City Research Online

City, University of London Institutional Repository

Citation: Trainba, Mahmud (2016). Estimation of the Developed Overvoltages at the

Entrance of a HV/MV Substation. (Unpublished Doctoral thesis, City, University of London)

This is the accepted version of the paper.

This version of the publication may differ from the final published
version.

Permanent repository link: http://openaccess.city.ac.uk/16690/

Link to published version:

Copyright and reuse: City Research Online aims to make research

outputs of City, University of London available to a wider audience.
Copyright and Moral Rights remain with the author(s) and/or copyright
holders. URLs from City Research Online may be freely distributed and
linked to.

City Research Online: http://openaccess.city.ac.uk/ [email protected]

Estimation of the Developed Overvoltages
at the Entrance of a HV/MV Substation

By

MAHMUD A. TRAINBA
[email protected]

A thesis submitted to
CITY , UNIVERSITY OF LONDON
for the Degree of
DOCTOROF PHILOSOPHY

SUPERVISOR: Dr. LAMBROS EKONOMOU

[email protected]

Department of Electrical and Electronics Engineering

School of Engineering and Mathematical Sciences
City, University of London
United Kingdom

20th February 2016

 CITY , UNIVERSITY of LONDON

ABSTRACT
External overvoltages can cause several damages to a HV/MV substation, leading to
insulation breakdowns. The appropriate protection of the substation against external
overvoltages is critical, in order to ensure the efficient, reliable and safe operation of the
system. Shielding wires, mast and surge arresters are the main parts of a lightning protection
system.

The current PhD project deals with the effective lightning protection of HV/MV substations,
examining the impact of various factors on the magnitude of the developed overvoltages at
different positions of the system. The influence of the equivalent circuit model of the various
components to the calculated overvoltages is also examined. Moreover, the substation outage
rate due to lightning strokes is calculated, considering shielding failures and back-flashovers.
In addition, special issues are discussed, i.e. induced overvoltages and installation of arresters
in parallel.

In comparison with other studies, main contribution of the thesis is that does not focus only
on the role of the grounding resistance, but examines the dominant influence of other
parameters; the improvement of the lightning performance of the substation can be achieved
by the appropriate adjustment of various factors of the system. The current work highlights
also the advantages and drawbacks of each equivalent circuit model, providing a guide to other
researchers in order to select the appropriate models. As far as the risk assessment analysis is
concerned, innovation of the performed study constitutes the inclusion of the arresters‟ failure
rate to the total substation failure rate, since their possible failure results in malfunction of the
nominal operation of the system. The induced overvoltages arriving at the entrance of the
substation are also calculated, examining the role of the lightning hit position, the waveform
of the lightning current, emphasizing to the role of the installation position of the arresters.
Finally, the need for good matching of the voltage – current characteristics of the arresters is
revealed, otherwise, the expected equal sharing of the injected lightning current will not be
achieved. The current Thesis indicates that the combination of the arresters in parallel does
not influence the expected overvoltages, but has to do mainly with the absorbed energy by the
arresters.

Keywords: Lightning Overvoltages, Power Substations Models, Power Transformer, Surge Arresters,
Transmission Lines.

PhD Thesis
Page [1 of 211]
 CITY , UNIVERSITY of LONDON

ACKNOWLEDGEMENTS

First and foremost, I would like to express my sincere gratitude to my PhD

supervisor Dr. LAMBROS EKONOMOU for his enthusiastic, constant
guidance, valuable advices and encouragement in the completion of the project.

My most sincere gratitude and appreciation go to my wife for her patience,

continuous encouragement and support over the past difficult years.

I am deeply indebted to my parents, brothers and sisters who gave me

continuous support and encouragement throughout my life.

In addition, I would like to show appreciation to friends and staffs from the
university who have assist me throughout the course of my studies.

PhD Thesis
Page [2 of 211]
 CITY , UNIVERSITY of LONDON

DECLARATION

I understand that all my project work must be my own unaided work.

If I make use of material from any other source, I must clearly identify it as such

in any interviews, reports.

I understand that my reports must be written unaided in my words, apart

from any quoted material that I must identify clearly in the correct manner.

I understand the work, which I shall present for assessment, must be

work carried out by myself only during the project period.

PhD Thesis
Page [3 of 211]
 CITY , UNIVERSITY of LONDON

TABLE OF CONTENTS

No. DESCRIPTION PAGE

ABSTRACT 1
ACKNOWLEDGEMENT 2
DECLARATION 3
TABLE OF CONTENTS 4
LIST OF FIGURES 8
LIST OF TABLES 20
LIST OF SYMBOLS 21

CHAPTER 1
INTRODUCTION
1.1 Overview 25
1.2 Research Aims and objects 26
1.3 Research Contribution 27
1.4 Thesis Outlines 28
1.5 List of Authors‟ Publications 30

CHAPTER 2
LIGHTNING OVERVOLTAGES
2.1 Lightning Fundamentals 31
2.2 Lightning Parameters (Wave Shapes, Peak Values, Times etc.) 33
2.2.1 Lightning Current Waveform 33
2.2.2 Peak Current 34
2.2.3 Current Rise Time 36
2.2.4Leader Approach Angle 38
2.3 Lightning Damage 39
2.4 Ground Flash Density (GFD) – Keraunik Level (T) 39
2.5 Conclusions 41
References

PhD Thesis
Page [4 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 3
LIGHTNING PROTECTION OF SUBSTATIONS
3.1 Various Types of Substations 44
3.2 Basic Parts of Substations and Their Electrical Characteristics 46
3.2.1 Transmission Lines 46
3.2.2 Circuit Breakers 55
3.2.3Current Transformers 57
3.2.4Power Transformer 57
3.2.5Disconnect Switches 59
3.2.6Ground Switches 60
3.2.7 Cables 60
3.2.8 Busbars 61
3.3 Configuration of Substations 62
3.4 Grounding of Substations 64
3.4.1 Need for Grounding 64
3.4.2 Design of a Grounding System 65
3.5 Need for Lightning Protection of Substations 69
3.6 Development of Overvoltages in Substations 70
3.7 Design of Lightning Protection System 75
3.7.1 The fixed Angles Method 75
3.7.2 Empirical Curves Method 75
3.7.3 The Electrogeometrical Model 77
3.8 Surge Arresters 80
3.8.1The Physical Model 85
3.8.2 The IEEE Model 85
3.8.3 The Pinceti–Gianettoni Model 86
3.8.4 The Fernandez–Diaz Model 87
3.9 Conclusions 89
References

PhD Thesis
Page [5 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 4
DESCRIPTION OF THE EXAMINED SYSTEMS
4.1 Introduction 94
4.2 Topology of the Examined Substations 94
4.3 Characteristics of the Substations‟ Components 97
4.3.1 Transmission lines 97
4.3.2 Cables 99
4.3.3 Power Transformers 99
4.3.4 Surge Arresters 99
4.4 Simulation Models of the Substations‟ Components 100
4.4.1 Circuit Representation of Overhead Transmission Lines and Cables 100
4.4.2 Circuit Representation of Towers 102
4.4.3 Circuit Representation of Surge Arresters 102
4.4.4 Circuit Representation of Transformers, Busbars and Circuit Breakers 103
4.4.5 Circuit Representation of Grounding System 103
4.4.6 Circuit Representation of Insulators 104
4.5 Conclusions 105
References

CHAPTER 5
CALCULATION OF THE DEVELOPED OVERVOLTAGES
5.1 Introduction 108
5.2 Overvoltages in Case of Direct Lightning Hit 108
5.3 Overvoltages in Case of Lightning hit on the ground wire 116
5.4 Impact of the Used Models of the Components to the Obtained Results 130
5.5 Impact of the Number of the Incoming Transmission Lines 142
5.6 Calculation of Developed Overvoltages for Interconnected Substations 144
5.7 Conclusions 148
References

CHAPTER 6
LIGHTNING RISK ASSESSMENT OF SUBSTATIONS
6.1 Introduction 152
6.2 Calculation of Outage Rate due to Lightning Hits on Phase Conductors 152
6.3 Calculation of Outage Rate due to Lightning hits on ground wires 153
6.4 Failure Rate of Arresters 154
PhD Thesis
Page [6 of 211]
 CITY , UNIVERSITY of LONDON

6.5 Calculation of the Total Substation Outage Rate 155

6.6 Results 156
6.7 Conclusions 157
References

CHAPTER 7
SPECIAL ISSUES OF LIGHTNING PROTECTION OF
SUBSTATIONS
7.1 Introduction 179
7.2 Induced Overvoltages 179
7.2.1 Literature review 179
7.2.2 The LIOV code 185
7.2.3 Configuration of the system under examination 187
7.2.4 Results 187
7.3 Installation of Parallel Arresters 194
7.4 Conclusions 198
References

CHAPTER 8
CONCLUSIONS,CONTRIBUTION AND FUTURE WORK
8.1 Introduction 204
8.2 Conclusions 204
8.2.1Estimation of the Expected Overvoltages at Different Positions of the
204
Substation
8.2.2The influence of the used Models of the Substations Components 205
8.2.3Outage Rate of the Substation due to Lightning Strokes 206
8.2.4Induced Overvoltages 206
8.2.5Parallel Arresters 207
8.3 Contribution 208
8.4 Future Work 210

PhD Thesis
Page [7 of 211]
 CITY , UNIVERSITY of LONDON

LIST OFFIGURES

Figure No. DESCRIPTION PAGE

CHAPTER 2
LIGHTNING OVERVOLTAGES

Figure 2.1 Cloud-to-Ground Lightning Flashes 32

Figure 2.2 Development of a Downward Negative Lightning Flash 33
Figure 2.3 Lightning Current Waveform 34
Figure 2.4 Distribution Function of the Peak Value of Lightning Flash 35
Figure 2.5 Stroke Current Range Probability for Strokes to Flat Ground 36
Figure 2.6 Current Waveforms of Negative Lightning First and Subsequent Stroke 37
Figure 2.7 World Map of Isokeraunik Level 40
Figure 2.8 Lightning Imaging Sensor of Optical Transient Density 41

CHAPTER 3
LIGHTNING PROTECTION OF SUBSTATIONS
Figure 3.1 Arrangement of a Typical Substation 45
Figure 3.2 Configuration of High Voltage Transmission Lines 46
Figure 3.3 Cut-off of Steel-Reinforced Aluminum Conductor 47
Figure 3.4 Overhead Line PI Model 48
Figure 3.5 J.Marti model 49
Figure 3.6 Tower Models 50
Figure 3.7 Typical Towers of (a) 150 kV 51
Figure 3.8 Equivalent Circuit Model of a Power Transformer 58
Figure 3.9 Simplified Network Model with Wave Impedance of a High Voltage
Overhead Line and the Terminating Transformer Impedance 59
Figure 3.10 Disconnect Switch 60
Figure 3.11 Cut-off of a MV Cable 61
Figure 3.12 Basic configurations of HV/MV substations 63
Figure 3.13 Cases of Electric Shock 65
Figure 3.14 Wenner‟s Method 66

PhD Thesis
Page [8 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.15 Configuration of a Typical Substation 71

Figure 3.16 Direct Lightning Stroke on Phase-Wire of a High Voltage Transmission 72
Line
Figure 3.17 Direct Lightning Stroke on Phase-Wire of a High Voltage Transmission 73
Line
Figure 3.18 Overvoltage Distribution and Insulation Breakdown Probability 74
Figure 3.19 Fixed Angles for Shielding Wires 76
Figure 3.20 Single Lightning Mast Protecting Single Ring of Object 77
Figure 3.21 Areas Protected by Multiple Masts 77
Figure 3.22 Electrogeometrical Model: Representation of Shield Wires and Phase 79
Conductors
Figure 3.23 Principle of the Rolling Sphere 79
Figure 3.24 Multiple Shield Mast Protection for Stroke Current Is 80
Figure 3.25 Protection by Shield Wires and Masts 80
Figure 3.26 Metal Oxide Surge Arrester cut 82
Figure 3.27 Voltage – Current Characteristic of a Metal Oxide Surge Arrester 82
Figure 3.28 Power Transformer Protected by a Surge Arrester 84
Figure 3.29 Developed Overvoltages at the Terminals of the Arrester and the 84
Transformer
Figure 3.30 Physical Model 86
Figure 3.31 IEEE Model 86
Figure 3.32 Pianceti–Gianettoni Model 87
Figure 3.33 Fernandez–Diaz Model 88

CHAPTER 4
DESCRIPTION OF THE EXAMINED SYSTEMS
Figure. 4.1 Configuration1. Simple Topology and Simulation Model of a 150/20kv 95
Substation
Figure. 4.2 Configuration.2;Complex Topology Consisting of Two Interconnected 97
Transformers
Figure. 4.3 Tower of 150kV Transmission Line 98
Figure. 4.4 Grounding System of the Tower 98
Figure. 4.5 Two-port Equivalent Circuit Model of a Transmission Line 101
Figure. 4.6 High frequency model of the HV transformer side 103
Figure. 4.7 Insulation Flashover of a 150 kV Transmission Line 104

