaches
a
r
e
i
n
t
r
od
uce
d
ai
m
i
ng t
o
optim
ize
their chose
n
object
ive functions,
hence
provi
di
ng
globally optim
a
l solutions
[8]. Recently, new
tech
n
i
qu
e
b
a
sed
on
i
m
m
u
n
ity
alg
o
r
ith
m
,
n
a
mely
th
e Artificial I
m
m
u
n
e
Syste
m
(AIS) has b
een
im
p
l
e
m
en
ted
for so
lv
i
n
g ED prob
lem
s
in
o
r
d
e
r to
m
i
n
i
m
i
z
e
th
e
fu
el
co
st
g
e
n
e
ration
with
co
nsid
erati
o
n
o
f
so
m
e
con
s
train
t
s
[9]. T
h
ere
f
ore
,
recent studi
e
s ar
e ins
p
ire
d
to m
e
rge conve
n
tional
methods and a
d
vance
d
optim
ization
tech
n
i
qu
es
fo
r b
e
tter
an
d
faster o
p
tim
iza
tio
n
ap
pro
ach
es.
Th
is stu
d
y
in
ten
d
e
d
to
in
tro
d
u
ce a n
e
w
h
e
uristic alg
o
r
ithm wh
ich
was
an
i
m
p
r
ov
em
e
n
t to
th
e Meta
Heu
r
i
s
t
i
c
Evol
ut
i
ona
ry
Pr
og
r
a
m
m
i
ng (NM
E
P) t
echni
que
. The p
r
o
p
o
sed t
echni
que
was i
m
pl
em
ent
e
d t
o
sol
v
e
eco
no
m
i
c an
d env
i
ron
m
en
tally co
n
s
train
e
d p
r
ob
lem
s
u
til
izin
g
sing
le
ob
j
ective fun
c
tio
n. In
add
itio
n, th
e
per
f
o
r
m
a
nces of t
h
e ne
wl
y
devel
o
ped t
ech
n
i
que we
re com
p
are
d
wi
t
h
t
h
at
pro
v
i
d
e
d
by
t
h
e M
e
t
a
-EP an
d AI
S
al
on
g wi
t
h
B
a
se t
echni
q
u
e.
The be
st
sol
u
t
i
ons
were i
d
ent
i
fi
ed base
d o
n
t
h
e m
i
nim
u
m
tot
a
l
gene
rat
i
o
n
cost
,
least to
tal p
o
l
l
u
tio
n and
sm
al
lest to
tal system
lo
sses.
2.
R
E
SEARC
H M
ETHOD
2.
1. E
c
on
omi
c
Di
sp
atc
h
2.
1.
1. Ob
jecti
v
e
F
uncti
on
The
o
v
eral
l
res
earch
m
e
t
hodo
l
ogy
i
n
v
o
l
v
e
d
i
n
t
h
e
E
nvi
ro
n
m
ent
a
l
Econ
o
m
i
c
Load
Di
s
p
at
ch (
EEL
D)
was classified
in
to
fo
ur stag
es. Th
e
first task
was t
o
ach
iev
e
th
e obj
ective o
f
th
e st
u
d
y
wh
ich
was t
o
estab
lish
a new t
e
c
hni
q
u
e pa
rt
i
c
ul
arl
y
t
o
sol
v
e
EEL
D o
p
t
i
m
i
zat
i
on pr
obl
em
. The
researc
h
ap
p
r
oach
was t
o
d
e
si
gn a
new
o
p
t
i
m
i
zat
ion t
e
c
hni
que t
a
ki
n
g
s
o
m
e
i
n
spi
r
at
i
o
n f
r
om
t
h
e M
e
t
a
-EP
m
u
t
a
t
i
on st
rat
e
gy
. T
h
e de
vel
opm
ent
also
in
cl
u
d
e
d
i
d
en
tifying
su
it
ab
le obj
ectiv
e
fun
c
tion
s
wh
ich
were sign
ifi
can
t to
EEL
D
pr
o
b
l
e
m
al
ong
wi
t
h
so
m
e
co
n
s
trai
nts as d
i
scu
ssed in
th
e fo
llowin
g
section
[10]. In orde
r t
o
achieve t
h
e
re
search objective, the
devel
opm
ent of the ne
w si
ngle objective
technique
wa
s to be accom
p
lished. T
h
e
perform
a
nce of the
devel
ope
d t
ech
ni
q
u
e was eval
uat
e
d an
d com
p
are
d
wi
t
h
ot
h
e
r t
echni
ques
n
a
m
e
l
y
t
h
e AIS
and M
e
t
a
-EP
al
on
g
with Base technique. T
h
e
de
velope
d
techniques
were teste
d
on t
h
e sta
ndard IEEE 26
a
nd 57 bus syste
m
in
o
r
d
e
r to
m
i
n
i
mize th
e to
tal fuel co
st,
em
ission
dispe
r
se
d a
n
d system
losses.
2.
1.
1.
1.
T
o
t
a
l
Genera
ti
o
n
C
o
st
Mi
ni
mi
z
a
t
i
on
Prin
ci
p
a
lly, an i
m
p
o
r
tan
t
objectiv
e fu
n
c
tion o
f
ED
was t
o
ob
tain
th
e m
i
n
i
m
u
m
en
tire co
st
d
u
ring
p
o
wer syste
m
o
p
e
ration
id
en
tified
to
b
e
a to
tal g
e
n
e
ratio
n
co
st
m
i
n
i
mizatio
n
.
Th
is
o
b
j
ectiv
e fun
c
tio
n
is
prese
n
ted in m
a
them
atical form
ula
tion as i
n
equation
(1).
dol
l
a
r per ho
u
r
(
$
/
h
)
(1
)
Whe
r
e,
C
i
(
P
gi
) is th
e co
st of
g
e
n
e
ration
for
u
n
it
i
,
P
gi
is the po
wer
g
e
n
e
rated
b
y
un
it
i
,
α
i
,
b
i
,
c
i
a
r
e t
h
e cost
coefficient for
the unit
i
, a
n
d
C
Total
is th
e sum
fu
n
c
tio
n of each
g
e
n
e
rating
u
n
it
o
f
N
g
.
2.
1.
1.
2.
T
o
t
a
l
E
m
i
ssi
on Mi
n
i
mi
z
a
ti
on
The
next
es
sen
t
i
a
l
object
i
v
e
f
unct
i
o
n
was a t
o
t
a
l
em
i
ssi
on r
e
duct
i
o
n
w
h
i
c
h wa
s di
s
p
er
se
d by
t
h
erm
a
l
gene
rat
o
r as
gi
ven
by
e
q
uat
i
o
n
(2
).
t
o
n
pe
r
ho
ur (
t
on/
h)
(2
)
Whe
r
e,
E
Total
i
s
t
h
e s
u
m
func
t
i
on
fo
r eac
h
gene
rat
i
n
g em
issi
on
u
n
i
t
o
f
N
g
,
γ
i
,
β
i
,
α
i
,
ε
i
,
λ
i
are
the em
ission
coefficient for
the unit
i
, a
n
d
P
gi
i
s
t
h
e
po
wer
ge
nerat
e
d
by
u
n
i
t
i
.
