Internati
o
nal
Journal of Ele
c
trical
and Computer
Engineering
(IJE
CE)
Vol
.
5
,
No
. 5, Oct
o
ber
2
0
1
5
,
pp
. 91
8~
92
8
I
S
SN
: 208
8-8
7
0
8
9
18
Jo
urn
a
l
h
o
me
pa
ge
: h
ttp
://iaesjo
u
r
na
l.com/
o
n
lin
e/ind
e
x.ph
p
/
IJECE
Loss Mi
nimizati
on of Powe
r Distribution Network using
Different Types of Distributed Generation Unit
Su Hl
aing Wi
n*,
P
y
one Lai
Swe **
* Department of
Electrical Power
Engin
eering
,
M
a
ndalay
Techno
logical University
, M
y
anmar
** The Repub
lish of th
e Union
o
f
M
y
anmar, M
y
anmar
Article Info
A
B
STRAC
T
Article histo
r
y:
Received
May 9, 2015
Rev
i
sed
Ju
l 10
,
20
15
Accepte
d
J
u
l 21, 2015
A Radial Dis
t
rib
u
tion network is
im
portant in po
wer s
y
s
t
em
are
a
becaus
e
o
f
its simple design and r
e
duced cost
. Reduction of s
y
s
t
em losses an
d
improvement of
voltage pro
f
ile
is one
of
th
e key
aspects in
po
wer s
y
stem
operation. Distributed gen
e
rators
are be
n
e
fic
i
a
l
in
reducing losses
effec
tiv
e
l
y
in distribu
tion s
y
stems as com
p
ared
to
other
methods of loss reduction.
S
i
zing and lo
ca
ti
on of DG s
ources
places
an
im
portant ro
le in r
e
du
cing los
s
e
s
in distribu
tion
n
e
twork. Four
types of
DG ar
e
c
onsidered
in th
is paper
with
one DG ins
t
all
e
d for m
i
nim
i
ze t
h
e tot
a
l r
eal
and
reac
tiv
e power
los
s
e
s
.
The
objective of th
is methodolog
y
is to calcu
late size and to
identif
y
the
corresponding o
p
timum location
for DG pl
acement for
minimizing the
total
real
and
rea
c
t
i
v
e
power
los
s
e
s
and to
improve
voltag
e
profile
in pr
imar
y
dis
t
ribution s
y
s
t
em
. It can obtai
n m
a
xim
u
m
loss
reduction for each of four
t
y
p
e
s of optim
al
l
y
p
l
a
ced DGs.
Optim
al
sizing o
f
Distributed G
e
neration can
be cal
cula
ted us
ing exac
t los
s
form
ula and an effici
ent approa
c
h
is
us
ed to
determ
ine
the o
p
tim
um
location
for Distribut
ed
Genera
tion Pla
cem
ent.
T
o
demonstrate th
e performance
of the
proposed approach 36
-bus radial
distribution s
y
st
em
in Belin Sub
s
tation in M
y
an
m
a
r was tested a
nd valida
t
ed
with diff
erent
s
i
zes
and
th
e r
e
s
u
l
t
was
dis
c
us
s
e
d
.
Keyword:
Di
st
ri
b
u
t
e
d ge
nerat
i
o
n
Ex
act lo
ss form
u
l
a
Fo
ur
t
y
pes of
DG
Optim
al location and
sizing
Power l
o
ss m
i
nim
i
zation
Copyright ©
201
5 Institut
e
o
f
Ad
vanced
Engin
eer
ing and S
c
i
e
nce.
All rights re
se
rve
d
.
Co
rresp
ond
i
ng
Autho
r
:
Su
H
l
aing
W
i
n,
Depa
rt
m
e
nt
of
El
ect
ri
cal
Po
w
e
r E
ngi
neeri
n
g, Mandalay Te
chno
log
i
cal Univ
ersity
Em
a
il: su
h
l
aing
wi
n
24@g
m
ai
l.co
m
1.
INTRODUCTION
In rece
nt years
,
environm
ental concerns, fue
l
cost
uncertai
n
ties, liberalization of electri
city
markets
and a
dva
nces
i
n
t
echn
o
l
o
gy
have r
e
sul
t
e
d
i
n
i
n
creasi
n
g
DG
uni
t
s
i
n
d
i
st
ri
but
i
o
n sy
st
em
. Thi
s
t
r
end has
o
f
fered
g
r
eat
op
portun
ities b
u
t created
sev
e
ral ch
allen
g
e
s in
p
l
ann
i
ng
and o
p
e
ration
s
o
f
d
i
stribu
tio
n syste
m
s.
The prim
ary
purpose of DG units
is
e
n
ergy
injection;
howeve
r,
strategi
cally placed a
n
d operate
d
DG
units
can y
i
el
d se
ver
a
l
ot
her
be
ne
fi
ts to
u
tilities. A typ
i
cal ex
am
p
l
e o
f
su
ch
ben
e
fi
t is th
e app
licatio
n
o
f
DG un
its
for lo
ss redu
ctio
n
[1–3
]. Vo
lt
ag
e and
lo
ad
ab
ility en
h
a
n
c
emen
t, reliab
ilit
y i
m
p
r
o
v
e
m
e
n
t
an
d
netwo
r
k
up
grad
e
defe
rral
a
r
e ot
h
e
r bene
fi
ts [4
,
5
]
.
