THESIS

COST-EFFECTIVE INSULATION COORDINATION DESIGN

OF 115 KV TRANSMISSION LINE FOR LIGHTNING BACK-
FLASHOVER

PHONESAVANH SENGMANIVANH

GRADUATE SCHOOL, KASETSART UNIVERSITY

Academic Year 2021
 THESIS APPROVAL
GRADUATE SCHOOL, KASETSART UNIVERSITY

DEGREE: Master of Engineering (Electrical Engineering)

MAJOR FIELD: Electrical Engineering
DEPARTMENT: Electrical Engineering
TITLE: Cost-Effective Insulation Coordination Design of 115 kV
Transmission Line for Lightning Back-Flashover

NAME: MR. PHONESAVANH SENGMANIVANH

THIS THESIS HAS BEEN ACCEPTED BY

THESIS ADVISOR
(Assistant Professor Komsan Hongesombut, Ph.D.)
DEPARTMENT HEAD
(Assistant Professor Nithiphat Teerakawanich, Ph.D)

APPROVED BY THE GRADUATE SCHOOL ON

DEAN
(Associate Professor Srijidtra Charoenlarpnopparut, Ph.D.)
 THESIS

COST-EFFECTIVE INSULATION COORDINATION DESIGN OF 115 KV

TRANSMISSION LINE FOR LIGHTNING BACK-FLASHOVER

PHONESAVANH SENGMANIVANH

A Thesis Submitted in Partial Fulfillment of

the Requirements for the Degree of
Master of Engineering (Electrical Engineering)
Graduate School, Kasetsart University
Academic Year 2021
 C

ABSTRACT

PHONESAVANH SENGMANIVANH : Cost-Effective Insulation

Coordination Design of 115 kV Transmission Line for Lightning Back-Flashover.
Master of Engineering (Electrical Engineering), Major Field: Electrical
Engineering, Department of Electrical Engineering.
Thesis Advisor: Assistant Professor Komsan Hongesombut, Ph.D.
Academic Year 2021

This thesis presents a method to design the insulation coordination for the
Electricité Du Laos (EDL) transmission line due to the back-flashover phenomenon
under cost constraints. Although the transmission line is initially well designed and
calculated, it can reduce the effectiveness of protection because of a lifetime,
contamination, and heavy operation. As a result, the withstand capability cannot keep
at complete as in the beginning. Therefore, improvements need to be made to ensure
the insulation coordination is always readily available and sufficiently effective. In
this research, the transmission line, lightning surge, and the different protection are
modeled on PSCAD/EMTDC software. Then, the total costs of the transmission line
design are calculated. This thesis aims to achieve the lowest cost of design. Therefore,
the multi-objective optimization method is used to find the best design satisfying the
transmission line design’s technical and economic aspects.

_________________ _______________________________ ____ / ____ / ____

Student's signature Thesis Advisor's signature
 D

ACKNOWLEDGEMENTS

ACKNOWLEDGEMENTS

This thesis would not be possible without direct and indirect supports from these
people. Firstly, I appreciate my advisor, Assistant Professor Komsan Hongesombut, for
providing support and guidance; his knowledge and experience were extremely helpful in
this research. Secondly, I am eternally grateful to Assistant Professor Dulpichet
Rerkpreedapong from Kasetsart University and Associate Professor Pramuk
Unahalekhaka from the Rajamangala University of Technology Suvarnabhumi, have
taken their valuable time to become the examination committees and kindly given the
advice to make this thesis more complete.

I am so thankful to officers at the electrical engineering department, Kasetsart

University, for their help and the PEAKU5 group for the best friendship throughout study
there. Special thanks go to my parents for their love, support, and encouragement
throughout my life.

Finally, I am so grateful to the Provincial Electricity Authority (PEA) and

Electricité Du Laos (EDL) for supporting a scholarship in this research.

PHONESAVANH SENGMANIVANH
 TABLE OF CONTENTS

Page

ABSTRACT .................................................................................................................. C
ACKNOWLEDGEMENTS .......................................................................................... D
TABLE OF CONTENTS .............................................................................................. E
LIST OF TABLES ........................................................................................................ G
LIST OF FIGURES ...................................................................................................... H
INTRODUCTION ......................................................................................................... 1
OBJECTIVE .................................................................................................................. 4
LITERATURE REVIEW .............................................................................................. 5
MATERIALS AND METHODS ................................................................................. 13
Materials .................................................................................................................. 13
Methods ................................................................................................................... 13
1. Power system network ......................................................................................... 15
2. PSCAD/EMTDC modelling and simulation........................................................ 15
2.1. Electromagnetic transient modelling for Nammang3-Nabong .................. 15
2.1.1 Transmission line model.................................................................. 16
2.1.2 Tower model.................................................................................... 19
2.1.3 Footing resistance model ................................................................. 24
2.1.4 Insulator and back-flashover model ................................................ 25
2.1.5 Line terminal model ........................................................................ 29
2.1.6 Substation equipment "Autotransformer" model ............................ 29
2.1.7 Line surge arrester model ................................................................ 30
2.1.8 Lightning stroke model.................................................................... 33
2.2 Method of simulation .................................................................................. 36
2.2.1 Simulation variable settings ............................................................ 36
2.2.2 Simulation criteria ........................................................................... 36
 F

2.2.3 Project simulation settings ............................................................... 37

3. Economical objective ........................................................................................... 37
4. Multi-objective optimisation................................................................................ 40
RESULTS AND DISCUSSIONS ................................................................................ 42
CONCLUSIONS AND RECOMMENDATION ........................................................ 46
Conclusions.............................................................................................................. 46
Recommendation ..................................................................................................... 47
LITERATURE CITED ................................................................................................ 48
APPENDIX .................................................................................................................. 51
CURRICULUM VITAE .............................................................................................. 56
 LIST OF TABLES

Page

Table 1 The fault statistics of the central-1 transmission system .................................. 2

Table 2 Characteristics of transients in power systems ................................................. 5
Table 3 Tower dimension parameters .......................................................................... 21
Table 4 Calculated result for tower dimension ............................................................ 22
Table 5 Tower footing resistance parameters .............................................................. 24
Table 6 Initial V-I values for A0 and A1 ...................................................................... 32
Table 7 Basic impulse insulation levels (WADHWA, 2009) ...................................... 37
Table 8 List and price of EDL 115 kV transmission line of financial proposal .......... 39
Table 9 Calculation results of total investment costs................................................... 39
Table 10 Calculation result of lightning overvoltage and total cost ............................ 44
Table 11 Calculation result of weighting coefficients ................................................. 45
 LIST OF FIGURES

Page

Figure 1 Planned power system diagram in the year 2017 ............................................ 1

Figure 2 Summary of cause failure ................................................................................ 3
Figure 3 Lightning stroke points .................................................................................... 5
Figure 4 Procedure method for optimal insulation coordination design...................... 13
Figure 5 Single line of Nammang3 hydropower plant – Nabong substation............... 15
Figure 6 Completed transmission line model purposed for analysis ........................... 16
Figure 7 Overhead transmission line segment ............................................................. 17
Figure 8 Transmission line configuration windows..................................................... 17
Figure 9 Frequency dependent (phase) model ............................................................. 18
Figure 10 Tower dimension ......................................................................................... 18
Figure 11 Geometry data input for conductor.............................................................. 19
Figure 12 Multistory of transmission line tower model............................................... 19
Figure 13 Bergeron model and manual entry of data .................................................. 23
Figure 14 Transmission tower model .......................................................................... 23
Figure 15 Test of footing resistance............................................................................. 25
Figure 16 Tower footing resistance model .................................................................. 25
Figure 17 Insulator string flashover model .................................................................. 26
Figure 18 Back flashover logic diagram ...................................................................... 26
Figure 19 Insulator string model .................................................................................. 27
Figure 20 Back flashover control model ...................................................................... 27
Figure 21 Voltage-time of insulator string with 1.168 m and 1.606 m ....................... 28
Figure 22 Line impedance terminal ............................................................................. 29
Figure 23 Winding capacitance terminal ..................................................................... 29
Figure 24 Lightning arrester circuit ............................................................................. 30
Figure 25 IR graph for A0 and A1 ................................................................................ 31
Figure 26 Line surge arrester model ............................................................................ 32
 I