2
()
ig
i
i
i
g
i
i
g
i
CP
b
P
c
P
1
()
g
N
Total
i
gi
i
CC
P
22
1
(
)
(
1
0
)
e
xp(
)
g
N
Total
i
g
i
i
i
i
i
i
g
i
i
EP
P
P
Evaluation Warning : The document was created with Spire.PDF for Python.
In
d
onesi
a
n
J
E
l
ec En
g& C
o
m
p
Sci
ISS
N
:
2
5
0
2
-
47
52
S
u
s
t
a
inab
le Enviro
nmen
ta
l Eco
nomic Dispa
t
ch
Op
timiz
at
i
o
n
w
i
t
h
Hy
bri
d
Met
ahe
uri
s
t
i
c
…
(
M
.R.M.
Ri
d
z
ua
n)
16
3
2.
1.
1.
3. T
o
t
a
l
Sys
t
em L
o
ss
Mi
ni
mi
z
a
ti
on
Ano
t
h
e
r sign
ifican
t o
b
j
ectiv
e fun
c
tio
n
was to
id
en
tify the to
tal syste
m
lo
ss
m
i
n
i
m
i
z
a
tio
n
.
Th
is
ob
ject
i
v
e
f
unct
i
on i
s
f
o
rm
ul
at
ed as
i
n
e
q
uat
i
o
n
(
3
)
.
Megawatt (MW)
(3
)
Whe
r
e,
P
loss
is the sum
functi
on of each
ge
ne
rating
unit
N
g
,
P
gi
i
s
t
h
e po
we
r ge
nerat
e
d by
uni
t
i
, and
P
load
is th
e
real
p
o
w
er l
o
a
d
dem
a
nd by
u
n
i
t
i
.
2.
1.
2. C
o
ns
tra
i
nts
In
ob
tain
ing
t
h
e resu
lts fo
r who
l
e
o
b
j
ect
iv
e fu
n
c
tion
,
th
e fo
llo
wi
n
g
eq
u
a
lity and in
equ
a
lity
ope
rat
i
o
nal
co
nst
r
ai
nt
s
m
u
st
be
un
de
r t
h
ei
r
l
i
m
i
t
a
t
i
ons u
s
i
n
g e
quat
i
o
ns
(
4
)
an
d
(5
).
2.
1.
2.
1. E
qual
i
ty
C
o
ns
trai
n
t
Formul
a
Megawatt (MW)
(4
)
Whe
r
e,
P
load
i
s
sy
st
em
l
o
ad de
m
a
nd a
n
d
T
loss
is to
tal syste
m
lo
sses.
2.
1.
2.
2. I
n
equ
a
l
i
t
y
C
o
ns
tr
ai
nt F
o
rm
ul
a
Megawatt (MW)
(5
)
Whe
r
e,
P
min
is th
e m
i
n
i
m
u
m
r
eal p
o
wer g
e
neratio
n of
u
n
it,
I
an
d
P
min
i
s
t
h
e m
a
xim
u
m
real
po
wer
ge
ne
r
a
t
i
o
n
o
f
un
it
i
.
2.
2. Me
th
od
ol
og
y
Development
of New Meta
Heuristic Evol
utionar
y
Pr
ogramming Technique
(NMEP
)
The f
u
ndam
e
nt
al
of NM
E
P
was a c
o
m
b
i
n
at
i
on bet
w
ee
n M
e
t
a
-EP an
d
AIS t
e
c
h
ni
q
u
e
s
wi
t
h
s
o
m
e
alg
o
rith
m
m
o
d
i
ficatio
n to im
p
r
o
v
e
t
h
e
o
r
ig
in
al techn
i
qu
e
wh
ile
produ
cing
a b
e
tter so
l
u
tio
n for
EELD
pr
o
b
l
e
m
i
n
po
wer
sy
st
em
. The
di
ffe
re
nces
fr
om
ot
her t
e
c
hni
que
s i
n
cl
ud
ed t
h
e
m
odi
fi
cat
i
on
occu
rri
n
g
i
n
t
h
e
Gaus
si
an m
u
t
a
t
i
on p
r
oces
s and t
h
e cl
oni
ng
pr
ocess i
n
o
r
d
e
r t
o
m
i
nim
i
ze t
h
e t
o
t
a
l
gene
rat
i
on c
o
st
, em
i
ssi
o
n
and
sy
st
em
l
o
sses. E
v
ery
t
ech
ni
q
u
e
wo
ul
d
b
e
si
m
u
l
a
t
e
d t
h
r
o
u
g
h
t
h
e
sam
e
com
m
on para
m
e
t
e
r sh
ow
n i
n
Ta
bl
e
1 and the re
sul
t
s were com
p
ared to
determ
ine the be
st
sol
u
t
i
on
fo
r t
h
e e
c
on
om
i
c
di
spat
ch p
r
o
b
l
e
m
[11]
. Thi
s
technique
was
conducted i
n
t
h
e laboratory
using M
A
TL
AB
sim
u
lation bas
e
d on
standard IEEE 26
a
n
d
57 bus
syste
m
.
The si
n
g
l
e
o
b
j
ect
i
v
e fu
nct
i
o
n
i
nvol
ve
d si
x a
nd se
ve
n co
nt
r
o
l
gen
e
rat
o
r u
n
i
t
s
i
n
orde
r t
o
opt
i
m
i
ze
t
h
e
resul
t
s
o
f
t
h
e si
ngl
e o
b
ject
i
v
e
funct
i
ons f
o
r st
anda
rd I
EEE
26 an
d 5
7
b
u
s
sy
st
em
respec
t
i
v
el
y
.
Nat
u
ral
l
y
, t
h
e
main
p
r
o
c
esses o
f
NMEP are in
itializa
tio
n, fitn
ess, m
u
tatio
n
,
clon
ing
an
d
selection
pro
cess to
ob
tain
th
e
resul
t
s
[
12]
. H
o
we
ve
r, som
e
m
i
nor p
r
oce
s
s
e
s wo
ul
d be a
dde
d t
o
i
m
prove t
h
e res
u
l
t
s
of t
h
e t
ech
ni
q
u
e
i
n
t
h
e
in
itializat
io
n
,
Gau
s
sian
m
u
ta
tio
n
and
clon
in
g
p
r
o
cess
which
wou
l
d
m
a
k
e
th
is tech
n
i
que rare fro
m
th
e rests.
