Indonesian J
ournal of Ele
c
trical Engin
eering and
Computer Sci
e
nce
Vol. 2, No. 1,
April 201
6, pp. 40 ~ 48
DOI: 10.115
9
1
/ijeecs.v2.i1.pp40
-48
40
Re
cei
v
ed Fe
brua
ry 3, 201
6; Revi
se
d Ma
rch 18, 201
6
;
Accepte
d
March 26, 201
6
Determining Basic Cost of Electricity on the
Regionalization System
Yusra Sabri
1
, Hermagas
a
n
tos Zein*
2
1
Electrical En
gi
neer
ing, ST
EI,
Institute T
e
c
hnolo
g
y
of Band
u
ng, IT
B Bandu
ng, Indo
nesi
a
2
Energ
y
Co
nve
r
sion En
gin
eer
i
ng De
pt. State Pol
y
t
e
ch
nic of Band
un
g, POLBAN Band
un
g, Indones
ia
*Corres
p
o
ndi
n
g
author, e-ma
i
l
:
y
u
sra.s
abri@
yma
il.com
1
, he
rmaga
_s@
y
a
h
oo.co.id
2
A
b
st
r
a
ct
In electric
bus
i
ness, costs tha
t
support
prod
u
c
tion ca
lle
d w
i
th al
low
abl
e e
l
e
c
tric cost or b
a
s
ic cost
of el
ectricity. T
herefor
e, the
b
a
sic
c
o
st of
el
ectricity h
a
s to
get
more
atte
ntion
seri
ous
ly
from the
e
l
ect
r
ic
compa
n
y in
or
der to b
e
trusted by c
onsu
m
ers. In
deter
mi
nin
g
of the cos
t
s
must be tra
n
spar
ence
an
d
d
o
the o
p
ti
mi
z
i
n
g
in c
a
lcu
l
atin
g
process
bec
au
se of r
e
ve
nues
from custo
m
e
r
s hav
e to
be
abl
e to c
o
ver
th
e
basic c
o
st of el
ectricity so that
the el
ectric co
mp
any
does
n
o
t go b
ankr
upt. T
h
is pa
per w
i
l
l
create c
onc
e
p
t
and
for
m
ul
atio
ns for
deter
mi
nin
g
th
e b
a
sic
cost of
el
ectricity in
a
re
gi
on. T
her
e
are
a few
differe
n
t
character
i
stics of each reg
i
on
w
ill infl
ue
nce t
he costs. The basic
cost of el
ectricity can be
different amon
g
regi
ons th
at w
e
re ca
use
d
by
ge
ner
ati
ng, i
n
vestme
nt a
nd
losses
in t
he r
egi
on. L
a
ter
h
a
s b
een
do
ne
a
nu
meric
a
l si
mu
latio
n
to simple
pow
er system
and t
he res
u
lts
are matchin
g
w
i
th desired ex
pectatio
n
s.
Ke
y
w
ords
: ele
c
tric busin
ess, allow
a
ble cost
,
create conce
p
t, cost
formu
l
ati
ons
Copy
right
©
2016 In
stitu
t
e o
f
Ad
van
ced
En
g
i
n
eerin
g and
Scien
ce. All
rig
h
t
s reser
ve
d
.
1. Introduc
tion
Electri
c
bu
si
ness be
gin
s
from natura
l
m
onopoly, integrate
d
sy
stem, cont
rol
l
ed by
govern
m
ent. In last three decade
s, sev
e
ral countri
es have chan
g
ed their ele
c
t
r
ic bu
sin
e
ss into
forms mo
re
efficien
cy an
d tran
sp
aren
ce, i.e.
com
petitive syste
m
[1, 2, 3,
4, 5]. Anoth
e
r
developin
g
o
f
the integ
r
at
ed
syst
em i
s
se
parated t
o
be
a fe
w
region
s th
at i
s
calle
d ele
c
tric
regio
nali
z
atio
n [6], and every regio
nal ha
ve indepe
nde
nce o
w
n a
u
th
ority.
Regi
on authority will be obligated to meet deman
d wi
th optimal electri
c
cost. So, every
regio
n
can b
e
distin
ct its
electri
c
co
st that suita
b
le
with the regi
o
n
co
ndition. I
n
this
context
,
the
regio
n
s
of surplu
s e
n
e
r
g
y
can
sell t
heir e
n
e
r
gy
to the regi
o
n
s of min
u
s energy. In the
regio
nali
z
atio
n context, this pap
er p
r
o
p
o
se
s dete
r
mi
nation ba
si
c
co
st of electri
c
ity, or co
st. The
cal
c
ulatio
n u
s
ed
in
determining th
e
cost
will a
d
o
p
ts el
ectri
c
and
economi
c
fo
rmulatio
n
s
.
Furthe
rmo
r
e,
this pa
per creates a
cost
con
c
e
p
t
an
d formulatio
ns
for a regio
n
. The co
ncept of
the co
st p
r
op
ose
d
ha
s b
e
e
n
sim
u
lated
o
n
the
simple
power
syste
m
and th
e results have
met
the
desi
r
ed exp
e
c
tation
s.
2. Rese
arch
Metho
d
2.1. Terminolog
y
and Concep
t
A. Terminolog
y
There are three terms that
need
to be considered here, nam
ely
cost, price and the tariff.
The thre
e thing is so
met
h
ing differe
nt from one a
nd other, an
d they will be explained
the
following.
