Arc-wall and arc-electrode interactions are dominant phenomena.

Pressure (ba)
.........
Computed solution Indeed, we estimate that in a medium voltage circuit breaker more than
Experiments 75 percent and in a low voltage circuit breaker more than 90 percent
of the arc energy is absorbed by walls and electrodes.
1.0 We have developed a model which is built on an energy balance at
the boundary between the flow and the wall (or the electrode). In the
gas SF6, we solve the Navier Stokes equations and, in the wall material,
we account for thermal conduction.
0.5 With the above model, a computation has been done on a puffer
circuit breaker geometry. In this kind of breaker, a piston creates an
over pressure creating a flow along the electrical arc. The flow is guided
by a loaded teflon nozzle. During the switching phenomenon, the
electrical arc induces ablation of the nozzle material. Figure 3 presents
Time (ms)
comparisons of the pressure in the puffer chamber.
0.0 3.(0

Figure 1. Comparisons of the pressure in the expansion volume

Transformers

96 WM 224-6 PWRD, T-PWRD October 1996

Arc Voltage (V)
50
Tinme (ms)
3.0 6.0 9.0
Figure 2. Comparisons of the arc voltage (i=8 kA, 50 Hz)
Pressure (ba) ~ experimenEs
6 9
Figure 3. Comparison of pressure in a puffer circuit breaker
We assume that a part of the spectrum of the emitted light is absorbed
in the cold gas and that another part isn't absorbed (this part is called
the thin radiation). The absorption in the cold gas is defined using an
exponential law, determined through experimental measurements.
To validate the model, a thermal expansion medium voltage circuit
breaker has been simulated. The switching principle of thermal expan-
sion circuit breaker is the following: the electrical arc heats the gas in
the expansion volume, press then increases; at the current 0, this
over-pressure blows the arc. The validation consists in a comparison
between experiments and calculations for the pressure in the expansion
volume together with the arc voltage. The peak of current is 8 kA and
the arc's time 9 ms (Figure 1 and Figure 2). The good agreement
between experiments and computations shows that macroscopically the
energy exchanges are well taken into account.
EHV Transformer Bank Unbalance:
Practical Issues and Solutions
R. Gutman (Senior Member, IEEE; American Electric Power
Service Corporation, Columbus, OH), L.B. Wagenaar (Senior
Member, IEEE; American Electric Power Service Corporation)
An unconventional tap configuration has been implemented in
AEP's South Canton 765/345/34.5 kV, 3-500 MVA transformer bank
to counter the combined effects of impedance mismatch within the
bank and natural unbalance of the extra-high-voltage (EHV) transmis-
sion system. These effects manifest themselves as unequal phase volt-
ages and loadings and as elevated neutral and tertiary circulating
currents in the bank. This paper describes the South Canton bank
unbalance, the concerns it raised, and the actions taken by AEP to
control its possible adverse effects on the integrity and operation of the
South Canton Station facilities.
South Canton Bank Unbalance: Heavy loadings with signifi-
cantly unbalanced phase currents were observed in the South Canton
bank of three single-phase 765/345/34.5 kV, 500 MVA transformers
following their return to service with a new, mismatched transformer
in the Phase 1 position. On-site measurements taken by field personnel
on several occasions, supplemented by three-phase readings from the
AEP data acquisition system, confirmed the presence of this unbal-
anced operating condition. The condition raised concerns about possi-
ble overloading of one or more phases of this bank, which provides
principal supply of electric service for Canton, Ohio and the surround-
ing communities. In addition, unbalance effects had been observed in
the neutral and tertiary connections of the bank, where current flows
were reaching relatively high magnitudes. Sustained exposure to high
neutral currents was deemed undesirable due to potential safety prob-
lems in the station, accelerated corrosion of the station grounding mat
and its neutral connections, and possible interference with nearby
telecommunication lines.
The new South Canton single-phase transformer represents a con-
siderable mismatch at that station; its high-to-low reactance, tertiary
reactance, and summer-normal rating are about 30 percent higher, 50
percent higher, and 10 percent lower, respectively, compared with the
transformers operating in Phases 2 and 3. The South Canton trans-
former bank is fed radially from Kammer Station through an 80-mile
765 kV line which, by itself, is a source of unbalance because of its
flat nontransposed configuration.
Several strategies were considered for controlling the South Canton
bank unbalance, including energization of a single-phase 100 MVAr
spare shunt reactor located at Kammer, changing tap(s) on one or more
phases of the South Canton bank, inserting an air-core reactor in the
bank's tertiary delta, and opening the delta and/or neutral connections.
All of these methods were analyzed for their effectiveness in reducing
the bank unbalance, but the last two were dismissed due to increased
equipment risk. Consideration was also given to changing the assigned

