Professional Documents
Culture Documents
Abstract: The daVa of four networks that can be used in carrying out comparative studies with
methods for transmission network expansion planning are gwen. These networks are of various
types and different levels of complexity. The main mathematical formulations used in transmission
expansion studies-transportation models, hybrid models, DC power flow models, and disjunctive
models are also SummdnSed and compared. The main algorithm families are reviewed-both
analytical, cOmbinatOnd1 and heuristic approaches. Optimal solutions are not yet known for some
of the four networks when more accurate models (e.g. the DC model) are used to represent the
power flow equations-the state of the art with regard to this is also summarised. This should sewe
as a challenge to authors searching for new, more efficient methods.
w1
g,= 50
3.4 87-bus system
This system is a reduced version of the Brazilian north-
northeastern network: the reduced model has 87 buses, 183
right-of-ways for the addition of new circuits, and a total
demand of20316MW for plan PI and 2974XMW for plan
P2. All relevant data about this case can he found in Tables
3 and 4. There is no limit for the number of circuit additions
in each right-of-way.
This system shows a high degree of complexity due to the
large number of islanded buses in the initial network. In
order to nm the cases considering generation redispatch, it
is necessary to consider generation limits: it is suggested to
consider the following: &[I . 3 g k ] , that is, the maximum
generation level is equal to the largest integer contained in
the product of the current generation by 1.3 (Le. a 30%
margin).
4 Illustrative example
From-To n," Reactance f,.MW Cost, io3 From-To n,: Reactance 3,. MW Cost. io3
PU US$ P.U US$
1-2 1 0.40 100 40 26 0 0.30 100 30
1-3 0 0.38 100 38 3-4 0 0.59 82 59
1-4 1 0.60 80 60 3s 1 0.20 100 20
1 5 1 0.20 100 20 34 0 0.48 100 48
1-6 0 0.68 70 68 4-5 0 0.63 75 63
2-3 1 0.20 100 20 46 0 0.30 100 30
2-4 1 0.40 100 40 54 0 0.61 78 61
2-5 0 0.31 100 31
Table3 Generation and load data for 87-bus system
30
Table4: lconriiiiredd)
From-To 8, Reactance
p.u.
J,, MW cost. io3
US;dollar:
From-To "; Reanance
p.u.
x,,MW cost.
US$
io3
31
Table4 (confinued)
From-To n?, Reactance x,,MW Cost, lo3 From-To n: Reactance x,, MW cost. io3
D.U. US;dollar: D.U. US$
0 5 93 5 1.25
0 5 96 5 5.45
0 5 rl 5 0.30
0 5 r2 5 2.40
0 5 r4 5 1.60
0 5 rs 5 2.40
n12,n14, nis, n23, n24: n26, n35?n e integer
81,82, 8 3 , 84, 85, &unbounded
f i 2 , f 1 4 r f ~fu,f24.f26,f35,f46unbounded
~r
lfL 5 n12
lf;41
5 0.8 n14
I&l 5 nls
lfi31 5 n23
Ifi4l 5 n24
If&5 n26
If& 5 n3s
If&l5n 4
33
For the disjunctive model, let us assume that the
maximum number of circuit additions is of two circuits,
except for right-of-way 2 4 , where the limit is four circuits.
I f l s l 5 1.0
Under these assumptions, the disjunctive model can be lfl3l i1.0
formulated as follows.
Minimise 5 1.0
u =4o(YI2 + ~ $ 2 )+ 6 0 ( ~ ; 4+ d 4 ) + 20(yfs + d s ) If!sl 5 1.0
+ 2o(Yj3 + Y;3) + 40(Ji4 f $4)
+ 30(,& + .& + $6 + p
Ifl215 Y;, P = 1:2
lf;;l~Yz?: P=l,2
If4Pnl5&; P=I;2
0 i~3 5 1.25
0 5 9 6 5 5.45
0 5 ri 5 0.30
0 5 r? 5 2.40
0 5 r4 5 1.60
0 5 rs i2.40
J$ E {O, I}; Lf>h f a n d B, unbounded
The transportation model has the five alternative optimal
solutions shown in Table 5. The first three solutions are also
optimal solutions for the hybrid model. Only, the first
solution in the Tabk is the optimal solution for the DC
model..