Price i
s
the value dete
r
min
ed by the su
p
p
ly
and dem
a
nd. In determ
i
ning the p
r
ice is not
dire
ctly rel
a
te
d to
pro
d
u
c
tion
co
sts. In
a ma
rket
system, wh
en
o
c
curs
su
rplu
s su
pply the
n
the
price will go
down, and vice versa. In this case
the sup
p
lier
coul
d get profit or loss. Thu
s
the
con
c
e
p
t of the price sh
oul
d be in the form of
the busi
ness ma
rket, su
ch a
s
com
petition.
Tariff state
s
the re
al pri
c
e
cha
r
ge
d to consum
e
r
s. Usually the ta
ri
ff is influence
d
by the
ability
of
(eco
nomic and so
cial) of
co
nsu
m
ers, co
rpo
r
ate profits
an
d the gove
r
n
m
ent's
su
bsid
ies,
and al
so influ
enced by the
political will i
n
doing t
he d
e
ci
sion
-ma
k
in
g. In determi
ning the ele
c
t
r
ic
co
st is gene
rally defined
by compro
mise (r
e
c
on
ciliation) bet
ween t
he pa
rties concerne
d.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 1, April 2016 : 40 – 48
41
Thereby, the tariff is not only influence
d
by purely
tech
nical but also influe
nced by the political,
eco
nomi
c
an
d so
cial.
While the
cost is just a techni
cal nature, i.
e. all costs that sup
p
o
r
ts the produ
ction of
electri
c
al
ene
rgy from
power pla
n
ts to t
he re
gion. In
this case, the cost
s a
r
e
not su
ppo
rting
prod
uctio
n
p
r
oce
s
s o
r
u
n
a
llowable
co
st, su
ch
a
s
non-te
ch
nical
ene
rgy lo
sses,
will not
be
involved. So, all co
sts th
at sup
port the
p
r
odu
ction
pro
c
e
ss i
s
d
e
fin
ed a
s
ba
si
c cost of ele
c
tri
c
ity
in this pap
er.
B. The Con
c
ept of
Cos
t
As for the
costs th
at will
be dete
r
mi
ned he
re
are for the re
gion
s of gen
eration,
transmissio
n, distributio
n, medium volta
ge and lo
w voltage dist
rib
u
tion su
ch a
s
shown by Figure
1. The
Figu
re
sh
ows th
e m
odel
of the el
ectri
c
p
o
wer
system th
at h
a
s t
w
o type
s
of po
wer pla
n
t
s,
namely IPP powe
r
plant, in
depe
ndent p
o
we
r produ
cti
on,
with the co
st of
purch
ase, IPP price,
has
bee
n det
ermin
ed b
e
fo
re, a
s
fixed p
r
odu
ction
costs, and
po
wer plants with
p
r
odu
ction
costs
cha
nge eve
r
y time. The power pl
ants can b
e
con
necte
d to th
e ultra/extra
/high voltage
s or
distrib
u
tion voltage both p
r
imary an
d seco
nda
ry.
Furtherm
o
re, the co
st can b
e
determin
e
d
on
each con
s
um
er, wh
ethe
r the con
s
ume
r
s of ultra, ex
t
r
a, high, p
r
im
ary or
se
con
dary voltage
s. In
addition, th
e
electri
c
po
we
r
system
ca
n
also b
e
divid
ed into
several region
s, fo
r exam
ple
Ja
va,
Madu
ra a
nd
Bali system
or JAMA
LI system, di
vide
d into 4 regi
ons,
Ja
karta
-
Tange
ra
ng,
We
st
Java, Ce
ntral
Java and Ea
st Java.
Figure 1. Model com
p
lete
electri
c
po
we
r system
2.2 Electric
Po
w
e
r Sy
stem
Referrin
g to Figure 1 abo
ve, the electri
c
po
wer
syst
em by functio
n
can be divi
ded into
three sectio
n
s
, namely:
Gene
ration
Tran
smi
ssi
on
Distri
bution
Trans
miss
ion Cost
o
f
ultra
/
ex
tra Voltag
e
IPP Price
Di
st
ribu
ti
on
Co
st
o
f
Med
i
um
V
o
lta
ge
Trans
miss
ion Cost
of High
Vol
t
age
Consumer
s o
f
E/
U v
o
ltage
Consumer
s o
f
Low
Voltage
Consumer
s o
f
Medium Vol
t
age
Consumer
s o
f
High Vol
t
age
Dis
t
ribut
i
on Cos
t
o
f
Low
Voltage
C
o
s
t
C
o
s
t
C
o
s
t
C
o
s
t
IPP Price
IPP Price
B
G
(
P)
=
c
+b
P
+
ap
2
B
G
(
P)
=
c
+
bP
+
ap
2
B
G
(
P)
=
c
+
bP
+
ap
2
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
Det
e
rm
inin
g B
a
sic
Co
st
of
E
l
ect
r
ic
it
y on
the Regio
nali
z
ation Syste
m
(Yusr
a
Sab
r
i)
42
A. Gener
a
tion
Powe
r pla
n
t can
be differentiated into
two gr
oup
s, private,
IPP,
and the
state
,
like PT
PLN in
Ind
o
nesi
a
, comp
anie
s
. Especially for IPP,
ele
c
tri
c
ity produ
ction
co
sts state
d
in
the
purcha
s
e
pri
c
e. IPP can b
e
co
nne
cted t
o
variou
s pl
a
c
e
s
in the
ele
c
tri
c
po
we
r system, either
on
the transmission or dist
rib
u
tion net
wo
rks. Co
st on the powe
r
plan
t terminal is the prod
uctio
n
co
sts befo
r
e
energy tran
smitted to a grid.