PhD Thesis
Page [9 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 5
CALCULATION OF THE DEVELOPED OVERVOLTAGES
Figure 5.1 Lightning Hit on Phase Conductor of an Overhead Transmission Line 109
Figure 5.2 Overhead Transmission Line Protected by Surge Arresters 109
Figure 5.3 Power Transformer Protected by Surge Arresters 110
Figure 5.4 The developed overvoltages at positions B (UB) and D (UD) in 111
function with the length of the cable (BD) in case of lightning hit on
phase conductor of the connected line (no arresters)
Figure 5.5 The Developed Overvoltage at Position B (UB) in function with the 114
Distance CD and the cable length (BD) in case of a Lightning hit on
the Phase Conductor of the Connected Transmission Line (case a)
Figure 5.6 The Developed Overvoltage at Position B (UB) in function with the 114
Distance CD and the cable length (BD) in case of a Lightning hit on
the Phase Conductor of the Connected Transmission Line (case a)
Figure 5.7 The Developed Overvoltage at Position B (UB) in function with the 115
Distance CD and the cable length (BD) in case of a Lightning hit on
the Phase Conductor of the Connected Transmission Line (case b)
Figure 5.8 The Developed Overvoltage at Position B (UB) in function with the 115
Distance CD and the cable length (BD) in case of a Lightning hit on
the Phase Conductor of the Connected Transmission Line (case b)
Figure 5.9 Lightning hit on the Metallic Tower of an Overhead Transmission Line 116
Figure 5.10 Installation of Surge Arresters at an Overhead Transmission Line 117
Figure 5.11 The Developed Overvoltage at Position D (UD) In function with 118
Tower Footing Resistance (Lightning hit on the Ground Wire of the
Connected Transmission Line, No Arresters Installed)
Figure 5.12 The Developed Overvoltage at Position B (UB) in function with the 119
Tower Footing Resistance (Lightning hit on the Ground Wire of the
Connected Transmission Line, No Arresters Installed)
Figure 5.13 The Developed Overvoltage at Position D (UD) in function with the 120
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=300m, case a)
Figure 5.14 The Developed Overvoltage at Position D (UD) in function with the 120
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=600m, case a)

PhD Thesis
Page [10 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.15 The Developed Overvoltage at Position D (UD) in function with the 121
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=1000m,case a)
Figure 5.16 The Developed Overvoltage at Position D (UD) in function with the 121
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=2000m,case a)
Figure 5.17 The Developed Overvoltage at Position B (UB) in function with the 122
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=300m, case a)
Figure 5.18 The Developed Overvoltage at Position B (UB) in function with the 122
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=600m, case a)
Figure 5.19 The Developed Overvoltage at Position B (UB) in function with the 123
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=1000m,case a)
Figure 5.20 The Developed Overvoltage at Position B (UB) in function with the 123
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the connected Transmission Line, BD=2000m,case a)
Figure 5.21 The Developed Overvoltage at Position D (UD) in function with the 124
Distance CD, the length of the cable (BD) and the Tower Footing
Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case a)
Figure 5.22 The Developed Overvoltage at Position D (UD) in function with the 124
Distance CD, the length of the cable (BD) and the Tower Footing
Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case a)
Figure 5.23 The Developed Overvoltage at Position D (UD) in function with the 125
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=300m, case b)
Figure 5.24 The Developed Overvoltage at Position D (UD) in function with the 125
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=600m, case b)
Figure 5.25 The Developed Overvoltage at Position D (UD) in function with the 126
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=1000m,case b)

PhD Thesis
Page [11 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.26 The Developed Overvoltage at Position D (UD) in function with the 126
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=2000m,case b)
Figure 5.27 The Developed Overvoltage at Position B (UB) in function with the 127
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=300m, case b)
Figure 5.28 Figure 5.28: The Developed Overvoltage at Position B (UB) in 127
function with the Distance CD and the Tower Footing Resistance
(Lightning hit on the Ground Wire of the Connected Transmission
Line, BD=600m, case b)
Figure 5.29 The Developed Overvoltage at Position B (UB) in function with the 128
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=1000m,case b)
Figure 5.30 The Developed Overvoltage at Position B (UB) in function with the 128
Distance CD and the Tower Footing Resistance (Lightning hit on the
Ground Wire of the Connected Transmission Line, BD=2000m,case b)
Figure 5.31 The Developed Overvoltage at Position B (UB) in function with the 129
Distance CD, the length of the cable (BD) and the Tower Footing
Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case b)
Figure 5.32 The Developed Overvoltage at Position D (UD) in function with the 129
Distance CD, the length of the cable (BD) and the Tower Footing
Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case b)
Figure 5.33 The developed Overvoltage at Position B (UB) for various Tower 133
Models
Figure 5.34 The developed overvoltage at Position D (UD) for various Tower 133
Models
Figure 5.35 The developed overvoltage at Position B (UB) for various surge 134
Arrester Models
Figure 5.36 The developed overvoltage at Position D (UD) for various Surge 135
Arrester Models
Figure 5.37 The developed Overvoltage at Position B (UB) for various Grounding 135
Resistance Models
Figure 5.38 The developed Overvoltage at Position D (UD) for various Grounding 136

PhD Thesis
Page [12 of 211]
 CITY , UNIVERSITY of LONDON

Resistance Models
Figure 5.39 The developed Overvoltage at Position B (UB) for various Insulator 137
Models
Figure 5.40 The developed Overvoltage at Position D (UD) for various Insulator 137
Models
Figure 5.41 The developed Overvoltage at Position B (UB) for various Tower 138
Models
Figure 5.42 The developed Overvoltage at Position D (UD) for various Tower 139
Models
Figure 5.43 The developed Overvoltage at Position B (UB) for various Surge 139
Arresters Models
Figure 5.44 The developed Overvoltage at Position B (UB) for various Surge 140
Arresters models
Figure 5.45 The developed Overvoltage at Position B (UB) for various Grounding 140
Resistance Models
Figure 5.46 The developed Overvoltage at Position D (UD) for various Grounding 141
Resistance Models
Figure 5.47 The developed Overvoltage at Position B (UB) for various Insulator 141
Models
Figure 5.48 The developed Overvoltage at Position D (UD) for various Insulator 142
Models
Figure 5.49 Voltage at the Entrance of the Substation in function with the Distance 144
between the Lightning Hit Position and the Substation
Figure 5.50 Overvoltage at Position A (UA) in function with Grounding Resistance 145
for Direct (case (I)) or Indirect (case (II))
Figure 5.51 Overvoltage at Position B (UB) in function with Grounding Resistance 145
for Direct (case (I)) or Indirect (case (II))
Figure 5.52 Overvoltage at Position C (UC) in function with Grounding Resistance 146
for Direct (case (I)) or Indirect (case (II))
Figure 5.53 Overvoltage at Position D (UD) in function with Grounding Resistance 146
for Direct (case (I)) or Indirect (case (II))
Figure 5.54 Overvoltage at Position E (UE) in function with Grounding Resistance 147
for Direct (case (I)) or Indirect (case (II))
Figure 5.55 Overvoltage at Position E (UF) in function with Grounding Resistance 147
for Direct (case (I)) or Indirect (case (II))

PhD Thesis
Page [13 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 6

LIGHTNING RISK ASSESSMENT OF SUBSTATIONS

Figure 6.1 Substation Outage Rate in case of lightning hit on the phase conductor 157
of the connected line in function with the length of the cable (BD) and
the distance between the arrester and the transformer (CD) (Case a)
Figure 6.2 Substation Outage Rate in case of lightning hit on ground wire of the 157
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=300m (Case a)
Figure 6.3 Substation Outage Rate in case of lightning hit on ground wire of the 158
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=600m (Case a)
Figure 6.4 Substation Outage Rate in case of lightning hit on ground wire of the 158
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=1000m (Case a)
Figure 6.5 Substation Outage Rate in case of lightning hit on ground wire of the 159
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=2000m (Case a)
Figure 6.6 Substation Outage Rate in function with the grounding resistance (R) 159
and the distance between the arrester and the transformer (CD)
considering BD=300m (Case a)
Figure 6.7 Substation Outage Rate in function with the grounding resistance (R) 160
and the distance between the arrester and the transformer (CD)
considering BD=600m (Case a)
Figure 6.8 Substation Outage Rate in function with the grounding resistance (R) 160
and the distance between the arrester and the transformer (CD)
considering BD=1000m (Case a)
Figure 6.9 Substation Outage Rate in function with the grounding resistance (R) 161
and the distance between the arrester and the transformer (CD)
considering BD=2000m (Case a)
Figure 6.10 Failure Rate of the arrester installed at position B in function with the 161
PhD Thesis
Page [14 of 211]
 CITY , UNIVERSITY of LONDON

Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=300m (Case a)
Figure 6.11 Failure Rate of the arrester installed at position B in function with the 162
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=600m (Case a)
Figure 6.12 Failure Rate of the arrester installed at position B in function with the 162
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=1000m (Case a)
Figure 6.13 Failure Rate of the arrester installed at position B in function with the 163
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=2000m (Case a)
Figure 6.14 Failure Rate of the arrester installed at position D in function with the 163
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=300m (Case a)
Figure 6.15 Failure Rate of the arrester installed at position D in function with the 164
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=600m (Case a)
Figure 6.16 Failure Rate of the arrester installed at position D in function with the 164
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=1000m (Case a)
Figure 6.17 Failure Rate of the arrester installed at position D in function with the 165
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=2000m (Case a)
Figure 6.18 Substation Outage Rate including Arresters Faults in function with the 165
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=300m (Case a)
Figure 6.19 Figure 6.19: Substation Outage Rate including Arresters Faults in 166
function with the Grounding Resistance and the distance between the
arrester and the transformer (CD) considering BD=600m (Case a)
Figure 6.20 Substation Outage Rate including Arresters Faults in function with the 166
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=1000m (Case a)
Figure 6.21 Substation Outage Rate including Arresters Faults in function with the 167
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=2000m (Case a)

PhD Thesis
Page [15 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.22 Comparison Substation Outage Rate including and not including 167
Arresters Faults in function with the Grounding Resistance and the
distance and the length of the cable (BD) considering CD=0m (Case a)
Figure 6.23 Substation Outage Rate in case of lightning hit on the phase conductor 168
of the connected line in function with the length of the cable (BD) and
the distance between the arrester and the transformer (CD) (case b)
Figure 6.24 Substation Outage Rate in case of lightning hit on ground wire of the 168
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=300m (case b)
Figure 6.25 Substation Outage Rate in case of lightning hit on ground wire of the 169
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=600m (case b)
Figure 6.26 Substation Outage Rate in case of lightning hit on ground wire of the 169
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=1000m (case b)
Figure 6.27 Substation Outage Rate in case of lightning hit on ground wire of the 170
connected line in function with the grounding resistance (R) and the
distance between the arrester and the transformer (CD) considering
BD=2000m (case b)
Figure 6.28 Substation Outage Rate in function with the grounding resistance (R) 170
and the distance between the arrester and the transformer (CD)
considering BD=300m (case b)
Figure 6.29 Substation Outage Rate in function with the grounding resistance (R) 171
and the distance between the arrester and the transformer (CD)
considering BD=600m (case b)
Figure 6.30 Substation Outage Rate in function with the grounding resistance (R) 171
and the distance between the arrester and the transformer (CD)
considering BD=1000m (case b)
Figure 6.31 Substation Outage Rate in function with the grounding resistance (R) 172
and the distance between the arrester and the transformer (CD)
considering BD=2000m (case b)
Figure 6.32 Failure Rate of the arrester installed at position D in function with the 172

PhD Thesis
Page [16 of 211]
 CITY , UNIVERSITY of LONDON

Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=300m (case b)
Figure 6.33 Failure Rate of the arrester installed at position D in function with the 173
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=600m (case b)
Figure 6.34 Failure Rate of the arrester installed at position D in function with the 173
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=1000m (case b)
Figure 6.35 Failure Rate of the arrester installed at position D in function with the 174
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=2000m (case b)
Figure 6.36 Substation Outage Rate including Arresters Faults in function with the 174
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=300m (case b)
Figure 6.37 Substation Outage Rate including Arresters Faults in function with the 175
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=600m (case b)
Figure 6.38 Substation Outage Rate including Arresters Faults in function with the 175
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=1000m (case b)
Figure 6.39 Substation Outage Rate including Arresters Faults in function with the 176
Grounding Resistance and the distance between the arrester and the
transformer (CD) considering BD=2000m (case b)
Figure 6.40 Comparison Substation Outage Rate including and not including 176
Arresters Faults in function with the Grounding Resistance and the
distance and the length of the cable (BD) considering CD=0m (case b)

CHAPTER 7
SPECIAL ISSUES OF LIGHTNING PROTECTION OF
SUBSTATIONS
Figure 7.1 Geometry for the Calculation of Overvoltages Induced on an 183
Overhead Power Line by an Indirect Lightning Return Stroke.
Figure 7.2 Induced Overvoltage for various distances between the Measurement 184
Point and the Lightning Hit Position
Figure 7.3 Number of Induced Voltages with Magnitude exceeding the BIL in 184