Distribu
ted
Gen
e
ration
is an e
m
erg
i
ng
tech
no
log
y
in
th
i
s
n
e
w era an
d it
provides clean
electric
po
we
r. Di
st
ri
b
u
t
e
d
Gene
rat
i
on s
h
oul
d be
l
o
cat
ed at
o
r
near a
n
el
ec
t
r
i
cal
l
o
ad C
e
nt
re.
In
st
al
l
a
t
i
on
of
Distributed
Ge
neration at opt
i
m
a
l pl
aces provi
des the cle
a
n electric power t
o
the c
u
s
t
om
er [6]. Di
ffe
rent
i
ssues
have
be
en m
e
nt
i
oned
t
o
defi
ne
t
h
e
Di
st
ri
but
e
d
ge
n
e
ratio
n
m
o
re im
p
o
rtan
tly .So
m
e o
f
th
e issu
es
of
DG
is Distribu
ted
Gen
e
ration
Ratin
g
,
t
h
en o
t
h
e
r are Tech
no
log
y
, sizing
, sitti
n
g
, m
o
d
e
o
f
op
eration
,
Distrib
u
t
ed
Gene
rat
i
o
n pe
net
r
at
i
o
n [7]
.
D
i
st
ri
but
ed
Gen
e
rat
i
on i
s
a sm
all g
e
n
e
rating
u
n
it lo
cated
in th
e effectiv
e p
o
i
n
t
of
the electric power system
near to the l
o
ad c
e
ntre.
DG system
s a
r
e sm
all powe
r
source
s that connect
to
d
i
strib
u
tion
syst
e
m
s. W
ith
t
h
e in
creasing
dem
a
nd for el
ectrical power
and t
h
e t
echni
cal, econom
i
c, and e
n
vironm
enta
l con
s
traints in
th
e con
s
tru
c
tio
n
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN
:
2
088
-87
08
I
J
ECE Vo
l. 5
,
N
o
. 5
,
O
c
tob
e
r
20
15
:
918
–
9
28
91
9
of
new
po
we
r pl
ant
s
an
d
new
t
r
ansm
i
ssi
on l
i
n
es, D
G
can e
ffi
ci
ent
l
y
resp
o
nd t
o
sy
st
em
requi
rem
e
nt
s. DG ha
s
pre
d
om
inant specifications.
Th
ere are a
n
u
m
b
er o
f
DG tech
no
lo
g
i
es av
ailab
l
e in
th
e mark
et t
o
d
a
y and
few are still
in
research
and de
velopm
ent stage
.
Some curre
ntly a
v
ailable techno
l
o
gi
es are
re
ci
pr
ocat
i
n
g e
ngi
nes,
m
i
cro t
u
r
b
i
n
es,
co
m
b
u
s
tio
n g
a
s tu
rb
in
es,
fu
el cells, pho
tov
o
ltaic, an
d wi
n
d
tu
rb
in
es. Each on
e
o
f
th
ese tech
no
log
i
es
h
a
s it
s
ow
n ben
e
fi
t
s
a
nd
cha
r
act
eri
s
t
i
cs. Am
ong
al
l
t
h
e D
G
,
di
es
el
or
gas
reci
p
r
ocat
i
n
g e
ngi
n
e
s an
d g
a
s t
u
r
b
i
n
es
m
a
ke up m
o
st
of t
h
e ca
paci
t
y
i
n
st
al
l
e
d so far. Si
m
u
l
t
a
neousl
y
,
new
D
G
t
ech
nol
ogy
l
i
k
e
m
i
cro t
u
r
b
i
n
e i
s
bei
n
g i
n
t
r
od
uc
ed an
d an
ol
d
e
r t
echn
o
l
o
gy
l
i
k
e reci
pr
ocat
i
ng en
gi
ne i
s
bei
n
g im
pro
v
e
d
[
8
]
.
Fuel
cel
l
s
are
t
echn
o
l
o
gy
o
f
t
h
e
fut
u
re.
H
o
w
e
ver
,
t
h
e
r
e a
r
e
som
e
prot
o-t
y
p
e
dem
onst
r
at
i
o
n
pr
o
j
ect
s.
I
n
t
h
i
s
pa
per
exact
l
o
ss f
o
r
m
ul
a
m
e
t
hod i
s
used t
o
cal
cul
a
t
e
opt
i
m
al
DG
uni
t
’
s si
ze
and
pr
o
p
e
r
location. So that the real power lo
sses, react
ive power los
s
e
s were mini
mized and the c
o
rres
ponding voltage
pr
ofi
l
e
val
u
es
were
i
m
prove
d
.
T
h
e
pr
op
ose
d
ap
pr
oac
h
has
been
t
e
st
ed
o
n
36
-
bus
di
st
ri
bu
t
i
on sy
st
em
i
n
B
e
l
i
n
Sub
s
tatio
n in
Myan
m
a
r. Resu
lt ob
tain
fro
m
th
is appro
a
c
h
clearly explains the
optim
al location and siz
i
ng
of
DG th
ere b
y
min
i
mizes th
e p
o
wer lo
sses
of th
e syste
m
. Th
e rem
a
in
d
e
r of th
is p
a
p
e
r is
stru
ctured
as fo
llo
ws
Sect
i
on
II
p
r
e
s
ent
s
P
r
obl
em
Fo
rm
ul
at
i
on,
Sect
i
on
II
I e
x
pl
ai
ns a
b
out
P
r
o
p
o
sed
M
e
t
h
od
ol
o
g
y
,
Sect
i
o
n
I
V
descri
bes a
b
ou
t
Sim
u
l
a
t
i
on R
e
sul
t
s
an
d
Di
sc
ussi
o
n
s,
a
n
d
fi
nal
l
y
Sect
i
o
n
V e
xpl
ai
n
s
a
b
o
u
t
C
o
ncl
u
si
on
.