The fl
ow c
h
ar
t
of t
h
e w
h
ol
e
pr
ocess
of
N
M
EP t
echni
qu
e i
s
sho
w
n i
n
Fi
gu
re 1
.
The
m
a
i
n
and a
ddi
t
i
onal
pr
ocess
are
di
s
c
usse
d i
n
det
a
i
l
s bel
o
w.
In
itializatio
n
pro
cess was a prim
ary
p
r
o
cess in
NMEP
. In
th
is o
p
tim
izat
io
n
,
th
e NMEP started
with
som
e
rand
om
num
bers
of
pa
r
a
m
e
t
e
r set
.
An
im
port
a
nt
p
r
o
c
ess of
N
M
EP
w
a
s selection
po
pu
latio
n. Th
e
sp
eed
o
f
op
timizatio
n
d
e
p
e
nd
ed
o
n
th
e nu
m
b
er
o
f
so
lu
tion
s
i
n
t
h
e po
pu
lation
.
In th
is sim
u
latio
n
,
t
h
e sets of sa
m
p
les
use
d
we
re 2
0
r
a
nd
om
num
bers, an
d t
h
e
n
t
h
e
ran
d
o
m
num
bers we
re ge
ne
rat
e
d re
pre
s
ent
i
n
g t
h
e r
eact
i
v
e
po
w
e
r
t
o
be di
sp
at
che
d
by
ge
nerat
o
r
i
n
po
wer sy
st
em
. Fi
ve gene
r
a
t
i
ng u
n
i
t
s
na
m
e
l
y
Pg1, P
g
2
,
Pg
3, P
g
4
,
Pg
5, P
g
6
,
Pg
8, P
g
9,
Pg
12
an
d P
g
26
as t
h
e act
ual
gene
rat
o
rs s
u
c
h
as
1,
2,
3,
4
,
5
,
6
,
8
,
9,
1
2
a
nd
2
6
re
sp
ect
i
v
el
y
gene
rated
react
ive powe
r acc
ording to
standard IEEE 26
a
nd 57 bus syste
m
for t
h
is study. In the
cons
traints
p
a
rt at th
e in
itial, stag
e o
f
the p
r
o
cess, so
me co
n
s
trai
n
t
s were set to
mak
e
NMEP
g
e
n
e
rated
on
ly ran
d
o
m
n
u
m
b
e
rs th
at satisfied
p
r
ed
et
erm
i
n
e
d
co
nd
itio
n
s
b
u
t
th
e
valu
es m
u
st b
e
less th
an
th
e in
itial NMEP valu
es.
Th
ese con
d
ition
s
were to
im
p
r
ov
e t
h
e fitness. Th
e co
m
p
ati
b
le NMEP m
a
x
i
m
u
m
an
d
NMEP m
i
n
i
m
u
m
v
a
lu
es
were set
n
o
t
g
r
eat
er t
h
an
1.
05
p.
u. an
d
not
l
e
ss t
h
an
0.
95
p.
u. f
o
r b
u
s v
o
l
t
a
ge l
i
m
i
t
aft
e
r co
nd
uct
i
n
g s
e
vera
l
1
g
N
lo
s
s
g
i
lo
a
d
i
P
PP
1
g
N
g
i
l
oa
d
l
os
s
i
P
PP
mi
n
m
a
x
gi
PP
P
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
250
2-4
7
5
2
I
ndo
n
e
sian
J Elec Eng
& Com
p
Sci, V
o
l. 11
,
No
.
1
,
Ju
ly
20
18
:
1
61–
168
16
4
t
e
st
s [1
3]
. Thi
s
con
d
i
t
i
on
ens
u
res t
h
at
any
v
i
ol
at
i
on
of t
h
e
sy
st
em
can be
avoi
ded al
o
ng
wi
t
h
an i
m
pro
v
em
ent
in
th
e vo
ltag
e
p
r
o
f
ile.
The eval
uation proce
ss was the firs
t pro
cess to
p
r
o
d
u
ce th
e resu
lts o
f
fitn
ess o
r
kno
wn
as sin
g
l
e
o
b
j
ectiv
e fun
c
tio
n
.
Fro
m
th
e si
m
u
latio
n
,
th
e fitn
ess was to rep
licate o
r
ig
i
n
al p
opu
latio
n
to
n
e
w
p
opu
latio
n
s
.
Loa
d
fl
ow programm
er was
conducted t
o
calcu
l
ate th
e fit
n
ess. In
o
r
d
e
r
to
co
m
p
lete the fitn
ess pro
c
ess, th
e
load
flo
w
pr
og
ram
m
e
r fr
om
the m
a
in NM
EP p
r
o
g
r
a
m
m
e
r ha
d
to
b
e
called
to
g
e
t th
e
resu
lts of fitn
ess. The
fitn
ess
was cal
cu
lated
acco
r
d
i
n
g
to
equ
a
tion
s
(1
) t
o
(3).
Fig
u
re
1
.
Th
e
Flo
w
Ch
art of
Mu
lti Obj
ective Fun
c
tion
a
NMEP Tech
n
i
q
u
e
The
n
, t
h
e m
u
t
a
t
i
on
pr
ocess
was a
n
i
m
port
a
nt
p
r
ocess
fo
r
t
h
i
s
pa
pe
r
be
cause t
h
i
s
p
r
o
cess di
f
f
e
r
ed
fr
om
ot
her t
e
c
hni
que
s t
h
at
c
o
n
d
u
ct
ed
som
e
m
odi
fi
cat
i
o
n
s
o
n
t
h
e
pr
o
g
r
am
i
ng o
f
t
h
e
Ga
ussi
an
m
u
t
a
t
i
on
pr
ocess
.
T
h
e m
u
t
a
t
i
on pr
oce
ss
wa
s pr
o
duci
n
g
n
e
w
ge
ne
ra
t
i
on or
kn
ow
n
as
o
ffs
p
r
i
n
g us
i
n
g
eq
uat
i
o
ns (
6
)
t
o
(8
).
(6
)
''
,,
exp(
(
0
,
1
)
(
0
,
1
)
)
ij
i
j
j
NN
Evaluation Warning : The document was created with Spire.PDF for Python.
In
d
onesi
a
n
J
E
l
ec En
g& C
o
m
p
Sci
ISS
N
:
2
5
0
2
-
47
52
S
u
s
t
a
inab
le Enviro
nmen
ta
l Eco
nomic Dispa
t
ch
Op
timiz
at
i
o
n
w
i
t
h
Hy
bri
d
Met
ahe
uri
s
t
i
c
…
(
M
.R.M.