PhD Thesis
Page [17 of 211]
 CITY , UNIVERSITY of LONDON

function with the BIL for a MV line

Figure 7.4 Figure 7.4 Configuration of the System under Examination 187
Figure 7.5 Induced Overvoltages at the position B for lightning hit on position A 188
for various current waveforms (case a)
Figure 7.6 Induced Overvoltages at the position D for lightning hit on position A 188
for various current waveforms (case a)
Figure 7.7 Induced Overvoltages at the position B for lightning hit on position K 189
for various current waveforms (case a)
Figure 7.8 Induced Overvoltages at the position D for lightning hit on position K 189
for various current waveforms (case a)
Figure 7.9 Induced Overvoltages at the position B for lightning hit on position L 190
for various current waveforms (case a)
Figure 7.10 Induced Overvoltages at the position D for lightning hit on position L 190
for various current waveforms (case a)
Figures 7.11 Induced Overvoltages at the position B for lightning hit on position A 191
for various current waveforms (case b)
Figure 7.12 Induced Overvoltages at the position D for lightning hit on position A 191
for various current waveforms (case b)
Figure 7.13 Induced Overvoltages at the position B for lightning hit on position K 192
for various current waveforms (case b)
Figure 7.14 Induced Overvoltages at the position D for lightning hit on position K 192
for various current waveforms (case b)
Figure 7.15 Induced Overvoltages at the position B for lightning hit on position L 193
for various current waveforms (case b)
Figure 7.16 Induced Overvoltages at the position D for lightning hit on position L 193
for various current waveforms (case b)
Figure 7.17 Voltages – Current Characteristics of Two Different Metal Oxide 195
Gapless Surge Arresters
Figure 7.18 Two Arresters at the Entrance of a Substation in Paralleled for each 195
Phase
Figure 7.19 Voltage – Current Characteristic of two Arresters in Parallel (Case 1) 196
Figure 7.20 Voltage – Current Characteristic of two Arresters in Parallel (Case 2) 196
Figure 7.21 Voltage – Current Characteristic of two Arresters in Parallel (Case 3) 197
Figure 7.22 Sharing of the Lightning Current (Cases 1, 2, 3) 197
Figure 7.23 The developed Overvoltages at Position B and D of the examined 198

PhD Thesis
Page [18 of 211]
 CITY , UNIVERSITY of LONDON

Substation (Cases 1, 2, 3)
Figure 7.24 Substation Outage Rate and Arresters Failure Rate (Cases 1, 2,3) 198

PhD Thesis
Page [19 of 211]
 CITY , UNIVERSITY of LONDON

LIST OF TABLES

NO. DESCRIPTION PAGE

CHAPTER 2
LIGHTNING OVERVOLTAGES
Table 2.1 Mean Value and Standard Deviation of the Peak Value and Steepness of 38
Various Lightning Flashes
CHAPTER 3
LIGHTNING PROTECTION OF SUBSTATIONS
Table 3.1 Tower Surge Impedance for Single Vertical Lossless Line Models 52
Table 3.2 Current Dependent Tower Grounding Resistance Models 54
Table 3.3 Leader Progression Models 55
Table 3.4 Transformer Ratio for different Windings Connections 57
Table 3.5 Range of Earth Resistivity 66

CHAPTER 4
DESCRIPTION OF THE EXAMINED SYSTEMS
Table 4.1 Electrical Characteristics of Power Transformer 99
Table 4.2 Electrical Characteristics of Surge Arresters 100

CHAPTER 5
CALCULATION OF THE DEVELOPED OVERVOLTAGES
Table 5.1 Used Tower Models for the Sensitivity Analysis 131
Table 5.2 Used Surge Arresters Models for the Sensitivity Analysis 131
Table 5.3 Used Grounding Resistance Models for the Sensitivity Analysis 131
Table 5.4 Used Insulator Models for the Sensitivity Analysis 132

PhD Thesis
Page [20 of 211]
 CITY , UNIVERSITY of LONDON

LIST OF SYMBOLS

SYMBOL DESCRIPTION

P(I) Probability that the peak current in any stroke will exceed I
I Specified crest current of the stroke in kiloamperes
i Peak value of the lightning current
φ Angle between the conductor and the vector of the magnetic field.
GFD Ground Flash Density (GFD) – Keraunik Level (T)
Ng Number of flashes to earth per square kilometer per year
T Average annual keraunic level, thunderstorm days
h Tower height,
vT Surge propagation velocity equal to the speed of light.
Ig Limiting current to initiate sufficient soil ionization (kA).
ρ Soil resistivity (Ωm).
r Equivalent radius of the tower footing (m).
n Number of grounding electrodes.
s Characteristic distance (m).
VFO Flashover strength (kV).
D Insulator string length (m).
t Elapsed time after lightning stroke (μs).
V Instantaneous voltage across the insulator string (kV).

*
600 (kV/m) and 1.3 k (m2.kV-2.s-1) , respectively, for cap and pin
E0
insulators.
**
E0 Value according to insulator type and string length.
Etouch Touch voltage (V).
Factor depended on resistivity ρs of the earth beneath the surface
Cs
material in (V-m) .
ρs Resistivity of the earth beneath the surface material (Ωm)
ρs Resistivity of the earth beneath the surface material in (Ωm).
hs Thickness of the surface material in (m).
ts Duration of shock current in (sec).

PhD Thesis
Page [21 of 211]
 CITY , UNIVERSITY of LONDON

Rg Substation ground resistance in (Ω).

A Area occupied by the ground grid in (m2).
LT Total buried length of conductors in (m).
h Depth of the grid in (m).
R1 Ground resistance of grid conductors in (Ω).
R2 Ground resistance of all ground rods in (Ω).
Rm Mutual ground resistance between the group of grid conductors, R1,
and group of ground rods, R2 in (Ω).
Lc Total length of all connected grid conductors in (m).
a' Conductors buried at depth h (m).
a' For Conductor on earth surface (m).
2a Diameter of conductor in (m).
A Area covered by conductors in (m2).
k1, k2 Coefficients
Lr Length of each rod in (m).
2b Diameter of rod in (m).
nR Number of rods placed in area (A).
W and L Dimensions of the substation (width and length)
N Number of flashes to earth per square kilometer per year
A and b Constants dependent on the termination point
I Prospective stroke current (kA).
DC Shielding failure exposure distance.
D Vertical distance between points C and A
I Current through the arrester.
V Voltage across the arrester.
α Non-linearity exponent (measure of non-linearity).
k Constant depended on the arresters type.
c Propagation velocity of the voltage wave.
S Front Steepness of the lightning overvoltage.
LIWV Standard lightning impulse withstand voltage of the device to be
protected (kV).
Uπ Lightning impulse protective level of the arrester (kV)
d Length of arrester column in meters.
n Number of parallel columns of metal oxide disks.
Vn Arrester‟s rated voltage,

PhD Thesis
Page [22 of 211]
 CITY , UNIVERSITY of LONDON

Vr(8/20) Residual voltage for a 8/20 10 kA lightning current.

Vr(1/T2) Residual voltage for a 1/T2 10 kA lightning current.
n Scale factor
L΄1 Percentage increase of the residual voltage.
Vr(8/20) Residual voltage for a 8/20 lightning current.
Vr(1/T2) Residual voltage for a 1/T2 lightning current with the nominal
amplitude.
l Per unit length inductance of the line.
c Per unit length capacitance of the line.
τ Transport delay of the voltage wave.
v Propagation velocity of the wave.
d Length of the line.
r The per unit length resistance of the line.
r Tower base radius (m).
a Winding ratio
d Length of the cable.
C Propagation speed.
i Lightning current.
L Inductance of the connection wires of the arrester.
Rs Grounding resistance of the arrester.
Utrasformer Voltage at the terminals of the transformer.
Uarrester Residual voltage of the arrester.
U Residual voltage of the arrester
E Peak value of the incoming voltage surge.
n Number of the incoming lines.
Vmax Overvoltage at the lightning hit position.
l Distance between the lightning hit position and the substation.
k Empirical coefficient between 0.0001 and 0.0004 (kV km)-1
SFRX Shielding failure rate of the connected line (shielding failures/year) .
XSF Distance between the substation entrance and the lightning hit position
(km).
SFR Shielding failure rate of the incoming overhead line(shielding
failures/100km/year) .
F(I) Probability of lightning crest current being greater than I.
ICSF Minimum value of the lightning current hitting the phase conductor of

PhD Thesis
Page [23 of 211]
 CITY , UNIVERSITY of LONDON

the connected transmission line at a distance XSF from the substation

entrance (kA) .
IMSF Maximum shielding failure current of the connected line.
NSX Rate of lightning strikes to shield wires of the connected line
(strikes/year).
XBF Limit distance between the substation entrance and the lightning hit
position(km) .
u(t) Residual voltage of the arrester (kV)
i(t) Value of the discharge current through the arrester (kA).
ArrFR Probability that an arrester fails due to lightning stroke on a phase
conductor.
ArrFR Probability that an arrester fails due to lightning stroke on the overhead
ground wire.
IA(Tt) Minimum stroke peak current in kA required to damage the arrester,
when lightning hits on a phase conductor, depending on each time-to-
half value (Tt).
IB(Tt) Minimum stroke peak current in kA required to damage the arrester,
when lightning hits on the overhead ground wire, depending on each
time-to-half value (Tt).
f(IP) Probability density function of the lightning current peak value.
g(Tt) Probability density function of the time-to-half value of the lightning
current.
T Rise time of the incident waveform
PT Total failure probability of an arrester.
AT Arrester total failure rate in failures per year per line.
NL Number of lightning flashes to a line per 100km per year
ht Tower height in m, g is the horizontal spacing in m, between the
ground wires.
g The horizontal spacing between the ground wires (m).
Ng Ground flash density in flashes per km2per year and Lis the line length
(km).

PhD Thesis
Page [24 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 1
INTRODUCTION

1.1 Overview

The protection of High-voltage/Medium-voltage (HV/MV) substations against atmospheric

overvoltages and the improvement of their lightning performance are technical issues of great
importance, since they are related to safe, uninterrupted and high quality power supply.
HV/MV substations are critical part of the electrical power systems, that step down the HV
level to a MV level of the distribution system. Power transformers, incoming overhead
transmission lines, cables, circuit breakers, current transformers, disconnect switches, ground
switches constitute the basic parts of a typical substation. External overvoltages can cause
several damages to a substation, leading to insulation breakdowns; insulators, switching
devices, cables and transformers are intensively stressed by lightning phenomena and their
failure creates several serious malfunctions, interruptions and dangers. Moreover, the safety of
the personnel must be considered in priority to avoid accidents. In addition, lightning surges
may also cause dangerous electromagnetic interference problems to low voltage systems and
especially to electronic devices. Hence, the study of the lightning repercussions and the design
of an appropriate protection system against lighting is a crucial issue, since substations are
complex installations of high investment cost.
The adoption of the appropriated protection measures according to the basic guidelines of the
International Standards can contribute to the limitation of the developed overvoltages and the
reduction of the expected lightning faults. The design of the lightning protection system has to
take into consideration the stochastic nature of the external overvoltages phenomena and the
various technoeconomical factors and restrictions of a substation. Added to this, the
installation or not of surge arresters at various positions of the substations has a decisive
impact on the magnitude of the developed overvoltages and, consequently, the expected
outage rate. The protection of substations against the deleterious effects of lightning may be
achieved by using highest insulation levels, taking into account the financial cost, or by
installing overhead ground wires in order to intercept the lightning flashes. The
implementation of metal oxide gapless surge arresters can also contribute to the improvement
of the lightning performance of the installation, especially in regions with high soil resistivity.
The position of the lightning hit, the geometrical characteristics of the external lightning
PhD Thesis
Page [25 of 211]
 CITY , UNIVERSITY of LONDON

protection system, the grounding resistance, the basic insulation level are parameters that
affect the intensity of the lightning impact.
Moreover, the incoming surges from the connected overhead transmission lines have to be
considered. Considering that an external lightning protection system (overhead ground wires)
combined with the achievement of low values of the grounding resistance of the substation
offer an adequate protection against direct lightning hits, the incoming surges consist the main
danger for the insulation of the installation. For this reason, the tower footing resistance has a
key role for the lightning performance of the substation. Added to this, the length of the cable
and the appropriate placement of the arresters have a decisive impact on the magnitude of the
developed overvoltages and, consequently, the expected failure rate.