2.
PROBLEM FORMUL
ATION
The o
b
j
ect
i
v
e
of
pr
obl
em
i
s
to fi
n
d
t
h
e l
o
ca
t
i
on o
f
D
G
s a
nd i
t
s
si
ze fo
r
t
y
pe-1
, t
y
pe-
2
,
t
y
pe-3 an
d
type-4 DG t
o
minimize the real and
reactive power lo
sses
an
d to im
p
r
o
v
e
th
e
vo
ltag
e
p
r
o
f
ile. Real
Power loss
in
th
e
syste
m
can
b
e
calcu
lated
b
y
equ
a
tion
(1
),
g
i
v
e
n
th
e syste
m
o
p
e
rating
co
nd
itio
n,
N
1
i
N
1
j
j
i
j
i
ij
j
i
j
i
ij
L
Q
P
P
Q
β
Q
Q
P
P
α
P
(1
)
Whe
r
e,
j
i
j
i
ij
ij
δ
δ
cos
V
V
r
α
(2
)
j
i
j
i
ij
ij
δ
δ
sin
V
V
r
β
(3
)
The m
i
ni
m
i
ze
reactive
power loss will
have si
gni
ficant i
m
pact
on voltage
stability of the power
sy
stem
. The re
active p
o
we
r l
o
ss
fo
rm
ula is give
n
by
eq
uat
i
on
(
4
),
N
j
N
k
jk
jk
ζ
γ
11
k
j
k
j
k
j
k
j
L
Q
P
P
Q
Q
Q
P
P
Q
(4
)
Whe
r
e,
k
j
k
j
jk
jk
V
V
X
cos
(5
)
k
j
k
j
jk
δ
δ
sin
V
V
X
jk
(6
)
P
i
is
real power fl
ow at bus i in M
W
Q
i
is reactive power
fl
ow
at
bus i in MVAR
P
j
is real
power fl
ow at bus j in M
W
Q
j
is r
eactiv
e po
wer
f
l
ow at
bu
s
j
in
MVAR
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN:
208
8-8
7
0
8
Lo
ss Min
i
miza
t
i
o
n
o
f
Po
wer
Distrib
u
tio
n Netwo
r
k
u
s
ing
Differen
t
Typ
e
s
o
f
Distribu
ted
…
(
Su Hl
ai
ng
Wi
n)
92
0
R
ij
is resistance of the line co
nnecting
bus i
and j in
Ohm
s
X
ij
is Reactance of the
line c
o
nn
ecting
bus i
and j in
Ohm
s
V
i
and
V
j
a
r
e
bus
v
o
ltage
m
a
gnitu
de at
b
u
s i
an
d
j in
P
U
δ
i
and
δ
j
are
bus voltage
angle at bus i and j
2.
1. T
y
pe
-1
D
G
Photo
voltaic, m
i
cro turbines, fuel cells
which
are i
n
tegrated to m
a
in
grid wit
h
t
h
e
help
of
co
nv
er
ter
s
/inver
t
er
s ar
e
g
ood ex
am
p
l
es o
f
typ
e
1.
For type-1
DG, the
optim
al siz
e
of
DG
at eac
h
bus
i f
o
r m
i
nim
i
zing l
o
sses
is give
n
by
e
q
u
a
tion
(7
).
N
i
j
1,
j
j
ij
j
ij
i
ii
ii
Di
DGi
Q
β
P
α
Q
β
α
1
P
P
(7
)
P
i
=P
DGi
- P
Di
(8
)
R
eal po
wer
los
s
re
ductio
n
by
DG
,
100%
P
P
P
PLR
Loss
DG
Loss
Loss
(9
)
2.
2. T
y
pe
-2
D
G
Syn
c
hro
nou
s
m
o
to
r
s
su
ch
as g
a
s
turbine
s
are exam
ples for
type2.
For type-2
DG, the
optim
al siz
e
of
DG
at eac
h
bus
i f
o
r m
i
nim
i
zing l
o
sses
is give
n
by
e
q
u
a
tion
(9
),
N
i
j
j
j
ij
j
ij
P
Q
,
1
i
ii
ii
Di
DGi
P
β
α
1
Q
Q
(1
0)
Di
DGi
i
Q
Q
Q
(1
1)
Reactive powe
r loss
re
d
u
ctio
n by
D
G
,
100%
Q
Q
Q
QLR
Loss
DG
Loss
Loss
(1
2)
2.
3. T
y
pe
-3
D
G
DG
units that are base
d o
n
s
y
nch
r
o
n
ous m
achine
fall in type 3. The com
b
ination of both P
DGi
and
Q
DGi
injected a
t
the sam
e
bus
i can
pro
duce
the
optim
al size o
f
Ty
pe
3
D
G
,
2
2
DGi
Q
DGi
P
DGi
S
(1
3)
2.
4. T
y
pe
-4
D
G
The
DG will supply real
power and in t
u
rn will
absorb reactive power.
In
case of
t
h
e wind
t
u
rbines,
induction generator is
used
to produce real
power and
the reactive power
will be co
nsum
ed in the
process.
2
04
0
5
0
DG
P
.
.
DG
Q
(1
4)
The optim
al
sizes
at
vari
ous locations ha
ve been cal
culated f
o
r
dif
f
ere
n
t
ty
pes of
D
G
a
nd t
h
e los
s
es
are calculated
with optim
a
l
sizes for eac
h ca
se. The ca
se
with
m
i
ni
m
u
m
l
o
sses is selected as optim
al
location
fo
r eac
h ty
pe
DG
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN:
2
088
-87
08
I
J
ECE Vo
l.