In an ele
c
tri
c
power
syste
m
, there a
r
e
severa
l state
power pl
ants
con
n
e
c
ted to
the grid
in variou
s reg
i
ons. To dete
r
mine the am
ount of
powe
r
generated b
y
generato
r
s at a certain lo
ad
will be determined ba
se
d on the optima
l
powe
r
flow
so the produ
cti
on co
st of power pl
ants to be
minimal. Cost
determi
nation based
on t
he optimal
power flow
will
us
e the cost
characteri
stic of
power pla
n
t in the followin
g
equatio
n.
2
)
(
i
i
i
i
i
i
i
P
a
P
b
c
P
B
(1)
Whe
r
e,
B
i
(
P
i
) is ba
si
c cost pe
r hou
r of generatin
g
i.
c
i
, b
i
and
a
i
are ge
ne
ratin
g
con
s
tant
s
P
i
is active po
wer p
r
o
d
u
c
ed
by generatin
g i.
Suppo
se the
r
e are
N bu
se
s, total syste
m
load is P
sys
and total powe
r
of IPP is P
IPP
then the
formulatio
n o
p
timization i
s
max
min
i
max
1
i
max
min
1
2
V
:
Voltage
4.
0
:
Line
3.
P
:
Load
2.
:
Generator
1.
:
Subject to
min
:
Objective
i
i
ij
ij
IPP
sis
Loss
N
i
i
i
i
N
i
i
i
i
i
i
V
V
S
S
P
P
P
P
P
P
P
c
P
b
c
B
(2)
Nee
d
to
rem
e
mbe
r
that t
he lo
ad
s will
vary ea
ch ti
me so the
d
a
ily load
du
ration
curve
will b
e
different eve
r
y day. Experiences
sh
ow that for
wo
rking
days th
e curve
s
a
r
e
almo
st simil
a
r,
whe
r
ea
s fo
r
red d
a
tes an
d
holid
ays th
e
curve
a
r
e
also
quite
simil
a
r. In the
cost
cal
c
ulatio
n
will
use an
nual lo
ad duration curve of
the next year, because the co
st
will apply for the next yea
r
.
The
ann
ual l
o
ad d
u
ration
curve
used i
s
t
he ave
r
ag
e
o
f
the d
a
ily loa
d
du
ration
cu
rve in
on
e ye
ar.
Whe
r
e thi
s
curve is e
s
tim
a
ted from the
results of
loa
d
fore
ca
st within the one y
ear a
head. T
hen,
be stated th
at load every h
ourly is n
o
t chang
e so
th
e
r
e are 24 va
ri
ations of the l
oad in o
ne d
a
y.
Furthe
rmo
r
e, it can be calculated cost e
a
ch p
o
wer pl
ant based on
estimates of
the annu
al load
duratio
n cu
rv
e throug
h eq
uation 3.
24
1
)
(
24
1
)
(
j
i
j
i
i
G
P
B
P
B
(3)
B. Transmissio
n
Tran
smi
ssi
on
ca
n b
e
con
s
iste
d of
hig
h
, ex
tra
and
ultra
-voltag
e
s. T
he tran
smissio
n
cau
s
in
g
co
st
is d
ue to
inve
stment of th
e
tran
sm
i
ssi
on
network. In t
he fa
cts, the
comp
one
nts
o
f
the tra
n
smi
ssion
wa
s b
u
ilt in the
time
s that a
r
e
not
sa
me, the
r
e
are ol
d
com
pone
nts,
so
me
comp
one
nts
are ne
wly bu
ilt and some
compo
nent
s have been retrofitted. These all mu
st be
cal
c
ulate
d
un
der ann
ual
in
vestment co
sts
ba
se
d
on t
he e
c
on
omi
c
prin
ciple. It should
be
not
ed
that the outlived compon
ents of the economi
c
life will be
valued zero in determini
ng the
transmissio
n
co
st. The e
c
o
nomic
cal
c
ul
a
t
ion re
sults
m
u
st be
able t
o
determine t
he an
nual
co
s
t
of each
com
p
onent involve
d
in the tran
smissi
on net
work. Thi
s
is v
e
ry importa
nt beca
u
se in the
cost cal
c
ul
ation in
a region will
be determined
by t
he transmi
s
si
on comp
onents in the region.
But whe
n
th
e compo
nent
s a
r
e
in the
two
regio
n
s,
each regio
n
will b
ear half
of cost
of t
he
comp
one
nts.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 1, April 2016 : 40 – 48
43
C. Distribu
tion
Whe
r
ea
s the
distrib
u
tion
of distingui
sh
ed to
be two
grou
ps, n
a
m
e
ly the distri
bution of
medium
and
low voltag
es.
Cal
c
ulatio
n
of the an
nual
co
st of thi
s
distrib
u
tion
will be id
entica
l
to
the cal
c
ulatio
n of the annu
al co
st of tr
an
smissio
n
that have bee
n de
scribe
d abov
e.
