ial size=-1 color=black> Below is a cache of http://www.eeh.ee.ethz.ch/pscc02/papers/s35p02.pdf. It's a snapshot of the page taken as our search engine crawled the Web.
The web site itself may have changed. You can check the current page or check for previous versions at the Internet Archive. Yahoo! is not affiliated with the authors of this page or responsible for its content.
SWITCHING TRANSIENTS ON LONG TRANSMISSION LINES WITH TAPPED TRANSFORMERS
SWITCHING TRANSIENTS ON LONG TRANSMISSION LINES WITH
TAPPED TRANSFORMERS
Liliana Oprea, Victor Popescu
FICHTNER GmbH&Co KG
Stuttgart, Germany
opreal@fichtner.de;

popescuv@fichtner.de
Abstract The paper presents the results of a study on
the probability distribution laws for switching overvoltages
on high voltage transmission lines with tapped transform-
ers. Three distribution functions have been considered:
normal, logarithmic normal and Weibull. In order to esti-
mate the agreement between the chosen distribution func-
tions and the empirical one, Kolmogorov-Smirnov and
c
2
goodness-of-fit tests have been applied. The paper includes
guidance to the choice of transformer ratings, winding
connections and adequate protection philosophies.
Keywords: switching transients tapped transform-
ers, single-pole autoreclosing, protection
1 INTRODUCTION
Many power systems include so-called energy corri-
dors consisting of long transmission lines between
remote production sites and bulk load centers as metro-
politan areas or major industrial sites. For obvious envi-
ronment and right of ways reasons, the technical solu-
tions for efficient use of the energy corridors are a major
optimization target. In this context, especially in devel-
oping countries, it is often required to drop along the
energy corridor relatively small amounts of power for
the supply of scattered loads. Depending on their size,
there is a choice between two conventional solutions.
-
Multi-terminal line: insertion in the main High
Voltage Overhead Line of a simple switching sub-
station (in-out); this solution is typically suitable for
a relatively large load site, including eventually
some own generation.
-
Tapped distribution transformers: directly con-
nected step-down transformers of relatively small
capacity as compared to the main line, supplying at
distribution voltage load sites without own genera-
tion. Transformer taps are often the only practical
conventional solution for supplying from the inter-
connected grid loads situated far away from the
main energy nodes. A similar concept is applied in
wide industrial areas like oil fields, where an over-
head distribution backbone supplies a large number
of small capacity tapped transformers.
Whatever the selected solution is, one of the main
application requirements is that the transformer or sub-
station tapping should not reduce the reliability of the
main transmission line, which, if interrupted for example
during peak load scenarios can be the cause of a severe
grid-wide incident.
In case of the multi-terminal lines with generation at
all terminals, the protection and autoreclosing philoso-
phy must take into account the infeed and outfeed ef-
fects; a large number of publications are dedicated to the
adequate selection of the protection functions and relay
settings [1].
Depending on the winding connection and its size, a
tapped transformer can be also a source of considerable
current for single-pole faults on the main line. In par-
ticular, there are concerns about the impact of the trans-
former tapping on the single-pole autoreclosing (SPAR)
of the main line. It is known that the galvanic and mag-
netic coupling of the faulty and sound phases through
the tapped transformers windings during the SPAR cycle
impairs on the secondary arc extinction time. However,
there is a clear need for more systematic information and
guidance to the choice of the tap equipment in order to
reduce or mitigate the adverse effects.
Based on the results obtained within an on-going
multi-national transmission back-bone project, the paper
shows the methodology applied for quantifying the im-
pact of the transformer tapping on switching overvolt-
ages during SPAR and provides design criteria for the
choice of transformer ratings, winding connections and
adequate protection philosophies.
2 PROBABILITY DISTRIBUTION LAWS
FOR SWITCHING OVERVOLTAGES ON
HIGH VOLTAGE TRANSMISSION LINE
For exemplifying the methodology, reference is made
to the simulation of switching overvoltages performed
for a single circuit 220 kV, 275 km OHL transmission
line. The statistical study is done by taking into account
a normal distribution for the breaker closure time. The
simplified system representation is given in Figure 1.
In determining the probability distribution for the
single pole switching overvoltages, three different prob-
ability distribution laws are considered: normal, loga-
rithmic normal and Weibull.