5
,
No.
5
,
Octob
e
r
20
15
:
918
–
9
28
92
1
3.
PROP
OSE
D
METHO
D
OL
OGY
A com
puter
program
has been writte
n in MATLAB to calculate the op
timal sizes of DG at vari
ous
bus and approxi
m
ate
total losses with
DG at
different locations i
d
entify th
e best location.
A Newt
on
-R
a
phs
o
n
algo
rithm
based loa
d
flo
w
pr
o
g
ram
is use
d
to
sol
v
e the
loa
d
flo
w
pr
oblem
.
The
36
-
bus
dis
t
ributio
n sy
ste
m
is extracted
fr
om
230/
33/
1
1
k
V
B
e
lin
Su
b
s
tation in
Ky
a
uks
e area
o
f
Myanm
a
r. This ra
dial distribution system
c
onsists
of
six
m
a
in feede
r
. T
h
e total r
eal a
n
d reacti
v
e power loa
d
s
are
20
0M
W an
d
19
8 M
V
A
r
,
r
e
spectively
.
T
h
e B
e
lin distri
b
u
tion
sy
stem
, is as s
h
o
w
n in
Figu
re
1.
I
n
com
i
ng line
of B
e
lin station is
23
0
kV
line. Th
ere are thirty five outg
oing
lines. Their rated
voltage is
33 kV line. Analytical
m
e
thod is applied for opti
mal size a
nd l
o
cation
of
distri
bute
d
gene
ration
.
Figu
re
1.
Sin
g
l
e
line dia
g
ram
of Belin Substation in Myanmar
4.
SIM
U
LATI
O
N
RESULTS
AN
D DIS
C
US
SION
S
The Pr
o
pose
d
M
e
tho
dol
ogy
is tested on 3
6
-
bus
dist
ribution syste
m
in
Belin Substation in Myan
m
a
r
with di
ffe
rent
sizes are sim
u
lated in MAT
L
A
B envir
onm
ent to calculate
the opti
m
u
m
DG sizes
for
various
bus
es a
nd a
p
pr
oxim
a
te total
po
we
r loss
wit
h
D
G
at
vari
o
u
s locatio
n.
T
h
e line pa
ram
e
ters a
nd l
o
ad
d
a
ta are
listed in
Appendix. The
data
ar
e
b
a
sed on
100
MVA.
4.
1.
Op
ti
mum
Si
z
e
Al
l
o
ca
ti
o
n
Based
on
th
e pr
opo
sed
ap
pr
oach
op
ti
m
u
m
s
i
zes o
f
DG
’s a
r
e determ
ined usin
g eq
uation
(7) an
d (
1
0
)
at vari
ou
s
no
de
s f
o
r
B
e
lin
distrib
u
tion
sy
stem
in M
y
an
m
a
r. In Belin
distribution system
optim
u
m
sizes of
DG
ran
g
in
g f
r
o
m
4.6 M
W
to
97
.2
M
W
an
d
1.
67
M
VAR
to
7
2
.
2
M
V
AR
.
Afte
r co
n
n
ecting
D
G
’s
o
n
e by
o
n
e
it is
i
m
portant t
o
note that t
h
e tot
a
l power lo
s
s
i
s
low
e
st at th
e corresponding buses.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN:
208
8-8
7
0
8
Lo
ss Min
i
miza
t
i
o
n
o
f
Po
wer
Distrib
u
tio
n Netwo
r
k
u
s
ing
Differen
t
Typ
e
s
o
f
Distribu
ted
…
(
Su Hl
ai
ng
Wi
n)
92
2
0
10
20
30
40
50
60
70
80
90
100
13579
1
1
1
3
1
5
1
7
1
9
2
1
2
3
2
5
2
7
2
9
3
1
3
3
3
5
Si
z
e
o
f
D
G
Bu
s
N
u
m
b
er
DG
Si
z
e
in
MW
DG
Si
z
e
in
MV
A
R
Figure
2. Optim
u
m
real and
reactive
p
o
w
e
r sizes
o
f
ty
pe 1 and
ty
pe 2 DG
at 36
b
u
s
distri
butio
n
sy
stem
at
Belin Substation in Myanm
a
r
Based
on this result of
DG si
zing
the optimal
location of DG
is de
term
ined
whe
r
e t
h
e
total real a
n
d
reactive
power loss is m
i
ni
mum
at th
e r
e
spectiv
e bu
s
e
s.
Fig
u
r
e
2
sho
w
s
the
optim
u
m
real and
reactive powe
r
sizes of ty
p
e
1
and ty
pe 2
D
G
at 36
- b
u
s B
e
li
n distri
b
u
tion
s
y
stem
. Figure
3 s
h
o
w
s t
h
e o
p
t
im
u
m
size of ty
pe 3
DG at va
rio
u
s
no
des f
o
r
36
bus
distri
bution syste
m
. As far as one locati
on is co
ncer
n
e
d, in a distri
b
u
tion
sy
stem
, Figure
3 w
oul
d give
the value o
f
DG size to
ha
ve a p
o
ssible
m
i
nim
u
m
total loss. The
ran
g
e o
f
optim
u
m
sizes of ty
pe 3
D
G
is fr
om
0.3
M
W
to
3.
9
M
W
.
Now, it is i
m
portant t
o
identify the location i
n
which the total
power l
o
ss
will be m
i
ni
m
a
l.
Figu
re
3.