Figure 27 Test surge waveform 8/20 µs with 10 kA ................................................... 33

Figure 28 Test line surge arrester discharges waveform 8/20 µs at 10 kA .................. 33
Figure 29 Standard 8/20 µs current waveform ............................................................ 34
Figure 30 Standard 1.2/50 µs voltage waveform ......................................................... 35
Figure 31 Module of lightning stroke .......................................................................... 35
Figure 32 Test surge waveform 1.2/50 µs at 10 kA..................................................... 35
Figure 33 Project setting for simulation....................................................................... 37
Figure 34 Overvoltage without improvement a systems ............................................. 42
Figure 35 Overvoltage with 10 Ω footing resistance and 1.606 m insulator length .... 42
Figure 36 Overvoltage with 96 kV line surge arrester and 10 Ω footing resistance.... 43
Figure 37 Overvoltage with 96 kV line surge arrester and 1.606 m insulator length .. 43
Figure 38 Overvoltage with combine all protection scheme ....................................... 44
 INTRODUCTION

Electricité Du Laos (EDL) is a state enterprise in Laos. EDL belongs to the

energy and mines ministry; it operates electricity to power consumption domestically
and exports the power to neighboring countries such as Thailand, China, Vietnam,
Cambodia, and Singapore. The power system of EDL has comprised generation
systems, transmission systems, and distribution systems. The transmission system of
EDL is divided into four parts as northern, central-1, central-2, and southern
transmission system areas. The line is an overhead transmission line (OHTL) at voltage
levels of 115 kV, 230 kV, and 500 kV with a total length of 9,657 km (Mines, 2018).
Meanwhile, EDL has many projects under construction based on the master plan of the
Lao government to make Lao PDR into a battery of ASEAN.

Figure 1 Planned power system diagram in the year 2017

 2

Although EDL is growing, EDL still encounter a problem of power quality.

According to EDL annual report, the power interruption is high. About 14 % of failures
in the transmission line causes by lightning; these faults result in a power interruption
for hours or days due to damage to the transmission equipment. The power disruption
affects production processes in industrial estates, transportation, affecting people,
causing high economic losses, which is one factor of investors deciding to start a
business. In addition, it is a big problem for EDL to establish customer confidence. To
solve the issues and make credulity with customers and investors, the system
improvements to increase protection efficiency, reduce the rate of failure, and make
reliability for customers is essential and challenging for EDL.

Table 1 The fault statistics of the central-1 transmission system

Cause types
No. Line description Natural
Tree Lightning Accident Animal Unknown
disaster
1 KhoksaAd - Nabong1 2 0 2 1 0 1
2 KhoksaAd - Nabong2 2 0 2 1 0 1
3 KhoksaAd - Viengkeo 2 2 0 1 0 2
4 Nammang3 – Lao steel factory 0 0 0 0 0 0
5 Nammang3 - Nabong Line1 1 2 4 0 0 1
6 Nammang3 - Nabong Line2 1 2 4 1 0 0
7 Naxaythong - Donkoi 0 0 0 0 0 0
8 Naxaythong - Namnguem1 Line1 0 2 0 0 1 3
9 Naxaythong - Namnguem1 Line2 2 2 1 0 1 3
10 Naxaythong - Phonsoung 2 1 1 0 0 6
11 Naxaythong - Pakthang 1 0 2 0 2 2
12 Naxaythong - Thangon 1 2 1 0 0 3
13 Naxaythong - Viengkeo 3 1 3 0 2 19
14 Phontong - Donkoi 4 2 0 1 3 6
15 Phontong - Nongkhai Line1 2 3 3 1 0 9
16 Phontong - Nongkhai Line2 0 3 3 1 0 5
17 Phontong - Naxaythong Line1 12 10 5 4 2 18
18 Phontong - Naxaythong Line2 6 4 4 2 2 7
19 Phontong - Naxaythong Line3 4 4 1 3 1 7
20 Phontong - Pakthang 2 2 1 2 1 6
21 Thanaleng - Donkoi 2 2 0 0 1 3
22 Thanaleng - KhoksaAd 1 2 1 1 0 3
23 Thanaleng - Nongkhai3 3 3 2 1 1 11
Total 53 49 40 21 17 116
 3

18%

38%

17%

6%
7% 14%

Tree Natural disaster Lightning Accident Animal Unknown

Figure 2 Summary of cause failure

This research analyzes the magnitude of lightning overvoltage on 115 kV

Nammang 3 hydropower plant (HPP) - Nabong substation and presents a method to
design and determine the appropriate protection scheme and precautions to be taken,
and practical guidelines. In this study, numerical electromagnetic analysis is used to
simulate the system for analysis and coordination insulation design; in such programs,
electrical equipment components can model in detail. In addition, this work has
considered the economic aspect based on saving investment.
 4

OBJECTIVE

1. Develop comprehensive modeling of an existing 115 kV transmission line

and the insulation coordination design for transmission line lightning protection based
on various risk factors by using PSCAD/EMTDC simulation program.

2. Calculate for transmission line cost of in each case protection design.

3. Determine the appropriate insulation coordination design for transmission

lines based on performance following various risk factors and cost constraints for
saving cost by using multi-objective optimization.
 5

LITERATURE REVIEW

Lightning is a natural phenomenon spectacular and dangerous; this

phenomenon cannot control and predictable also establishes damage to humans, living
things, and objects. The transmission line is one power equipment that encounters
lightning strikes because the structure is tall and installed out-of-doors. The lightning
phenomena on OHTL or nearby can cause temporary overvoltage to systems, known
as transients. The overvoltage due to lightning may be greater than other overvoltages,
high frequency, and shortest occurring time. Where lightning surges current through
the transmission tower with impedance arise traveling overvoltage in a system, the
voltage increase 4 – 6 times of phase voltage, time duration of transient is
0.1microsecond – 1.0 millisecond, and a frequency in the range 10 kHz – 3 MHz (Das,
2010).

Figure 3 Lightning stroke points

Table 2 Characteristics of transients in power systems

Character Time Duration

Lightning 0.1 µs – 1.0 ms
Origin Frequency Range
Lightning surges 10 kHz – 3 MHz
Representation Mainly for
Lightning overvoltage
Shape Designation
Fast front surges
 6

The lightning surges on OHTL depend on how surges are caused and describe
the shape of the lightning current wave as usual; they can be classified broadly into five
types: The first is lightning strikes directly to phase conductor without flashover. The
second is an attack to phase conductor followed by a flashover. The third is an attack
to the transmission tower or overhead ground wire (OHGW) followed by a back
flashover (BFO) to a phase conductor through an insulator string due to the ground
potential. The fourth is the third type BFO but followed by flashover along the ground
to a conductor of the previous or nearby tower, stepped if subsequent flashover to earth
occurs at more than one tower. Finally, the fifth is lightning that strikes the ground
nearby a transmission tower and induces through the tower footing system.

Insulation coordination indicates the performance of the ability to work of

lightning protection on transmission lines. Although it is well designed, the
performance can deteriorate due to lifetime, environmental, and operation factors. It is
requisite to know the magnitude of a lightning stroke to assign protection. The subject
of analysis to reduce a fault and increase efficiency has been of considerable interest to
electrical engineers for decades. Generally, the method of prevent against phase
conductors from direct lightning is usual to use OHGW. In the beginning, some
transmission lines in America began to use ground wire to avoid lightning strikes; a
conductor connects parallel to phase conductors and grounds at intervals. These
conductors mounted upper the tower. Besides being used for the transmission line, this
experiment applies to a factory for a while. The result of this method has good feedback
against lightning. In terms of indirect lightning protection, a lightning arrester (LA)
uses to reduce overvoltage in the system. Reducing the overvoltage caused by lightning-
induced using periodic grounding protection lines is another factor in the prevention
(Mershon, 1903) (Neall, 1905).