14th PSCC, Sevilla, 24-28 June 2002
Session 35, Paper 2, Page 1 1)
Note 1): isolator or circuit breaker
220 kV, 275 km
SENDING END
4000 MVA
RECEIVING END
3000 MVA
Figure 1: Simplified representation of the investigated system
The normal probability distribution function can be
written as follows [2]:
( )
( )
dU
U
dU
U
f
U
F
x
x
U
U
x
�
�
�
-
�
-
�
�
�
�
�
�
�
�
�
�
�
�
�
�
s
m
-
-
s
�
p
=
=
=
2
2
1
exp
2
1
(1)
where
m represents the overvoltage mean value and s
represents the standard deviation with respect to the
mean of the overvoltages U.
For the logarithmic normal distribution, a similar prob-
ability distribution function can be written:
( )
dU
U
U
F
x
U
x
�
�
-
�
�
�
�
�
�
�
�
�
�
�
�
�
�
s
m
-
-
s
�
p
=
2
ln
ln
2
1
exp
2
1
(2)
The parameters
m and s have the same meaning as for
the normal distribution.
The Weibull distribution can be analytically expressed
as:
( )
�
�
�
�
�
�
�
�
b
-
-
=
a
U
exp
U
F
1
(3)
where
a represents the shape factor and b the scale
factor.
For validation of the probability distribution laws,
two goodness-of-fit tests have been performed: Kol-
mogorov-Smirnov test [3] and
c
2
(Chi-Square) test [2].
3 SYSTEM MODELING
Most of the calculations were performed with
PSCAD-EMTDC software. Basic system data and as-
sumptions are briefly listed in the followings.
Sending end system
Short circuit contribution at the line terminals
4000 MVA (3-phase and single phase fault)
Receiving end system
Short circuit contribution at the line terminal
3000 MVA (3-phase and single phase fault)
Overhead transmission line
220 kV, 275 km, single circuit, 300/69 mm
2
OL-Al
frequency-dependent parameters, rated capacity 300
MVA
Tapped transformer
Rating 1, 5 and 10% of the line capacity
Standard values for the positive and zero-sequence
reactances
Connection group
Primary
YN
YN
D
YN
Secondary YN
D
YN
YN(d)
Performances of the 220 kV line protection
Typical fault clearing time for any fault location
along the line 100 ms (signal-aided, including circuit
breaker interruption time)
Single phase autoreclosing with several dead times up
to 1.5s
In line with the targets of this study, namely to high-
light the impact of the tapped transformers parameters
on the single-phase switching overvoltages, a reclosing
sequence for a fault in phase A at the sending end (ex-
pected maximum switching overvoltage at the receiving
end) was adopted for all simulations. In respect to the
tapped transformer, combinations between the four usual
winding connections and three transformer capacity
ratings have been investigated. All star points were
considered solidly grounded. The tertiary winding rating
is the usual 30 % of the through-load capacity.
Several positions of the tapped transformer along the
main line have been selected at 25%, 50% and 75% of
the line length from the sending end.
4

SIMULATION RESULTS
Of special interest is the magnitude of the switching
overvoltages and their probability distribution in respect
to the transformer rating and connection group.
To illustrate by example, Figure 2 presents the volt-
ages at the receiving line end in case of single phase
reclosing (phase A) with YN/YN transformer connected
in the middle of the line.
Table 1 gives the estimated parameter values of the
three probability distribution functions for the overvolt-
ages at the receiving end as function of the connection
group of the tapped transformer. Applying Kolmogorov-
Smirnov and
c
2
test, the agreement between the empiri-
cal and the estimated distribution has been investigated.
The empirical distributions are all obtained by simula-
tion.
For the line 1 in Table 1 (3 MVA YN/YN tapped
transformer), the histogram of empirical frequency dis-
tributions and the theoretical estimates are presented in
Figure 3: a- normal distribution, b- logarithmic normal
distribution and c- Weibull distribution.
14th PSCC, Sevilla, 24-28 June 2002
Session 35, Paper 2, Page 2 Figure 2:
Receiving end voltages at single-phase reclosing
Table 2 gives the estimated parameter values of the
three probability distribution functions for the overvolt-
ages at the receiving end of the line as function of the
rating