1.2 Research Aims and Objectives

The current PhD project deals with the effective lightning protection of HV/MV substations,
examining the impact of various factors on the magnitude of the developed overvoltages at
different positions of the system. The substation outage rate due to lightning strokes is also
calculated, considering shielding failures and back flashovers. Moreover, special issues are
discussed, i.e. induced overvoltages and installation of parallel arresters. In details, the basic
objectives of the presented work are summarized as following:
• Calculation of the developed overvoltages at different positions of the substation:
the expected overvoltages at different positions of the examined topologies are computed
and a sensitivity analysis will be performed, considering the role of the grounding
resistance, the length of the cable, the installation position of the arresters, the number of
the incoming transmission lines. The presented analysis concerns lightning hits on the
phase conductors and on the overhead ground wires of the connected transmission lines,
considering that the external LPS of the substation protects adequately the substation
against direct lightning hits. Compared with other similar studies, the current work reveals
the important role of the length of the cable and examines different scenarios for the
placement of the arresters. Scope of the thesis is to highlight that the lightning
performance of the substations depends on various factors, except from grounding
resistance.
• Examination of the role of the used equivalent circuit models of the substations
components: A sensitivity analysis is performed, in order to examine the influence of the
equivalent circuit model of the various components (i.e. tower, arresters, grounding and
insulators) to the calculated overvoltages. The international research literature provides
several models for the representation of the substations‟ equipment. The current work will
PhD Thesis
Page [26 of 211]
 CITY , UNIVERSITY of LONDON

examine the appropriateness and the effectiveness of the various models, examining their
impact on the simulation results. The expected outcomes are expected to become a guide
for other researchers, in order to select the appropriate models.
• Calculation of the outage rate of the substation due to lightning strokes: the outage
rate of the substation under study is computed, considering the role of the grounding
resistance, the length of the cable and the installed arresters. Novelty of the current study
is the inclusion of the failure rate of the installed metal oxide surge arresters to the total
substation outage rate, since the protective devices are part of the equipment of the system
and their damage consequences repair of replacement costs and malfunction of the
system.
• Calculation of the induced overvoltages: the current work examines the expected
induced overvoltages at the entrance of the substation under study, highlighting the
impact of various factors and parameters on the magnitude of the computed surges. The
magnitude and the consequences of the developed overvoltages is widely examined in the
literature for the transmission and distribution lines, but the theoretical and field
experience for HV/MV substations is limited.
• Examination of the effectiveness of arresters installed in parallel: In the current
work, two arresters for each phase at the entrance of a substation are combined in parallel
and the expected current that will pass through each arrester is computed, considering
three cases, depending on the difference between the voltage – current characteristic of
the arresters. Scope of this work is to indicate that the parallel installation of surge
arresters improves the lightning performance of the system and reduces the energy stress
of the arresters, only if specific requirements are satisfied.

1.3 Research Contribution

The contribution of the PhD project is described as following:
• The current work examines the lightning performance of substations, considering the
influence of various parameters (not only the grounding resistance, which is the most
common factor presented in the international research bibliography). The performed
study highlights the role of the installation position of the arresters, the length of the cable
that connects the incoming overhead transmission line and the transformer and the
number of the connected lines, indicating that low values of grounding resistance cannot
always ensure the adequate protection of the substation against lightning. In comparison
with other studies, main contribution of the thesis is that does not focus only on the role

PhD Thesis
Page [27 of 211]
 CITY , UNIVERSITY of LONDON

of the grounding resistance, but examines the dominant influence of other parameters; the
improvement of the lightning performance of the substation can be achieved by the
appropriate adjustment of various factors of the system.
• The obtained simulation results show that the impact of the used models for each
component of the substation is not significant, since cannot change dramatically the
estimation of the arising voltage surges. The current work highlights the advantages and
drawbacks of each equivalent circuit model, providing a guide to other researchers in
order to select the appropriate models.
• The calculation of the substation outage rate due to lightning strokes is presented,
considering the role of the grounding resistance, the protection distance of the arresters
and the configuration of the substations. Innovation of the performed study constitutes
the inclusion of the arresters failure rate to the total substation failure rate, since arresters
are part of the equipment of the substation and their possible failure results in malfunction
of the nominal operation of the system.
• The induced overvoltages arriving at the entrance of the substation are calculated,
including not only lightning hit on ground but considering different cases (hit on the mast,
lightning current characteristics etc.). A sensitivity analysis is performed, indicating that
induced overvoltages can be restrained by adopting appropriated protective measures. The
thesis examines the role of the installed arresters for the restrain of the expected induced
voltages.
• The parallel installation of surge arresters is a common practice in order to reduce the
absorbed energy by them and, consequently, their failure probability. However, the current
thesis reveals the need for good matching of the voltage – current characteristics of the
arresters; otherwise, the expected equal sharing of the injected lightning current will not be
achieved. The current thesis emphasizes to the fact that the combination of the arresters
in parallel does not influence the expected overvoltages, but has to do mainly with the
absorbed energy by the arresters.

1.4 Thesis Outlines

Scope of the current work is the study of the lightning performance of HV/MV substations
and the examination of the appropriate protective measures that should be adopted, in order
to restrain the developed overvoltages and reduce the expected failure rate. The outline of the
Thesis is described as following:
 Chapter.2 presents the basic characteristics of the lightning surges and describes the main
mechanisms of their development. The range of the lightning current, the steepness and
PhD Thesis
Page [28 of 211]
 CITY , UNIVERSITY of LONDON

the charge are factors that determine the consequent damages of a probable lightning hit.
In addition, the ground flash density and the keraunik level are defined.

 Chapter.3 deals with the main aspects of the lightning protection of substations, based on
the international Standards, the results of other researchers and the common practice.
Initially, the various types of substations are presented and their basic parts (transmission
lines, cables, switches, busbars, grounding system etc.) are analyzed. The need for lightning
protection of substation is highlighted and the mechanisms of the development of
overvoltages are analytically presented, considering lightning hit either on a phase
conductor or on overhead ground wire. The methods for the design of the lightning
protection system of a substation are also presented (fixed angles method, empirical
curves method, and electrogeometrical method). Finally, the main characteristics of metal
oxide gapless surge arresters are presented, emphasizing the role of their installation
position. Note, that for each component of the substation the most used equivalent circuit
models are presented.

 Chapter.4 presents the topologies of the examined systems and provides the
characteristics of each component of the substation.

 Chapter 5 the expected overvoltages at various positions of the substations under study
are computed, considering lightning hit on phase conductors or ground wires of the
incoming transmission lines. A sensitivity analysis is performed, in order to indicate the
impact of the grounding resistance, the installation position of the surge arresters and the
length of the MV cable. Furthermore, the influence of the applied simulation models of
the components to the obtained results is also examined.

 Chapter.6 performing the lightning risk assessment of substations and calculating the
outage rate of the substation due to direct or indirect lightning hit. The outage rate for
each case is computed according to the international Standards, examining the influence of
various parameters. The failure probability of the installed surge arresters is also included
to the extracted results.

 Chapter.7 deals with the induced overvoltages and the parallel combination of arresters.
The lightning hit position, the crest and the steepness of the injected lightning current are
the main factors that influence the developed induced overvoltages. The effectiveness of

PhD Thesis
Page [29 of 211]
 CITY , UNIVERSITY of LONDON

the combination of surge arresters is parallel is also examined, highlighting the role of the
voltage-current characteristics of each arrester.

 Chapter.8 summarizes the conclusions of the work.

 Chapter.9summarizes the contribution of the obtained results and proposes subjects for
future research.

1.5 List of Authors’ Publications

The following publications have produced based on the work presented in this Thesis:

1.5.1 Journal Publications

1. Trainba M., Ekonomou L., Estimation of the Developed Overvoltages at the Entrance of a
HV/MV Substation,IET Science Measurement & Technology, SMT-2016-0029,
(Under Review).
2. Trainba M., Ekonomou L., A Sensitivity Analysis for the more Effective Lightning Protection of
HV/MV Substations, ,Electrical Power & Energy Systems, IJEPES-2016-132, (Under
Review).

1.5.2 Conference Publications

1. Trainba M., Ekonomou L., Lightning Performance of a HV/MV Substation, 10th WSEAS
International Conference on Energy & Environment (EE '15), Budapest, Hungary, pp.
28-32, 2015.

PhD Thesis
Page [30 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 2

LIGHTNING OVERVOLTAGES

2.1 Lightning Fundamentals

Lightning phenomena are generated due to electrostatic discharges during thunderstorms and
are main cause of faults and damages in buildings and electrotechnical installations (Sabiha).
Various theories have been proposed in order to describe and explain the formation and the
separation of the developed charges in clouds and, consequently, the development of lightning
strokes. The electrical charges are separated due to air currents movements, resulting to the
development of potential difference between the regions of positive and negative charge,
which may vary between 100 and 1000MV (Abdel-Salam et al.).

Depending on the hit position of the atmospheric overvoltages, lightning strikes are
distinguished as following:
 Hits inside the cloud
 Hits between different clouds
 Hits on tall installations
 Hits that terminate on earth (IEEE Std 998-1996).

Moreover, atmospheric overvoltages can be characterized and classified by their polarization

and the direction of the propagation of the leader toward ground. Figure 2.1 depicts the cases
of lightning flashes between cloud and earth, as determined by the charge (positive or
negative) and the direction of the initial leader (cloud to ground or ground to cloud) (Uman
2008). Note that the majority of the recorded lighting flashes are negative (approximately
90%). It is worth mentioning that the rest 10% of positive strikes (i.e. positive cloud and
negative ground) are more violent.

The mechanism that describes the development of the lightning flashes includes a multi-step
process, presented in Figure 2.2 (IEEE Std 998-1996). Necessary requirement is the
development of leaders (upward and downward) due to the ionization of the air surrounding
the charge centers. Between the cloud and the earth a strong electric field is arisen and the
PhD Thesis
Page [31 of 211]
 CITY , UNIVERSITY of LONDON

dielectric strength of the air is reduced. The downward leader approaches the ground, since
the upward leader has the opposite direction from ground to cloud. When the two leaders are
joined, a lightning flash is occurred (return stroke). The return stroke, that follows the
“electric” path formed by the stepped leaders, consist the main lightning current; its current
median value is about 24kA and its velocity of propagation is 30000000m/s, i.e. 10% of the
speed of the light.

Downward flash – Negative Cloud Charge Upward flash – Positive Cloud Charge

Downward Flash – Positive Cloud Charge Upward Flash – Negative Cloud Charge

Figure 2.1: Cloud-to-Ground Lightning Flashes (Uman 2008)

PhD Thesis
Page [32 of 211]
 CITY , UNIVERSITY of LONDON

Current magnitudes associated with stepped leaders are small (in the order of 100 A) in
comparison with the final stroke current. The velocity of propagation is approximately
150000m/s, i.e. 0.05% of the light speed (Uman 2008), (IEC 62305-1 2006), (Anderson 1987).
After the first stroke, subsequent strokes can be developed along the same or in similar paths.
A positive lightning usually has only one strike. The downward negative flash is considered to
be the most important discharge process for practical engineering applications.

Figure 2.2: Development of a Downward Negative Lightning Flash (Abdel-Salam et al.)

2.2 Lightning Parameters (Wave Shapes, Peak Values, Times etc.)

2.2.1 Lightning Current Waveform

Figure 2.3 presents a typical double exponential waveform of a lightning current. The peak
current and the duration of the waveform play important role for a series of electrical
parameters.

PhD Thesis
Page [33 of 211]
 CITY , UNIVERSITY of LONDON

Figure 2.3: Lightning Current Waveform (IEEE Std 998-1996)

2.2.2 Peak Current

The peak value of the lightning current is one of the most significant parameters that
determine the developed overvoltages. The median value of strokes to overhead ground wires,
conductors, structures and masts is commonly considered to be 31 kA (Anderson 1987).
According to (Anderson 1987) the probability that a certain peak current will be exceeded in
any stroke is given as following:

PhD Thesis
Page [34 of 211]
 CITY , UNIVERSITY of LONDON

( ) ( )
( )

Where;
P(I)is the probability that the peak current in any stroke will exceed I
I is the specified crest current of the stroke in kiloamperes

In (Mousa, Srivastava 1989) is proposed that a median stroke current of 24 kA for strokes to
flat ground results to the optimum correlation with available field observations data. In this
case, the probability that a certain peak current will be exceeded in any stroke is given by the
equation:

( ) ( )
( )

Figure 2.4 is obtained by the implementation of eq.(2.1);Figure 2.5depicts the probability of

lightning flash to a flat ground will be within the ranges shown on the abscissa (IEEE Std
998-1996).

Figure 2.4: Distribution Function of the Peak Value of Lightning Flash (IEEE Std 998-1996)

PhD Thesis
Page [35 of 211]
 CITY , UNIVERSITY of LONDON

Figure 2.5: Stroke Current Range Probability for Strokes to Flat Ground (IEEE Std 998-1996)

2.2.3 Current Rise Time

The current rise time determines the induced surges. Various researchers have given
probability distributions, based on real recorded measurements. Berger‟s results in (BRITISH
STANDARD 6651 1999) give a median value of μs for the first stroke current and steeper
fronts of up to 100 kA/μs for subsequent strokes. Other researchers give 9–65 kA/μs
(median 24 kA/μs) for the first stroke, and l0–l62 kA/μs (median 40 kA/μs) for subsequent
strokes. In addition, BS6651 (Bouquegneau et al. 1986) gives (di/dt)max= 200 kA/μs.