Another experimental study of flashover phenomenon on OHTL due to direct

lightning strike to OHGW (Novizon, Seftiani, & Ahmad, 2019) has focused on tower
footing resistance. Based on modeling on EMTP-ATP software, the lightning struck
point was used to simulate and analyze the event, including transmission line tower,
grounding, lightning stroke, an insulator with flashover, and the focal point on this
method is the footing resistance. The area struck the point of approximately range from
 7

the substation. Similarly (Ametani & Kawamura, 2005) presented an insulation

coordination design due to lightning overvoltage in Japan for a 1,100 kV transmission
line. Tokyo Electric Power Company has used EMTP for simulation adopted JEC-
0102-1994. This model includes transmission sections, AC sources and matching
impedance terminal, lightning stroke and impedance, tower and cross-arm, tower
footing resistance, arc-horn flashover, corona wave deformation, arrester, and
substation equipment. Furthermore, (Malcolm & Aggarwal, 2014) proposed an
analytical method to reduce the fault due to BFO on transmission line double circuit
with two voltage levels of 69/138 kV in the event of direct lightning strikes on OHGW.
EMTP-ATP software uses to simulate the insulator string flashover and the impulse of
tower footing resistances and tower surge impedance.

A solution effort to reduce lightning faults from past to present has produced
many new methods. For instance, (Mobarakei, Sami, & Porkar, 2012) analyzed the
BFO mechanism and the high-efficiency parameter introduced, and these efficiency
parameters are determined to decrease back flashover rated. There are also simulations
involving the protection performance of a gas-insulated substation regarding the BFO
and evaluating the breakdown voltage and the effectiveness of LA. The proposed
methods to reduce BFO has suggested in the literature are as follows:

 The reduction of tower footing resistance is the most influential

parameter on back flashover rate. The discharge velocity of electrical
charges to the earth varies by increase or decrease of impedance.
 Use phase conductor arrangement for multi-circuit lines. In
horizontally-phase double-circuits of OHTL, the lower phase is usually
the most likely phase to flashover. When this phenomenon appears, it
decreases the BFO rate for an upper circuit.

Additionally, the LA analysis for OHTL due to BFO and shielding failure is
another condition considered; this approach focuses on estimating the arrester energy
of the BFO phenomenon due to shielding failure by using the PSCAD/EMTDC
program. The model uses to simulate the event of the discharged energy of LA. In this
case, the engineers are interested in LA only and behavior of the LA such as strokes to
 8

tower or OHGW, strokes to the phase conductors, and the effect of footing resistance
(Nor Hidayah Nor, Bakar, Mokhlis, & Illias, 2012).

The design for improving the transmission line performance due to lightning
overvoltage (Franc, Filipovic-Grcic, & Milardic, 2016) has presented a combination of
LA with different rated voltages for use in 110kV air-insulated substations connected
with HPP. It is a model on EMTP-ATP software.

Three different overvoltage protection schemes are:

 The first protection case determines LA with different rated voltage at

96 kV installed in the substations and installed on phase conductor at
144 kV.
 Based on the IEC standard, the second protection case is determined
with the same rated voltage for the substation and line at 102 kV.
 The third case determines LA at 102 kV installed in three towers
incoming substations at all phases: transformer bay and line bay. This
protection scheme analysis LA's rated voltage and energy sharing
between line bay and substation in multiple strokes.

From a survey of the existing, most research works are devoted to finding new
technical methods only, disregarding resultant costs of required infrastructure.
However, some researchers are interested in the economic viewpoint. For instance,
(Katic & M.S.Savic, 1998) presented an optimization of technic with concern financial
for lightning protection on overhead power distribution. This work estimated a
lightning flashover rate in both direct and indirect surges.

The performance design divides into two cases:

 First is to determine a protection scheme without OHGW.

 Second is to determine a protection scheme with OHGW.

The part of the economic analysis is the total value on the distribution line
consisting of the line building, the maintenance, energy loss, and outage costs.
 9

In the design of insulation coordination, (Williams, Vakilian, & Blackburn,

2015) have used ATP software to forecast the withstand in each transmission line
design followed by basic impulse insulation level (BIL) and a set of alternative
insulation designs using various surge protection options. This practice used insulators,
OHGW, and grounding following under constraints as required lightning withstand,
switching withstand, and insulation withstands enhancement (contamination of
insulator surface) based on discretion economic. Meanwhile, the financial calculation
has consisted of the failure costs and outage costs to be considered in selecting the most
appropriate point in the design.

 The failure costs are a summary of a BFO, shielding failure, switching

load energisation.
 The parameters for calculation of outage costs has divided as:
- The system costs comprise energy loss cost and an increase in
the expenditure of maintenance or repair is required a location of
failure.
- The cost of outage consumers depends on the impacts on utility
as straightway, indirect, economic.

In the protection design for the transmission line, (De Oliveira-De Jesus,
Hernandez-Torres, & Urdaneta, 2018) presented a way to specify the proper insulation
and the specific footing resistance of each supports for the outage rate requirement by
using EMTP software with constraints the lowest cost by optimization method. Finding
a suitable insulation design was divided into three parts: cost optimization, outage cost
of the design methodology, and expert system. The optimization and outage cost from
insulation and tower footing resistance costs. The expert system for comparison
analysis was used linear programming integer by a general algorithm.

As previously reviewed in the literature above, the methods for analyzing and
evaluating lightning protection performance and design have used computer simulation
to solve problems. The most favorite tool for analyzing the fast-front transient is the
electromagnetic transient program (EMTP) expanded in the USA. These programs have
three types: ATP, RV, and PSCAD. The EMTP instrument was adopted by the theory
 10

of horizontal electromagnetic wave propagation. However, numerical calculations of

inverse matrices require many computing resources.

The original EMTP established in the Bonneville Power Administration. Prof.

H.W. Dommel adopted the concept of an equivalent circuit consisting of a resistance
and a current source to represent any circuit element in a power system. The node input
matrix becomes a matrix. Conductivity, that is, the matrix becomes real. However, it
was not complicated. It is because the admission matrix is relatively small. Reducing
the matrix by the scattered matrix method is common in power/load flow and stability
analysis. Applied the concept of a comparative circuit consisting of an RLC and source
to indicate elements in a utility so that the node input becomes a matrix.

To manage parameters of transmission lines such as OHTL and underground

transmission lines. The method is known as Schneider-Bergeron. Subsequently, a
method for managing the frequency dependence of distributed wires is used in EMTP.
In the '80s, various electrical elements such as winders and clamps install in the EMTP.
One vital element was a subroutine called Transient Control System Analysis (TACS),
later revised and modified into a model. TACS and MODELS are a type of computer
language for managing circuits and mathematical that run attractively with EMTP core
routines to calculate a given power system's transient or dynamic behavior (Ametani,
2015).

Manitoba HVDC Research Centre has developed PSCAD/EMTDC; it consists

of two parts:

PSCAD or power systems computer aid design is a graphic user interface for
the EMTDC, analyze the results and manage data in a fully functional graphically. It
has plotting, controls, and gauge functions are included. Allows the user to modify
system parameters during simulation operations and watch the effects while the
simulation progresses. While running the simulation, users can edit parameters and see
the effects along the way. PSCAD has many objects tested in the master library. From
simple elements and control functions to more complicated models as rotating
machines, full FACTS devices. If the desired model is unavailable, PSCAD can provide
 11

a custom-modeling gateway. For instance, a custom model can be created by combining

existing models to create modules or building a basic model from scratch in a flexible
design environment (Manitoba HVDC Research Centre, 2018b).

Although the EMTDC and other EMTP-types were theoretically applying in

(Hermann W. Dommel, 1969), EMTDC was developed independently of these other
programs. Thus, this program has outstanding capabilities than other types as follows
(Manitoba HVDC Research Centre, 2018a):

 It has a GUI that is easy to understand and reduces preparation and

testing time.
 It has many series and parallel electrical elements mathematically
collapse to decrease the number of nodes and tributary.
 The optimized algorithm in this program is used to speed up the LDU
matrix dissection.
 The optimized switch sequencing algorithm guarantees fast and
efficient.
 It uses the numerical solution of a mobile grid-segregated electrical
network that is mathematically independent.
 Correction algorithms are used to perform switching. It permits this
event to arise at the instant switching. Even though this is between time
phases, it can operate in huge steps while maintaining precision results.
Additionally, a SNUBBER circuit is used to resolve numerical problems
naturally.
 It uses the Chatter removal algorithm to remove needless vibrations.
 This program does not limit circuit elements that can be integrated: any
number of components can be placed in series or parallel.
 The EMTDC device and signal source can figment impedance or non-
optimal.
 Users can quickly make models. From accessible to advanced, it has
provided a natural interface to all core program variables and storage
elements, allowing the user directly.
 12

 This program uses the new Fortran 90/95 standard, allowing dynamic
memory allocation at the start of each run.
 Starting the system with a snapshot file, this start-up technique is rapid
and implemented on massive systems. It is the only practical method for
displaying highly non-linear systems.
 Manitoba HVDC Research Centre provides full-time professional
support services for EMTDC.

