XI SIMPÓSIO DE ESPECIALISTAS EM PLANEJAMENTO DA
OPERAÇÃO E EXPANSÃO ELÉTRICA
XI SEPOPE
17 a 20 de Março 2009
March – 17th to 20th – 2009
XI SYMPOSIUM OF SPECIALISTS IN ELECTRIC OPERATIONAL
AND EXPANSION PLANNING
BELÉM (PA) - BRASIL
Non-Conventional AC Solutions Adequate for Very Long Distance Transmission
C. PORTELA
COPPE / UFRJ
Brazil
SUMMARY
There are several cases around the world in which the use of very important energy resources,
interesting in a strategic, economic and ambient impact point of view, imposes an adequate solution to
transmit large electric power, at very long distance. As an example, the natural medium term option for
the Brazilian electric sector is to base its growth in the hydroelectric resources of Amazon basin, with
moderate complementary generation based in other sources. This choice imposes an adequate solution
to transmit most of such energy at distances of the order of 2500 km.
In order to obtain an adequate transmission system, a specific analysis must be done, with careful
optimization, global, and considering a long-term point of view. It is not adequate to extrapolate
solutions developed for medium distances, of the order of a few hundred kilometers.
There are two types of solutions potentially interesting:
A- Transmission in alternating current (AC) based in non-conventional transmission lines (LNC), with
also a non-conventional conception of the transmission trunk.
B- Transmission in direct current (DC).
The solutions A and B are, both, essentially “point to point”, without prejudice of eventual
“adaptations”, of “subsidiary” type, and have quite different optimization constraints.
The correct comparison imposes a separate optimization of both solution types (A and B), and the
objective and quantitative comparison of results. In some conditions, a hybrid solution may be justified.
We have done studies of non-conventional solutions, deliberately abandoning the criteria of choosing
solutions similar to lines and compensation equipment of present systems, but considering very robust
criteria of physical validity, ambient impact and joint global optimization of transmission trunk. For
instance, very interesting solutions have been obtained, based in alternating current transmission trunks,
non-conventional, with unitary transmission capability from 2 GW to 12 GW, and electric length a little
higher than half wave length (at power frequency), without the need of reactive compensation, or with
very small reactive compensation, and without the need of intermediate substations. Such trunks can be
energized and de-energized switching a single circuit breaker, with moderate switching overvoltages,
have moderate losses, very good behaviour for load variations and for electromechanical stability of
interconnected networks, originate moderate electromagnetic field near the line, have low ambient
impact and have cost typically much lower than some recent transmission systems based in
conventional solutions (by example, the costs per unit of transmission capability of described trunks are
of the order of one fifth to one third of the costs of recent transmission systems).
In this article it is presented a discussion of mentioned non-conventional solutions, with emphasis in
conceptual aspects and in optimization and validation procedures. Also, examples will be presented, in
the power and parameters range potentially interesting for transmission electric power from Amazon
basin to main consumption regions of Brazil.
KEYWORDS
Very long distance transmission, Non-conventional AC solutions, Lines with a little more than half
wave length.
1
Type here the contact info of the main author as a footnote.
* Author contact. E-mail: portelac@ism.com.br, Phone: 55-21-24934201, Fax: 55-21-24934201
Adress: Rua Eng. Cesar Grillo, 249, Rio de Janeiro, RJ, CEP 22640-150, Brazil
_
1. Introduction
This article presents non-conventional transmission systems potentially convenient to transmit large
electric power, at very long distance, e.g. of the order of two to three thousand kilometers. The
development of such systems was originated when searching adequate solutions for the development of
Brazilian power system. The natural medium term option is to base its growth in the hydroelectric
resources of Amazon basin, with moderate complementary generation based in other sources. This
choice imposes an adequate solution to transmit most of such energy at distances of the order of
2500 km. The direct use of traditional solutions for typical transmission systems, developed for quite
different constraints, was naturally put in doubt and it was decided to search eventually different
solutions, founded in basic physical constraints, operational requirements and robust optimization and
validation procedures. The main purpose of this article is to present one of the results of the research
work, namely: transmission in alternating current (AC), based in non-conventional transmission lines
(LNC), with also a non-conventional conception of the transmission trunk, with electric length a little
higher than half wave length (at power frequency). This type of solution was studied for applications in
the medium term development of Brazilian network, to which some presented examples apply, although
their main features are valid for a large spectrum of similar conditions in other countries.