5 Solution techniques
NOR 284142 N ~
N 424800 C
WR ~
S ~
N ~
C
NRNN ~
vc ~
vc ~
vc
WRNN ~ vc - vc ~ vc
Table9: Results for 87-bus system
model, transportation model, hybrid model and disjunctive 1 MONTICELLI, A,. SANTOS. JR. A,. PEREIRA, M.V.F., CUN-
model. Finally, the best known solutions for each system HA, S.H., PARKER, B.J.. and PRACA. J.C.G.: 'Interactive
transmission network planning using a least-efbrt cntenan', IEEE
and each alternative model are presented-the importance Trom. Poiier Appur. Syxf.. 1982, 101. (IO), pp. 391W925
of these data is that for certain combinations of system/ 8 PEREIRA. M.V.F.. and PINTO. L.M.V.G.: 'Aoolicatian of
~~
..
Sensitivity 'analysis bf load supplying capability to interactive
model the optimal solutions are not yet, known and so the transmission expansion planning', lEEE Tmm Ponrr Appor. Syst.,
data should serve as a benchmark for further developments 1985, 104, (2). pp. 381-389
in the area. 9 VILLASANA. R.. CARVER. L.L.. and SALON, S.J.: 'Transmission
network o l a n n i n ~mine linear Droeramine'. IEEE Tram. Pon.er
Appar. S<.yt.. 1985. lWy(2). pp. 3491356
8 Acknowledgement 10 LEVI, V.A., and CALOVIC, M.S.: 'A new decomposition based
method for optimal expansion planning of large transmission
networks'. IEE Triuri P m v r Sy.st., 1991, 6, (3). pp. 937-943
This project was partly supported by the FundaqZo de I I PEREIRA. M.V.F.. PINTO. L.M.V.G.. OLIVEIRA. G.C.. and
Amparo i Pesquisa do Estado de SZo Paulo (FAPESP), CUNHA, S.H.F.. Stanford University, 'Composite gen&ion-trans-
and the Conselho Nacional de Desenvolvimento Cientifico
e Tecnologico (CNPq).
phmning', IEEE Trans. Power Sjs..'1994. 9,(4). pp. l886-18$4
13 ROMERO, R.. and MONTICELLI, A,: 'A hierarchical dccomposi-
References ..
tion amroach for transmission network exnansion dannine'. IEEE
Trrrris Power S j x . 1994, 9, (I). pp. 37S380'
SEIFU, A., SALON, S.. and LIST, G.: 'Optimaization of transmis- 14 GALLEGO.. R.A., MONTICELLI. A,. and KOMERO. R.: 'Tabu
sion line planning including security constrahs', IEEE P a m Power search algorithm for network synthesis', IEEE Tram Power. Sw..
S j s f . , 1989, 4. (4). pp. 1507-1513 2w0, 15, (2). pp. 49a-495
SHARIFNIA. A,. and AASHTIANI. H.Z.: 'Transmission network 15 OLIVEIRA. G.C.. COSTA, A.P.C.. and BINATO. S.: 'Large scale
planning: a method of synthesis of minimum cost secure networks', transmission network planning using optimization and 6eunstic
IEEE Trilnr Poapr A p p m Syrr. 1985, IW, (8). pp. 2026-2034 techniques', IEEE Twm. Power Sy.%. 1995. IO, (4). pp. 1828-1834
GRANVILLE. S.. and PEREIRA. M.V.F.: 'Analysis of the lineanred 16 LEVI. V.A.: 'A ncw -4-integer methodology for optimal
~ o w eflow r model in Benders decomoosition ...EPRI-nnort RP 2473-
k. Stanford University, 1985
CARVER, L.L.: 'Transmission network estimation using linear
programming'. IEEE Trom. Power A p p m S ~ I .1970, , 89, (7). pp.
1688-1697
HAFFNER. S.. MONTICELLI. A,. GARCIA, A,, MANTOVANI,
J.. and KOMEKO. R.: 'Branch and bound algorithm for transmission 'Transmission system expansion planning by extended genetic
system expansion planning using a transportition model'. IEE Pror algo6thm'. /E Proc.. Gener. Trumtn. Dim& 1998. 145. (3). pp.
Gener. TronFm Dirtrib.. 2000. 147. (3); pp. 149-156 329-335
DECHAMPS. C..~~ ~. and JAMOULLE. A,: 'Interactive comouter
~ ~~ 19 WEN, F.S., and CHANG, C.S.: 'Transmission network optimal
pro@-dm for planning the expansion of meshed transmission net- planning using the tabu search method'. E l e m Poarr Sy.yt. Rem.
works'. Inf. J. Elecrn Power Energy Sysl.. 1980, 2. (2). pp. 103108 1997.42, ( 2 ) . pp. 153-163
36