Technical analysis for an estimate of the magnitude and rate of lightning

withstand on the computer simulation is used widely; however, the economic analysis
is minimal that concerning solving the problem of this field.
 13

MATERIALS AND METHODS

Materials

1. Personal Computer
2. Microsoft Window 7 Operation
3. PSCAD/EMTDC Simulation Program
4. MATLAB
5. Microsoft Office 2013
Methods

The operational steps to select the appropriate insulation coordination design

illustrates in Figure 3, which divides into three objects. The step by step follows as:

Figure 4 Procedure method for optimal insulation coordination design

 14

1) The technical objectives are to analyze the overvoltage rate and magnitude
of a lightning strike on OHTL, design of insulation coordination in other case schemes
on PSCAD. The LA, insulator string, and tower footing resistance are the main factors
to assess lightning withstand based on the BIL standard, which of this step requires
iteration until the overvoltage is lower than BIL. The parameter used in PSCAD
modeling obtains from the operation & maintenance manual, drawing, and single line
diagram.

2) The economic objectives are to calculate the overall investment cost of the
design, including equipment material and installation, transportation adopted by actual
price of projects. In addition, other cost factors are taken into consideration, such as
maintenance and energy non-supply costs, which involve opportunity and economic
loss.

3) The last objective compares the technical and economic objectives for the
design over necessary but still efficiency; this step uses to find the optimal solution in
the design, then, the multi-objective optimization (MOOP) for decision.
 15

1. Power system network

The study focuses on the effects of the BFO phenomenon, modeling the 115-
kV transmission line system of Nammang 3 - Nabong. The line has connected between
northern areas to central-1 area transmission systems, as illustrated in Figure 5. The
method comprises HPP at Nammang3 with 20 MW generators, two units, supplied
power in nominal loads about 30 MW, and hard loads about 38 MW transmit power to
Nabong substation to supply power 115 kV for large customers and 22 kV for five
feeders.

Figure 5 Single line of Nammang3 hydropower plant – Nabong substation

2. PSCAD/EMTDC modelling and simulation

2.1. Electromagnetic transient modelling for Nammang3-Nabong

The modeling for the fast front surge transmission line implemented in
PSCAD for each component adopts theory (Imece et al., 1996). The complete model
has shown in Figure 6.

- Transmission line selection models

- Tower model

- Footing resistance model

- Insulator and BFO model

- Line terminal model

 16

- Substation equipment "Autotransformer" model

- Arrester model

- Lightning stroke model

Figure 6 Completed transmission line model purposed for analysis

The primary transmission line model has been defined into six segments with
seven transmission towers and a starting point from the line terminal until winding
capacitance. The OHGW and phase conductor used in a wire module are available in
the master library, assigned from top to bottom. Which these studies assume the
lightning has struck near the line incoming substation. The lightning designs to knock
on the OHGW. The lightning model connected in the top line of the last tower incoming
substation as shown above.

2.1.1 Transmission line model

The primary method for creating OHTL in PSCAD depends on usage

type. The "Remote to Ends" is used with components of two OHTL interfaces and
transmission line configuration, showing the imparting and harvesting of the line. The
module of Remote Ends method and frequency-dependent available in the PSCAD.
 17

Figure 7 Overhead transmission line segment

Setting up a TLine configuration ensures the segment name is the

same as line interface, steady-state frequency, and the number of conductors is a similar
and terminal style set Remote Ends.

Figure 8 Transmission line configuration windows

This frequency-dependent (phase) has been adopted by theoretical in

(Morched, Gustavsen, & Tartibi, 1999). It has the most accurate transmission
configurations, including the geometry of asymmetrical lines.
 18

Figure 9 Frequency dependent (phase) model

The transmission line configuration component was assigned

parameters inside the Tline configuration module: as Figure 11. The general line
geometry data input was entered based on the tower dimension illustrated in Figure 10.

Figure 10 Tower dimension

 19

Figure 11 Geometry data input for conductor

2.1.2 Tower model

Transmission towers are represented by a simplified circuit of

resistance, inductance, and tower surge impedance, as Figure 12. These parameters are
calculated from the actual tower structure, from the formulation proposed in reference
(Ishii et al., 1991). The tower model comprises four sections divided by cross-arms.
Each unit includes a tower surge impedance connected series with an R-L parallel and
mitigation of the wave velocity.

Figure 12 Multistory of transmission line tower model

 20

The tower parameter for calculating has given from Figure 11.

0.5
ZT  60log[cot{ }] (1)
tan(R/h )
Where:

r1h 2  r2 h  r3h1
R (2)
h1  h 2

ZT Average surge impedance tower, (Ω)

R Average tower radius
r1 Upper radius, (m)
r2 Mid radius of the tower, (m)
r3 Bottom radius of the tower, (m)
h Tower altitude, (m)
h1 Altitude bottom to mid, (m)
h2 Altitude mid to top, (m)

The traveling wave is essential for frequency-dependent visualization

of the tower's wave motion mitigation and deformation. In addition, the tower travel
time calculates characteristics of surge propagation velocity assigned to equal the light
velocity.

τ T  h/c (3)

Where:

h Tower altitude, (m)

c rapidity of lighthtning, (m/µs)
τT Tower travel time, (µs)

Distance of tower structure sections (X1, X2, X3 and X4) are

represented by resistance connect parallel inductance values with connected series
connection surge impedance (A. Ametani, 2005).
 21

Xi 1
Ri  2ZT ln( ) (4)
h α

Li  2τT R i (5)

Where:
i Number of the tower structure
α Attenuation factor is 0.89

Ri Damping resistance, (Ω)

Li Damping inductance, (µH)

Table 3 Tower dimension parameters

Tower parameters Symbols Values Units

Tower height bottom-mid h1 17.8 m
Tower height mid-top h2 11.6 m
Radius top r1 1.3 m
Radius mid r2 1.8 m
Radius bottom r3 6.433 m
Distance between cross arm X1 4.8 m
X2 3.4 m
X3 3.4 m
X4 17.8 m

The steps and equations used in a model for tower calculations:

Step 1: Calculation of average surge impedance tower

From equation (1)
Find the average radius and tower altitude from (2);
h  h1  h 2  17.8  11.6  29.4 m
And:
r1h 2  r2 h  r3 h1 1.3 11.6  1.8  29.4  6.433 17.8
R   6.2077
h 29.4
From (2) and (3) assign in (1);
0.5 0.5
ZT  60log[cot{ }]  60log[cot{ }]  135.56 
tan(R/h ) tan(6.2077/29.4)
 22

Step 2: Tower travel time calculation

h 29.4
τT    0.098 μs
c 300
Step 3: Damping resistance calculation
X1 1 4.8 1
R1  2ZTln( )  2  6.2077  ln( )  5.1583 
h α 29.4 0.89

X2 1 3.4 1
R2  2ZTln( )  2  6.2077  ln( )  3.6538 
h α 29.4 0.89
X3 1 3.4 1
R3  2ZT ln( )  2  6.2077  ln( )  3.6538 
h α 29.4 0.89

X1 1 17.8 1
R4  2ZTln( )  2  6.2077  ln( )  19.129 
h α 29.4 0.89
Step 4: Damping inductance calculation

L1  2τT R1  2  0.098  5.1583  1.0110 μH

L2  2τT R 2  2  0.098  3.6538  0.71614 μH

L3  2τT R 3  2  0.098  3.6538  0.71614 μH

L4  2τT R 4  2  0.098  19.129  3.7492 μH

Table 4 Calculated result for tower dimension

Tower parameters Symbols Values Units

Tower surge impedance ZT 135.56 Ω
Traveling time T 0.098 s
Damping resistance R1 5.1583 Ω
R2 3.6538 Ω
R3 3.6538 Ω
R4 19.129 Ω
Damping inductance L1 1.011 H
L2 0.71614 H
L3 0.71614 H
L4 3.7492 H
 23

Establishing tower model assigned parameters from Table 4, the cross

arm modes by transmission line "Direct Connection method" is the corridor properties
housed within a single component function. The Direct Connection method applied
with "Bergeron Model Options" is a straightforward, constant frequency model based
on traveling waves, and the "Manual Entry Data" component can only use with the
Bergeron model. The insulator string model presents next.