In Fig. 1 are represented the basic transmission distances, between new power stations in Amazon basin
and main Brazilian load centers, and the geographical distribution of hydroelectric power stations in 2003.
2. A Non-Conventional Very Long Distance Transmission System Alternative
We present now some of the main aspects of transmission systems based in “Non-Conventional Lines”
(LNC), three-phase, double three-phase or six-phase, defined with basis in the following criteria:
- Do not consider restrictions that result merely of usual solutions.
- To impose only restrictions related to basic physical constraints and to performance, security and
ambient impact.
- Physical parameters optimization according the specific operational functions and objectives of the
line, including costs, losses, operational reliability, transmission range and operation constraints, and
ambient impact, ponderated along the useful life of the transmission system and pertinent scenario
range.
It was identified a set of basic physical properties that allows to choose a limited number of parameters
with high correlation with several other physical, performance and cost parameters.
It is viable a robust optimization analysis, based in a moderate number of parameters and the specific
constraints of the considered transmission system.
In previous work, have been defined optimization and validation methodologies, according this type of
analysis [1-22].
The transmission at very long distances (of the order of 2000 km or more) has constraints much
different of “usual” transmission distances (till a few hundred kilometers). So, the simple extrapolation
of “usual” procedures, for very long distances, leads to inadequate or non-optimized solutions.
The defined methodologies were applied to a significant range of conditions and to a large number of
examples. This analysis has allowed an approximate definition of practically feasible transmission
powers, with prudent criteria, for transmission at very long distances.
Naturally, for each specific condition, it is necessary an optimization and validation analysis.
For transmission at very long distance (of the order of 2000 km or more), there are interesting solutions
based, approximately, in:
- Selection of transmission trunks that behave with an “electric length” a little longer than half wave
length (at power frequency).
- Point to point connection, without section switching.
- Null or very reduced reactive compensation.
- Non-conventional lines (LNC) with large transmission capability (in comparison with conventional
lines).
- Joint optimization of lines, network equipment and switching and protection criteria, detecting and
avoiding potentially critical conditions.
2
Brazil
Fig. 1 - Basic transmission distances, between new power stations in Amazon basin and main Brazilian
load centers, and geographical distribution of hydroelectric power stations in operation in Sept. 2003.
These solutions allow:
- Good performance of the transmission trunk in what concerns electromechanical stability.
- Good performance of the transmission trunk in what concerns switching overvoltages.
- Cost much lower than “traditional” solutions.
- Operational reliability much higher than “traditional” solutions.
- Ambient impact much lower than “traditional” solutions.
For these solutions, the characteristic power is, approximately, the limit of transmitted power
(differently of what happens with “short” distance lines), and, within some conditions range, the
maximization of characteristic power corresponds also, exactly or approximately, to:
- Maximization of transmitted power limit.
- Minimization of losses.
- Minimization of corona effect.
- Maximization of viable operation voltage.
- Minimization of reactive power in several operation conditions.
- Minimization of sustained overvoltages in several operation conditions.
- Minimization of switching overvoltages for several switching operations.
3
14
Pc
12
[GW]
10
8
6
4
2
600
800
1200
Uc [kV]
1000
Fig. 2 - Characteristic power, Pc , that can be obtained with prudent criteria with LNC three-phase
lines, in function of voltage, Uc , between phases, for voltage till 1250 kV . Curves in red and violet
represent the typical range of Pc that can be obtained, depending on specific conditions and options.
Fig. 3 – Schematic representation of a 2550 km, three-phase line, without reactive compensation,
switched from one extremity (all the line).
!
We present, now, some results for non-conventional lines (LNC), three-phase, of voltage till 1250 kV,
optimized for transmission at very long distance (of the order of 2000 km or more) [1-22].