Ri
d
z
ua
n)
16
5
(7
)
(8
)
Whe
r
e,
whe
r
e,
L
i
a
nd
L
oi
,
η
i,j
and
η
’
i,j
is th
e
i
th
co
m
p
one
nts of the respective vect
ors
,
N
(
0,1
) is the n
o
rm
al distributio
n
of
o
n
e di
m
e
nsi
onal
ran
d
o
m
num
ber wi
t
h
m
ean 0
an
d
1 a
nd
N
j
(
0,
1
)
i
n
di
cat
es t
h
e n
e
w
ra
n
dom
num
ber f
o
r eac
h
val
u
e o
f
j
.
Oth
e
r th
an
m
u
tatio
n
p
r
o
cess, th
e clon
al selectio
n
algo
r
i
t
h
m
r
e
pr
odu
ces those in
d
i
v
i
du
als
w
ith
h
i
gh
er
affinity and
se
lected thei
r i
m
pr
o
v
ed m
a
t
u
re
d
of
fsp
r
i
n
gs,
whe
r
e si
ngl
e
m
e
m
b
ers w
oul
d
be l
o
cal
l
y
o
p
t
i
m
i
zed
and
newc
om
ers yielded a
broade
r e
x
ploration of t
h
e
sea
r
ch s
p
ace.
Thi
s
cha
r
acteristic m
a
de
the
clona
l
sel
ect
i
on al
go
r
i
t
h
m
sui
t
a
bl
e for s
o
l
v
i
n
g m
u
l
t
i
-
m
odal
opt
im
izat
i
on pr
o
b
l
e
m
s
. The eff
ect
of
vary
i
ng t
h
e
nu
m
b
er
o
f
clon
es
g
e
n
e
r
a
ted
accor
d
i
ng
to th
e f
itn
ess (
a
f
f
i
n
ity) of
th
e ind
i
v
i
d
u
a
l
w
a
s i
n
v
e
sti
g
ated
i
n
th
is study. Th
e
cl
oni
n
g
i
s
exec
ut
ed i
n
M
A
TL
AB
p
r
og
ram
m
i
ng
usi
n
g e
quat
i
on
(
9
)
.
Clo
n
e
=
repm
at
(
A,
[
a,
b
])
(9
)
M
o
re
ove
r, t
h
e
sel
ect
i
on o
f
ra
nd
om
num
bers from
a co
m
b
i
n
at
i
on
of
fi
t
n
e
ss and
of
fs
pri
n
g i
n
o
r
de
r t
o
id
en
tify th
e n
e
w g
e
n
e
ration
is essen
tial to
p
r
odu
ce th
e op
ti
m
i
zat
io
n
v
a
lue fo
r
o
b
j
ectiv
e fun
c
tio
n
resu
l
t
. The
p
opu
latio
n
will b
e
rank
ed in ascend
i
ng
o
r
d
e
r
fro
m
th
e min
i
m
u
m
to
max
i
m
u
m
o
p
timizatio
n
v
a
lu
es. Th
is
ran
k
i
n
g
o
n
l
y
cove
re
d
20
ra
nd
om
sam
p
l
e
s of
fi
t
n
ess a
n
d
of
f
s
pri
n
g
beca
use
t
h
e
new
ge
ner
a
t
i
on
pr
o
duce
d
wa
s
base
d on
t
h
e or
i
g
i
n
al
n
u
m
b
er of
sam
p
l
e
s.
In the m
eantime, converge
nce
test was
co
ndu
cted
to
d
e
term
in
e th
e sto
p
p
i
ng
criteria o
f
th
e
optim
ization proces
s. The conve
rgen
ce c
r
iterion was s
p
ec
ified by the di
ffere
n
ce bet
w
ee
n the m
a
xim
u
m and
min
i
m
u
m
fitn
ess to b
e
less th
an
0
.
0
001
. If th
e co
nv
erg
e
n
c
e co
nd
ition
was
no
t sat
i
sfied
,
th
e m
u
tatio
n
,
to
urn
a
m
e
n
t
and
selection
p
r
ocess wou
l
d
b
e
rep
eated
u
n
til
co
nv
erg
e
n
ce criterio
n
was m
e
t u
s
ing
equ
a
tion
(10
)
.
maxi
mu
m
f
itness
–
mi
ni
m
u
m
f
itness
≤
0
.
00
01
(1
0)
Table
1. T
h
e
P
a
ram
e
ter Used
To
Produce t
h
e
Result for Sta
nda
rd IEEE
26 and
57 Bus
Sy
ste
m
Standar
d
I
E
E
E
26
Bus Sy
stem
No. of
Generato
r
Cost Coeff
i
cients (p.u.)
Generator Li
m
it
(M
W)
E
m
ission coef
f
i
cients (p.u.)
α
i
b
i
c
i
Min
Max
α
β
γ
ε
λ
1 240
7.
0
0.
0070
100
500
4.
091
-
5
.
543
6.
490
2.
0e-
4
2.
857
2 200
10.
0
0.
0095
50
200
2.
543
-
6
.
047
5.
638
5.
0e-
4
3.
333
3 220
8.
5
0.
0090
80
300
4.
258
-
5
.
094
4.
586
1.
0e-
6
8.
000
4 200
11.
0
0.
0090
50
150
5.
326
-
3
.
550
3.
380
2.
0e-
3
2.
000
5 220
10.
5
0.
0080
50
200
4.
258
-
5
.
094
4.
586
1.
0e-
6
8.
000
26
190
12.
0
0.
0075
50
120
6.
131
-
5
.
555
5.
151
1.
0e-
5
6.
667
Standar
d
I
E
E
E
57
Bus Sy
stem
No. of
Generato
r
Cost Coeff
i
cients (p.u.)
Generator Li
m
it
(M
W)
E
m
ission coef
f
i
cients (p.u.)
Min
Max
α
β
γ
ε
λ
1 115
2.
00
0.
0055
50
576
4.
091
-
5
.
543
6.
490
2.
0e-
4
2.
857
2 40
3.
50
0.
0060
10
100
2.
543
-
6
.
047
5.
638
5.
0e-
4
3.
333
3 122
3.
15
0.
0050
20
140
4.
258
-
5
.
094
4.
586
1.
0e-
6
8.
000
6 125
3.
05
0.
0050
10
100
5.
326
-
3
.
550
3.
380
2.
0e-
3
2.
000
8 120
2.
75
0.
0070
40
550
4.
258
-
5
.
094
4.
586
1.
0e-
6
8.
000
9 70
3.
45
0.
0070
10
100
6.
131
-
5
.
555
5.
151
1.
0e-
5
6.
667
12
150
1.
89
0.
0050
30
410
4.
258
-
5
.
094
4.
586
1.
0e-
6
8.