Figure 13 Bergeron model and manual entry of data

Figure 14 Transmission tower model

 24

2.1.3 Footing resistance model

The grounding system uses to prevent a human from hazard and

power equipment damage from overcurrent and overvoltage. A transmission tower
equipment connects to the ground connection of adequately minor impedance. In the
process of electromagnetic analysis transient, the tower footing resistance is an essential
condition that uses analysis because the tower footing resistance can define surge
propagation and magnitude of lightning, it depends on characteristics of soil resistive;
it can be expressed in equation (7):

1 E Oρ O
IO  (6)
2π R O2

ρO Soil resistive, (Ω/m)

EO Soil ionisation gradient, (kV/m)

Ro
Rf 
I (7)
1 ( )
IO
Where:
RO Grounding resistive, (Ω)

Rf Grounding resistive depend on soil, (Ω)

I Stroke to earth, (kA)

IO Soil ionisation current limit, (kA)

Table 5 Tower footing resistance parameters

Parameter for calculated

Tower parameters Symbols Values Units
Surge current I 40 kA
Tower footing resistance RO 10 Ω
Soil resistive ρO 300 Ω/m
Soil ionisation gradient EO 300 kV/m
 25

The surge current wave characteristic at 40 kA varies tower footing

resistance at 10 Ω has been calculated and plotted on MATLAB:

Figure 15 Test of footing resistance

The resistive model for impulse grounding is represented by the

variable RLC components connect series with the ground in a master library in PSCAD.

Figure 16 Tower footing resistance model

2.1.4 Insulator and back-flashover model

The insulator is equipment exposed to atmospheric conditions, which

is vital to withstand before the voltage stresses into phase conductor, especially as the
voltage and current from lightning or the return stroke of reflecting due to high ground
resistance. The insulators string for BFO was considered and developed in the '90s by
 26

(Imece et al., 1996). A simplified insulator circuit models as Figure 17; a capacitor is a
connected parallel breaker switch with BFO control.

Figure 17 Insulator string flashover model

The capacitor has represented the insulators with a suspension

insulator typical is porcelain disc has a capacitance value is around 80 pF/unit. The
breaker switch uses for simulating an arcing horn; the state is set at normally open,
operated by a BFO module.

Figure 18 Back flashover logic diagram

The BFO control module operates under the condition of voltage

withstand of the insulator by time. Therefore, if the comparison between string voltage
is less than the line voltage, the outputs generated the close signal into the breaker
switch to occur a BFO event.

As previously stated, a simplified model of insulator sting and circuit

breaker as Figure 17 and the BFO control modules as Figure 18 use passive elements
and block control diagrams.
 27

Figure 19 Insulator string model

The BFO control module has four inputs signal include (Vstring),
(Vline), and insulator length by meter and flashover (Fin). Vstring and Vline given from
the voltmeter arc-horn gap given from length of number disc insulators and phase
flashover is integer constant value.

Vstring and Vline calculation are based on 1/t 0.75 trigger output
voltage across the string by the range and the 1/t 0.75 as input parameters. The
comparator on the right compares the trigger voltage across and string value and
provides the output with a pulse. The AND component collects output value "1" by
pulse and phase flashover (Fin) value "1", which the output "BRK" formulate shut of
the breaker switch makes an outward induce across the insulator string which imitates
the BFO trigger across the between (ACt) to (ACb). The integer constant value "0" is
needed to reset the breaker switch to open.

Figure 20 Back flashover control model

 28

The voltage withstands efficacy of the insulator varies by time. The

volt-time characteristic represents a distance of flashover determined (8).

710L
Vv  t  400L  (8)
t 0.75
Where:
Vv  t Voltage across insulator, (kV)
L Arc horn gap, (m)
t Time, (s)

The insulator string design of 115-kV for the simulation increases the
number of disc insulators from 8 to 11, represented by length 1.168 m to 1.606 m. The
voltage-time curve between old and new designs has been calculated and plotted on
MATLAB:

Figure 21 Voltage-time of insulator string with 1.168 m and 1.606 m

 29

2.1.5 Line terminal model

The line terminal is the start point of the model connected to the line's
ground and ground. The ground impedances are established in the line terminal module
connected with the transmission line configuration component on the top left side, as
shown in Figure 6. OHGW assigned grounded through 426 Ω equivalent impedances
for two top interfaces, and the rest phase conductors are 354 Ω at all six interfaces.

Figure 22 Line impedance terminal

2.1.6 Substation equipment "Autotransformer" model

The last point of the model is similar to the line terminal model,
determined by the phase conductors connected with the autotransformer. An
autotransformer is indicated by its capacitances to the ground, assigned ground through
3,500 pF equivalent capacitances (Imece et al., 1996).

Figure 23 Winding capacitance terminal

 30

2.1.7 Line surge arrester model

A surge arrester is an alternative to protect the conductors against the

direct incidence of lightning strokes. Non-linear resistive circuits of A0 and A1 indicated
the V-I characteristic by two portions R-L. The two non-linear resistive are connected
parallel. R-L act as a filter for the fast-front surge. Non-linear A0 and A1 specify in
Table 6.

L0 L1

R0 R1
C A0 A1

Figure 24 Lightning arrester circuit

The inductance "L0" (µH) represented the inductance associated with

magnetic fields, resistance "R0" (Ω) used to stabilize, capacitance "C" (pF) represented
the positive to negative. "L1" (µH) and "R1" (Ω) act to filter A0 and A1. The formula for
calculating parameters in the model gives below:

L0  0.2d/n (9)

R 0  100d/n (10)

L1  15d/n (11)
R1  65d/n (12)
C  100n/d (13)

Where:

d dimension of LA "obtained from catalogue data", (m)

n Column of LA parallel
 31

Figure 25 IR graph for A0 and A1

The non-linear resistor is given from the graph in Figure 25, it use to
specify a parameter crossover between a voltage and current. This value is then
multiplied by (V10/1.6) for calculating a discharge voltage in A0 and A1.

The equation for A0 and A1:

Discharge kV = [V-I character for A0(i)]×(V10/1.6) (14)

Discharge kV = [V-I character for A1(i)]×(V10/1.6) (15)

The steps and equations used in a model for surge arrester calculations are as follows:

Surge arrester use in this work has selected PEXLIM R-Y, line surge
arrester rated voltage 96 kV, discharge voltage 249 kV, and d = 1.216 m [Appendix -
section 4].

Step 1: Calculate the RLC values for frequency-dependent arrester circuit.

L0  0.2d/n  (0.2  1.216)/1  0.2432 μH

R 0  100d/n  (100 1.216)/1  121.60 Ω

L1  15d/n  (15 1.216)/1  18.24 μH

R1  65d/n  (65 1.216)/1  79.04 Ω

C  100n/d  (100 1)/1.216  82.237 pF
 32

Step 2: Calculate the arrester voltages rated A0 and A1 .