In any case, the rate of change of current measured in (BRITISH STANDARD 6651 1999),
for the first and the subsequent strokes, is given by the equations:

(kA/μs) (First stroke) (2.3)

(kA/μs) (Subsequent stroke) (2.4)

In (BRITISH STANDARD 6651 1999)are also provided mean values for the current tail
duration and its statistical range:
 First stroke: mean 75 μs, range 30–200 μs
 Later strokes: mean 32 μs, range 6.5–140 μs.

PhD Thesis
Page [36 of 211]
 CITY , UNIVERSITY of LONDON

Figure2.6 depicts the waveforms of a negative lightning flash, including the first and the
subsequent strokes.

Figure 2.6: Current Waveforms of Negative Lightning First and Subsequent Stroke (Uman 2008)

PhD Thesis
Page [37 of 211]
 CITY , UNIVERSITY of LONDON

Table 2.1 presents the mean value and the standard deviation of the peak value and the
steepness of various lightning flashes.

Table 2.1: Mean Value and Standard Deviation of the Peak Value and Steepness of Various Lightning
Flashes (Berger et al. 1975), (CIGRE W.G. 1991), (Garbagnati et al. 1978)

Mean value Dispersion

Berger (Berger et al. 1975)
Positive Stroke I (kA) 35 1.21
Positive Stroke di/dt (kA/μs) 2.4 1.54
Negative Stroke I (kA) 30 0.53
Negative Stroke di/dt (kA/μs) 12 0.54
CIGRE WG (CIGRE W.G 1991)
Negative Stroke I (kA) 31.1 0.48
Negative Stroke di/dt (kA/μs) 24.3 0.60
Garbagnati et al (Garbagnati et al. 1978)
Negative Stroke I (kA) 33 0.25
Negative Stroke di/dt (kA/μs) 14 0.36

2.2.4 Leader Approach Angle

The lightning stroke direction from cloud to ground is considered in general in a vertical path
(Eriksson 1987), (Mousa, Srivastava 1989). However, in some cases an approach angle leader
must be considered, according to a probability distribution, given by the following equation
[Brown, whitehead1969]:

( ) { (2.5)

Where;
m=1 (Km=1/2), either
m=2 (K2=2/π) or
m= (case of vertical strokes).

PhD Thesis
Page [38 of 211]
 CITY , UNIVERSITY of LONDON

2.3 Lightning Damage

In general, the effects of lightning strikes depend mainly on the conductivity of the materials
and include electrical, thermal, electrodynamic, electromagnetic, electrochemical, acoustic and
physiological repercussions(Abdel-Salam et al. 2000). The lightning effects and the consequent
damages depend on the characteristics of the lightning current (see paragraph 2.2). The peak
current of the lightning flash and the duration of the waveform determine the thermal
consequences, which include increase of the temperature or melting of the conductors. In case
that lightning hits objects that present high resistance (e.g. trees, wooden structures etc.) the
produced thermal energy will be even higher, including also the danger of fire (IEC 62305-1
2006).Furthermore, mechanical effects can be caused by the lightning current, which are
influenced by the peak value and the duration of the lightning current as well as by the elastic
characteristics of the affected mechanical structure; in practice, both thermal and mechanical
effects occur simultaneously. Lightning affecting a substation can cause serious damages to the
insulation of the equipment, leading to potential interruption of the normal operation of the
system. The damages and failures may also extend to the surroundings of the substation and
even involve the local environment and the personnel. As far as the induced overvoltages is
concerned, their magnitude is proportional to the self-inductance multiplied by the steepness
of the lightning current(IEC 62305-1 2006).

2.4 Ground Flash Density (GFD) – Keraunik Level

Keraunic level is defined as the average annual number of thunderstorm days or hours for
a given locality. A daily keraunic level is called a thunderstorm-day and is the average number
of days per year on which thunder is heard during a day (24h). If thunder is recorded on any
one day more than one time, the day is still classified as one thunder-day (IEEE Std 998-
1996). The keraunic level (T) of various regions is given in appropriate maps (isokeraunik),
which represent by using lines the annual frequency of thunderstorm days in the world (World
Distribution of Thunderstorm Days, WMO No. 21, 1056). Figure 2.7 depicts the world map
of isokeraunik level (HUBER & SUHNER Group 2010).

PhD Thesis
Page [39 of 211]
 CITY , UNIVERSITY of LONDON

Figure 2.7: World map of Isokeraunik Level (HUBER & SUHNER Group 2010)

Ground flash density (GFD) is defined as the average number of strokes per unit area per unit
time at a particular location. The GFD to earth, a substation, or a transmission or distribution
line is, generally, roughly proportional to the keraunic level at the locality. Various researchers
have proposed different approaches for the calculation of GFD, given by the following
equations (ΙΕΕΕ No.3 1993), (IEEE Std 1243-1991 1997):
(2.6)

(2.7)

(2.8)
(2.9)
(2.10)
(2.11)
Where;
Ngis the number of flashes to earth per square kilometer per year
T is the average annual keraunic level, thunderstorm days (IEEE Std 998-1996).

The ground flash density can be directly recorded by using appropriate flash counters (such as
CIGRE counter system and CCIR counter system which are located in most of European and
Mediterranean countries (CIGRE W.G. 1991)). Otherwise, the ground flash density can be
estimated by using NASA satellite‟s observations of average optical transient density for the
geographical the region (NASA OTD Lightning 2008). Figure 2.8 presents a revised map from
the NASA/MSFC lightning imaging sensor of optical transient density after completion of
OTD mission and NSSTC analysis (Christian et al. 2003).
PhD Thesis
Page [40 of 211]
 CITY , UNIVERSITY of LONDON

Figure 2.8: Lightning Imaging Sensor of Optical Transient Density after Completion of OTD Mission and
NSSTC Analysis. (IC+CG) flashes per km2 per year (Christian et al. 2003).

2.5 Conclusions

The current chapter presents the fundamentals of lightning surges that consist the main cause
of stresses and damages of the equipment of the HV/MV substations. The knowledge of the
lightning mechanism and the characteristics of the lightning currents are necessary, in order to
obtain the appropriate protective measures. The peak current and the steepness of the
lightning current determine the expected magnitude of the overvoltage and the range of the
induced voltages. Furthermore, the duration of the lightning current waveform determines the
energy consumption of the arresters, influencing their failure probability. The ground flash
density and the keraunik level are also necessary information for lightning performance
studies.

PhD Thesis
Page [41 of 211]
 CITY , UNIVERSITY of LONDON

References

Abdel-Salam M., Anis H., El-Morshedy A., Radwan R., ―High Voltage Engineering, Theory and
Practice‖, Marcel Dekkienrc, New York,2000.
Anderson J.G., Chapter 12 of Transmission Line Reference Book 345 kV and Above, 2nd Ed. Rev.
PaloAlto, Calif.: Electric Power Research Institute, 1987.
Anderson R.B, Eriksson AJ. Lightning parameters for engineering application. Electra
No.69,CIGRE,1980,pp65-102.
Anderson R.et al., Lightning and Thunderstorm Parameters, in Lightning and Power Systems, London, IEE
Conf. Publ. n° 236, 1984
Berger K., Anderson R.B., Kroninger H., ―Parameters of lightning flashes‖, Electra, no. 41, pp. 23-37,
1975.
Bouquegneau C., Dubois M., Trekat J., Probabilistic analysis of lightning performance of high-voltage
transmission lines, Electr. PowerSyst. Res. 102 (1–2) (1986) 5–18.
BRITISH STANDARD 6651: ‗Code of practice for protection of structures against lightning‘, 1999
Chisholm W.A., Estimates of lightning ground flash density using optical transient density, IEEE PES
Trans. Distribution Conf. Exposition 3 (2003) 1068–1071.
Christian H.J., Blakeslee R.J., Boccippio D.J., Boeck W.L., Buechler D.E., Driscoll K.T., et al., Global
frequency and distribution of lightning as observed from space by the optical transient detector, J. Geophys. Res.
108 (D1) (2003) 4005–4019.
CIGRE Working Group 01 of Study Committee 33, ―Guide to procedures for estimating the lightning
performance of transmission lines", CIGRE Tech.brochures, no. 63, 1991.
Eriksson J.A., ―The incidence of lightning strokes to power lines‖, IEEE Transactions of Power Delivery,
vol. 2, no. 3, pp. 859-870, July 1987
Garbagnati E., Giudice E., Lopiparo G.B., ―Messung von Blitzstromen in Italien- Ergebnisse einer
statistischen Auswertung‖, ETZ, no. 11, pp. 664-668, 1978.
Heidler F., Zischank W., Flisowski Z., Bouquegneau Ch., Mazzetti C., ―Parameters of lightning current
given in IEC 62305 – background, experience and outlook‖, 29th International Conference on Lightning
Protection, 23-26 June 2008, Uppsala, Sweden
Hileman A. R., ―Insulation Coordination for Power Systems‖, CRC Press, 1999.
HUBER & SUHNER Group, ―Lightning Protection‖, Edition 2010.
IEC 62305-1, ―Protection against lightning – Part 1: General principles‖, 2006.
IEEE Std 1243-1991, ―Guide for Improving the Lightning Performance of Transmission Lines‖, June
1997
IEEE Std 998-1996, ―IEEE Guide for Direct Lightning Stroke Shielding of Substations‖, 1996
IEEE Working Group on Estimating the Lightning Performance of Transmission Lines, Estimating the
lightning performance of transmission lines. II. Updates to analytical models, IEEE Trans. PWRD 8 (3)
(1993) 1254–1267.
IEEE Working Group on Lightning Performance of Transmission Lines, A simplified method for estimating
lightning performance of transmission lines, IEEE Trans. PAS 104 (4) (1985) 919–927.
IΕΕΕ Working Group Report, ―Estimating Lightning Performance of Transmission Lines II – Updates to
Analytical Models‖, IEEE Transactions on Power Delivery, Vol.8, No.3, July 1993
Mousa A.M., Srivastava K.D., ―The implications of the electrogeometrical model regarding effect of height of
structure on the median amplitudes of collected lightning strikes‖, IEEE Transactions on Power Delivery, vol.
4, no. 2, pp. 1450-1460, 1989.
NASA OTD Lightning, 2008, http://thunder.nsstc.nasa.gov.
PhD Thesis
Page [42 of 211]
 CITY , UNIVERSITY of LONDON

Sabiha N.A., ―Lightning-induced overvoltages in medium voltage distribution systems and customer
experienced voltage spikes‖, Doctoral Dissertation, Aalto University, School of Science and Technology,
Faculty of Electronics, Communications and Automation, Department of Electrical Engineering.
Uman M.A., ―The Art and Science of Lightning Protection‖, Cambridge University Press, 2008
World Distribution of Thunderstorm Days, ―World Meteorological Organization‖, Geneva, Switzerland,
WMO No. 21, 1056.
Brown, Gordon W; Whitehead, Edwin R ,IEEE Transactions on Power Apparatus and Systems, 1969,
Volume PAS-88, Issue 5

PhD Thesis
Page [43 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 3

LIGHTNING PROTECTION OF SUBSTATIONS

3.1 Various Types of Substations

Electrical power systems are complex configurations that warrant the safe, economical and
reliable transmission of the electrical energy and its distribution to the consumers. In general,
electrical energy is transmitted and distributed by using ac high voltage, but in some cases
technical and economic criteria impose the use of dc high voltage. ac voltage is easier to be
transformed (changing the voltage level), since dc voltage presents lower power losses (but
conversion of ac to dc and vice versa is demanded).The various parts of a typical power
system are appropriately connected, in order to achieve high degree of continuity, maximum
reliability, and flexibility (Siemens 2008, Abdel-Salam et al.). So, basic role of the substations is
to connect the transmission lines, subtransmission feeders, generating units, and transformers
and to meet the above demands and restriction with the highest possible economy (Siemens
2008, Abdel-Salam et al.).

There are four major types of electric substations (Grigsby 2007):

 The switchyard substation: it connects the generators to the utility grid.
 The customer substation: it is the main source of electric power supply for one
particular business customer.
 The system station: it involves the transfer of bulk power across the network or
switching facilities.
 The distribution station: it provides the distribution circuits that directly supply the
customers.

The substations can also be categorized as following (Grigsby 2007, McDonald et al 2000):
 Outdoor type with air insulated substations (AIS) equipment
 Indoor type with air insulated substations (AIS) equipment
 Outdoor type with gas insulated substations (GIS) equipment
 Indoor type with gas insulated substations (GIS) equipment
 Mixed technology substations (AIS) & (GIS)
 Mobile substations
PhD Thesis
Page [44 of 211]
 CITY , UNIVERSITY of LONDON

Moreover, substations are distinguished according to the voltage level (Siemens 2008):
 Medium Voltage / High Voltage (MV/HV)
 High Voltage/High Voltage (HV/HV)
 High Voltage/ Medium Voltage (HV/MV)

Fig. 3.1 depicts the configuration of a typical substation with multiple bus-bars.