000
''
,,
,
((
0
,
1
)
)
ij
ij
ij
j
LL
N
''
,,
,
((
0
,
1
)
)
oi
j
o
i
j
i
j
j
LL
N
1
2
n
'
1
2
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
250
2-4
7
5
2
I
ndo
n
e
sian
J Elec Eng
& Com
p
Sci, V
o
l. 11
,
No
.
1
,
Ju
ly
20
18
:
1
61–
168
16
6
3.
R
E
SU
LTS AN
D ANA
LY
SIS
The
de
vel
o
pm
ent
o
f
a
Ne
w
M
e
t
a
He
uri
s
t
i
c Ev
ol
ut
i
o
nar
y
Pro
g
r
am
m
i
ng
(NM
E
P
)
al
g
o
ri
t
h
m
was
tested on standard
IEEE
26 a
nd
IEEE
57 bus syste
m
using
MATLAB sim
u
lation e
n
vironm
ent. The
results for
single
objective we
re c
onsi
d
e
r
ed in
t
h
ree
differe
n
t categori
e
s as
follows:
i)
SOCEELD = to
tal co
st m
i
n
i
mizatio
n
(fitn
ess)
wh
ile
t
o
tal e
m
issio
n
and
t
o
tal syste
m
lo
ss (o
b
s
erv
e
d)
ii)
SOEEELD = t
o
tal em
iss
i
o
n
min
i
mizatio
n
(fitn
ess) wh
ile to
tal co
st and
total syste
m
lo
ss (ob
s
erv
e
d
)
iii)
SOLEELD = t
o
tal syste
m
lo
ss m
i
n
i
mizat
io
n (fitn
ess)
wh
ile to
tal co
st and
t
o
tal em
iss
i
o
n
(o
b
s
erv
e
d)
Thi
s
part
i
c
ul
a
r
sol
u
t
i
o
n
hi
ghl
i
ght
e
d
t
h
e
res
u
l
t
s o
b
t
a
in
ed from
in
d
i
v
i
du
al
ob
j
ective fun
c
tio
n
s
(fitn
ess)
whi
l
e
o
b
se
rvi
n
g t
h
e
ot
her t
w
o
fu
nct
i
o
ns. Eac
h
sol
u
t
i
o
n p
r
es
ents the ac
hieve
m
ent for
si
ng
le o
b
j
ectiv
e
functio
n
that is execute
d 20 tim
es using i
d
en
tical optimization m
odel as a perfor
mance m
easurement [14]. The
results
were
di
scus
se
d i
n
det
a
i
l
s
i
n
t
e
r
m
of para
m
e
t
e
r appl
i
cat
i
on, t
h
e best
pos
si
bl
e ans
w
ers bet
w
een
o
b
ject
i
v
e
fun
c
tion
s
and
th
e ov
erall op
t
i
m
a
l so
lu
tio
n
s
were
d
e
term
in
ed
throug
h
t
h
e o
p
tim
ize fin
e
st v
a
lu
e
o
f
ob
jectiv
e
fun
c
tion
.
In ad
d
ition
,
t
h
e
AIS, Meta-EP an
d Base t
echniq
u
e
s were com
p
ared
to
v
e
rify th
e
q
u
a
lity of th
e
per
f
o
r
m
a
nce pro
p
o
sed t
e
c
hni
que
s sol
u
t
i
ons
.
Ho
wev
e
r, t
h
e
B
a
se t
echni
q
u
e base
d
on
Hadi
Saa
d
at
i
s
onl
y
sh
own
as fu
nda
m
e
n
t
al r
e
su
lts fo
r
bo
th
sta
n
d
a
rd
IEE
E
26
an
d
57
b
u
s
sy
st
em
s [15]
.
Table
2. T
h
e
O
p
tim
a
l Generat
i
ng
U
n
its f
o
r
MOC
EEEL
D a
m
ong the
Thre
e Techniques
Using Fi
xe
d a
n
d
Ran
d
o
m
W
e
i
g
h
t
s Valu
es
on
Stan
d
a
rd
IEEE 26
an
d 57
bu
s
SOEE
LD
Standar
d
I
E
EE
T
echniques
Pload
(MW
)
SOCEE
L
D
(
dollar
/
h)
SOEE
ELD
(to
n
/h
)
SOLE
ELD
(MW
)
Aggregate 1
Aggregate 2
Aggregate 3
Total
A
gg
re
g
ate
26 Bus
Syste
m
Base
1263.
0
1548
6.
19
1973
3.
95
13.
028
38
4
4
4
12
AI
S
1263.
0
1546
1.
30
1925
1.
01
12.
992
33
3
2
3
8
M
e
ta-E
P
1263.
0
1545
9.
30
1956
2.
67
12.
960
21
2
3
2
7
NM
E
P
1263.
0
1536
1.
53
1906
3.
08
12.
865
71
1
1
1
3
57 Bus
Syste
m
Base
1250.
8
6493.
7
5
2174
8.
59
36.
134
81
4 4
4
12
AI
S
1250.
8
6490.
0
9
2148
1.
10
34.
133
47
3 2
2
7
M
e
ta-E
P
1250.
8
6487.
2
6
2170
9.
62
35.
392
16
2 3
3
8
NM
E
P
1250.
8
6475.
0
6
2074
0.
36
34.
073
89
1 1
1
3
B
u
s V
o
ltage V
i
olations for Stan
dar
d
I
E
E
E
26 and 57 B
u
s System = 0.
95 p.u.
≤
V
≥
1.
05 p.
u
Th
e
NMEP
for SOCEELD
prov
id
ed
th
e solu
tio
n
t
o
redu
ce th
e to
tal g
e
neratio
n
co
st
by n
o
t
o
n
l
y
foc
u
si
n
g
o
n
t
h
e fu
el
ge
nerat
o
r b
u
t
al
s
o
i
n
cl
udi
ng
t
h
e
ot
he
r m
a
i
n
t
e
nance
cost
. B
a
se
d
on
Tabl
e
2, t
h
e l
o
wes
t
ent
i
r
e cost
was
at
onl
y
15
3
6
1
.
53
(d
ol
l
a
r/
h
)
u
s
i
ng
NM
EP i
m
pl
em
ent
a
t
i
on. I
n
ot
her
wo
r
d
s,
t
h
e NM
EP m
e
tho
d
sp
en
t ab
ou
t 873
985
.2
0
(do
llar/year) wh
ile 85
646
5.20
(do
llar/year) less than
AIS and
M
e
ta-EP resp
ectiv
ely.
Add
itio
n
a
lly, th
e NMEP also sh
owed
t
h
e outp
e
rform
so
l
u
t
i
on as c
o
m
p
ared wi
t
h
A
I
S a
n
d M
e
t
a
-EP t
echni
que
as sav
i
n
g
th
e p
r
ofit on
to
t
a
l g
e
n
e
ration
co
st ab
ou
t
1
3
1
662
.8
0 (d
o
llar/year)
fro
m
AIS wh
ile 106
872
.0
0
(d
ollar/y
ear) th
an M
e
ta-EP
o
n
5
7
bus sy
ste
m
. Hence,
t
h
e
NM
EP
was su
pp
ose
d
l
y
savi
n
g
t
h
e a
v
era
g
e
pr
ofi
t
abo
u
t
R
M
3
3
5
7
0
7
3
.