Arrester voltage rating = 249/1.6 = 155.62 kV

Defining of V-I characteristic for A0, A1 from relative IR graph and equation (14), (15)

For A0:
Discharge kV = 1.10×155.62 = 171.19 kV

For A1:
Discharge kV = 0.72×155.62 = 112.05 kV

Table 6 Initial V-I values for A0 and A1

Current V-I character for A0 V-I character for A1

(kA) (p.u), (kV) (p.u), (kV)
1e-7 1.10 171.19 0.72 112.05
1e-6 1.28 199.20 1.00 155.62
1e-5 1.33 206.98 1.08 168.08
1e-4 1.37 213.21 1.11 172.74
0.001 1.39 216.32 1.15 178.97
0.01 1.42 220.99 1.18 183.86
0.1 1.52 236.55 1.22 189.86
1 1.65 256.78 1.32 205.43
3.816 1.75 272.34 1.40 217.88
10 1.90 295.69 1.55 241.22
100 3.80 591.38 1.95 303.47

Step 3: Assign results from step 1 and step 2 in PSCAD

Figure 26 Line surge arrester model

 33

Figure 27 Test surge waveform 8/20 µs with 10 kA

Figure 28 Test line surge arrester discharges waveform 8/20 µs at 10 kA

2.1.8 Lightning stroke model

Generally, lightning strokes comprise multiple strokes. According to

the statistics of lightning phenomena, we can suppose that the first stroke of lightning
is more violent than several subsequent strokes. Thus, lightning transient analysis it is
contemplated the first time. The characteristic of lightning amplitude is determined by
their electric current, utmost sudden value, slant, and similar front/tail times. The
maximum sudden apex amplitude is numerically correlated with the slant or time to the
apex of the amplitude waveform. The slant rises as the apex amplitude increases.
 34

Current source control waveform models the lightning stroke by bi-

exponential in equation (16) function with standard 8/20 µs and 1.2/50 µs.

i  I m (e αt  e βt ) (16)

Where:

Im Current amplitude of lightning, (A)

α Constant represent for front-wave
β Constant represent for tail-wave
t time, (s)

α and β for 8/20 s and 1.2/50 s has been determined in reference (Das, 2010)

For the front wave of 8/20 µs, the origin starts from zero to 10 % and
leads to 90 %. The duration of the tail wave is determined time between virtual origins
to the time of 50 % point.

Figure 29 Standard 8/20 µs current waveform

For the front wave of 1.2/50 µs, the origin starts from zero to 30 %
and leads to 90 %. The duration of the tail wave is determined time between virtual
origins to the time of 50 % point.
 35

Figure 30 Standard 1.2/50 µs voltage waveform

Figure 31 Module of lightning stroke

Figure 32 Test surge waveform 1.2/50 µs at 10 kA

 36

2.2 Method of simulation

2.2.1 Simulation variable settings

The complete simulation transmission line model for improving

insulation coordination test the lightning strike phenomenon at 40 kA, value for BFO
phenomena with standard waveform 1.2/50 µs. The design uses three variables that
directly affect the transmission line withstand due to BFO: tower footing resistance,
insulator and LA.

This case study has created for five types:

Case (1): This case is used to compare with other cases; therefore no
protection design in this case.

Case (2): A new insulator string is designed to increase the number

of discs from 8 to 11 discs and improve tower resistance for incoming substations at 5-
10 Ω.

Case (3): This case uses the improved tower resistance following a
case (2). More is the adding a lightning protector of 96 kV for the tower front of the
substation in all phases.

Case (4): This case is the same design of insulator in case (2) and
lightning protector of case (3) is installed.

Case (5): This case combines three variable factors: tower footing
resistance, insulator, and LA.

2.2.2 Simulation criteria

The complete simulation transmission line model carried on

transmission line lightning withstand performance based on BIL standard. To ensure
the peak of voltage is lower than the BIL values.
 37

Table 7 Basic impulse insulation levels (WADHWA, 2009)

Voltage system Impulse voltage

(kV) (kV)
23 150
34.5 200
46 250
69 350
92 450
115 550
138 650
161 750
196 900
230 1,050
287 1,300
345 1,550

2.2.3 Project simulation settings

The runtime simulation has been specifying the time settings follow:

Figure 33 Project setting for simulation

3. Economical objective

Currently, the energy investment is growing exponentially, whether expanding

or improving transmission lines. The aim is to make transmission lines efficient and
reliable to customers. The effort to solve their problem calls for high investments to
develop this sector. However, economics is an essential factor. Improving existing
transmission lines for efficiency to have a longer lifetime is also very challenging to
reduce unnecessary investment. The economic analysis has been implemented by the
method suggested in (Katic & M.S.Savic, 1998). The formulation of overall costs is the
 38

total investment cost, given in equation (17). The various expenses used in the
calculation are derived from the bill of quantities and the bidder contractor in the actual
prices EDL.

CT = CI + CM - CL (17)

Where:

CT Transmission line overall charges, (USD)

CI Transmission line investment charges, (USD)

CM Transmission line maintenance charges, (USD)

CL Energy non-supply costs, (USD)

Investment charges are used to purchase equipment for the design, including
construction and portage costs. The leading equipment used includes an insulator with
arc-horn, LA, and grounding, detail presented in Table 8.

Maintenance charges are used to repair the damaged equipment when outage,
which has been different by causes of each case. Due to differences, the costs in this
section cannot be set equally. This section's cost is determined to be the same in all
cases to simplify calculations.

Energy non-supply is an interrupted power due to failure accounted for the

number of costs it should have received. This cost in this section is a multiple of the
energy supply by an hour and price, which this cost assumes the power interruption at
one hour in nominal load at all cases same as the maintenance.

CL = Energy non-supply × Prices (18)

Where:

ENS Energy non-supply, (kWh)

The cost of electricity per kilowatt-hour for a 115-kV transmission line has used
the prices during the rainy season. It is 5.5 Cent/kWh accounted for 0.06 USD/kWh.
 39

Table 8 List and price of EDL 115 kV transmission line of financial proposal

Supply Construction
component component
(SC) (CC)
No. Description Unit Qty
Unit Total Unit Total
price price price price
(USD) (USD)
115kV Lightning arrester with
1 pcs 6 892 5,352 418 2,508
counter & support structure
String suspension insulator
2 with arcing horn & armor rods set 6 164 984 23 1,122
fpr DARKE conductor
3 5×2m earthing rod Each 3 83 249 15 294
4 5×60m C/Tape Each 3 240 720 15 765
5 Additional earthing C/Tape m 3 4 12 1 15
6 Additional earthing rod m 3 11 33 8 57
7 Insulator 70kN set 36 10.07 362.52 1.12 402.84
8 Insulator 120kN set 36 7.10 255.60 0.79 284.04

# Calculations of investment costs for each cases design:

Total investment costs = Total SC prices + Total CC prices (19)

(a) The insulator string comprises lists numbers 2, 7 and 8.

Total investment costs = sum of SC prices + sum of CC prices = 1,808.88 USD

(b) Line surge arrester is list number 1.

Total investment costs = sum of SC prices + sum of CC prices = 7,860 USD

(c) Tower footing resistance comprises lists numbers 3, 4, 5 and 6.

Total investment costs = sum of SC prices + sum of CC prices = 1,131 USD

Table 9 Calculation results of total investment costs.

Case Type design Total investment costs

(1) - -
(2) (a) and (c) 2,939.88
(3) (b) and (c) 8,991
(4) (a) and (b) 9,668.88
(5) (a), (b) and (c) 10,799.88
 40

# Calculations of line maintenance cost:

The calculation of line maintenance cost assigns to the average of installation

costs because the behavior is the same construction.

# Calculations of cost energy non-supply:

The calculations of the cost of energy non-supply assume the power interruption
in one hour in all cases based on the same rate of power supply at 30 MW.

From (18) can be calculated as follows:

ENS = 30,000 kW × 1h = 30,000 kWh

CL = 30,000 kWh x 0.06 USD/kWh = 1,800 USD

# Calculation the total cost of transmission line for each cases design:

From (18) can be calculated as follows:

Case (2):
CT = CI + CM - CL = 2,939.88 + 60.23 – 1,800 = 1,010.75 USD
Case (3):
CT = CI + CM - CL = 8,991 + 60.23 – 1,800 = 7,251.23 USD
Case (4):
CT = CI + CM - CL = 9,668.88 + 60.23 – 1,800 = 7,739.75 USD
Case (5):
CT = CI + CM - CL = 10,799.88 + 60.23 – 1,800 = 8,870.75 USD

4. Multi-objective optimisation

Optimization is critical in many applications such as engineering, business

activities, and industrial design. The goal of optimization can be anything to reduce
energy consumption and costs, to maximize profits, productivity, efficiency, and
efficiency.