54 IEEE Power Engineering Review, October 1996

phase positions of the South Canton transformers by either electrical
reconnection or physical moves, but the high cost of this approach
made it difficult to justify.
Raising taps on Phase 2 and 3 transformers by one position to
746 kV, while holding the Phase 1 tap at 727 kV, was considered
sufficiently effective and secure, and was selected for implementa-
tion. A number of other tap arrangements were also studied, but
none were found as effective. The selected approach, while uncon-
ventional, reduces the neutral and delta currents in the South Canton
bank by about 50 percent and 10 percent, respectively. The added
advantages of this approach include lower reactive flow through the
bank and improved voltage profile on the 345 kV bus. The 765 kV
bus voltages at South Canton are more unbalanced as Phases 2 and
3 rise by about 1.5 percent each, but this is expected to pose no
operating problems even during light load conditions. Voltages at
Kammer rise by less than 0.5 percent. Load losses in the South
Canton bank are reduced by up to 70 kW, or 3-4 percent. Similar
performance improvements, although smaller in magnitude, exist
during other operating conditions.
Conclusion: As the remaining useful life of existing EHV trans-
formers declines over time and these transformers are replaced with
new units of different designs, purchased from new suppliers world-
wide, the use of dissimilar single-phase transformers in a common
bank has and will become more widespread. Experience shows that
unbalanced operation of transformers is technically feasible, pro-
vided that the capability of each of the single-phase transformers is
adhered to and care is taken to control their neutral and tertiary
circulating currents, including possible adverse effects of these
currents on station facilities.
It is noteworthy that the operating condition achieved at South
Canton through the use of unbalanced transformer taps would be
attainable with no tap changes, if transformer reactance mismatch were
only about one-third of that actually encountered. Further work in this
area could provide useful insights with regard to both acceptable manu-
facturing tolerances for single-phase EHV transformers and operational
flexibilities available to the electric utilities.
96 WM 225-3 PWRD, T-PWRD October 1996
An Artificial Neural Network Approach
to Transformer Fault Diagnosis
Y. Zhang (The Bradley Department of Electrical Engineering,
Virginia Tech, Blacksburg, VA), X. Ding (Virginia Tech), Y Liu
(Virginia Tech), PJ. Griffin (Doble Engineering Company,
Watertown, MA)
As a major apparatus in a power system, the power transformer is
vital to system operation. Techniques for diagnosis and incipient fault
detection are valuable.
A transformer is subject to electrical and thermal stresses. These
two stresses could break down the insulating materials and release
gaseous decomposition products. Overheating, corona and arcing are
three primary causes of fault related gases. Principally, the fault related
gases commonly used are hydrogen (H2), carbon monoxide (CO),
carbon dioxide (CO2), methane (CH4), acetylene (C2H2), ethane
(C2H6), and ethylene (C2H4). The analysis of dissolved gases is a
powerful tool to diagnose developing faults in power transformers.
Many diagnostic criteria have been developed for the interpretation of
the dissolved gases. These methods would find the relationship between
the gases and the fault conditions, some of which are obvious and some
of which may not be apparent (hidden relationships). However, much
of the diagnostics relies on experts to interpret the results correctly.
New computer-aided techniques can consistently diagnose incipient-
fault conditions for the novice and in some cases may provide further
insight to the expert. Expert system and fuzzy-set approach have been
developed to reveal some of the hidden relationships in transformer
fault diagnosis. Expert system derives the decision rules from the
IEEE Power Engineering Review, October 1996
previous experience while the fuzzy-set represents the decision rules
by using vague quantities. Artificial neural network method (ANN) has
also been used for this purpose since the hidden relationships between
the fault types and dissolved gases can be recognized by ANN through
training process.
A two-step neural-network classifier has been developed and care-
fully tested for transformer fault diagnosis using dissolved gas-in-oil
analysis. Several feature types have been evaluated and several neural-
net topologies have been considered. Patterns with different identifica-
tion sensitivity are detected by different ANNs to obtain the highest
accuracy for each pattern. The two-step approach makes ANN easier
to train and more accurate in detecting faults.
The results of this study show that the two-ANN approach is prom-
ising for transformer fault diagnosis even with limited sample data.
Theoretically, the ANN can be trained to represent any observable
phenomenon if there are sufficient data available. The more complex
a relationship is, the more training data are needed. The transformer
fault diagnosis could be very complicated. For example, it is desirable
to distinguish between faults of oil and cellulosic materials (paper),
different temperatures (low, medium, and high) for overheating, in oil,
or low energy and high energy sustained arcing. To deal with such a
complicated diagnosis problem, the available input data may not always
be enough. It can be a very effective way to construct different ANNs
for different pattern recognitions to obtain the highest diagnosis accu-
racy for each pattern.
The accuracy of the diagnosis can be improved with the increase
of training data. Another very important part of dissolved gas-in-oil
analysis is evaluation of trends - the increasing rates of gas generation.
Once this information is available, the diagnosis can be made more
reliable. Due to the limited trend data available, this work will be done
at a later stage.
Transmission and Distribution
95 ESMO, T-PWRD October 1996
Transmission Towers Inventory
VI. Kogan (Senior Member, IEEE; American Electric Power
Co., Columbus, OH), R.J. Gursky (Member, IEEE; American
Electric Power Co.)
The transmission network of an electric power utility constitutes a
critical part of the whole power system. It is clearly not desirable to
lose any major transmission lines for longer than a few weeks because
of the increased probability of loss of load for multiple contingencies.
Because new transmission towers are not readily available, it is impor-
tant to maintain an adequate inventory of towers to prevent long-term
outages.
This paper discusses a statistical procedure to analyze failure data
for transmission towers and to determine adequate levels of inventory.
The failure of a transmission tower is defined as a damaged condition
that prevents operation of the associated transmission line.
A standard approach to inventory assessment is to minimize the
expected cost of a reasonable number of spares. That cost is a sum of
two parts: one is associated with cost of owning spares, another is
associated with the cost of nonavailability of spares. Unavailability of
transmission towers may lead to the inability to supply power to the
utility's customers, which is unacceptable. Therefore, the cost of
nonavailability of spares is assumed to be so large that the above
mentioned standard approach is inapplicable.
As an alternate, this paper suggests a methodology based on
historical data of failures, using the homogeneous Poisson process
(HPP) as a mathematical model of the failure process. Such a choice
is based on the common failure causes of transmission towers,
including tornadoes, combinations of severe icing and wind, and
vandalism. Theoretically, these causes could lead to tower failure
or weakening of the structures. However, such weakening is not
5s