78
per y
e
ar o
n
2
6
bus
sy
st
em
and R
M
46
2
7
5
7
.
5
1 pe
r
y
ear fo
r 5
7
bu
s
sy
st
em
based on t
h
e
cur
r
ent
cu
rre
n
c
y
rates
(1
$ =
RM
3.
88
). T
h
e second
nece
ssary
objectiv
e fu
n
c
tion n
a
m
e
ly SOEEELD was to
obt
ai
n t
h
e l
eas
t
possi
bl
e t
o
t
a
l
em
i
ssi
on di
sp
ersed t
h
ro
u
gh
t
h
e envi
r
o
nm
ent
caused
by
t
h
e o
p
erat
i
o
n i
n
t
h
e
po
we
r sy
st
em
net
w
or
k. Acc
o
rdi
ng t
o
[
1
6]
and [
1
7]
, 1 t
on
of em
i
ssi
on of
coal
i
s
equal
t
o
t
h
e l
o
sses o
f
98
0
k
W
h.
C
o
nse
q
u
e
nt
l
y
, i
f
t
h
e
N
M
EP p
r
od
uce
d
t
h
e em
i
ssi
on a
b
o
u
t
2.
95
T
W
h
pe
r y
ear
,
henc
e, t
h
e
ave
r
age
savi
n
g
o
n
th
e lo
sses
of
electr
i
city w
a
s equ
i
v
a
len
t
to RM 130
143
988
.4
0 (
i
f ev
er
y
kW
h
is ch
arg
e
at 26
cen
t
) p
e
r
year
fo
r the 2
6
bus
sy
stem
. Sim
ilarly
for
5
7
b
u
s s
y
stem
, NM
EP
was also a
b
le to re
duce t
h
e average c
o
st about RM
32
3
6
9
4
1
7
6
.
4
0
per y
ear f
r
o
m
7.34 T
W
h
un
wa
nt
ed em
issi
on
di
spe
r
se
d pe
r y
ear. T
h
e f
o
l
l
o
wi
ng
ob
ject
i
v
e
fun
c
tion
was
min
i
mizin
g
to
tal syste
m
lo
ss o
r
SOLEELD i
n
so
l
v
ing
EELD prob
lem
.
Fo
r th
at reaso
n
, selectin
g
su
itab
l
e g
e
n
e
ratin
g
u
n
its
was a prio
rity to ach
iev
e
t
h
is
part
i
c
ul
a
r
o
b
j
e
c
t
i
v
e. As m
e
nt
i
one
d
pre
v
i
o
us
l
y
, t
h
e
pr
o
p
er ge
ne
rat
i
ng
uni
t
s
i
n
fl
ue
nced t
h
e
best
p
o
ssi
bl
e s
o
l
u
t
i
o
n. T
h
ere
f
o
r
e, t
h
e sum
of opt
i
m
al
generat
i
n
g
uni
t
s
wh
ich
co
n
t
ri
bu
ted
t
o
ach
ieve th
e b
e
st
po
ssib
le SO
LSEELD so
lu
tion
s
a
m
o
n
g
all
m
e
n
tio
n
e
d
op
ti
m
i
zatio
n
techniques
wa
s calculated a
n
d m
u
st be e
q
ual to
P
load
aft
e
r a
ddi
ng
t
h
e
P
loss
. Fro
m
th
e resu
lts ob
tained
, t
h
e
sm
a
llest syste
m
lo
ss w
a
s abo
u
t
12
.8
657
1
(
M
W
)
using
N
M
EP tech
n
i
qu
e. Ev
en
thoug
h, th
e lo
sses
g
a
in
ed
t
h
r
o
u
g
h
A
I
S a
n
d
M
e
t
a
-EP
w
e
re n
o
t
ob
vi
o
u
s
l
y
di
ffe
red
f
r
o
m
t
h
e pr
op
os
ed m
e
t
hod,
ab
out
8
2
7
.
8
2
M
W
a
n
d
1
109
.1
9
M
W
will b
e
lo
st th
ro
ugh
ou
t a year. Fu
rt
h
e
rm
o
r
e,
assu
m
i
n
g
in
dustrial tariff in
Malaysia
is at
av
erag
e
cost
abo
u
t
4
4
.
10 se
n/
k
W
h t
h
us, t
h
e NM
E
P
t
echni
q
u
e sav
e
abo
u
t
R
M
427
1
1
.
07
per y
ear fo
r 2
6
bu
s
sy
st
e
m
Evaluation Warning : The document was created with Spire.PDF for Python.
In
d
onesi
a
n
J
E
l
ec En
g& C
o
m
p
Sci
ISS
N
:
2
5
0
2
-
47
52
S
u
s
t
a
inab
le Enviro
nmen
ta
l Eco
nomic Dispa
t
ch
Op
timiz
at
i
o
n
w
i
t
h
Hy
bri
d
Met
ahe
uri
s
t
i
c
…
(
M
.R.M.
Ri
d
z
ua
n)
16
7
w
h
er
eas RM 26
614
2.75
p
e
r
year
o
n
57
bus syste
m
if
lo
s
s
es cap
tur
e
w
a
s co
n
s
i
d
er
ed
acco
rd
ing
p
e
r
ho
ur
i
n
or
der
t
o
get
e
q
ui
val
e
nt
res
u
l
t
s
as t
h
e
com
p
ari
s
on
res
u
l
t
s
.
In
or
de
r t
o
d
e
fi
ne t
h
e
best
p
e
rf
orm
a
nce of
si
ngl
e
ob
ject
i
v
e f
u
nct
i
ons i
n
sol
v
i
n
g EEL
D p
r
o
b
l
e
m
s
,
Tabl
e 2
was u
s
ed t
o
s
h
o
w
t
h
e fi
ne
st
res
u
l
t
of al
l
i
d
ent
i
f
i
e
d si
ngl
e
ob
j
ect
i
v
e fu
nct
i
o
n
s
am
ong t
hose
t
h
ree
techniques al
ong
with Base techni
que
for both standa
rd
IEEE 26 bus sy
st
e
m
and 57 bus system
. Thoroughly,
t
h
e NM
EP a
p
p
r
oac
h
was com
p
are
d
bet
w
ee
n
t
w
o
ot
he
r co
m
m
on opt
i
m
i
z
at
i
on t
ech
ni
q
u
e
s k
n
o
w
n as A
I
S a
n
d
M
e
t
a
-EP m
e
t
hods
an
d al
s
o
w
i
t
h
t
h
e ba
se
val
u
es as
rec
o
m
m
ende
d
by
Ha
di
Saadat
.