The multi-objective optimization (MOOP) problem is different from the single-

objective optimization problem. There is no need for an ideal solution to reduce the
functionality of multiple objects all at once. Frequently, objectives can conflict with
 41

each other. The best parameters of some objectives did not conduct to the suitability of
others. For instance, we need the best service on vacation. Meanwhile, we wish to pay
as minimum as possible. Therefore, the best service is the first objective. The cost will
unavoidably increase; however, this contradicts other objectives (to minimize costs).
Therefore, we must choose to compromise or achieve an equilibrium objective among
these often adversative goals. If these are not possible, we have to select the setting item
that should be taken first. More importantly, we have to compare different objectives
and reconcile. This often requires reformatting, and one of the most reach to formatting
is discovering a scalar function that shows a weighted intergrade of inclination of all
goals. Such scalar functions are often called preference or utility functions. An easy
way to create a scalar function is to use a weighted sum (Yang, 2014). This method
intergrades all MOOP functions into a single compound scalar objective function using
a weighted sum.

J  w1 j1  w 2 j2  ...  w m jm (20)

Where:
w Weighting coefficients

j Objective

The MOOP for this work assign the performance is the first objective “j 1”
determined the voltage reduction from section 2. The cost is the second objective, “j2”
from section 3. Weighting coefficients are determined into three cases: high risk,
moderate risk, and low risk, such as the transmission is installed in a mountainous area
that frequency attack by lightning and another factor is the difficulty of access to
maintenance, etc. From (20), the weight of performance of 80% and cost of 20%, the
weight of performance of 50% and cost of 50%, and the weight of performance of 20%
and cost of 80% are defined as follows:

J80/20 = w1j1 + w2j2 = 0.8j1+0.2j2

J50/50 = w1j1 + w2j2 = 0.5j1+0.5j2

J20/80 = w1j1 + w2j2 = 0.2j1+0.8j2

 42

RESULTS AND DISCUSSIONS

According to simulation criteria, the BIL of the 115-kV transmission lines, the
breakdown voltage is 550 kV. Therefore, the current amplitude for simulation and
design is 40 kA with 1.2/50 µs.

Figure 34 Overvoltage without improvement a systems

Case (1): Figure 34 shows that the overvoltage at 637 kV exceeds the BIL values
of 550 kV.

Figure 35 Overvoltage with 10 Ω footing resistance and 1.606 m insulator length

Case (2): The design of a new insulator length from 1.168 m to 1.606 m and
the grounding resistive specified at 5-10 Ω can reduce overvoltage at 445 kV.
 43

Figure 36 Overvoltage with 96 kV line surge arrester and 10 Ω footing resistance

Case (3): LA was installed at all phase conductors in the first incoming tower,
while the ground resistive used is the same as the case (2). It can reduce overvoltage at
355 kV.

Figure 37 Overvoltage with 96 kV line surge arrester and 1.606 m insulator length

Case (4): This case insulator and LA was chosen as a shield against the
lightning BFO phenomenon, both of which are determined based on case (2) and case
(3). It can reduce overvoltage at 360 kV. But it high than the case (3).
 44

Figure 38 Overvoltage with combine all protection scheme

Case (5): From the combination design between ground resistive, insulator, and
LA, the voltage was reduced to 340 kV. In this case, the voltage was reduced the most.

The results of the experiments in PSCAD and the calculation of overall charges
were summarised in Table 10. In the first case, the voltage is higher than the BIL shown
in the table at 637 kV. In the second case, the voltage is reduced to 445 kV, which
decreases to 192 kV from the first case. In this case, the total price is 1,010 USD. In the
third case, the voltage is reduced to 355 kV, which decreases to 282 kV from the first
case, and the total price is 7,251 USD. However, this case's voltage is lower than the
second case, but the budget is high. In the fourth case, the voltage is reduced to 360 kV,
which decreases to 277 kV from the first case, and the total price is 7,739 USD.
However, this case's voltage and budget are higher than the third one. Finally, in the
last case, the voltage is reduced to 340 kV, which decreases to 297 kV from the first
case, and the total price is 8,870 USD. Although the voltage of this case is lower than
all cases, the total prices are higher than in other cases.

Table 10 Calculation result of lightning overvoltage and total cost

Case Voltage (kV) Total prices (USD)

(1) 637 -
(2) 445 1,010
(3) 355 7,251
(4) 360 7,739
(5) 340 8,870
 45

In order to select an optimal solution, MOOP has been used in this process
which explains in section 4. Since the object has a different scale, it is impossible to
compare. Therefore, it is necessary to convert both objectives to a similar scale between
0-1 with normalization. The weighting coefficients of objective assign as follows: (a)
high risk: w1 = 80% and w2 = 20%, (b) moderate risk: w1 = 50% and w2 = 50%, and (c)
low risk: w1 = 20% and w2 = 80%.

Table 11 Calculation result of weighting coefficients

Case j1 j2 w1j1 w2j2 J= w1j1+ w2j2

a: 0.800 a: 0.023 a: 0.823
(1) 1.000 0.114 b: 0.500 b: 0.057 b: 0.557
c: 0.200 c: 0.091 c: 0.291
a: 0.638 a: 0.163 a: 0.802
(2) 0.797 0.817 b: 0.398 b: 0.409 b: 0.808
c: 0.159 c: 0.654 c: 0.814
a: 0.647 a: 0.174 a: 0.822
(3) 0.809 0.872 b: 0.404 b: 0.436 b: 0.841
c: 0.161 c: 0.698 c: 0.860
a: 0.611 a: 0.200 a: 0.811
(4) 0.764 1.000 b: 0.382 b: 0.500 b: 0.882
c: 0.152 c: 0.800 c: 0.953
 46

CONCLUSIONS AND RECOMMENDATION

Conclusions

The insulation coordination is critical for the transmission line to ensure that the
system can withstand much energy during abnormal conditions. This transmission line
is modeled according to PSCAD/EMTDC. This program can be used to gain insights
into the system's transient behavior studied during fault conditions; many functions
make sense in the design. Due to operation and this line being connected to a large
customer, the Nammang 3 HPP and Nabong substation network is chosen to study the
operation.

A power system network model focused on the BFO behavior in modeling the
transmission lines, dimensions, calculations, lightning, and protection scheme design
of their transmission line towers have been presented by frequency-dependent phase
model. The simulation results of initial parameters from these transmission line
networks have been compared with an economic viewpoint. Ground resistive,
insulators, and LA are integrated into this study on how the impact of transient behavior
affects the operation of the protection scenario.

Cost constraints are currently being considered to optimize the utilization of an

object with many methods and techniques. The concept MOOP weight sum mono
objective was used. It is a simplified calculation and fasts because it is not complicated.

This thesis proposes the methodology for selecting the most suitable design to
help reduce transmission line maintenance costs to increase the opportunity to distribute
electricity. Furthermore, it also helps reduce the failure caused by lightning that affects
industries and commercial factors. Different risks can choose each design at the total
prices based on effectively by the lower costs. This methodology can be applied in other
transmission lines of EDL for planning and improvement of the protection.
 47

Recommendation

Technical analysis reveal a lack of data on certain factors that may impair
technical aspects' technical performance: The transmission line parameter is the actual
grounding resistance variation of towers due to the soil ionization effect of surge wave
propagation. As a result, the simulation may not be as good as it should be. In addition,
lightning statistical data such as the number of lightning thunderstorms day per year for
calculating the number of cloud to ground flash densities are required. More complete
performance analysis can be done if the data from a lightning location system is
available.

Economic analysis, the parameters of economic analysis still need more data to
make more accurate decisions for appropriate such as transmission line maintenance
costs and transmission line outage costs in each type of risk cases due to this works
assumes that all cases have the same condition. In addition, determining the type of
customers affected by lightning is might one factor that affects the costs, for instance:
an industry, commercial, electric monorail, hospital, etc., that needs more than the
residential.
LITERATURE CITED

LITERATURE CITED

A. Ametani, T. K. ( 2005). A Method of a Lightning Surge Analysis Recommended in

Japan Using EMTP. IEEE Transactions on Power Delivery, 20 ( 2 April 2005
), 867 - 875. doi:10.1109/TPWRD.2004.839183
Ametani, A. (2015). Numerical Analysis of Power System Transients and Dynamics.

Ametani, A., & Kawamura, T. (2005). A method of a lightning surge analysis

recommended in Japan using EMTP. IEEE Transactions on Power Delivery,
20(2), 867-875. doi:10.1109/TPWRD.2004.839183
Das, J. C. (2010). Transients in Electrical Systems: Analysis, Recognition, and
Mitigation: McGraw-Hill Professional.