2.3 Cos
t
Co
mponents
As a
con
s
e
q
uen
ce of the
function
s of
gene
ra
tion, transmi
ssion
a
nd di
stributio
n in an
electri
c
po
we
r sy
stem, the
co
st will
ap
p
ear
on
each
of these fun
c
tions. T
hey
wi
ll emerge i
n
t
h
e
power
plant
as
co
sts of t
he fuel,
asse
ts an
d O&
M,
or op
eratio
n
and
maint
e
n
ance, wherea
s in
the transmission and di
stri
bution networks
will em
erg
e
cost
s of asset, O&M and loss e
nerg
y
.
Thus th
ere
wi
ll five cost co
mpone
nts, na
mely:
1) Fuel
2) Ass
e
ts
3) Ope
r
atio
n & Maintenan
ce (O & M)
4) Lo
ss of en
ergy
5) Buying and selling
of inter-regi
on
A. Fuel
Fuel is
only found in th
e p
o
we
r pla
n
t on
ly. Th
is co
mp
onent i
s
a mi
xture of different types
of fuel, fuel
m
i
x, whe
r
e
po
wer pl
ants in
volved in
ele
c
tric
po
wer sy
stem
s u
s
in
g
variou
s fu
els,
like
water, co
al, diesel/MFO a
nd
n
a
tural
ga
s.
Becau
s
e
o
f
the pri
c
e
of
ea
ch type
of fuel an
d
co
st-
cha
r
a
c
teri
stic of each po
wer plant are not sa
me a
n
d
then it needs to be opti
m
ized. Th
us
here
need o
p
timization metho
d
to determin
e
the optimal fuel
mix in se
rving total loa
d
of system from
all power pla
n
ts co
nne
cte
d
to the electric po
wer
syst
em.
The be
st opti
m
ization
met
hod until n
o
w is t
he optim
a
l
powe
r
flo
w
method. Con
c
eptu
a
lly,
this meth
od i
s
very
sup
e
ri
or b
e
ca
use al
l con
s
trai
nts
have be
en a
c
comm
odate
d
in it. But it is less
robu
st in it
s
operation. Th
e ro
bu
stne
ss will de
pen
d
on the m
e
tho
d
used, for e
x
ample
simpl
e
x,
linear, q
uad
ratic, and inte
rior p
o
int met
hod
s. In the pra
c
tice
of the interio
r
poi
nt method, [7
], is
sup
e
rio
r
be
ca
use it is faste
r
and mo
re ro
bust.
The very ro
b
u
st metho
d
is economi
c
disp
atch, ED. Howeve
r, this metho
d
doe
s not
involve con
s
traints of voltage an
d transmiss
io
n
line so th
e issue of losse
s
ca
n’t be
accomm
odat
ed. As the
co
nse
que
nces,
it may not op
eration
a
l be
cause of if the
ED results a
r
e
applied
will possible be a
few
constraints violat
ed,
such as the transmi
ssion li
ne overloaded.
Becau
s
e
of this meth
od is very robu
st, in pra
c
tice th
e losse
s
can
be predi
cted,
i.e. 2.5%, and
the possibility
of the load is not t
oo larg
e, it is advisa
b
le to use thi
s
metho
d
in determi
ning t
he
co
st for ea
ch
comp
one
nt of the robu
st gri
d
.
B. Ass
e
t
Asset i
s
the
entire inve
stments from
powe
r
pla
n
t
to distributi
on. Providin
g asset
depe
nd
s al
so
on th
e
relia
bility level de
sire
d. Th
e m
o
re
asset
s
wi
ll be
more
re
liable, b
u
t it
will
cau
s
e the hi
gher
co
sts. F
o
r example t
w
o gen
erat
in
g unit workin
g on a syste
m
would b
e
more
reliabl
e wh
en
using o
ne un
it and so on.
Relatio
n
ship
betwe
en a
s
set cost p
r
ovisi
on and
reliabi
lity
is sho
w
n by
Figure 2. On the Figure, for
relia
bility approa
chin
g one re
quires additional very
much asset in order to improve the reli
ability of a ve
ry small.
Co
st
R
e
lia
b
i
li
t
y
c
o
s
t
Ass
e
t
co
st
Figure 2. Reli
ability and asset co
sts
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
Det
e
rm
inin
g B
a
sic
Co
st
of
E
l
ect
r
ic
it
y on
the Regio
nali
z
ation Syste
m
(Yusr
a
Sab
r
i)
44
Gene
rally, procu
r
em
ent of
assets
are
b
a
se
d
on th
e level of relia
bi
lity to anticip
ate the
comp
one
nt failure
s, such
as n
-
1, this
mean
s t
hat d
e
spite
of the
failure of o
n
e
comp
one
nt, the
system is n
o
t colla
pse. In the rule
s of the ec
o
nomi
c
, the asset
s
are
generally determin
ed by h
o
w
much
dep
re
ssion eve
r
y year. Th
e de
pre
ssi
on i
s
d
epen
dent on
investment
s, interest rates,
con
s
tru
c
tion
and econo
mi
c life of the a
s
set. The iss
ue of these a
s
sets is the d
e
sig
n
desi
r
ed
to
power sy
ste
m
based on t
he quality of servi
c
e ele
c
tricity desired.
It depend
s o
n
the ability of th
e
financi
a
l gove
r
nme
n
t in the resp
on
sibility to supply
the electri
c
ity afforda
b
le by the comm
unitie
s
.