In Fig. 2 it is represented the characteristic power, Pc , that can be obtained with prudent criteria, with
LNC three-phase lines, in function of voltage, Uc , for voltage till 1250 kV . In Fig. 3 it is represented,
schematically, a three-phase line, with 2550 km, at a 60 Hz network, without reactive compensation, which
can be switched from one extremity (all the line). As examples of this type of line, in Fig. 4, it is shown
some bundle geometries and structures’ variants of two lines, of 1000 kV and 800 kV, with characteristic
power, Pc , 8.6 GW and 4.8 GW. Sustained overvoltage in one terminal (opened) is 1.017 pu (referred to
the other terminal voltage). The three types of structures shown for the 800 kV line apply to different
conditions, namely in what concerns wind. It is also shown, in the same Fig. 4, the power, P (in GW),
and the reactive power, Q (in Gvar), at line terminal 1, with voltage U1 = 1000 kV or 800 kV
(respectively for the two example lines), in function of phase difference, α , and ratio, R , between
modules, of voltages at terminals 2 and 1 (as indicated in Fig. 3). A phase range (of α − π) of the order of
± 0.2 radian allows to vary the transmitted power in a range ± Pc .
It is out of the scope of this article a discussion of all aspects of the electrical and mechanical behaviour
of this type of solution, that have been studied, in order to validate it as a convenient and a potentially
optimum alternative for several real conditions !
of long distance transmission, including the medium
term new main transmission trunks adequate for Brazil electrical network expansion. So, we have
chosen a few results, illustrating some important aspects of the example lines’ behaviour.
A first example refers to switching overvoltages. In order to separate and characterize the fundamental
line behaviour, avoiding the influence of network transient behaviour and of specific soil electric
parameters, chosen examples consider the line energization from an “infinite” busbar, with the opposite
terminal open and with simultaneous switching on of the three phases. Results are presented in Fig. 5.
This example shows that it is obtained a remarkable low value of switching overvoltage, energizing the
2550 kV line with a very simple switching procedure, in a single step.
A second example refers to the possibility of, for single-phase faults, to open only the fault phase, in both
4
y [m]
Uc = 1000 kV
U1 = 1000 kV
Pc = 8.6 GW
Pc = 8.6 GW
→
R
Guyed structure, 1000 kV line
y [m]
x [m]
α − π [rad]
Uc = 800 kV
U1 = 1000 kV
Pc = 4.8 GW
Pc = 8.6 GW
→
R
y [m]
α − π [rad]
x [m]
Guyed structure, 800 kV line
Uc = 800 kV
U1 = 800 kV
Pc = 4.8 GW
Pc = 4.8 GW
→
R
Chainette type structure, 800 kV line
y [m]
α − π [rad]
x [m]
Uc = 800 kV
U1 = 800 kV
Pc = 4.8 GW
Pc = 4.8 GW
→
R
Cross-rope type structure, 800 kV line
x [m]
α − π [rad]
Fig. 4 – Examples of bundles and phase geometry, and line structures, for two 2550 km LNC lines, with
nominal voltages, Uc , 1000 kV and 800 kV , and characteristic power, Pc , respectively 8.6 GW and
4.8 GW . It is also shown, in this Fig. 4, the power, P (in GW), and the reactive power, Q (in Gvar), at
line terminal 1, with voltage U1 = 1000 kV or 800 kV (respectively for the two example lines), in
function of phase difference, α , and ratio, R , between modules, of voltages at terminals 2 and 1.
line extremities, and to use procedures that assure a high probability of extinction of secondary arc, in
faulted phase, in a reasonably short time, allowing its fast reclosure, with a small network disturbance. As
the large majority of line faults are single-phase faults, originated by lightning, and such faults, with
successful fast reclosure, have little impact in the network, this possibility allows an important increase in
5
u1a
u2b
u2a
t [ms]
!
t [ms]
t [ms]
Fig. 5 – Examples of switching overvoltages of the example line of 2550 km , 1000 kV, in function of time.
u1a – Applied voltage, in one phase, switched in terminal 1 at the “infinite” busbar, at t = 0 , in pu of
phase voltage amplitude at the infinite busbar, Û . For the nominal voltage 1 MV at the busbar, it is
Û = 2/3 . 1 MV = 816.5 kV.
u2a – Transient voltage, in the same phase, in the other line extremity, with no pre-insertion resistor at
the circuit breaker, and without surge arresters at terminal 2 . Peak value of | u2a | is 1.88 pu . Power
frequency voltage amplitude at terminal 2 , with no load, is 1.017 by the voltage amplitude at terminal 1 .
u2b – Transient voltage, in the same phase, in the other line extremity, with a pre-insertion resistor of
77.5 Ω in the switching circuit breaker, connected during 1 cycle (16.7 ms). Peak value of | u2b | is 1.13
pu (923 kV for nominal voltage at terminal 1 ). For the nominal voltage 1 MV at the busbar, the energy
dissipated in the pre-insertion resistor is 11.5 MJ .