O
p
tim
u
m
size of t
y
pe 3
D
G
at
3
6
b
u
s
distrib
u
tio
n sy
stem
at B
e
lin S
ubstatio
n i
n
M
y
anm
a
r
Figu
re 4
sh
o
w
s the ba
r re
pre
s
entation
of
o
p
t
im
al si
zes of ty
pe 4
DG at all buses
fo
r 3
6
bus
sy
stem
.
From
figure, it can
be
obse
r
ved that the DG size do no
t f
o
l
l
ow a
re
gular
m
a
nner a
n
d th
e size is inde
p
e
nde
nt
of
locatio
n
of
b
u
s.
0
5
10
15
20
25
30
35
40
0
10
20
30
40
50
60
70
80
90
10
0
Bu
s
N
u
m
b
e
r
O
p
t
i
mu
m D
G
S
i
z
e
(
M
W
)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN:
2
088
-87
08
I
J
ECE Vo
l.
5
,
No.
5
,
Octob
e
r
20
15
:
918
–
9
28
92
3
Figu
re
4.
O
p
tim
u
m
size of ty
pe
4
DG
at 3
6
bus
distri
buti
o
n
sy
stem
in B
e
lin Su
bstatio
n
4.
2. Sel
ecti
o
n of
Optimum Location
W
i
t
h
the
help of
optim
u
m
DG sizes
obtained at
v
a
r
i
ous n
o
d
e
s it is best en
oug
h
to
f
i
nd
ou
t the
optim
u
m
location t
h
at woul
d leads to c
a
lculate the leas
t total power losses. In Be
lin
d
i
stributio
n sy
st
em
bus
24 is located
as optim
a
l
place
m
e
nt of DG. After
DG
place
m
e
nt total real and
reactive
powe
r losses are
reduce
d and average real power loss is reduc
e
d from
0.27
6
7
M
W
to 0.
2
0
30 M
W
a
nd a
v
erage reactiv
e po
we
r
loss is reduced to the range
o
f
0.4
7
63 M
V
A
R
to 0.0
4
91 M
VAR
.
Ta
ble 1
shows the
result of avera
g
e re
al and
reactive
power loss after conn
ecting
DG at B
e
lin distribu
tion system
in Myanm
a
r.
Table
1. Result of a
v
era
g
e
rea
l
an
d reactive power l
o
ss afte
r connecting DG
at
Belin
distri
butio
n
sy
stem
Bus Nu
m
b
er
at which
DG Connect
ed
Size of
DG in
MW
Size of
DG in
MVAR
Average Real P
o
wer Loss
in MW
Average React
i
ve P
o
w
er Loss
in MVAR
1 0
0
2.
095
0.
3301
2 97.
2
72.
2
2.
929
1.
371
3 8.
94
1.
67
0.
3405
0.
0253
4 21.
1
20.
7
0.
2635
-
0
.
6205
5 32.
3
10.
3
0.
5433
0.
1644
6 32.
3
24.
1
0.
195
-
0
.
6637
7 17.
6
7.
17
0.
2107
0.
1118
8 6.
09
5.
64
0.
2184
0.
0494
9 5.
59
5.
57
0.
2202
0.
0329
10
7.
83
3.
58
0.
2949
0.
0562
11
9.
6
3.
68
0.
2980
0.
0633
4
12
9.
42
3.
6
0.
2930
0.
0680
3
13
12.
5
7.
35
0.
4003
0.
0167
14
11.
4
11.
1
0.
2673
-
0
.
0836
15
9.
09
3.
81
0.
2954
0.
0396
9
16
9.
62
8.
36
0.
2070
-
0
.
1005
17
10.
4
5.
2
0.
3167
0.
045
18
10.
5
4.
71
0.
2180
0.
1062
19
6.
96
2.
84
0.
3188
0.
0691
5
20
10
4.
95
0.
316
0.
0191
3
21
10.
1
4.
74
0.
3134
0.
0282
6
22
10
4.
62
0.
3178
0.
1055
23
10.
5
4.
8
0.
0109
8
0.
0964
6
24
6.
43
2.
7
0.
2030
0.
0491
25
6.
29
3
0.
4881
0.
0989
2
26
20.
1
14.
8
0.
2801
0.
1374
27
9.
06
3.
5
0.
3204
0.
0774
28
9.
05
4.
46
0.
2014
0.
0722
29
6.
99
3.
47
0.
2337
0.
0347
30
93.
2
35.
3
4.
004
-
2
.
19
31
55.
8
18.
7
2.
202
0.
1365
32
7.
8
4.
18
0.
1782
0.
0353
4
33
8
4.
01
0.
1759
0.
0501
5
34
8.
1
4.
36
0.
2028
0.
0573
9
35
8.
5
3.
31
0.
2284
0.
0651
2
36
4.
6
2.
75
0.
1025
-
0
.
1268
0
5
10
15
20
25
30
35
40
0
10
20
30
40
50
60
B
u
s
N
u
m
ber
O
p
t
i
mu
m D
G
S
i
z
e
(
M
W)
Evaluation Warning : The document was created with Spire.PDF for Python.
IJECE
ISS
N
:
2088-8708
Lo
ss Min
i
miza
t
i
o
n
o
f
Po
wer
Distrib
u
tio
n Netwo
r
k
u
s
ing
Differen
t
Typ
e
s
o
f
Distribu
ted
…
(
Su Hl
ai
ng
Wi
n)
92
4
Fi
gu
re 5 s
h
ow
s t
h
e real
p
o
we
r l
o
sses
of t
h
e
sy
st
em
are reduced
by
o
p
t
i
m
a
l
pl
acem
e
nt
of t
y
pe 3
D
G
,
typ
e
4
DG
an
d with
ou
t DG.