Becau
s
e
of t
he el
ectri
c
po
wer sy
stem i
s
ma
nag
ed
b
y
PT. PLN a
nd o
w
n
ed by
the g
o
vern
m
ent.
Obviou
sly this syste
m
is a
monop
oly system by the governme
n
t.
C. O&M
In the ma
nag
ement of
ele
c
tri
c
ity sho
u
l
d
pr
ovide th
e
co
st of O&M
in o
r
de
r to
g
e
t good
perfo
rman
ce.
The
cost i
s
intende
d to l
ong
ope
ratio
n
s
so
the
r
e i
s
n
o
p
r
obl
em
in the
future
. It
con
s
i
s
ts of fixed an
d varia
b
l
e co
sts. Fixe
d co
st
will
be
related to
em
ployee
s, mai
n
tenan
ce, fixed
and life
cycle
,
and in
su
ran
c
e.
While th
e
variable
cost
s con
s
ist
of startup / sta
r
t
down, oil, fue
l
additives an
d
chemi
c
al
s. Determin
ation of the cost
is
influen
ced by
the real co
nd
ition of the fie
l
d
in the form of histori
c
al d
a
ta.
But in the
reality it is
diffic
u
lt to find th
is d
a
ta
so it mu
st do
the ap
proa
che
s
in
determi
ning t
he co
st. Besi
des, it is also
influenc
ed b
y
the overall operation pla
n
s. The influ
e
n
ce
of these co
st
s to the overall electri
c
p
o
we
r syst
e
m
is so small
with varie
s
b
e
twee
n 2-5%
, in
pra
c
tice i
s
often take
n 4%.
D. Energ
y
Losses
Energy lo
sse
s
are dete
r
mined o
n
ly on the
net
work, b
o
th on
the tran
smi
ssi
on a
nd
distrib
u
tion n
e
tworks. Whil
e
the
po
we
r plant
i
s
con
s
i
dere
d
sup
p
lying to
the
net
work with
ne
t
power. Cost
cal
c
ulation
of the ene
rg
y losses
can
be de
rived
from Figu
re
3, whi
c
h is t
h
e
transmissio
n line with the current flows to the con
s
um
ers a
nd g
r
id.
C
o
nsum
e
r
s
Gr
i
d
P
G
P
D
k
P
R
ρ
G
ρ
D
j
Figure 3. Model of determi
ning the cost
of energy lo
sse
s
From
Figu
re
3, it can
be
shown that th
e ene
rgy p
r
o
ductio
n
, in o
ne ho
ur, i
s
e
qual to th
e cost
s
received o
n
t
he bu
s
con
s
u
m
ers in the
same time.
If e
nergy
co
st is
same
for all
region
s, i.e.
ρ
e =
ρ
G
, the cost o
f
energy losses will b
e
)
P
(P
ρ
P
ρ
R
e
G
e
(4)
Whe
r
e,
e
is energy co
st
P
G
is
ac
tive power into line
P
R
is line losses
P is active po
wer at receipt
the con
s
ume
r
bus
So the co
st of energy lo
sse
s
in one h
our
it is
R
e
R
P
ρ
B
(5)
Then, from th
e con
s
um
er b
u
s the seen p
o
we
rs fl
o
w
in two dire
ction
s
, to the consu
m
ers, D
k
, and
the gri
d
, D
j
, which meet eq
uation
P = D
k
+
D
j
. Furthermore, loss cost
w
ill be bear by
consum
ers
and dete
r
min
ed ba
sed o
n
the followi
ng a
ppro
a
ch.
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 1, April 2016 : 40 – 48
45
e
r
k
Dk
R
P
D
B
(6)
E.
Bu
y
i
ng and
Selling bet
w
een T
w
o
Re
gions
In developin
g
the ele
c
tric
busi
n
e
ss
and
meet
the de
mand
s of cu
stome
r
s, e
a
ch regi
on
can
sell / buy electrical e
nergy from o
t
her re
gion
s. Surely, for re
gion
s as exp
o
rter mu
st h
a
ve
surplu
s ene
rgy and vice
versa. Thi
s
ill
ustratio
n
ca
n
be see
n
in Figure 4, wh
ere regio
n
2 buys
energy from
region
1. Befo
re tra
n
sactio
n
,
cost
i
n
regio
n
1 a
nd regio
n
2 a
r
e
re
spe
c
tively
ρ
1 an
d
ρ
2. Powe
r i
s
transfe
rred to
regi
on1 i
s
P
tr
a
n
. While the
power
pro
d
u
c
ed
by re
gio
n
1 itself is P
G1
.
So the balan
ce of powe
r
in the regio
n
1 is
1
1
1
L
Trans
G
P
D
P
P
(7)
Whe
r
e D
1
and
P
L1
are demand an
d losse
s
in regi
on
1, resp
ectivel
y
.
Figure 4. Tra
n
sa
ction
s
bet
wee
n
two re
g
i
ons
So, cost in th
e regi
on 1
wil
l
chan
ge influ
enced by
the
co
st differen
c
es b
e
twee
n the two
regi
on
s,
transmissio
n
co
st an
d l
o
sses
co
st. So
cost afte
r the
t
r
an
sa
ction
on
re
gion
1
ca
n
be
calculate
d
as
follows
.
Loss
Trans
Trans
G
new
D
P
P
1
2
1
1
1
(8)
Whe
r
e
ρ
tran
and
ρ
loss
are transmi
ssion
cost that conn
ects bot
h the regio
n
s a
nd the co
st of losse
s
cau
s
e
d
by the power into region
1.