I
[kA]
I
I
[A]
[A]
α - π [rad]
U
[kV]
α - π [rad]
U
U
[kV]
[kV]
α - π [rad]
A
α - π [rad]
α - π [rad]
B
α - π [rad]
C
Fig. 6 – Example of a 2550 kV, 1000 kV line behaviour, for single-phase fault in an unfavourable point.
α – Phase angle between line terminal busbars.
I , U – Sustained fault current and phase-ground voltage at fault point, neglecting arc impedance.
A – Line terminals connected to busbars at both extremities, in the three phases.
B – Line terminals of fault phase open at both extremities, and closed in the other two phases.
C – Same as B and with connected auxiliary circuits, to reduce secondary arc current.
line and network reliability. Fig. 6 shows the result of a procedure that allows an important reduction of
secondary arc current and recovery voltage, that assures a high probability of success of single phase
opening and reclosure, for single phase faults, in the presented type of very long transmission trunks.
Some other procedures, using secondary arc physical behaviour, are also promising, and are being
investigated. We notice that conventional type long transmission trunks, with massive compensation, do
not allow successful single phase open and reclosure, due to physical reasons, which have been identified.
6
3. Some Basic Economical and Cost Aspects
We present, now, some brief comments about economical and cost aspects of non-conventional
transmission systems of the type presented in this article.
The indication of absolute specific costs, in a general context, would have a large error margin, due to
the many aspects that affect the “indicated cost” and are, in a large proportion, independent from the
technical effective cost, according an objective set of clear conditions. So, we present only some
approximate relative costs, taking as a basis some recent transmission trunks of about 1000 km, with
1 GW maximum transmitted power. Such systems are based in “conventional” transmission systems for
lengths of a few hundred kilometers, and use massive reactive compensation, in the transmission trunks
and in the interconnected networks. It is out of the scope of this article to discuss such projects, that are
referred only because they were object of recent decisions and options and reflect what was the
“chosen” reality in a specific context.
For this comparison, it is considered as “unitary cost” the “technical cost” of a line (Case A), with
2550 km, similar to a line of a relatively recent project, 500 kV, 1 GW, excluding intermediate
substations and reactive compensation, along the line and in the interconnected networks, required to
allow line operation. Such cost was decomposed in representative parcels, from which, with uniform
criteria, were estimated the costs of example trunks presented in item 2. , with 2550 km, one of
1000 kV, 8.6 GW (Case B), and other of 800 kV, 4.8 GW (Case C). The result is indicated in Table I.
Table I – Cost comparison of:
- Transmission system similar to a recent project based in conventional systems (Case A).
- Example of non-conventional system of 1000 kV, P = 8.6 GW (Case B).
- Example of non-conventional system of 800 kV, P = 4.8 GW (Case C).
Total costs, C
Cost parcel
Relative costs, c = C / P
CA(Case A) CB (Case B) CC (Case C) cA (Case A)
P = 1 GW P = 8.6 GW P = 4.8 GW P = 1 GW
cB (Case B) cC (Case C)
P = 8.6 GW P = 4.8 GW
cB / c A
cC / cA
C1
1.00
2.56
1.80
1.00
0.298
0.375
0.298
0.375
C2
0.70
0.00
0.00
0.70
0.000
0.000
0.000
0.000
Ct = C1 + C2
1.70
2.56
1.80
1.70
0.298
0.375
0.175
0.221
C1 Line cost, excluding intermediate substations and reactive compensation, along the line and in
C2
interconnected networks, to allow line operation.
Cost of intermediate substations and reactive compensation, along the line and in interconnected
networks, to allow line operation.
