Lightning interference occurs mainly on overhead lines and has been a
problem since the earliest days of the electricity supply industry. Overvoltages
which occur on the lines, travel toward the terminal or substation and
can cause damage, particularly to expensive equipment such as transformers.
In view of their importance, cost and the difficulty of making internal
repairs, the protection of large transformers against lightning overvoltages
is usually given special consideration. Lightning activity in South East
Asia, especially in Malaysia, ranks as one of the highest in the world.
Tenaga Nasional Berhad Research (TNBR) Malaysia has recorded as high as
320 kA lightning impulse current in Malaysia using their lightning detection
network system (LDNS). Every year, million dollars worth of damage is
caused by the devastating effects of lightning including to electrical
power systems. The transmission line trip in Malaysia is majorly caused
by lightning, which is about 70%.
Therefore, thorough knowledge on insulation coordination studies is urgently
needed strategic planning and protection of the expensive assets especially
in the substation section. Lightning overvoltages are fast front overvoltages
with times to crest from 0.1-20 msec. For substations, shield failures,
backflash and induced overvoltages generate surge voltages that impinge
on the substation equipment.
Lightning induced voltages are generally below 400 kV and are important
only for lower voltage systems. The incoming surges caused by the backflash
are more severe than that caused by shielding failures. As these surges
travel from the stroke terminating point to the station, corona decreases
front steepness and the crest magnitude. The shield wire has significant
impact on the wave propagation. A shield wire grounded at each tower makes
the propagation velocity of the ground mode wave component very close
to the conductor mode component. The magnitude of the surges caused by
a backflash ranges from 70 to 120% of the positive polarity critical flashover
voltage (CFO) of the line insulation. The front steepness is a function
of the conductor size, the distance between the location of the backflash
and the station (IEEE Power Engineering Society, 1999).
The objectives of this study are to model the high voltage substation
and perform the analysis on prediction of the level of current that causes
the transformer to breakdown and determine the effect of surge arrester
placement at the substation. This will be done by comparing the voltage
level measured close to the transformer with the suggested basic insulation
level (BIL) value used by the utility.
MODELING OF THE SYSTEM
The main emphasis of this research is to model a high voltage substation
for the lightning surge analysis. This modeling must include the tower,
power line, tower footing resistance, lightning, substation equipments
and insulation coordination.
|| Substation model for case studies
|| Key parameters used for modeling the system
||Comparison of capacitors value between TNB calculation
and IEEE recommendations
Figure 1 shows the modeling arrangements
at the substation, which is adopted for the case studies and based on
real configurations of the TNB`s 132 kV substation. The lightning strike
is placed at the tower close to the substation. The distance between the
tower and the substation entrance is 50 m. Point E1 measures the entrance
voltage induced by the lightning and point E2 is the point-of-connection
(POC) of the surge arrester, where the voltage is expected to be clamped
before passing through the capacitive voltage transformer, labeled as
CCVT. Whilst points E3 and E4 are the second surge arrester,
SA2 and the power transformer, labeled as CTX, respectively.
Further of specific details relating to the model are described in Table
Table 2 describes the comparison of capacitor value
between TNB calculation approach and IEEE recommendation base on 115 kV
US substation system model. For this study, TNB calculation approach of
capacitor values was adapted to model the system as it more or less agreed
with the value recommended by IEEE WG 3.4.11 (1992) and for the actual
analysis. The distance between each substation equipments are as below:
Tsub1=3.0 mTsub5=4.0 mTsub9=14.5
Tsub2=3.5 mTsub6=4.5 mTsub10=3.0
Tsub3=3.5 mTsub7=3.0 mTsub11=3.0
Tsub4=3.0 mTsub8=3.0 mTsub12=5.0
SURGE ARRESTER MODELING
Several models of arrester had been described elsewhere in literature (IEC,
1993; Martinez and Castro-Aranda, 2004; IEEE WG 3.4.11, 1992). Most of the arrester
model must include two nonlinear resistances A0 and A1 as shown in Fig.
2, with other combination of the components. However for different approach,
it is basically using different type of lumped parameter arrangement. The frequency-dependent
surge arrester model which was recommended by IEEE WG 3.4.11 (1992). is used
in this work. This model is shown in Fig. 2 and it was reported
as the most accurate representation based on single phase line model (Goudarzi
and Mohseni, 2004). Adjustment procedure of parameters is described by IEEE
WG 3.4.11 (1992).
|| IEEE frequency-dependent model
RESULTS AND DISCUSSION
Surge arrester breakdown current: There are two cases are considered
under the critical conditions; when the surge arrester 1 (SA1) is not
operated and also when both surge arresters (SA1 and SA2) are not operated.
The idea is to demonstrate the effect of floating surge arrester (missing
of copper conductor connected between surge arrester and substation earthing)
due to the vandalism cases as reported by the utility company in recent
Table 3 shows the data for the case where no SA1 is
installed. As the currents increase, the voltage level also increases.
The BIL used by TNB Malaysia for the 132 kV rated transformer is 550 kV.
Therefore, the probability of the capacitive voltage transformer, CCVT
damage can be estimated when the lightning current reaches 144 kA.
Table 4 shows the data for the case when both surge
arresters are not operated. This is the worst case scenario that could
possibly happen involving the case of vandalism on the surge arresters.
For the case of capacitive voltage transformer, CCVT, it is
estimated that the probability of the damage is at the current of 33 kA,
whilst for the case of power transformer, CTX, current of 31
kA can already cause the breakdown on the equipment.
Effect of surge arrester placement: Table 5
demonstrates the effect of surge arrester placement at the substation.
For the first case, SA2 is placed 8 m before power transformer, CTX,
instead of the real placement which is just 5 m. For the distance of 8
m, Table 5a shows that the voltages level at point E4
is slightly higher compared to the result in Table 5c
for the original placement. Whilst for the SA2 placed at 11 m away from
Tx, voltages level at point E4 are also increased, as shown
by the data in Table 5b. Having the differences for
only few kilovolts, the results perhaps very difficult to be judged. However,
this is very good analysis in determining a proper insulation coordination
|| Case of no SA1 is installed
|| Case of both surge arresters are not operated
|| Effect of surge arrester placement
In this case, having the surge arrester located at the proper
location is very crucial and without having all the related knowledge,
it is very difficult in making a decision.
Detail modeling guidelines and parameters for high substation are successfully
presented. Results for the first part have clearly shown that the impact
of lightning surge can be very dangerous even at low value of current
if there is no surge arresters are in operating or used for protection.
Overall results have demonstrated the importance of having a right location
of surge arrester placement as it is crucially needed in order to optimize
the substation performance in term of its reliability and cost effective.
In other words, this surge arrester must be placed as close as possible
to the equipment to be protected, as fail to do so will cause a significant
damage to the equipment.
The authors would like to express their sincere gratitude to the Engineering
Department (Transmission and Substation) of the Tenaga Nasional Berhad
for their cooperation and kind supply of various technical data.