Fi
gu
re
5.
R
eal
po
we
r l
o
ss
es at
b
u
ses
D
G
Ty
p
e
4,
D
G
Ty
pe
3 a
n
d
wi
t
h
o
u
t
DG
Fig
u
re 6
sho
w
s th
e to
tal reactiv
e p
o
wer l
o
sses o
f
th
e syst
e
m
fo
r d
i
fferen
t lo
catio
n
s
o
f
DG wit
h
t
h
e si
ze
obt
ai
n
e
d
fr
om
t
h
e pre
v
i
o
us sect
i
o
n.
Ba
se
Ca
s
e
‐
1
‐
0.
5
0
0.
5
1
1.
5
2
2.
5
3
3.
5
4
1
3
5
7
9
11
13
15
17
19
21
23
25
27
29
31
33
35
R
e
a
c
t
i
v
e
P
o
w
e
r
L
o
sse
s
(
M
W
)
Bu
s
N
u
m
b
e
r
Ba
se
Ca
s
e
DG
Ty
p
e
4
DG
Ty
p
e
3
Fi
gu
re
6.
R
eact
i
v
e p
o
w
er
l
o
ss
es at
b
u
ses
D
G
Ty
pe
4,
D
G
T
y
pe 3
an
d
wi
t
h
out
D
G
Tab
l
e
2
sh
ows
th
e op
tim
al lo
c
a
tio
n
an
d th
e op
ti
m
a
l size o
f
t
y
p
e
-1
DG
at 36
b
u
s
d
i
st
ribu
tio
n system
in
Myan
m
a
r
.
0
1
2
3
4
5
6
7
1
3579
1
1
1
3
1
5
1
7
1
9
2
1
2
3
2
5
2
7
2
9
3
1
3
3
3
5
Real Power
Losses
Bus Number
Base
Case
DG
Type
4
DG
Type
3
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN:
2
088
-87
08
I
J
ECE Vo
l.
5
,
No.
5
,
Octob
e
r
20
15
:
918
–
9
28
92
5
Table
2.
O
p
tim
a
l DG
Ty
pe
-1
Placem
e
nt of
the Sy
stem
DG installed
DG T
y
pe 1
DG T
y
pe 1
DG T
y
pe 1
DG T
y
pe 1
L
o
cation 24
30
5
33
PDG size,
M
W
6.
43
93.
2
32.
3
8
Ploss,
MW 0.
1098
0.
2186
0.
5433
0.
1759
Ploss r
e
duction,
%
60.
31%
39.
31%
22.
89%
15.
14%
Table
3 shows
the optim
al pla
c
e
m
en
t and the opti
m
a
l size of type-2
DG installed into t
h
e
syste
m
.
Table
3.
O
p
tim
a
l DG
Ty
pe
-2
Placem
e
nt of
the Sy
stem
DG installed
DG T
y
pe 2
DG T
y
pe 2
DG T
y
pe 2
DG T
y
pe 2
L
o
cation 24
30
5
33
QDG size
,
M
W
2.7
35.3
10.3
4.01
Qloss,
MW
0.
0491
-
1
.
72
0.
1644
0.
0501
5
Qloss r
e
duction,
%
39.
69%
52.
65%
47.
84%
40.
59%
Table 4 shows
the optim
al
location and the opti
m
a
l si
ze for
ty
pe-3
DG at 3
6
b
u
s distri
buti
on sy
stem
in Myanm
a
r.
Table
4.
O
p
tim
a
l DG
Ty
pe
-3
Placem
e
nt of
the Sy
stem
DG installed
DG T
y
pe 3
DG T
y
pe 3
DG T
y
pe 3
DG T
y
pe 3
L
o
cation 24
30
5
33
PDG size,
M
W
6.
97
93.
7
32.
3
8.
94
Ploss,
MW
0.
1949
0.
1283
0.
0704
8
0.
1385
Ploss r
e
duction,
%
29.
56%
9.
429%
83.
17%
24.
86%
Table 5 s
h
ows
the o
p
tim
a
l location
(
bus
n
u
m
b
er) an
d si
ze (P
DG
size and
Q
DG
size are real and
reactive ca
pacity of
DG i
n
st
alled, respectively) fo
r
different types
of DG installed int
o
t
h
e system
.
Table
5.
Optim
a
l DG Placem
e
n
t of t
h
e Syste
m
for DG
Place
m
e
nt
DG installed
DG T
y
pe 1
DG T
y
pe 2
DG T
y
pe 3
DG T
y
pe 4
L
o
cation
24
30
5 33
PDG size,
M
W
6.
43
32.
3
5.
71
QDG size
,
MVA
r
35.3
Ploss,
MW
0.
1098
4.
004
0.
0704
8
0.
0964
Qloss,
M
V
Ar
0.
0491
1
-
1
.
72
0.
0184
5
0.
0781
7
Ploss r
e
duction,
%
60.
31%
52.
65%
83.
17%
50.
49%
Qloss r
e
duction,
%
74.
39%
66.
5%
94.
16%
34.
39%
These ta
bles also
prese
n
t the i
n
form
ation of
real
(
P
lo
ss redu
ctio
n)
an
d r
e
activ
e po
wer
loss r
e
d
u
c
tion
(Ql
o
ss
reducti
o
n) i
n
term
of
perce
n
tage
, a
n
d total
real
(Ploss) a
n
d reacti
v
e
powe
r loss
(Qloss) i
n
term
of M
W
and
MVAR, re
spectively.