Al
l
re
sul
t
s
we
re e
v
al
uat
e
d
usi
n
g t
h
e ag
gr
egat
e fu
nct
i
o
ns
appr
oac
h
i
n
o
r
de
r t
o
ap
pr
ov
e t
h
e best
perf
orm
a
nce sol
u
t
i
on am
ong t
h
e
m
. Thi
s
im
ple
m
entatio
n was
m
easured by decl
aring
the fi
rst
winn
er with th
e smallest ag
g
r
egate v
a
lu
e am
o
n
g
th
e
com
p
arat
i
v
e t
e
chni
que
s. B
a
se
d
on
t
h
e
t
a
bl
e,
t
h
e m
i
ni
m
u
m
overall c
o
st
(SOCEEL
D), t
h
e sm
allest e
m
ission
am
ount (
S
O
E
E
ELD
) and total loss of t
h
e system
(SOLEEL
D)
were re
pres
ented
by Aggr
egate 1,
Aggr
e
g
ate 2
and
Aggregate 3 res
p
ectively. Overa
ll, the
NMEP m
e
thod won the
first
place
for Aggregate
1, Aggre
g
ate 2
and
A
g
g
r
e
g
at
e 3 f
o
r b
o
t
h
st
anda
r
d
IE
EE
26
and
5
7
bus
syste
m
s. Moreover, the t
o
ta
l aggre
g
ate col
u
mn al
s
o
sho
w
e
d
t
h
e l
o
west
val
u
e t
h
a
t
resul
t
e
d i
n
t
h
e pr
o
pose
d
m
e
t
h
o
d
as t
h
e
ex
cel
l
e
nt
t
echni
q
u
e am
ong t
hos
e t
h
re
e
opt
i
m
i
zati
on t
echni
que
s.
I
n
ad
di
t
i
on,
t
h
e
NM
EP al
s
o
ha
d t
h
e
fe
wer a
m
ount
s i
n
t
h
e
i
d
ent
i
f
i
e
d
o
b
se
rvat
i
o
n
q
u
a
n
tities d
u
rin
g
th
e si
n
g
l
e
ob
j
ective so
lu
ti
o
n
. In
sho
r
t,
NMEP is th
e
mo
st su
itab
l
e tech
n
i
q
u
e
p
a
rticu
l
arly in
resol
v
i
n
g t
h
e SOEE
LD i
s
s
u
es am
ong
ot
he
r t
w
o c
o
m
m
on t
echni
q
u
es i
n
t
e
r
m
s of t
h
e t
o
t
a
l
generat
i
o
n cost
m
i
nim
i
zation, least e
m
ission
productio
n and s
m
allest syste
m
losses for
bo
th standa
rd
IEEE 26 and 57 bus
syste
m
s.
4.
CO
NCL
USI
O
N
In c
oncl
u
si
o
n
,
t
h
e gr
o
w
t
h
of
ener
gy
dem
a
nd an
d i
n
a
d
e
q
u
acy
of ene
r
gy
reso
u
r
ce are re
qui
red
fo
r
a
secure
d l
o
a
d
d
i
spat
ch.
Neve
r
t
hel
e
ss, t
h
e p
r
essur
e
fr
om
publ
i
c
awa
r
ene
s
s cont
ri
but
es t
o
t
h
e re
qui
rem
e
nt
fo
r
red
u
ct
i
o
n i
n
t
o
xi
c wast
e em
i
s
si
on p
r
od
uce
d
by
t
h
e po
we
r p
l
ant
s
. Th
us t
h
e
devel
o
pm
ent
of n
e
w o
p
t
i
m
izat
i
o
n
tech
n
i
qu
e
n
a
mely NMEP is ai
m
i
n
g
fo
r
eco
no
m
i
cal lo
ad
with
ou
t com
p
ro
misin
g
th
e
well-b
e
i
n
g o
f
th
e
envi
ro
nm
ent
.
Al
l
reco
gni
ze
d
si
ngl
e o
b
j
ect
i
v
e sol
u
t
i
o
n
s
ar
e com
p
ared a
m
ong t
h
e basi
c
M
e
t
a
-EP an
d cl
assi
cal
m
e
thod
by H
a
di Saa
d
at re
spectively. T
h
e best po
ssi
bl
e solution
for
these
indivi
dual objective
EELD
pr
o
b
l
e
m
s
i
s
obt
ai
ned
by
t
h
e NM
EP m
e
t
h
o
d
.
The
r
e
f
o
r
e, t
h
e
NM
EP
m
e
t
hod
i
s
h
i
ghl
y
rec
o
m
m
e
nde
d
p
a
rticu
l
arly in
so
lv
i
n
g Env
i
ron
m
en
tal Eco
nomic Lo
ad
Dispatch
pro
b
l
em
s.
AC
KN
OWLE
DG
MENT
The
resea
r
ch
i
s
fi
na
nci
a
l
l
y
sup
p
o
rt
e
d
by
t
h
e t
eam
of t
h
i
s
pr
o
j
ect
f
r
o
m
Advanc
e
D
i
gi
t
a
l
Si
gnal
Pro
cessin
g
La
bo
rato
ry
(A
D
SP Lab
)
.
Spe
c
i
al thanks also go to the F
acul
t
y
of El
ect
ri
cal
Engi
nee
r
i
ng o
f
Un
i
v
ersiti Tekn
ik
al Malaysia Melak
a
(UTeM) and
Mi
n
i
stry o
f
High
er
Edu
catio
n
Mal
a
ysia (MOHE) for
giving the c
o
ope
ration a
n
d funding t
o
ensure
the
fe
asibility and
success of t
h
is resea
r
ch na
m
e
ly
RA
G
S
/
1
/201
5
/
TK0
/
FK
E/
0
3
/
B0
094
. Th
eir su
ppo
r
t
is
gr
atefu
lly ack
now
ledg
ed.
REFERE
NC
ES
[1]
Li, C., De Bosio, F., Chen
, F., Chaudhar
y
, S
.
K., Vas
quez, J.C., & Guerr
e
ro, J.M.
“Economic Dispatch for
Operating Cost
Minimization U
nder Rea
l
-Tim
e
Pricing in Dro
op-Controlled
DC Microgrid”.
I
EEE Journal
o
f
Emerging and S
e
lected
Topi
cs in
Power Electron
i
cs
. 2017
; 5(1)
:
587-595.
[2]
Muhtazarudd
in,
M.N.B., Bani
, N
.
A., Aris, S.A
.
M., K
a
idi
,
H.M.,
Fatah, A.Y.A., J
a
m
i
a,
J
.