De Oliveira-De Jesus, P. M., Hernandez-Torres, D., & Urdaneta, A. J. (2018). Cost-

Effective Optimization Model for Transmission Line Outage Rate Control Due to
Back-Flashover Phenomena. Electric Power Components and Systems, 46(16-
17), 1834-1843. doi:10.1080/15325008.2018.1525775
Franc, B., Filipovic-Grcic, B., & Milardic, V. (2016). Lightning overvoltage
performance of 110kV air-insulated substation. Electric Power Systems
Research, 138, 78-84. doi:https://doi.org/10.1016/j.epsr.2015.12.002

Hermann W. Dommel. (1969). Digital Computer Solution of Electromagnetic

Transients in Single and Multiphase Networks. IEEE TRANSACTIONS ON
POWER APPARATUS AND SYSTEMS,, VOL. PAS-88,(NO. 4).
doi:10.1109/TPAS.1969.292459

Imece, A. F., Durbak, D. W., Elahi, H., Kolluri, S., Lux, A., Mader, D., . . . Tarasiewicz,
E. (1996). Modeling guidelines for fast front transients. IEEE Transactions on
Power Delivery, 11(1), 493-506. doi:10.1109/61.484134

Ishii, M., Kawamura, T., Kouno, T., Ohsaki, E., Shiokawa, K., Murotani, K., & Higuchi,
T. (1991). Multistory transmission tower model for lightning surge analysis.
IEEE Transactions on Power Delivery, 6(3), 1327-1335.
doi:10.1109/61.85882

Katic, N. A., & M.S.Savic. (1998). Technical and Economical Optimisation of

Overhead Power Distribution Line Lightning Protection. IEE Proceedings -
 49

Generation, Transmission and Distribution, 145(3), 239 - 244.

doi:10.1049/ip-gtd:19981897

Malcolm, N., & Aggarwal, R. K. (2014, 31 March-3 April 2014). An analysis of

reducing back flashover faults with surge arresters on 69/138 kV double circuit
transmission lines due to direct lightning strikes on the shield wires. Paper
presented at the 12th IET International Conference on Developments in Power
System Protection (DPSP 2014).

Manitoba HVDC Research Centre, a. d. o. M. H. I. L. (2018a). EMTDC USER'S GUIDE

V4_6_0.

Manitoba HVDC Research Centre, a. d. o. M. H. I. L. ( 2018b). PSCAD USER'S GUIDE

V4_6_0.

Mershon, R. D. (1903). The Grounded Wire as a Protection against Lightning.

Transactions of the American Institute of Electrical Engineers, XXII, 331-336.
doi:10.1109/T-AIEE.1903.4764367

Mines, M. o. E. a. (2018). Electricity Statistic 2018. Retrieved from

Mobarakei, S. T., Sami, T., & Porkar, B. ( 2012). Back Flashover Phenomenon Analysis
in Power Transmission Substation for Insulation Coordination. 2012 11th
International Conference on Environment and Electrical Engineering.
doi:10.1109/EEEIC.2012.6221567

Morched, A., Gustavsen, B., & Tartibi, M. (1999). A universal model for accurate
calculation of electromagnetic transients on overhead lines and underground
cables. IEEE Transactions on Power Delivery, 14(3), 1032-1038.
doi:10.1109/61.772350

Neall, N. J. (1905). Performance of lightning-arresters on transmission lines.

Proceedings of the American Institute of Electrical Engineers, 24(11), 1001-
1031. doi:10.1109/PAIEE.1905.6741785
Nor Hidayah Nor, H., Bakar, A. H. A., Mokhlis, H., & Illias, H. A. ( 2012, 2-5 Dec.
2012). Analysis of arrester energy for 132kV overhead transmission line due to
back flashover and shielding failure. Paper presented at the 2012 IEEE
International Conference on Power and Energy (PECon).

Novizon, N., Seftiani, Y. M., & Ahmad, M. H. ( 2019, 22-24 July 2019). Flashover
Phenomenon on 150kV Transmission Line Due to Direct Lightning Strike on the
 50

Ground Wire. Paper presented at the 2019 16th International Conference on

Quality in Research (QIR): International Symposium on Electrical and Computer
Engineering.

WADHWA, C. L. (2009). Electrical Power Systems: New Age International (UK) Ltd.

Williams, A., Vakilian, M., & Blackburn, T. R. (2015). A Method for Optimal
Coordinated Insulation Design of Transmission Line. Australian Journal of
Electrical and Electronics Engineering, 7(3), 211-224.
doi:10.1080/1448837x.2010.11464273

Yang, X.-S. (2014). Chapter 14 - Multi-Objective Optimization. In X.-S. Yang (Ed.),

Nature-Inspired Optimization Algorithms (pp. 197-211). Oxford: Elsevier.
 APPENDIX
1. Transmission line parameters

No Description Value Unit

1. System information
1.1 Voltage 115 kV
1.2 Frequency 50 Hz
1.3 Length 28.87 Km
1.4 Number of circuit 2
2. Phase conductors
2.1 Type ACSR
2.2 Size 240 mm2
2.3 Diameter 25.65 mm
2.4 DC resistance 0.0326 Ω
2.5 Mid-span sag 5 m
3. Overhead ground wire
3.1 Type ACSR
3.2 Size 240 mm2
3.3 Diameter 25.65 mm
3.4 DC resistance 0.0326 Ω
3.5 Mid-span sag 3 m
4. Ground rod
4.1 Diameter 25.4 mm
4.2 Length 2 m
4.3 Impulse resistance 5-100 Ω
 52

2. Calculation surge grounding on MATLAB

Ro = 10;
Eo = 300;
p = 300;
t = (0:1:50)*1e-6;
a = 1.3e4;
b = 4.4e6;
k = 1.02;
Im = 40;
I = Im.*k.*(exp(-a.*t)-exp(-b.*t));
Io = (Eo*p)./(2*pi*Ro^2);
Rf = (Ro./(sqrt(1+(I./Io))));
plot(t,Rf,'r','linewidth',1.5)
set(gca,'FontSize',15)
xlabel('Time (sec)','FontSize',15)
ylabel('Footing Resistance (ohm)','FontSize',15)
legend('Footing resistance 10 ohms')
grid on
 53

3. Calculation insulator string voltage-time on MATLAB

L=1.168;
K1=400*L;
K2=710*L;
t=(0:1:50)*1e-6;
V=K1+(K2./(t.^0.75))
t1=t;
v1=V;
L=1.606;
K1=400*L;
K2=710*L;
t=(0:1:50)*1e-6;
V=K1+(K2./(t.^0.75))
t2=t;
v2=V;
plot(t1,v1,'r','linewidth',1.5); hold on
plot(t2,v2,'b','linewidth',1.5);
set(gca,'FontSize',15)
xlabel('Time (sec)','FontSize',15);
ylabel('Voltage (Volt)','FontSize',15);
legend('Insulator length 1.168m','Insulator length 1.606m')
grid on
 54

4. Procedure for select line surge arrester of ABB

Classified of rated voltage from Table A4-1

Us = 115 kV, system earthing is effective, fault duration less than ≤ 1s.
Ur ≥ 0.74 × 115 kV
Ur ≥ 85.10 kV
From Table A4-2 author select an arrester at rated voltage 96 kV

Table A4-1 Arrester rated voltage

Table A4-2 PEXLIM R-Y Guaranteed protective data 123 – 170 kV

 55

Table A4-3 PEXLIM R-Y Technical data for housing

Figure A4-1 Drawing technical data housing

CURRICULUM VITAE

CURRICULUM VITAE

NAME Phonesavanh Sengmanivanh

DATE OF BIRTH 17 February 1993

BIRTH PLACE Vientiane capital, Lao PDR

ADDRESS Nongbone Village, Xaysettha Dist, Vientiane, Lao PDR

EDUCATION 2016 - National University of Laos, B.Eng, (Electrical

Engineering)
WORK EXPERIENCE 2018 - Present: Electricité du Laos
2017 - 2018: EDL- Generation Public Company
PUBLICATION ECTI-CON 2021

SCHOLARSHIP Academic cooperation project between the Provincial

Electricity Authority (PEA) and Electricité Du Laos (EDL)