In the assumptions of Table I (for Case B and Case C, presented in item 2):
- The total costs of presented examples, with transmission capacity 8.6 GW and 4.8 GW, at 2550 km,
are only about 51 % and 6 % higher than the total transmission cost of a transmission system with
1 GW transmission capacity, similar to a recent transmission system based in conventional solutions
(for the same line length, in compared costs).
- The relative costs (per unity of transmission power capacity, and for the same line length), of the
presented examples:
- Are of the order of 30 % and 38 % of the relative cost of comparison line, based in conventional
solutions, excluding intermediate substations and reactive compensation (along the line and in
interconnected networks, to allow the line operation).
- Are of the order of 18 % and 22 % of the relative cost of comparison line, based in conventional
solutions, including intermediate substations and reactive compensation (along the line and in
interconnected networks, to allow the line operation).
Anyhow, the presented non-conventional solution allows transmission costs at very long distances (of
the order of 2000 to 3000 km) much lower than costs obtained with AC transmission systems based in
solutions developed for traditional transmission systems, with transmission distances of a few hundred
kilometers.
7
4. Application of the Presented Alternative to Madeira Transmission System
A few years ago [1-22], we examined the use of the presented alternative for the transmission of
hydroelectric energy from Amazonian region to main consumption centers and it appeared quite
interesting, as a natural AC alternative, that should be adequately studied and compared with the DC
alternative, in order to optimise such transmission. Some presentations were done, with participation of
persons of the electric sector. Also, some presentations were done to agencies related to electric sector
development, but, although no technical argument was presented against the alternative, it was always
said that there was no time to consider it, because options at to be done within, e.g., “two weeks”. Much
more than “two years” elapsed, with the successive “two weeks” argument, and the alternative was, in
fact, ignored, by “responsible agencies”, for the Madeira Transmission System. In the mean time, it was
studied and discussed, by several people, and the obtained and presented results, informed to
responsible agencies, and presented and discussed in several forums, showed that the alternative was
competitive and evidenced important advantages, in comparison with alternatives studied and chosen by
such agencies.
It is out of the scope of this article a detailed discussion of results, for the Madeira Transmission
System, of an alternative of the type presented in this article. We present, only, some of its important
aspects, and its comparison with the alternatives chosen by responsible agencies.
In Fig. 7 it is represented the configuration of the proposed alternative, and are indicated some of its
parameters and characteristics. This configuration is much simpler than the alternatives proposed, e.g.,
in document R1 that was available at agencies’ internet sites.
For instance, the final configuration of the AC alternative, according page 31 of R1 EPE’s document,
has three AC 765 kV lines, nominal power 3 x 2.1 GW = 6.3 GW, six intermediate substations, total
length 2375 km, and:
- Inductive compensation: fixed 21.75 Gvar; switched 3.05 Gvar.
- Capacitive compensation: fixed 69.33 % of longitudinal lines inductance, switched 2.2 Gvar.
- Controlled compensation: –0.84 / +1.75 Gvar, more reactive power exchanged with terminal
networks and power generators.
It was concluded that, for Madeira Transmission System, the alternative presented in this article is
competitive with all the alternatives presented by responsible agencies. To situate the subject, in
Table II it is presented a summary of the cost comparison of the proposed alternative with the
alternatives proposed by EPE, in R1, for Madeira Transmission System.
It was also done a comparison of the presented alternative (of a little more than half wave length) with
the three R1 alternatives, for the Madeira Transmission System, in what concerns technological risk,
Fig. 7 − Madeira Transmission System. Configuration of the alternative of trunks with a little more than
half wave length, 800 kV AC lines, with nominal power 2 x 4.85 GW = 9.7 GW, without intermediate
substations, total length 2500 km, with 8 Bittern cables per phase. The transmitted power may vary
continuously between – 9.7 GW and +9.7 GW . There is no inductive compensation (neither fixed, nor
switched), no capacitive compensation (neither fixed, nor switched), and no controlled reactive
compensation.
8
Table II – Cost comparison of the alternative of trunks a little longer than half wave length, with
LNC lines, with the three alternatives of R1, for the Madeira Transmission System
- Case a – Transmission system in AC, of the presented type, based in two 800 kV lines.
- Cases b/c – Transmission system in DC of R1 (Case b, considering information of published articles and
public EPE documents, and Case c, considering relative comparison costs, published by EPE, of
Cases c, d, e). Case b and Case c are assumed “analogous”.