Figure 3.1: Arrangement of a Typical Substation (Siemens 2008)

Depending on the insulation material that will be used, three basic designs are possible
(Siemens 2008):

 Air-Insulated Substations (AIS): AIS are preferred for voltage levels up to 800 kV,
wherever space and environmental restrictions are not severe. Disadvantage of outdoor
AIS is their exposure to environmental effects

 Gas-Insulated Indoor or Outdoor Switchgear (GIS): They are appropriate to be

installed in urban or industrial regions, due to their compact design and their small
dimensions. For this reason, they are preferred in cases that overhead high voltage
transmission lines are not compatible.

 Mixed Technology (Compact/Hybrid Solutions): Except from the above

conventional designs, there are also compact solutions available that can be realized with
Air-Insulated and/or Gas-Insulated components (Siemens 2008).

Medium and low Voltage Switchgears are used for residential, commercial and industrial
purposes, and are out of scope of the current work.
PhD Thesis
Page [45 of 211]
 CITY , UNIVERSITY of LONDON

3.2 Basic Parts of Substations and their Electrical Characteristics

3.2.1 Transmission Lines

The electrical energy is transmitted to the substation through overhead high voltage
transmission lines. A typical transmission line consist of the tower, the phase conductors, the
insulators, the overhead ground wires and the grounding systems. Figure 3.2 presents various
configuration of high voltage lines that are used in the current analysis; it is worth mentioning,
that each type of tower results to differentiations of the applied electrogeometrical model,
since the geometry of the line changes.

ground wires

insulator

Figure 3.2: Configuration of High Voltage Transmission Lines

(a) 150kV Single Circuit, (b) 400kV Double Circuit(Christodoulou et al. 2010)

Conductors are the basic components of the overhead line, since they carry the transferred
electrical energy; for this reason, their materials and geometrical characteristics are of great
importance, in order to ensure reliable and economical transmission of the power. The most
PhD Thesis
Page [46 of 211]
 CITY , UNIVERSITY of LONDON

used conductive material is Aluminum, due to its very advanced characteristics, i.e.
advantageous price, low weight and the achievement of minimum cross-sections (high
conductivity). On the other hand, Aluminum is a very corrosive metal with limited mechanical
strength. In order to reinforce the mechanical behavior of the conductors, compound
conductors with a steel core (so-called aluminum conductor steel-reinforced (ACSR)) are used.
Figure 2.3 presents the cut-off of various types of steel-reinforced aluminum conductors.

St
Al

Figure 3.3: Cut-off of Steel-Reinforced Aluminum Conductor (McDonald, J. et al 2000)

Experience field data give an average lifetime of ACSR of years. The electrical resistance of
the conductors is mainly depended on:
 The Material
 The Cross Section
 The Length

The temperature and the skin effect should also be taken into consideration. A current density
of 0.5 to 1.0 A/mm2 based on the aluminum cross-section has proven to be an economical
solution in most cases. Furthermore, during the design of a transmission line and the selection
of the conductor‟s electrotechnical characteristics high voltage gradients at the conductors‟
surface have to be avoided, since the resulting corona effects cause visible partial discharges,
radio interferences, audible noise and energy losses. To this direction, ac voltage gradient
should not exceed 15 to 17kV/cm. The effects of the Corona phenomenon are restrained by
using multiple sub-conductors, in order to increase the equivalent cross-section of the
conductor. This aspect is important for lines with voltages of 245 kV and above. As far the
mechanical characteristics concern, the conductor has to be withstanding the environmental
conditions and stressing loads by wind and ice.

PhD Thesis
Page [47 of 211]
 CITY , UNIVERSITY of LONDON

Concerning the above characteristics of the transmission lines, the appropriate circuit
representation is a critical issue, in order to guarantee the quality and the reliability of the
lightning performance studies. To evaluate influence of various transmission lines models
application on simulation results in insulation coordination studies, PI model, and frequency-
dependent JMarti model are usually chosen to analysis.

Overhead lines models are based on PI circuit consist of linear RLC components. Typical
transmission line PI model is presented in Figure 3.4.

L R

C C

Figure 3.4: Overhead Line PI Model (Martinez & Castro-Aranda 2005)

PI (or π) model is a simplified model for transient analysis studies, appropriate mainly for
transmission lines of short length. In fact, PI model constitutes a discrete approximation of
the constant distributed parameter model. PI model uses lumped parameters, which can result
to false oscillation. The above disadvantage can be limited by connecting in parallel R-L
elements. JMarti frequency-dependent model is more appropriate for the study of lightning
performance of high voltage transmission lines, since it improves the reliability and the
accuracy of the obtained results by approximating the characteristic admittance and the
propagation constant by rational functions, in order to convert from mode domain to phase
domain. JMarti model offers higher accuracy compared with PI model. In JMarti model, both
the characteristic impedance Z as well as the propagation function are calculated using modal
characteristics (calculated in defined frequency range for a constant transformation matrix).
Thus, JMarti model computes the characteristic admittance and propagation constant by
rational functions. JMarti model uses a constant transformation matrix in order to convert
from mode domain to phase domain (although in case of modeling transmission line it does
not matter, still it can have influence for cable line case). Model requires defined frequency
PhD Thesis
Page [48 of 211]
 CITY , UNIVERSITY of LONDON

where the transformation matrix is calculated (this frequency should be dominant is the later
transient study), and a steady state frequency for calculation of the steady state condition. The
JMarti model needs in some cases modification of the default fitting data.

Fig 3.5 depicts the basic representation of each end of the multi-phase J.Marti model
(Dommel et al. 1986, IEEE Fast Front Transients Task Force 1996, Marti 1982).

Figure 3.5: J.Marti model (IEEE Fast Front Transients Task Force 1996)

The circuit representation of line towers is a more complex procedure in comparison to the
other parts of the line. The tower is divided in various segments, which are represented by
vertical lossless single-phase frequency-independent distributed parameter line in combination
with lumped circuit elements (IEEE W. G. P1243 1996, Hara & Yamamoto 1996, Ametani et
al. 1994, Rondon et al. 2005, Gutierrez et al. 2003, Ishii et al. 1991, Yamada et al. 1995,
Motoyama et al. 1998, Baba & Ishii 2000, Ametani & Kawamura 2005, Oettle E. E. 1988,
Chisholm & Janischewskyj 1989, CIGRE W.G. 33.01 1991, Weck 1988, Chowdhuri 1996,
Mikropoulos et al. 2010) (Figure 3.5). Tower equivalent circuit models can be distinguished as
(Fig. 3.6):

PhD Thesis
Page [49 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.6: Tower Models

(a) Single Vertical Lossless Line Models (IEEE W. G. P1243 1996, Hara & Yamamoto 1996,
Ametani et al. 1994, Rondon et al. 2005, Gutierrez et al. 2003, Ishii et al. 1991, Yamada et al. 1995,
Motoyama et al. 1998),
(b) Multi-Conductor Models (Baba & Ishii 2000, Ametani & Kawamura 2005)
(c) Hara et al. Multi-Conductor Model (Hara & Yamamoto 1996),
(d) Multistory Models (Chisholm & Janischewskyj 1989, CIGRE W.G. 33.01 1991, Weck 1988),
(e) Baba & Ishii Multistory Model (Chowdhuri 1996).

 Single Vertical Lossless Line Tower Models: the towers are represented bare regarded
as cylinder or cone shape constructions and their surge impedance is given in Table 3.1.
These equations are obtained by theoretical estimations or laboratory measurements.
Note, so, that the propagation of the overvoltage wave is performed with a velocity equal
to 85% of the light speed (IEEE W. G. PAS-104 1985).

PhD Thesis
Page [50 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.7: Typical Towers of :(a) 150 kV, (b) 400 kV Double-Circuit Lines; the length of insulator strings
are 1.86 (m) and3.62 (m)for the 150 kV and 400 kV line, respectively (Mikropoulos et al. 2010)

 Multi-conductor Tower Models: the surge impedance of each segment is estimated by

using electromagnetic field analysis. The proposed models represent each segment by
single vertical lossless line with different surge impedance, considering also the effect of
bracing and cross arms. The propagation of the overvoltage wave is performed with
a velocity equal the light speed (IEEE W. G. PAS-104 1985).

PhD Thesis
Page [51 of 211]
 CITY , UNIVERSITY of LONDON

Table 3.1: Tower Surge Impedance for Single Vertical Lossless Line Models

No Model Expression
1
Jordan (Jordan 1934) ( )

2 Wagner &Hileman (Wagner & Hileman

(√ )
1960)
3 Sargent &Darveniza (Sargent & Darveniza
(√ )
1969)*
4 Sargent &Darveniza (Sargent & Darveniza
√
1969)** (√ )
IEEE WG (IEEE W. G. PAS-104 1985)
5 Menemelis& Chum (Menemenlis & Chun
√
1982)
6
Chisholm et al. (Chisholm et al. 1985)** ( ( ( )))

7
Chisholm et al. (Chisholm et al. 1985)* ( ( ( )))

8
Chisholm et al. (Chisholm et al. 1985)*** ( ( ))

9
√ ( ( ( ))
Flash ver. 1.7 (IEEE W. G. P1243 1996)
√ )

10 Hara &Yamatomo (Hara & Yamamoto

(√ )
1996)

* Cylindrical Tower ** Conical Tower *** Waisted Tower

Where;
r is the tower base radius(m),
r1 is the tower top radius(m),
r2is the tower radius at waist(m),
h is the tower height(m),
h1is the height from the base to waist(m),
h2is the height from the waist to top(m),
T=(r1h2+r2h+rh1)/h2 ,
PhD Thesis
Page [52 of 211]
 CITY , UNIVERSITY of LONDON

r3is the cylinder radius representing the tower(m) (Sargent & Darveniza 1969).

Multistory Tower Models: each segment is represented by lossless line and R-L circuit
(Figures 3.5(d) and 3.5(e)). Ri and Li respectively, can be calculated for models (Ishii et al.
1991, Yamada et al. 1995, Motoyama et al. 1998) (Figure 3.6(d)) by using the following
equations:

√
(3.1)

√ (3.2)

(3.3)

Where;
his the tower height(m) ,
h1, h2, h3are depicted in Figure 3.5,
ki= 2, and
vT is the surge propagation velocity equal to the speed of light.

As far the modeling of the grounding resistance of each tower is concerned, a concentrated
earthing system can be represented as a constant resistance equal to the low current and low
frequency grounding resistance (IEEE W. G. PAS-104 1985, Ametani & Kawamura 2005).
Moreover, a concentrated tower grounding system can be represented by a current dependent
grounding resistance, R(I), in order to take into account soil ionization (Oettle E. E. 1988,
Chisholm & Janischewskyj 1989, CIGRE W.G. 33.01 1991, Weck 1988, Chowdhuri 1996,
Korsuntchev 1958, Yasuda et al. 2001).

Table 3.2 presents the most used current dependent tower grounding resistance models.

PhD Thesis
Page [53 of 211]
 CITY , UNIVERSITY of LONDON

Table 3.2: Current Dependent Tower Grounding Resistance Models

Model Expression

Oettle (Oettle E. E. 1988)

Chisholm et al. (Chisholm & Janischewskyj

1989) (a)
Chishlom et al. (Chisholm & Janischewskyj
1989) (b)

( )
CIGRE WG (CIGRE W.G. 33.01 1991) √
from Weck (Weck 1988)

Chowdhuri (Chowdhuri 1996)

from Korsuntcev (Korsuntchev 1958)

( )

( )
Yasuda et al. (Yasuda et al. 2001) √

Where;
Ig is the Limiting Current to initiate sufficient Soil Ionization(kA).
ρis the Soil Resistivity(Ωm) .
ris the equivalent Radius of the Tower Footing(m).
n is the Number of Grounding Electrodes.
Π1=R.s/ρ, Π2=(I.ρ/s2.Eo)

PhD Thesis
Page [54 of 211]
 CITY , UNIVERSITY of LONDON

s is the Characteristic Distance (Chowdhuri 1996)(m).

Finally, when the developed overvoltage across the insulators of the line exceeds their
dielectric strength, then a flashover is occurred. The flashover strength VFO (kV) determined
by the voltage-time characteristic of the insulator strings (Darveniza et al. 1975, IEEE W.G
no. 3 1993). The flashover strength VFO(kV) is given as (IEEE W.G no. 3 1993):

VFO = (400 + 710 / t 0.75). D (3.4)

Where;
VFO is The flashover strength (kV)
D is the insulator string length(m),
tis the elapsed time after lightning stroke(μs).

Except from the above approach, leader progression models (Weck 1981, Pigini et al. 1989)
are used for circuit modeling of the insulators. Leader progression models are governed by the
equation of Table 3.3, which are used for the computation of the leader length, L, at each time
instant. Flashover occurs when L becomes equal to the insulator string length, D(Mikropoulos
et al. 2010).