3. Simulation and Analy
s
is
3.1 Simulation
Here sim
u
lation cal
c
ulatio
n of electri
c
cost de
scribe
d
above will u
s
e a sim
p
le system in
Figure 5. This system i
s
d
i
vided into two regio
n
s, i.e
.
region 1 on
the top and regio
n
2 on the
belo
w
. The
r
e
are fo
ur tran
smissi
on lin
es
con
n
e
c
ting th
e two
regi
on
s and o
ne IPP
is in the
re
gion
2. Data of
system
can b
e
see
n
on th
e
se
ction A con
s
ist of d
a
ta of
gene
ration,
cha
nnel, volt
age
limits, and others.
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
Det
e
rm
inin
g B
a
sic
Co
st
of
E
l
ect
r
ic
it
y on
the Regio
nali
z
ation Syste
m
(Yusr
a
Sab
r
i)
46
Figure 5. Five bus sy
stem
A. Dat
a
Table 1. Gen
e
rating
Gene
rating
c
b
a
P
mi
n
(pu)
P
ma
x
(pu)
G1
0
1774,125
0,171
0,3
1,0
G2
0
15
0,0015
0,05
0,25
G3
0
850
0.03
0,02
0,1
G4
0
382,788
0,3808
0,05
0,5
IPP 0
1850
0
0,05
0,2
Table 2. Tran
smissio
n
line
Bus-i Bus-j
R
(pu)
X
(pu)
Y
(
pu)
S
ma
x
(pu)
1 2
0,15
0,26
0,030
0,75
1 3
0,17
0,30
0,035
0,75
1 4
0,17
0,30
0,035
0,75
2 3
0,14
0,15
0,020
0,75
2 4
0,15
0,18
0,020
0,55
3 4
0,13
0,12
0.015
0,60
Table 3. Bus
voltage
No. bus
Region
V
mi
n
(pu)
V
ma
x
(pu)
1 R-1
0,9
1,05
2 R-1
0,9
1,05
3 R-2
0,9
1,05
4 R-2
0,9
1,05
Table 4. Loa
d
foreca
st in M
W
hour
load
hour
load
hour
load
1 90
9
110
17
160
2 80
10
115
18
145
3 90
11
130
19
143
4 110
12
120
20
145
5 115
13
100
21
145
6 120
14
110
22
130
7 110
15
120
23
100
8 90
16
130
24
90
Notes :
Ratio of free d
a
y
s to w
o
rk da
y
s
is 0.8
Ratio of Saturd
a
y
to
w
o
rk da
y
s
is 0.85
One year equal
t
o
365
B. Calculation Resul
t
s
Table 5. Opti
mal fuel mix and en
ergy lo
sses
Item /
Region
F
uel cost
[RP/kWh]
Ener
gy
-
l
osses cost
[RP/kWh]
R-1
834.90
22.76
R-2
863.40
17.81
Table 6. Co
st
of energy losse
s
at distrib
u
tion
Item /
Region
Primary
[RP/kWH]
Secondar
y
[RP/kWH]
R-1
35.30
21.51
R-2
24.36
13.66
Table 7. Investment and
O
&
M cost
s
Item
Region
Power plant
Inves/O
&
M
Transmission
Inves/O
&
M
MV
Inves/O
&
M
LV
Inves/O
&
M
R-1
30.87/4.14
10.92/2.09
18.14/2.62
22.07/4.19
R-2
38.21/5.32
14.37/4.04
17.97/4.76
37.93/8.98
Note:
in million Rp/da
y
MV: medium voltage
LV: low
voltage
Evaluation Warning : The document was created with Spire.PDF for Python.
ISSN: 25
02-4
752
IJEECS
Vol.
2, No. 1, April 2016 : 40 – 48
47
Table 8. Basi
c co
st of elect
r
icity of each
regio
n
Region
Power plant
Transmission
MV
LV
R-1
869.91
905.69
961.75
1009.52
R-2
906.93
943.16
990.24
1050.81
Note: in Rp/kWh
Table 9. Ene
r
gy con
s
u
m
pt
ion of each re
gion
Region
Transmission
MV
LV
Sub total
R-1
97.36/20.0
146.05/30.0
243.41/50.0
486.82/100
R-2
71.10/15.0
189.60/40.0
213.30/45.0
474.01/100
TOTAL
168.47/17.5
335.65/34.9
456.72/47.5
960.83/100
Note: in kWh/%
Table 10. Ye
arly re
capitul
a
tion of co
st and reve
nue
COST
: [Million Rp]
REVENUE
: [Million Rp]
Fuel
: 813254.1
From HV
: 155241.9
Investment
: 115987.7
From MV
: 328214.0
O&M
: 24086.0
From LV
: 469871.9
Total
: 953327.8
Total
: 953327.8
3.2 Analy
s
is
The method
developed
has be
en ab
le to perfo
rm calculation
of the basic co
st of
electri
c
ity, for a regi
on o
n
each of its vo
ltage leve
l. T
h
is i
s
sho
w
n
by simulatio
n
for the
system
bus
5. The
simulation results sho
w
ed t
hat the co
st
of fuel ea
ch
regio
n
is ve
ry depend
ent
on
existing po
we
r plants in the
region itself, and if this
reg
i
on is still la
ck of power, the
n
own
e
r of the
regio
n
ca
n purcha
s
e en
ergy from other with surp
l
u
s
energy. Accordin
gly with the formulatio
ns
above, the
en
ergy
co
st
can
vary fr
om o
n
e
region
to
a
nother an
d
al
so th
e
co
st of
losse
s
i
n
curred
each
re
gion, as sho
w
n by
the
Table
s
5 and
6.