The results in these tables
sh
ow the significant real and r
eactive power loss re
duction after
installation of
DG in the syst
e
m
. In
addition, optim
al s
i
ze a
nd location
of
DG are also changed with
different
type of
DG installed in the syste
m
.
D
G
ty
pe
1 ca
n re
d
u
ce the
re
al and
reactive
po
we
r loss
by
60
.3
1%
& 7
4
.
3
9
%
com
p
are
d
to 5
2
.
6
5%
& 6
6
.
5
%
,
83
.1
7%
&
94
.1
6%,
an
d
50
.4
9%
&
3
4
.
3
9
%
f
o
r
D
G
ty
pe
2,
D
G
t
y
pe 3
,
a
n
d
D
G
ty
pe 4,
res
p
ecti
v
ely
.
a.
Op
ti
mum B
u
s V
o
l
t
age
Pr
of
i
l
e
The
Voltage a
t
vari
ous buse
s should
be m
a
intain
ed
withi
n
the
accepta
ble li
m
its to m
eet out t
h
e
powe
r syste
m
dem
a
nd. But t
h
e Bus
voltage
m
a
y reach the
perm
issible lim
it when DG i
s
not c
o
nnecte
d
to the
Evaluation Warning : The document was created with Spire.PDF for Python.
I
J
ECE
I
S
SN:
208
8-8
7
0
8
Lo
ss Min
i
miza
t
i
o
n
o
f
Po
wer
Distrib
u
tio
n Netwo
r
k
u
s
ing
Differen
t
Typ
e
s
o
f
Distribu
ted
…
(
Su Hl
ai
ng
Wi
n)
92
6
d
i
str
i
bu
tio
n sy
ste
m
o
r
th
e
bu
s
vo
ltag
e
m
a
y lack
d
u
e
to
som
e
distur
b
a
nces.
F
o
r
thi
s
reas
o
n
Distr
i
buted
Gene
ration shoul
d be
placed a
nd size
d at the releva
nt bus location in
radial Distribution system
so that the
bus
v
o
ltage
p
r
ofile
gets im
prove
d.
T
h
e im
pro
v
em
ent of
v
o
ltage
pr
ofile
of
the sy
stem
fo
r
diffe
r
e
n
t ty
pes
of
DG
installed is
sh
ow
n i
n
Fi
g
u
r
e
7.
Figu
re 7.
Va
riation of
v
o
ltage pr
of
ile fo
r 36
b
u
s di
stribution system
5.
CO
NCL
USI
O
N
Diffe
re
nt ty
pes of
D
G
installed in the
syste
m
have di
ffe
rent im
pacts o
n
reactive power loss
reduction. Opt
i
m
a
l s
i
ze and location
of DG are also
c
h
ange
d wit
h
di
ffe
rent ty
pe o
f
DG installed
in the
syste
m
. Exact
loss
form
ula is used t
o
determine optim
al
s
i
ze and the location
for ty
pe
-
1
, ty
pe
-
2
, ty
pe
-3
an
d
type-4
DGs. T
h
e four types of
DGs e
ffectively reduced
t
h
e real powe
r
loss, reactive
powe
r loss a
nd
voltage
pr
ofile a
r
e als
o
im
pro
v
e
d
.
DG ty
pe
3 has
the
m
o
st effe
ctive way
of
r
e
duci
ng
real and
reactive p
o
w
er los
s
, f
o
llo
win
g
by
D
G
ty
pe 1, D
G
ty
p
e
2, an
d D
G
ty
pe 4,
res
p
ectiv
ely
.
It is
also i
n
teresting to note th
at DG ty
pe 3
has the hi
ghe
st
loss re
d
u
ction
since it can
ge
nerate
bot
h rea
l
and
reactiv
e
po
we
r an
d
DG
ty
pe 4
has the
lowest lo
ss re
duct
i
o
n
since it consum
es reactive
powe
r i
n
the
syste
m
.
The m
e
thodol
ogy of finding prope
r
location and si
ze of DG in order to
m
i
nim
i
ze real and reactive
po
we
r losses i
n
ra
dial distri
butio
n sy
stem is
m
o
re effective in term
of
v
o
ltage im
provem
e
nt than
that o
f
finding prope
r location and
si
ze
of
DG in
order to m
i
nim
i
z
e
r
eal a
n
d react
ive power losses.
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
0.
9
2
0.
9
3
0.
9
4
0.
9
5
0.
9
6
0.
9
7
0.
9
8
0.
9
9
1
1.
0
1
Bu
s
N
u
m
b
e
r
V
o
l
t
ag
e (
p
.
u
)
Ba
s
e
C
a
s
e
DG
T
y
p
e
1
DG
T
y
p
e
2
DG
T
y
p
e
3
DG
T
y
p
e
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
SN:
2
088
-87
08
I
J
ECE Vo
l.
5
,
No.
5
,
Octob
e
r
20
15
:
918
–
9
28
92
7
Ap
pe
ndi
x A
Line
Param
e
ters a
n
d
L
o
ad
Da
ta of
3
6
-
b
us
Di
stributio
n
Sy
stem
Line Para
m
e
te
rs
Load Data
Fro
m
Bus
No.
To Bu
s N
o
.
R
(
p
.u
)
X
(
p
.u
)
Bu
s N
o
.
P
(
p
.
u
)
Q
(p
.
u
)
1 2
0.
0002
0.
0003
1.
200
198
2 3
0.
0021
0.
0039
2.
198
196.