Table 3.3: Leader Progression Models

Model Differential Equation

CIGRE (CIGRE W.G. 33.01

1991) ⁄ [ ⁄( ) ]
Weck (Weck 1981)*

Pigini et al (Pigini et al. 1989)** ⁄ [ ⁄( ) ] [ ⁄ ]

Where;
V is the Instantaneous Voltage across the Insulator String(kV).
*
E0is 600 for Cap (kV/m) &1.3 for Pin Insulators k (m2.kV-2.s-1).
**
E0Value according to Insulator Type and String Length.

3.2.2 Circuit Breakers

Circuit-breakers are the basic part of AIS and GIS switchgear. Main requirements that have to
be fulfilled include (Siemens 2008):
 Reliability during opening and closing
 Adequate quenching ability during rated and short-circuit currents
PhD Thesis
Page [55 of 211]
 CITY , UNIVERSITY of LONDON

 High-performance, reliable, maintenance-free operating mechanisms.

Figure 3.7 depicts a typical circuit breaker. Note, that circuit-breakers are assembled together
with all individual electrical and mechanical components of an AIS installation on site. Main
electrical characteristics of the circuit-breakers are: (Siemens 2008, Martinez-Velasco 2015)
 Rated Voltage: it is defined as the maximum level of the system voltage the circuit-
breaker is designed for. Rated voltage determines the dielectric stresses is the major
dimensioning criterion.
 Rated Insulation Level: it is the capability of the circuit-breaker to withstand various
types of voltages (nominal voltages, switching surges, atmospheric overvoltages). The
defined dielectric strength outcomes by impulse voltage (1.2/50μs) and power frequency
voltage (50Hz) tests and concerns the developed phase-to-ground and phase-to-phase
voltages and across the open contact gap.
 Rated Normal Current: it is the current that the circuit-breaker can continuously carry.
The current passing through the components determines the increase of the temperature,
which must not exceed the defined limits.
 Rated Peak Withstand Current: it is the peak value of the short-circuit current during
a compensation process that the circuit-breaker (closed) can carry. Rated peak withstand
current consist an index for the electrodynamic load of the device.
 Rated Short-Circuit Making Current: it is defined as the peak value of the making
current in case of short-circuits at the terminals of the switching device. Comparing with
the rated peak withstand current, the rated short-circuit making current is more violent,
since dynamic forces may act against the contact movement.
 Rated Breaking Current: it is the load breaking current in normal operation.
 Rated short-Circuit Breaking Current: it is defined as the breaking current (rms) in case
of short-circuits at the terminals of the circuit-breaker. (Siemens 2008, Martinez-Velasco
Juan 2015).

Circuit breakers can be represented as ideal switches, ignoring dynamic phenomena and losses.
Alternatively, more detailed circuit-breaker equivalent circuit models can be applied, in order
to improve the accuracy of the simulation procedures. The most common models are
(Siemens 2008):
 Time Controlled Switch: this model specifies the opening and closing times of the
circuit-breaker, including also a current margin in order to simulate current chopping.

PhD Thesis
Page [56 of 211]
 CITY , UNIVERSITY of LONDON

 Statistical Switch: appropriate distribution functions are applied in order to determine

the switching operation of the device. The mean time and standard deviation are necessary
inputs. Usually a few hundred simulations are run to determine the statistical distribution
of interest such as the maximum relay currents.
 Controlled Switch: a control signal opens or closes the switch following a given strategy
(Martinez-Velasco Juan 2015).

3.2.3 Current Transformers

Inductive ring type current transformers are installed inside or outside the substations
enclosure, in order to record the current magnitude. Appropriate shielding against electric
fields have to be performed, since high operating voltages and developed transient
phenomena produce intense fields that may interfere to the reliable operation of the current
transformers. The measured currents, converted to digital signs, are transmitted by using fiber
optics to the control and protective relays (Grigsby 2007, McDonald 2003).The reliability of
the current transformer depends on the electrical and insulation design, the manufacturing and
processing technology used and the specific physical arrangement. The partial discharge
performance under in-service conditions is a key determining factor in the life expectancy and
long-term reliability of an instrument transformer (Siemens 2008).

3.2.4 Power Transformer

Power transformers constitute the basic component of the power substations, changing the
voltage level of the transferred electrical energy. Substations power transformers are three
phase and are classified according to their windings connection and their insulation (oil or
solid dielectrics). The transformer ratio for any case is given in Table 3.4:

Table 3.4: Transformer Ratio for different Windings Connections

Windings connection Transformer ratio
Yy aYy=VLine-phase,primary/Vline-phase secondary
Yd aYd= aYy
Dy aDy= /aYy
Dd aDd=aYy

The rated power, the rated frequency, the vector group, the insulation level, the insulation
type, the rated primary and secondary voltages are the main specification data of the power

PhD Thesis
Page [57 of 211]
 CITY , UNIVERSITY of LONDON

transformers are (Siemens 2008, Grigsby 2007, McDonald et al 2000, McDonald 2003, IEC
60071-2 1996).As far as the circuit representation of a power transformer, Figure 3.8 depicts
the basic equivalent circuit model.

Figure 3.8: Equivalent Circuit Model of a Power Transformer

Where;
n :Transformer‘s Ratio,
Rm and Lm: core loss resistance and magnetization inductance to consider the no load current containing the
core loss and magnetization current,
R1and R2 the primary and secondary winding resistances including conductors skin effects,
C1and C2: series capacitances of the primary and secondary windings,
C12: parallel capacitance between two windings,
Cg1and Cg2: capacitances between each winding and earth,
L1and L2 : self-inductances of windings,
M: Mutual inductance between two windings)(Rahimpour & Bigdeli 2009).

However, the above basic circuit model does not adequately represent transient phenomena.
High frequency models of power transformers are more appropriate for lighting performance
studies, providing more accurate analysis and more reliable results. There are several
approaches for power transformers circuit modeling, proper to reproduce transient interaction
effects between the different high voltage power facilities. Figure 2.9presents a simplified
model of a transformer terminating a high voltage overhead line by its impedance. It is worth
mentioning that single ended impedance representations do not consider coupling effects
between phases and between low voltage and high voltage side, i.e. surge transfer is neglected
(W. G. 02 of Study Committee 33 Report 39, 2000).

PhD Thesis
Page [58 of 211]
 CITY , UNIVERSITY of LONDON

u(t)

Figure 3.9:Simplified Network Model with Wave Impedance of a High Voltage Overhead Line and the
Terminating Transformer Impedance. Transfer of arriving overvoltages to the Secondary Side is not
considered(W. G. 02 of Study Committee 33 Report 39, 2000).

The main proposed computer modeling techniques concerning transformers‟ circuit

representation can be categorized as (Francisco de Le6n, Semlyen Adam 1994):
 Modeling based on self and mutual inductances.
 Modeling based on leakage inductances.
 Modeling based on the principle of duality.
 Modeling based on measurements.
 Analysis based on electromagnetic fields.

3.2.5 Disconnect Switches

Disconnect switches are used to isolate equipment or to redirect current in a substation. A

movable contact opens or closes a gap between stationary contacts, when activated by an
insulating operating rod that is itself moved by a sealed shaft coming through the enclosure
wall (Fig. 3.10).

The stationary contacts have shields that ensure the necessary distribution of the developed
electric field, in order to restrain surface electrical stresses. The moving contact velocity is
relatively low (compared to a circuit breaker moving contact) and the disconnect switch can
interrupt only low levels of capacitive current or small inductive currents. (Grigsby 2007,
McDonald et al 2000, McDonald 2003, IEC 60071-2 1996, Joint W. G. 33/23.12 1998).

PhD Thesis
Page [59 of 211]
 CITY , UNIVERSITY of LONDON

contacts

floor plan

Figure 3.10: Disconnect Switch (McDonald et al 2000)

3.2.6 Ground Switches

Scope of ground switches is to ground and short-circuit switchgear parts, cables and overhead
lines. Their design is similar to that of vertical-break di-connectors. Ground switches are
interlocked with di-connectors, in order to avoid grounding during nominal operation of the
system. (Grigsby 2007, McDonald et al 2000, McDonald 2003, IEC 60071-2 1996, Joint W.
G. 33/23.12 1998).

3.2.7 Cables

HV and MV cables are used for the transmission and the distribution of the electrical power,
in cases that the overhead transmission or distribution is impossible or dangerous. Basic parts
of a HV or MV cable are:
 The conductor (Al or Cu)
 The semi-conductor layer for the restriction of the developed non uniform electric field
 The insulation (commonly XLPE)
 The metallic screen (Cu)
 The external insulation layer (commonly PVC)

PhD Thesis
Page [60 of 211]
 CITY , UNIVERSITY of LONDON

Fig. 3.11 presents the cut-off a typical MV cable.

conductor
semi-conductor
metallic
screen insulation

external layer
Figure 3.11: Cut-off of a MV Cable(McDonald et al 2000)

Cables can be vulnerable to lightning surges due their connection with overhead lines, which
may be hit by external overvoltages. The condition of the insulation is the main factor for the
lightning performance of the cable that influences the possibility of a breakdown. The testing
procedures of the insulation layer include series and type tests during and after the
construction of the cable (e.g. ac and impulse 1.2/50μs High Voltage Stress) and tests after the
installation (e.g. Damped AC Voltages (DAC) and Very Low Frequency(VLF) tests). The
joints between the different parts of the cables are usually susceptible points during the
development of surges.

3.2.8 Busbars

Scope of busbars is to conduct substantial currents in limited space, ensuring, simultaneously,

the continuity of power supply. The flexible design of the busbars, their electrical
characteristics, their reliable operation and their withstand capability against strong
electromechanical forces are of great importance, in order to warrant the uninterrupted
operation of the system (Grigsby 2007, IEC 60071-2 1996). The compliance with demanded
distances between live parts is also a critical issue, in order to ensure the necessary withstand
insulation levels against internal and external over voltages. In general, busbars are made of
aluminum and copper (ratio 1/3), which combine adequate conductivity and mechanical
strength. Note that, for substations up to 200kV, lighting strikes is the dominant factor that
determined the insulation level and the design of the protection system. For higher nominal
operating voltages the role of the switching phenomena prevails(Abdel-Salam et al., Grigsby
2007, IEC 60071-2 1996).

PhD Thesis
Page [61 of 211]
 CITY , UNIVERSITY of LONDON

3.3 Configuration of Substations

Substations are interconnected, creating a meshed network and increasing the reliability of the
power system, since the interconnection provides alternate paths for the power flow in case of
outages. For this reason, the circuit configuration of the network substations has to be
appropriately designed carefully, in order to avoid power supply interruption (Siemens 2008,
Abdel-Salam et al., Grigsby 2007). The main busbar configurations are (Siemens 2008,
Grigsby 2007):

 Single-Busbar
 Double-Busbar with Double Breaker
 Double-Busbar with Single Breaker
 Main and Transfer Busbar
 Ring Bus
Fig. 3.12 presents the basic configurations of HV/MV substations.
line line

BUS 1
Circuit Breaker

Disconnecting Switch
BUS

BUS 2

(a) (b)

BUS 1

BUS 2

Bus Tie Breaker

line line

PhD Thesis
Page [62 of 211]
 CITY , UNIVERSITY of LONDON

(c)

line line

Tranfer Bus

Bus Tie Breaker

Main Bus

(d)

line

line

(e)
Figure 3.12: Basic configurations of HV/MV substations
(a) Single-Busbar (Abdel-Salam et al.), (b) Double-Busbar with Double Breaker (Abdel-Salam et al.), (c)
Double-Busbar with Single Breaker (Abdel-Salam et al.), (d) Main and Transfer Bus Bars (Abdel-Salam et
al.), (e) Ring Bus (Abdel-Salam et al.)

PhD Thesis
Page [63 of 211]
 CITY , UNIVERSITY of LONDON

3.4 Grounding of Substations

3.4.1 Need for Grounding

The appropriate grounding of a substation and the achievement of low grounding resistance
values are of great importance, in order to ensure the safety of the human and the installation.
A proper grounding system offers a path and diverts the fault currents to earth, without
exceeding the dielectric restrictions of the equipment; simultaneously, grounding system
protects personnel against the dangers of electric shock under fault conditions (McDonald
2003).

The initial design is the first step for a safe and economical grounding system. Basic parts of
the earthing system are the ground grid, consisted of conductors and rods. In addition, the
grounding system includes all of the interconnected grounding facilities in the substation area,
including the ground grid, overhead ground wires, neutral conductors, underground cables,
etc. (McDonald 2003).

Various factors have a critical impact on the developed voltages in and around the substation
area. The soil resistivity, the kind, the magnitude and the duration of the fault (short-circuit,
lightning, switching overvoltages), the characteristics of the rods influence the effectiveness of
the grounding systems and have to be taken into consideration during the design of the
substation. For this reason the appropriate study and design of the grounding system and the
measurement of the grounding resistance values after its construction are necessary, in order
to achieve the optimum lightning performance of the installation. In case that the above
parameters contribute to high potential gradient at the earth surface, the grounding system
may be able to offer the necessary safety and protection, despite its capacity to carry the fault
current in magnitudes and durations permitted by protective relays.