J., Muha
m
m
a
d-Sukki, F.
and Abu-Bakar
,
S.H. “Distribution
power loss minimization via distributed
generation, capa
citor and networ
k
reconfigur
ation
”
.
Indonesian
Jou
r
nal of
Electrica
l
Eng
i
neering
an
d Computer Science
. 2017; 5(3):
488-495.
[3]
Arriffin,
A
.
M.
,
Othma
n
,
M.
M.
, Ka
ma
ruz
a
m
a
n
, A.
A.
M.
, Mu
s
i
rin, I
.
,
Y
a
h
y
a
,
A
., &
L
a
tip
, M
.
F
.
A
.
“
S
tochas
t
i
c
Approach of Vo
ltag
e
Optim
iz
ati
on to
Maximize Power Saving
in a Building
”.
I
ndonesian Jour
nal of
Electrica
l
Engineering and
Computer Scien
c
e
. 2017; 8(1): 2
68-272.
[4]
Qu, B.Y.
, Zhu
,
Y.S., Ji
ao,
Y.
C., W
u
,
M.Y.,
Sugantha
n
,
P.
N., &
Li
ang,
J.J. “
A
surve
y
on m
u
lti-obje
c
t
i
v
e
evolution
a
r
y
algorithms for the solution
of
th
e environm
ental/economic
dispatch prob
lems”.
Swarm and
Evolutionary Co
mputation
. 2018
; 38: 1-11.
[5]
Pourbabak, H., Luo, J., Ch
en, T., &
Su, W. “A novel consensus-based distri
buted algorithm for economic dispatch
based on
local es
timation
of pow
er mismatch”.
I
E
EE Transactions
on Smart Grid
. 2017.
[6]
H
a
jiabb
as
, M
.
P
.
, N
a
z
a
ri-H
eris
,
M
., M
a
dad
i
, S
.
,
& M
oham
m
a
di
-Ivatloo
,
B. “
T
h
e
util
iz
ation of
quantum
inspire
d
com
putation
a
l
in
tell
igent
in pow
er s
y
st
em
s opti
m
i
zation
”
. In
Qu
antum Computin
g: An En
vironment for In
tellig
ent
Large Scale Rea
l
App
lica
tion.
20
18: 489-505.
[7]
D
u
, D
., W
a
ng
, C
., D
u
, X
., Y
a
n
,
S
., R
e
n, X
.
, S
h
i,
X
., & H
e
in, J
.
R.
“
D
is
tance-grad
i
e
nt-bas
ed v
a
riog
ram
and K
r
iging
to ev
aluate cob
a
lt-rich
crust depo
sits on seamounts”
.
Ore Geology
Reviews
. 2017; 8
4
: 218-227.
[8]
P
a
ttan
a
ik, J
.
K
.,
Bas
u
, M
.,
& D
a
s
h
, D
.
P
.
“
R
evie
w
on applic
atio
n and comparis
on of metaheuristic techniqu
es to
m
u
lti-are
a
econo
m
i
c dispat
ch pro
b
lem
”
.
Protectio
n and Control of Modern Pow
e
r
Systems
. 2017; 2
(
1): 17.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
250
2-4
7
5
2
I
ndo
n
e
sian
J Elec Eng
& Com
p
Sci, V
o
l. 11
,
No
.
1
,
Ju
ly
20
18
:
1
61–
168
16
8
[9]
Yang, Z., Xian
g, J., & Li, Y. “Dis
tributed consensus based
supply
–de
mand
balance algorithm for economic
dis
p
atch probl
e
m
in a s
m
art
grid with switching
graph”
.
IEEE Transactions on Industrial Electr
onics
. 2017;
64(
2):
1600-1610.
[10]
Ajenikoko, G.A., & Olabode,
O.E. “Optimal Po
wer Flow with
Reactiv
e Power
Compensation f
o
r Cost and
Loss
M
i
nim
i
zation
o
n
Nigerian P
o
w
e
r Grid S
y
s
t
em
”.
Indon
esian Jo
urnal of Electrica
l Eng
i
neering
and Informatics
(
I
JEEI)
. 2017; 5
(
3): 236-247
.
[11]
Ridz
ua
n,
M.
R.
M.
,
Ha
ssa
n,
E.
E., Abdullah,
A.R., Bahaman,
N
.
, & Ka
dir, A.
F.
A. “NMEP ba
se
d Gaussia
n
Muta
tion
Process on Opti
mizing Fitness
Function for MOEED”.
Interna
tional Journal on Adv
anced Science, Engineering
and Information
Technology
. 201
7; 7(5): 1840-18
46.
[12]
Biethahn
, J., &
Nissen, V.
“Evo
lutionar
y
algor
ithms in manageme
nt application
s
”.
Springer Science
&
Business
Media
. 2012.
[13]
H
a
s
s
a
n, E.E
., Ra
hm
an, T.K
.
A
.
, Z
a
kari
a, Z
., & Ba
ham
a
n,
N. “The Improved of BFOA fo
r Ensuring
the Sustainable
Economic Dispatch”. In
Applied
Mechanics and
Materials
. 2015: 785: 83-87
[14]
Hassan, E.E.,
Zakaria, Z.,
& R
a
hm
an, T
.
K.A.
“
I
m
p
roved Adaptive
Tum
b
ling
Bac
t
eri
a
l For
a
ging Optim
iz
ati
o
n
(ATBFO) for emission constrained
economic dispatch problem”. In
Proceedings of the World Congress on
Engineering
. 20
12; 2: 1-4.
[15]
Ar
y
a
ni, N.K., Soeprijanto, A
.,
Negara
, I.M.Y., & S
y
ai’in
,
M.
“Economic Dispatch using Quan
tum Evolution
a
r
y
Algorithm in Electr
i
cal Power Sy
stem
involving
Distributed Generators”
.
In
tern
ational Journal
of Electrica
l
and
Computer Engin
eering (
I
JEC
E
)
.
2017; 7(5): 2365
.
[16]
Li,
R.
, Zh
an,
W
., & H
a
o,
Z.
“
A
rtifici
a
l Im
m
une Partic
le
Swarm
Optim
ization Algo
rithm
Based on
Clo
n
al
Selec
tion”
.
Bole
t
í
n Técn
ico
. 2017
; 55(1): 158-164.
[17]
Cus
h
m
a
n-Rois
in, B.
, G
a
cic
,
M
.
,
P
oulain,
P
.
M
.
,
& A
r
teg
i
ani
,
A.
“Phy
sical ocean
ograph
y
of the Adriati
c
S
e
a
:
pa
s
t
,
present and
fu
tur
e
”
. Springer S
c
ience
&
Bu
siness Media
. 2013.
Evaluation Warning : The document was created with Spire.PDF for Python.