T
h
e
s
e
differen
c
e
s
d
epen
d on th
e
cha
r
a
c
teri
sti
cs
of generation
cost
s in ea
ch r
egion, p
u
rcha
sing of
powe
r
from
other re
gio
n
s an
d the line
resi
stan
ce. T
y
pically, a re
gion h
a
s
a cheap
er e
n
e
r
g
y
price is
as
a su
rplu
s
reg
i
on, as
sh
own by
those tabl
es,
i.e. region 1. While regio
n
2 is
as a min
u
s that buys
energy from region 1.
If it is involve
d
the co
st of investment a
n
d
O
& M in Table 7, it is o
b
tained the b
a
si
c co
st
of ele
c
tricity i
n
Tabl
e 8. F
r
om this table
loo
k
s
the
co
st of fuel
is v
e
ry do
minant
. Gene
rally, the
co
st of fuel compon
ent is
arou
nd 80%
of total co
st. This indi
cate
s that fuels i
s
ab
solute to
be
optimize
d
for redu
cin
g
the basi
c
co
st
of electri
c
it
y. By optimizi
ng fu
els th
en the
p
r
inci
ple
s
of th
e
least cost in operating the po
wer system will be able to be met.
With the
p
r
o
posed
metho
d
, the
ba
sic
co
st of
ele
c
tricity ca
n b
e
determi
ned
i
n
ea
ch
voltage level
of the
con
s
u
m
er,
wheth
e
r it is
co
nne
ct
ed to th
e net
works of
high
voltage, MV
or
LV, which is sho
w
n by T
able 8. Thi
s
table ca
n be
a con
c
e
r
n for sta
k
e
hold
e
rs
be
cau
s
e
by
kno
w
in
g the
electri
c
co
st in the re
gion,
then they
ca
n take
ce
rtai
n step
s to re
duce the ele
c
tri
c
co
st, such as
addin
g
power plants for th
e
minus regio
n
s
or repai
ring
netwo
rks.
In addition, the con
c
e
p
t of the pro
p
o
s
ed
me
thod en
su
res a bal
an
ce
betwe
en prod
uction
co
sts with
rev
enue
s, whi
c
h
is sh
own by Table 10.
4. Conclusio
n
The co
st cal
c
ulatio
n con
c
ept
p
r
op
osed
ha
s
cov
e
red
all
com
pon
e
n
t
co
st
s t
hat
is
only
affect produ
ct
ion, allo
wabl
e
co
st, from p
o
we
r pl
a
n
t
s
t
o
co
ns
ume
r
s,
whi
c
h in
clu
d
e
s t
h
e
co
st
s
of
fuel, asset, O&M, loss e
nergy in the transmi
ssio
n and buying /
selling bet
ween re
gion
s. The
cal
c
ulatio
n in
the si
mulati
on
states th
a
t
the fuel
c
o
s
t
in
th
e su
pp
ly o
f
e
l
ec
tr
ic
po
w
e
r
is
ve
r
y
domina
n
t, i.e. more than 8
0
%. So, the
cal
c
ulatio
n
using optimization to get
the minimal co
st
is
the very pre
c
i
s
e.
The co
st sho
u
ld
b
e
cal
c
ul
ated
fo
r
o
ne year
a
hea
d.
This is cau
s
e
d
by the
p
r
ice
ch
ang
es
for the f
u
ture
is ve
ry un
ce
rtain, e
s
pe
cia
lly for
fuel
pri
c
e
s
. Be
side
s, load
fore
ca
st can
al
so
b
e
inaccu
rate fo
r a few ye
ars
ahea
d. The
p
r
opo
se
d cost
con
c
e
p
t ha
s
demon
strated
its su
cce
s
s in
the sim
u
latio
n
. This i
s
p
r
o
v
en by the result
s of
the
cal
c
ulatio
n o
n
the Ta
ble
s
5-1
0
. He
re
wa
s
Evaluation Warning : The document was created with Spire.PDF for Python.
IJEECS
ISSN:
2502-4
752
Det
e
rm
inin
g B
a
sic
Co
st
of
E
l
ect
r
ic
it
y on
the Regio
nali
z
ation Syste
m
(Yusr
a
Sab
r
i)
48
determi
ned t
he co
st that doe
s not onl
y in each
re
gion but al
so
individual co
nsum
er in e
a
c
h
regio
n
, i.e. th
e con
s
um
ers
of the
high, medium an
d lo
w voltage
s.
Referen
ces
[1]
Ilic MD, Galian
a
F
and F
i
nk L.
Power System Restructuring
. Klu
w
er. USA. 1
998.
[2]
Josko
w
P an
d
T
i
role J. Relia
bilit
y a
nd Com
petitive El
ectri
c
it
y
Market.
CSEM WP 129
. Universit
y
of
Cliforn
i
a En
erg
y
Institute. 20
0
4
.