- Case d – AC “conventional” transmission system, of R1, based in 765 kV lines.
- Case e – Hybrid transmission system, of R1, based in a DC bipole and in a “conventional” AC
transmission system, with 500 kV lines.
Comparison taking as basis (unity of relative cost) the Case a
Total cost (investment) [109 R$]
Case a
Case b=c
Case d
Case e
5.34
7.02
9.48 to 9.87 8.55 to 8.78
Difference of total cost (investment) [109 R$]
Dif. a-a Dif. b-a = c-a Dif. d-a
Dif. e-a
0
1.68
4.14 to 4.53 3.21 to 3.44
Relative cost (investment)
Case b=c
Case d
Case e
1.31
1.78 to 1.85 1.60 to 1.64
Difference of relative cost (investment)
Dif. a-a Dif. b-a = c-a Dif. d-a
Dif. e-a
0
0.31
0.78 to 0.85 0.60 to 0.64
Case a
1
reliability and execution time. For all this aspects, the presented alternative has also shown important
comparative advantages.
The proposed alternative presents, also, important advantages, especially in comparison with the DC
alternative, in what concerns operational and evolutionary flexibility. Namely:
- Allows the connection of each of the trunk terminals to a single busbar, or to several near busbars (at
distances till about 200 km).
- Allows interconnections in trunk intermediate points, maintaining the same functional and operational
conception. For instance, for Madeira Transmission System, it is viable to have several connections,
with the power in each one limited to a few hundred MW.
- May evolve, if and when that may become justified, for a set of lines, incorporated in a network, some
of them with eventual series or shunt reactive compensation.
It is important to notice that, when doing the comparison of the proposed methodology with the
alternatives proposed by EPE, in R1 document, for Madeira Transmission System, we have found
several aspects of such alternatives that imply in important technological risks, including:
- Several consequences of alternatives’ conceptual aspects, which imply in important risk of operational
problems, with reliability consequences, and whose “correction” is not trivial and may result in the
necessity of improvement measures, with eventual large costs and delays. An example: conceptual
options, of phases and insulators arrangement, and of structures, for 765 kV and 500 kV AC lines.
- Lack of specification precautions, for several important technological aspects of the R1 alternatives,
namely related to the very long length of the transmission system, to the important relative variation
of available generation power of Madeira power stations and to single pole operating constraints.
- Problems occurred in power transmission equipment that have not been adequately solved or
corrected by the electric sector, but that must be duly studied and object of efficient actions that assure an
adequate quality of important transmission systems. An example: transformers of converter stations.
Several of the mentioned aspects and problems have been presented and discussed in a public seminar
and are described in documents available in internet [21-22]. Its detailed presentation and discussion is
not within the scope of this article. As example, we comment here, only, one of such problems.
In Fig. 8 we indicate, briefly, the frequency behaviour of a 2500 km transmission trunk, comparing the
proposed alternative with an AC transmission trunk of the type presented in R1 responsible agencies’
document. It is evidenced that the R1 type has parameters with several poles and zeroes that originate low
damping transients and that, associated to active or non linear effects in network, may originate several
resonance or almost resonance phenomena. This behaviour occurs in other recent Brazilian transmission
systems and, typically, does not allow single-phase opening and fast reclosing for single-phase faults. This
restriction implies in an important negative effect in transmission trunk and electrical network reliability.
The comparison of Fig. 8 shows that this problem does not occur in transmission trunks of the proposed
type, for which a large majority of line faults may be eliminated, without significant effects.
9
2500 km
Im [Z e]
Im [Z e]
[Ω]
[Ω]
f [Hz]
Transmission trunk, with 2500 km, and
its equivalent π scheme, for nonhomopolar components or modes.
Imaginary part of elements Ze and Ye2
of equivalent π scheme:
A - For the presented transmission
trunk (without reactive compensation).
B - For R1 type AC transmission trunk
(with 100 % shunt and 70 % series
compensation).
f [Hz]
Im [Ye2]
Im [Ye2]
[S]
[S]
f [Hz]
f [Hz]
A
B
Fig. 8 − Frequency behaviour of a 2500 km transmission trunk, comparing the presented alternative
with an AC transmission trunk of the type presented in R1 responsible agencies’ document.
