Chapter- 2

Problem identification

2.1 Introduction

The performance of a power system is mainly depending on the performance of

transmission lines. Therefore continues operation of transmission lines without
sudden outages is utmost important for the performance in the view point of power
delivery as well as system stability. Lightning effects on transmission lines are one of
the major reasons which lead to sudden line outages. As described in the section 1.5
of the previous chapter, the back flashover events are the dominant reason of line
outages.

The selected 220kV Biyagama-Kotmale transmission line is a double circuit line

which delivers the largest amount of power generated by Mahaweli Hydropower
Complex to the main load centre at Colombo. Sudden outage of these two circuits
creates an excess power generated at Kotmale end and tends to flow through the other
two 132kV lines feeding to Badulla and New Anuradhapura Substations connected to
Kotmale and Rantembe Substations respectively. Out of these two lines, the new
Anuradhapura line has higher impedance due to its higher line length compared to
Rantembe-Badulla line. Therefore, Rantembe-Badulla line is more vulnerable to trip
off due to overloading. As a result the Kotmale to New Anuradhapura 132kV line will
get overload and resulting in loss of all lines connected to the Victoria generation end.
Ultimately this creates a sudden loss of around 300MW to the system within few
seconds of time which leads the system towards total failure.

Therefore it is utmost important to avoid any double circuit failures by improving the
lightning performance of the selected Biyagama- Kotmale transmission line to avoid
total failures and associated severe financial losses.

2.2 Preliminary studies

According to the past performance records of this transmission line, it has been
noticed that the failure of this transmission line has great influence towards a total
failure of the system. Out of those, most of the line outages were due to the effect of

Page 115
lightning, since most ofthem were recorded in the months, April to June and October
to November where the lightning is frequent.

2.2.1 The relationship between monthly Isokeraunic level and line failures

According to the study [4] it has been found that there is a clear relationship between
the monthly Isokeraunic level (IKL) variations with the monthly average failures of
this line. Tables and graphs showing the "monthly failures" and "IKL variation with
monthly failure variation" respectively are reproduced here including few more
recently available data. Table 2.1 shows the monthly line failures from 2004 to 2009
whereas the Figure 2.1 illustrates the relationship of IKL level with the monthly
transmission line failures.

.... ....
........ ........ ~
~
....
....
~ ~

~
e
~

-....
.....
=:::...... ....
t;
-
= .....= 5
.... ~
.... ~ ~
5
-
.... =
- -. .... :::: ....
(.1
~
....
;.....
::=
""":
~~
~ ~
~

--r. ~
~

~
~
::=
-= --r.= :;r. '~ z6
~
Cfo .....
:::.. ~
(.1
~

~
(.1
~
~
.....
0
(-;

nu[51 2 :'\ -. -. ---~ u - - .., 26 (-).., 6) -. -"'-.

~""" _:-".
-- 29 _..,
"l -
__...,.

2004 (} (I II () (l (_) 0 (I (I 1 (I (l 01
-. -.
2005 (l 1 (I _.., 1 ..) 1 I) (I (I ) (I 14
2006 () {) u (I (I (I 0 (j 1 (I (J (J 01
2007 (l (I (I (I (I (I n n (I (I (\ n 00
2008 1 (I u 1 "l
,1 !
- (J 0 (I (I u 0 07
..,_..,
2009 (I (I (I
-•
(J !) 1
(J (l (I
- -1 11
Total
- -
1 1 0 4 7 5 1 0 1 3 7 4 34

Table 2.1 -Monthly line failures and IKL

Comparison of monthly line failures with IKL

14

~ 12 -Monthly
~
:;2 10 ave rage "lKL"
...0 8 -e-Monthly line
..."'
Q)

6
failures
..:!
·;;
.... 4
0
0 2
z
0
Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec
Month

figure 2.1 -Comparison of monthly line failures with IKL

Page I 16
2.2.2 Line sections having higher probability of insulator failures

It has been found in the previous study [4], that there are two line sections of the
selected transmission line have higher probability of insulator damages or flashovers
marks according to the breakdown and maintenance records. Those are:

1. The line section having towers, no.33 to 50 and

2. The line section having towers, no.85 to 95 from Kotmale side

Further it has been found that the selected transmission line is crossing over the
132kV Athurugiriya - Polpitiya transmission line at the above two line sections [4].
Therefore the above two line sections have been selected for the study described in

this report.

2.3 Back flashover effects on transmission lines

As described in section 1.5, back flashover events occur when the lightning strikes on
either tower or shield wires. These strikes produce waves of currents and voltages
travelling on the shield wires called travelling waves and reflections occurs at every
points where impedance discontinuities. Accordingly surge voltages can be developed
across line insulators exceeding the Critical Flashover Voltage (CFO) where
flashovers occur from tower to line called back flashovers. The following list shows
the parameters those affect the line Back Flashover Rate (BFR).

a) Ground flash density (See equation 1-1)

b) Surge impedances of the shield wires and towers

c) Coupling factors between conductors

d) Power frequency voltage

e) Tower and line height

f) Span length

g) Insulation strength

h) Footing resistance and soil composition

According to a previous study [4] the BFR for the selected Biyagama-Kotmale
transmission line is calculated as 4.37per year, whereas there is no direct flashover
failures can be expected. Therefore it is concluded that the total expected failures are

Page 117
only due to the back flashover failures of this line which is 4.37per year. Further the
calculated expected failure rate 4.37per year is justified by the observed value which
is only 4.6per year [4].

2.3.1 Earth faults at power frequency voltage due to back flashover events

An ionization path forms between the Arc horn gaps, when the air insulation between
the gaps is breakdown due to a back flashover event. This ionization path acts as a
conductive path to form an earth fault condition even at the power frequency voltages.
When a transmission line protection system is provided with auto-reclosing facility,
the circuit breaker will be reclosed automatically with a set time delay (500ms) after a
back flashover trip event to avoid permanent line outage. An earth fault can be
developed at power frequency voltage if the ionization path is persists at the moment
of first reclosing operation. Therefore the reclosing operation will be blocked and the
circuit breaker will be at opened position (breaker lockout) leaving the transmission
line at dead condition. This type of line outages can develop severe system
instabilities and even total failures. Such events have been reported in the selected
Biyagama-Kotmale transmission line in the past history of operation.

Therefore this issue has been addressed in this report by proving a software based
analysis approach to provide solutions through a new concept of providing protection
called transmission line mounted arresters (TLA).

2.4 Prevention of Back flashover events

Improving tower earthing resistance is the key way of avoiding back flashovers.
However it is not practicable as well as not economical when the towers are located at
hilly areas where the soil conditions are very bad. Unbalanced or improved line
insulation is another way of preventing back flashovers. However this is also not an
economical way due to the requirement of additional insulator discs as well as this
may need modifications in the towers. Therefore it is found that the most economical
and effective way of preventing back flashovers is to install Transmission Line
Arresters at selected tower locations.

Page I 18
2.5 Project objectives

The objectives of this study are:

I. Modeling and simulation of 220kV Biyagama-Kotmale power transmission

line in EMTP software (PSCAD) for lightning back flashover analysis

2. Conducting sensitivity analysis of line model for back flashover effects

3. Selection and modeling of Transmission Line Arresters (TLA) for EMTP

~\m\.\\~\\Q~~

4. Simulation and performance analysis of line model combined with selected

TLAs

\ Page 119
\;