99
2 4
0.
0195
0.
0365
3.
1.
19
1.
142
4 32
0.
1332
0.
1540
4.
8
6.
5
4 33
0.
1332
0.
1540
5.
6.
17
5.
37
2 5
0.
0167
0.
0313
6.
8.
6
6.
7
2 6
0.
0181
0.
0339
7.
2.
06
2.
045
6 34
0.
0444
0.
0513
8.
2.
06
2.
045
6 35
0.
3553
0.
4109
9.
2.
04
2.
03
6 36
0.
8883
1.
0272
10
.
2.
91
2.
75
2 7
0.
0017
0.
0031
11.
2.
71
2.
15
2 8
0.
0014
0.
0026
12.
2.
54
2.
03
2 9
0.
0028
0.
0052
13.
2.
8
2.
6
2 10
0.
0003
0.
0005
14.
2.
4
2.
3
2 11
0.
0003
0.
0006
15.
2.
4
2.
3
2 12
0.
0023
0.
0043
16.
2.
98
1.
59
2 13
0.
0006
0.
0010
17.
2.
84
2.
613
2 14
0.
0005
0.
0008
18.
2.
4 2.
13
2 15
0.
0004
0.
0008
19.
2.
45
2.
338
2 16
0.
0119
0.
0221
20.
2.
45
2.
338
2 17
0.
0009
0.
0016
21.
2.
54
2.
15
2 18
0.
0009
0.
0017
22.
2.
5
2.
037
2 19
0.
0047
0.
0088
23.
2.
39
2.
225
2 20
0.
0020
0.
0038
24.
2.
57
2.
31
2 21
0.
0016
0.
0031
25.
4.
2
3.
15
2 22
0.
0017
0.
0031
26.
8.
4
6.
3
2 23
0.
0011
0.
0020
27.
2.
2
2.
15
2 24
0.
0011
0.
0021
28
.
2.
02
2.
015
2 25
0.
0181
0.
0339
29.
2.
01
2.
007
2 26
0.
0002
0.
0004
30.
28.
53
23.
75
2 27
0.
0004
0.
0008
31.
25.
68
29.
613
2
28
0.
0021
0.
0039
32.
2.
6
1.
95
2 29
0.
0053
0.
0099
33.
2.
83
2.
123
2 30
0.
0174
0.
0325
34.
4.
83
3.
623
2 31
0.
0178
0.
0334
35.
5.
83
5.
27
36.
2.
53
2.
373
ACKNOWLE
DGE
M
ENT
The a
u
thors a
r
e dee
p
ly grateful to U
Ky
aw
W
i
n a
nd
Daw
Kyi Hta
y
for their s
u
pports and
encouragem
ent to attain
he
r
destination
without any troubl
e and all
th
e
per
s
on
s
wh
o shar
e th
e tr
oub
le o
f
the
author on any situation
i
n
trying this
paper.
REFERE
NC
ES
[1]
Niknam T, Tah
e
ri SI, Agh
aei J,
Ta
batabaei S, N
a
y
e
ripour
M. A
modi
fi
ed honey
bee mating op
timization
algor
ithm
for multiobjectiv
e pl
acement of
r
e
newable
energ
y
resour
ces. Appl
Energ
y
2011
; 88
(12): 4817–30.
[2]
Taher N.
A ne
w HBMO algorithm
for m
u
ltiobjec
tive d
a
il
y
V
o
lt/Var
control
i
n
distribution s
y
stem
s consideri
ng
distributed
gen
e
r
a
tors. Appl En
er
g
y
2011; 88(3)
:
778–88.
[3]
Ma
rtinez
-Roja
s
M,
Sumpe
r
A,
Gomis-
Bellmunt O, Sudrià-Andreu A. Reactiv
e
power dispatch in
wind farms usi
ng
particle swarm optimization
tech
ni
que
and f
easib
le solu
tions sear
ch. Appl En
erg
y
2011; 88(12)
: 4
678–86.
[4]
Manfren M, Caputo P, Costa G
.
Parad
i
gm shift in urb
a
n
energ
y
s
y
stems th
rou
gh distributed g
e
neration: Meth
ods
and models. Appl En
erg
y
2011
; 8
8
(4): 1032–48.
[5]
Bakos GC. Distributed power generation:
a c
a
s
e
s
t
ud
y
of s
m
all s
cal
e P
V
power
plant in Greece.
Appl Energ
y
20
09;
86(9): 1757–66.
[6]
W
.
El-Khatt
am
,
M.M.A.Salam
a
,
“
D
istributed G
e
nera
tion
Te
ch
n
o
logies, defin
itions and be
n
e
fit
s
”, E
l
ec
tri
c
P
o
wer
S
y
stems Resear
ch, Vol. No.71
,
p
p
. 119-128
, 200
4.
[7]
Thom
as Acker
m
ann, Goran A
ndersson,
L
e
nna
rt Soder,
“
D
istributed
Gener
a
ti
on: a
defin
ition
”
,
El
ectr
i
c
Power
S
y
stems Resear
ch, Vol. No. 57, p
p
.195–204, 2001
.
[8]
Carm
en L
.
T.
B
o
rges, Dj
alm
a
M. Falc
ao,
“
O
ptim
al Distr
i
bute
d
Genera
tion
al
loca
tion for
re
li
abili
t
y
,
losses a
n
d
voltag
e
improvement”, Electr
i
ca
l power
and En
er
g
y
s
y
stems, Vol. No. 28
, pp
. 413-
420, 2006
.
Evaluation Warning : The document was created with Spire.PDF for Python.