5. Conclusions
Very interesting solutions have been obtained, based in alternating current transmission trunks, nonconventional, with unitary transmission capability from 2 GW to 12 GW, and electric length a little
higher than half wave length (at power frequency), without the need of reactive compensation, or with
very small reactive compensation, and without the need of intermediate substations.
An example of its application is the natural medium term option for the Brazilian electric sector, leading
to base its growth in the hydroelectric resources of Amazon basin, with moderate complementary
generation based in other sources. This choice imposes an adequate solution to transmit most of such
energy at distances of the order of 2500 km.
The proposed solution type has been carefully studied and was object of robust validation procedures. It
was presented in several forums and was proposed, a few years ago, to responsible agencies, to study it
as an alternative to compare objectively with other ones for Madeira Transmission System. However,
although no technical contrary argument was given, the alternative was, in fact, ignored, by
“responsible agencies”.
It has also shown very important advantages, namely in comparison with the “traditional” type solution,
that has been applied in recent transmission systems in Brazil, and with all the alternatives included in
R1 document, prepared by responsible agencies, for Madeira Transmission System, as discussed briefly
in this article and, with more detail, in [1-22].
It is important to notice that, when doing the comparison of the proposed methodology with the
alternatives proposed in R1 document, for Madeira Transmission System, we have found several
important aspects of such alternatives that imply in important technological risks, including [21-22]:
- Several consequences of the conceptual aspects of such alternatives, which imply in important risk of
operational problems, with reliability consequences, and whose “correction” is not trivial and may
result in the necessity of improvement measures, with eventual large costs and delays.
- Lack of specification precautions, for several important technological aspects of the R1 alternatives,
namely related to the very long length of the transmission system, to the important relative variation
of available generation power of Madeira power stations and to single pole operating constraints.
- Problems occurred in power transmission equipment that have not been adequately solved or
corrected by the electric sector, but that must be duly studied and object of efficient actions that assure
an adequate quality of important transmission systems.
10
BIBLIOGRAPHY [*]
[1] C. Portela – “Brazilian Electric System – Realities and Options” (in Portuguese) – Symposium Prof. Carlos
Portela 70 years; The Science in Electrical Engineering, 60 p., December 2005 (reproduced in [21])
[2] C. Portela, M. Tavares – “Modeling, Simulation and Optimization of Transmission Lines. Applicability and
Limitations of Some Used Procedures” – Transmission and Distribution 2002, IEEE - PES Society, 38 p., Invited
speech, http://www.ieee/pesTD2002, São Paulo, Brazil, March 2002
[3] C. Portela, M. Aredes – “Very Long Distance Transmission” – Proceedings 2003 International Conference on AC
Power Delivery at Long and Very Long Distances, pp. 385-394, Novosibirsk, Russia, September 2003
[4] C. Portela, M. Tavares – “A New Methodology to Optimize Large EHV-Transmission System Transient Studies”
– Proceedings 2003 International Conference on AC Power Delivery at Long and Very Long Distances, pp. 444-454,
Novosibirsk, Russia, September 2003
[5] M. Aredes, E. Emmerik, R. Dias, C. Portela – “Facts Applied to Very Long Distance Transmission Lines” –
Proceedings 2003 International Conference on AC Power Delivery at Long and Very Long Distances, pp. 395-403,
Novosibirsk, Russia, September 2003
[6] M. Aredes, E. Sasso, E. Emmerik, R. Dias, C. Portela – “The GTO-Controlled Series Capacitor Applied to HalfWave Length Transmission Lines” – Proceedings International Conference on Power Systems Transients (IPST’ 2003),
pp. 1-6, New Orleans, United States, September/October 2003
[7] C. Portela, M. Tavares – “Six-Phase Transmission Line . Propagation Characteristics and New Three-Phase
Representation” – IEEE Transactions on Power Delivery, vol. 8, nº 3, pp. 1470-1483, July 1993
[8] C. Portela, G. Moreno, M. Tavares – “Advantages of Six-Phase Transmission as Compared with Three-Phase
Transmission Concerning Environmental Impact” – Eighth International Symposium on High Voltage Engineering,
ISHVE, Yokohama, Japan, Proceedings, art. 91.04, vol.4, pp. 349-352, August 1993
[9] C. Portela – “Some Aspects of Very Long Lines Switching” – CIGRE SC 13 Colloquium 1995, Florianópolis, art.