[3]
Menteri Perta
m
ban
gan
dan
Energ
i
Indo
nes
ia. Po
w
e
r
S
e
ctor Restructuri
n
g Polic
y. T
he Governme
nt’
s
Polic
y
Pr
ogram for Po
w
e
r
Sector Rec
o
ver
y
.
R
e
struct
urin
g, Reg
u
l
a
tory Refor
m
a
nd Priv
ate
Particip
ation
. 1
989.
[4]
Hermag
a
sa
nto
s
Z
,
Yusra S, Ali M. Implement
ati
on of
Electricit
y Co
m
petitio
n F
r
ame
w
ork
w
i
t
h
Econom
ic Disp
a
tch Direct Method.
T
E
LKOMNIKA Indones
i
a
Journ
a
l of El
ectrical En
ggi
n
e
rin
g
. 2012
;
10(4): 62
5-6
3
2
.
[5]
Hermag
a
sa
nto
s
Z
,
Yusra S, Er
w
i
n
D. Deter
m
inin
g LBMP
throu
gh Optim
a
l Po
w
e
r F
l
o
w
i
n
the E
l
ectri
c
Po
w
e
r Bus
i
nes
s.
T
E
LKOMNIKA Indon
esia J
ourn
a
l of Electr
ical En
ggi
neri
n
g
. 2014; 1
2
(7): 508
6-50
95
[6]
Che
ng Hs
iao,
et.al., Model
in
g Ontario r
egi
ona
l el
ectricit
y s
y
stem d
e
ma
nd us
ing
a mi
xe
d fi
xe
d an
d
rand
om co
effici
ents ap
pro
a
ch,
Reg
i
on
al Sc
ie
nce a
nd
Urba
n
Econo
mics
. 1
989; V
o
l. 1
9
, Issue 4,
pp.
565
–5
87
[7]
Yan
X
and
Quinta
na V.H. I
m
provin
g An
Interi
or P
o
int B
a
sed OPF
D
y
namic A
d
justm
ents of St
e
p
Sizes an
d T
o
lerance.
IEEE Transaction on Power System
. 199
9; 14(2).
[8]
Hermag
a
sa
nto
s
Z
and Er
w
i
n D. Cost Allocati
on of T
r
ansmissi
on L
o
sses in El
ec
tric Marke
t
Mechanism.
T
E
LKOMNIKA Indo
nesi
a
Jour
nal of Electric
al
Enggi
neri
n
g
. 2
012; 10(
2): 211
-218.
[9]
El-Kei
b AA
a
nd
X M
a
. Ca
lculati
n
g
Short
-
Run M
a
rgi
n
a
l
Costs of Act
i
ve a
n
d
Re
ac
tive Po
w
e
r
Producti
on.
IEEE Transaction on Power System
. 19
97; 12
(2).
[10]
Vojdani AF, et.al. T
r
ansmission Access Is
ues
, IEEE
Transac
tion on Po
w
e
r
S
y
stem, 1996; 11(1).
[11]
W
ood AJ a
nd
W
o
llen
ber
g BF
.
Pow
e
r Gener
ation Op
erati
o
n an
d Co
ntrol
.
John W
i
l
e
y &
Sons, Ne
w
York, Second
Editio
n. 199
6.
[12]
Sch
w
e
ppe F
C
,
et.al.
Spot Pricing of Electric
ity
. Klu
w
e
r
. USA
.
1998.
[13]
Hermag
a
sa
nto
s
Z
.
Studi Biaya Pelay
a
n
an J
a
rin
gan T
r
a
n
s
m
is
i: Dal
a
m K
o
ntek Ko
mp
etitif
. Proccedin
g
SST
E-1. 2000
[14]
Hermag
a
sa
nto
s
Z
.
Simulas
i
kinerj
a
jar
i
ng
an
transmisi d
a
la
m ke
mampu
a
n
n
ya
meny
al
urk
an day
a d
a
r
i
suatu pe
mba
n
g
kit.
Prosidi
ng
ISSN: 1979-9
1
1
X
,
IST
AKPRIND. Yog
y
ak
arta. 2008.
[15]
Hermag
a
sa
nto
s
Z
.
A Ne
w
Me
thod for
Deter
m
inin
g Po
w
e
r
Decom
positi
o
n
in El
ectric S
y
s
t
em. T
he12t
h
Internatio
na
l C
onfere
n
ce o
n
QiR (Quality in
Rese
arch)
. Bal
i
. 4-7. 2011
[16]
W
a
kefiel
d RA. A
T
r
ansmissi
on Servic
es Costing F
r
ame
w
ork.
IEEE Tran
saction on Power System
.
199
7; 12(4).
[17]
Hirsh
RF.
Pow
e
r Loss: the origins of der
eg
ulati
on an
d re
structuring i
n
the Am
erica e
l
ectric utility
system
. T
he MIT
Press Cambridg
e. Massach
usetts. Lond
on
. Englan
d. 200
1.
[18]
Hammons, et.
a
l. Europ
e
a
n
Polic
y o
n
Ele
c
tric
it
y
Infrastructure, Interco
nnecti
ons, an
d Electrici
t
y
Exch
an
ges.
IEEE Transaction on Power System
. PES Summ
er Meeting
. 1997.
[19]
Gedra T
W
. On T
r
ansmission
Con
gestio
n
an
d Prici
ng.
IEE
E
T
r
ansacti
on
on P
o
w
e
r System
. 19
99
;
14(1).
Evaluation Warning : The document was created with Spire.PDF for Python.