2.2, 12 p., September 1995
[10] C. Portela, S. Gomes – “Analysis and Optimization of Non-conventional Transmission Trunks, Considering New
Technological Possibilities” – VI SEPOPE, Symposium of Specialists in Electric Operational and Expansion Planning,
SP-092, pp. 1-6, Salvador, BA, Brazil, May 1998
[11] E. Watanabe, M. Aredes, C. Portela – Electric Energy and Environment: Some Technological Challenges in
Brazil – chapter of book Energy and Environment - Technological Challenges for the Future - editors Y. H. Mori, K.
Ohnishi - Springer - ISBN 4-431-70293-8 Springer-Verlag, p. 10-40, 2000
[12] C. Portela – “Six-Phase Transmission Systems . Functional Characteristics and Potential Application Domains”
(in Portuguese) – X SNPTEE, Curitiba, Brazil, art. CTBA/GSP/27, 8 p., October 1989
[13] C. Portela, M. Tavares – “Behavior and Optimization of a Six-Phase Transmission System for Normal Operation
and Transient Phenomena” (in Portuguese) – XI SNPTEE, Rio de Janeiro, Brazil, art. RJ/GAT/20, 6 p., October 1991
[14] S. Gomes, C. Portela, C. Fernandes – “Principles and Advantages of Using LPNEs and Comparative Results” (in
Portuguese) – XIII SNPTEE, Florianópolis, SC, Brazil, art. FL/GLT/23, 6 p., October 1995
[15] C. Portela – “A Computational System for Optimization of Non-Conventional Transmission Lines (LNC)” (in
Portuguese) – XIV SNPTEE, Belém, Pará, Brazil, 6 p., October 1997
[16] C. Portela – “Present Situation and Perspectives of Brazilian Electric Sector – Perturbations and Risks of
Panaceas and Wishes to Control Complex Physical Systems with Rules of Speculative Games” (in Portuguese) – 58th
Annual Meeting of SBPC – Technological Symposium – Power Systems Control – The Electric Energy and the Brazil
Development – Florianópolis, SC, Brazil, pp. 1-28, available at www.labspot.ufsc.br and in [21], 16-21 July 2006
[17] C. Portela. – Overvoltages and Insulation Coordination (in Portuguese), Vols. I, II, III - Vol. I, 349 p. , Vol. II,
304 p. , Vol. III, 140 p. - edition COPPE/UFRJ, Rio de Janeiro, 1982
[18] Portela, C. – Transient Regimen (in Portuguese), Vol. I, II, III, IV - Vol. I, 357 p. , Vol. II, 365 p. , Vol. III, 318 p. ,
Vol. IV, 280 p. - edition COPPE/UFRJ and ELETROBRÁS, Rio de Janeiro, 1983
[19] C. Portela, J. Silva, M. Alvim – “Non-Conventional AC Solutions Adequate for Very Long Distance
Transmission ‒ An Alternative for the Amazon Transmission System” – IEC/CIGRE UHV Symposium Beijing
2007-07-18/21, article 2-2-5, 29 p. – Beijing, China, 18-21 July 2007
[20] C. Portela, M. Alvim – “Non-Conventional AC Solutions Adequate for Very Long Distance Transmission ‒
An Alternative for the Amazon Transmission System” (in Portuguese) – Seminar Transmission of Electric Energy
at Long Distance, 30 p. – Recife, Brazil, 05 October 2007
[21] Documents related to the presented alternative, including pdf versions of part of previous references –
http://www.pee.ufrj.br/labs/corona/portela/lnc.html
[22] Documents presented and discussed in a Seminar about the presented alternative – Amazonian Hydroelectric
Resources – Non-conventional Alternatives for the Transmission Trunks - COPPE/UFRJ, 21 July 2008
http://www.pee.ufrj.br/labs/corona/portela/seminario20080721/seminario.html
[*] – Part of the indicated references are reproduced, in pdf version, in [21] and [22].
Acknowledgement
The author thanks the support received from CNPq National Council of Scientific and Technological Development
(Brazil) in part of research work in which this article is based.
11