During typical ground fault conditions, the developed potential gradients along the earth
surface may be high enough to endanger a person in the area. Moreover, hazardous voltages
may arise between grounded structures or equipment frames and the nearby earth (Grigsby
2007). Figure 3.13 presents the cases that a person can be exposed to potential in a substation.

PhD Thesis
Page [64 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.13: Cases of Electric Shock (Grigsby 2007).

3.4.2 Design of a Grounding System

Soil resistivity is a very critical factor that determines the soil structure and the value of the
grounding resistance. The international literature gives various estimations of the soil
resistivity, based on soil types (clay, loam, sand, shale, etc.).
However, the soil resistivity can change significantly during the year, due to variation of
environmental factors (moisture, temperature, chemical content, etc.). For this reason, the
measurement of the soil resistivity is necessary, in order to obtain an accurate aspect of the
soil quality.
The most widely used test for determining soil resistivity data was developed by Wenner and
is called either the Wenner method or four-pin method (Figure 3.14). Using four pins or
electrodes driven into the earth along a straight line at equal distances of a, to a depth of b,
current is passed through the outer pins while a voltage reading is taken with the two inside
pins. Based on the resistance R, as determined by the voltage and current, the apparent
resistivity can be calculated using the following equation (assuming b<<a) (Grigsby 2007):
ρ=2π.a.R (3.5)

PhD Thesis
Page [65 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.14: Soil Resistivity Measurement ( Wenner‘s Method )

Table 3.5 presents indicative values of soil resistivity for various types of earth. Note that for
more precise analysis and more accurate results, the multi-layer model is used, considering the
variation of the soil quality for different depths.

Table 3.5: Range of Earth Resistivity (IEEE Std 80 2000)

Type of earth Average resistivity (Ω·m)
Wet organic soil 10
Moist soil 102
Dry soil 103
Bedrock 104

Interpretation of the apparent soil resistivity based on field measurements is difficult. Uniform
and two-layer soil models are the most commonly used soil resistivity models. The objective
of the soil model is to provide a good approximation of the actual soil conditions.
Interpretation can be done either manually or by the use of computer analysis. There are
commercially available computer programs that take the soil data and mathematically calculate
the soil resistivity and give a confidence level based on the test.

The achieved resistance of the constructed grid is mainly determined by the resistivity and the
area of the site. Adding more conductors or changing the depth of the grid does little to lower
PhD Thesis
Page [66 of 211]
 CITY , UNIVERSITY of LONDON

the resistance. The effect of ground rods depends on the location and depth of the ground
rod with respect to the soil resistivity. The effects of ground rods on the resistance can be
substantial, although it is sometimes difficult to determine the effects. In uniform soil, it is
difficult to determine if the addition of more conductors or the addition of ground rods will
affect the overall resistance the most. In most cases though, the addition of ground rods has
a greater impact because the ground rods discharge current into the earth more efficiently than
the grid conductors. Assuming a two-layer soil model with a lower resistivity soil in the lower
layer, ground rods can have a substantial impact on the resistance of the grid. The more the
ground rods penetrate into the lower resistivity soil, the more the rods will reduce the grid
resistance. These rods also add stability since the variations in soil resistivity due to moisture
and temperature are minimized at lower depths. The effects of moisture and temperature on
the soil resistivity can be quite dramatic. Ground rods placed on the outside of the grid have
a greater impact than those placed in the interior of the grid because of current density
(Grigsby 2007, Joint W. G. 33/23.12 1998, IEEE Std 998-1996).

The importance of the lower ground grid resistance is reflected in the ground potential rise
(GPR) and actual touch and step voltages. Lowering the resistance of the grid normally
reduces the GPR, although not necessarily proportionally. Lowering the resistance may
somewhat increase the grid current because the change is the current split between all the
ground current return paths. Another way to decrease the resistance is to install counterpoises
(Grigsby 2007, IEC 60071-2 1996, Joint W. G. 33/23.12 1998, IEEE Std 998-1996).

As far as the circuit representation of the grounding resistance, the simplest approach is the
use of a constant resistance equal to the low current and low frequency grounding resistance,
R0(IEEE W. G. PAS-104 1985, Ametani & Kawamura 2005). The calculation of the
grounding resistance of the substation, in case of uniform soil, can also be performed by using
the following equation (IEEE Std 80 2000):

√ ( )

Where;
Rg is the substation ground resistance in (Ω).
ρ is the soil resistivity in (Ω·m).
A is the area occupied by the ground grid in (m2).

PhD Thesis
Page [67 of 211]
 CITY , UNIVERSITY of LONDON

Laurent (Laurent 1951) and Niemann (Nieman 1952), based on the previous approach,
propose the following equation for the estimation of the grounding resistance:

√ ( )

Where;
LT is the total buried length of conductors in (m).
Sverak (Sverak 1984) modified equation (3.9), considering the effect of grid depth:

[ ( ] ( )
√
√ )

Where;
h is the depth of the grid in (m).

Schwarz (Schwarz 1954) proposes the following set of equations, in order to compute the
grounding resistance of a system consisting of horizontal (grid) and vertical (rods) electrodes
(Rüdenberg 1945, Sunde 1968).

( )

Where;
R1 is the ground resistance of grid conductors in (Ω).
R2 is the ground resistance of all ground rods in (Ω).
Rm is the mutual ground resistance between the group of grid conductors, R1, and group of
ground rods, R2 in (Ω).

PhD Thesis
Page [68 of 211]
 CITY , UNIVERSITY of LONDON

Ground resistance of the grid is given as:

[ ( ) ] ( )
√

Where;
ρ is the soil resistivity in (Ω·m).
Lc is the total length of all connected grid conductors in (m).
a' is for conductors buried at depth h in (m), or
a' is a for conductor on earth surface in (m).
2a is the diameter of conductor in (m).
A is the area covered by conductors in (m2).
k1, k2 are the coefficients [see Figure 25(a) and (b)]

Ground resistance of the rod bed is given as:

[ ( (√ ) ]] ( )
√

Where;
Lr is the length of each rod in (m).
2b is the diameter of rod in (m).
nR is the number of rods placed in area (A).

Mutual ground resistance between the grid and the rod bed is given as:

[ ( ) ] ( )
√

3.5 Need for Lightning Protection of Substations

Τhe study of the lightning repercussions and the design of an appropriate protection system
against lighting is a crucial issue, since substations are complex installations of high investment
cost. Moreover, the safety of the personnel must be considered in priority, in order to avoid
accidents. In the current section, the dangers arisen to a network substation by an atmospheric
discharge and the consequences of the lightning flashes to the normal operation of an
electrical system will be presented.

PhD Thesis
Page [69 of 211]
 CITY , UNIVERSITY of LONDON

External overvoltages can cause several damages to a substation, leading to insulation

breakdowns. When the incoming surge exceeds the insulation level of the equipment a
breakdown is occurred, resulting to serious damages and interruption of the power supply.
Insulators, switching devices, cables and transformers are the main parts of the installation
that are intensively stressed by lightning phenomena and their failure creates several serious
malfunctions, interruptions and dangers.

In case of lightning hit on the incoming overhead transmission line a shielding failure or a
backflashover, i.e. insulation failure of the ceramic of glass insulator, can be occurred,
depended on the ground wires position, the peak current of the lightning flash and the
grounding resistance. Such an insulation fault results to the development of a travelling wave,
which will be directed to the entrance of the substation. The high voltage cables can also be
influenced by the incoming surges; when the arisen overvoltages exceed the dielectric strength
of the XLPE insulation a fault is occurred. The most vulnerable parts are joints and
termination positions. Furthermore, a lightning surge can cause breakdown of the
transformers insulators or the transformers dielectrics, resulting to thermal and
electromechanical effects (fire, short-circuits, mechanical damages etc.).

The above indicate the need of appropriate lightning protection system, in order to avoid
lightning outages and restrain the repairing costs of the equipment. The design of the lightning
protection system has to take into consideration the stochastic nature of the external
overvoltages phenomena and the various techno-economic factors of a substation. The
striking point, the geometrical characteristics of the external lightning protection system and
the grounding system, the basic insulation level are factors that influence the severity of the
lightning impact. Protection of substations against the deleterious effects of lightning may be
achieved by using highest insulation levels, taking of course into account the financial cost, or
by installing overhead ground wires in order to intercept the lightning flashes. Moreover, surge
arresters can contribute to the improvement of the lightning performance of the installation,
especially in regions with high soil resistivity.

3.6 Development of Overvoltages in Substations

Figure 3.15 presents the configuration of a typical substation. A direct or indirect lightning hit
on an overhead transmission line can create a significant overvoltage, which can influence the
normal operation of the other parts of the equipment (transformer, cables, etc.).

PhD Thesis
Page [70 of 211]
 CITY , UNIVERSITY of LONDON

Configuration (a)

Configuration (b)

Figure 3.15: Configurations (a) & (b) of Typical Substations (Piotr Oramus, Marek Florkowski, 2014)

High voltage transmission lines must have high insulation levels, in order to prevent
breakdown between phases or between phase and earth during operating or fault conditions.
Phases are electrically separated by air, since ceramic or glass insulators offer electrical
separation between phases and metallic towers. The inherently high insulation levels for
transmission lines render them less susceptible to lightning damage in comparison to medium
voltage or low voltage lines, and so the annual outage rate is in general low; however, the
repercussions of a transmission line failure can be much more costly (McDonald et al. 2000,
Joint W. G. 33/23.12, no. 176, 1998).

PhD Thesis
Page [71 of 211]
 CITY , UNIVERSITY of LONDON

When a lightning stroke hits a phase conductor of a high voltage transmission line, the
lightning current, i, is separated in two equal parts, each of them is traveling in opposite
direction. Owing to the line characteristic impedance, Z, the two impulse lightning currents
create two equal voltage traveling waves, as it is shown in Figure 3.16.
The value of the voltage traveling waves is (eq. 3.16):

( )

If the magnitude of the voltage traveling waves exceeds the rated basic insulation level (BIL)
of the line, then a breakdown of the line insulation occurs. In case of line insulation
breakdown, two voltage traveling wavesu1 and u2 will be arisen, which are directed to the next
towers (Figure 3.16).

Taking into account the geometrical characteristics and the insulators type of the tower, there
is a critical lightning current value for each voltage level of the transmission line, above which
the developed overvoltage is greater than the line BIL (Christodoulou et al. 2012).

Figure 3.16: Direct Lightning Stroke on a Phase-Wire of a High Voltage Transmission Line
(HVTL)[Shield Failure](Christodoulou et al. 2012)

PhD Thesis
Page [72 of 211]
 CITY , UNIVERSITY of LONDON

When a lightning hits the tower structure or the overhead ground wire, the lightning impulse
discharge current, flowing through the tower and tower footing resistance, produces potential
differences across the line insulation. If the line insulation strength exceeds a defined value, a
flashover happens, i.e. it appears a backflashover.

Since the tower voltage, u3 (Figure 3.17), which appears after a back-flashover, is highly
depended on the tower resistance, it follows that footing resistance is an extremely important
factor in determining lightning performance (IEEE Std 1243-1997).

Figure 3.17: Indirect Lightning Stroke on a High Voltage Transmission Line(HVTL)

[Back-flashover] (Christodoulou et al. 2012)

It is worth mentioning, that any overvoltage surge appearing on a transmission line due to
internal or external disturbances will propagate in the form of a traveling wave toward the
ends of the line. During its travel the overvoltage surge will, in general, experience attenuation
and distortion inflicted upon it by earth resistance and line losses, and corona discharges
(Abdel-Salam et al.).

The above analysis highlights the importance of the insulation level for the adequate lightning
performance of the network. Main factors that influence the efficiency of an insulating item
PhD Thesis
Page [73 of 211]
 CITY , UNIVERSITY of LONDON

are the voltage characteristics (peak value, duration) that stress the item. Moreover, the
strength of the insulation must be coordinated with the stresses that may occur on the power
system. To this direction, the insulation coordination between the various components of the
installation has to been insured, in order to limit the lightning outages and improve the
reliability of the system. In details, insulation coordination is defined as the correlation of the
insulation of electrical equipment with the characteristics of protective devices such that the
insulation is protected from excessive overvoltages. In practice, it means that an insulation
level is determined for each voltage level to which the equipment is designed and tested.

Moreover, proper grounding, shielding and surge arresters‟ implementation can contribute to
the limitation of the developed overvoltages under the BIL (Wadhwa 2007).

An important aspect of external and internal overvoltages and breakdown withstand capability
of the insulating materials is their stochastic nature. For this reason, the installation has to be
protected against the worst probable (not just possible) condition. Otherwise, over
dimensioning of the insulation leads to useless costs. This, however, would involve some level
of risk failure, but it is desired to accept some level of risk of failure than to design a risk-free
but a very costly system (Wadhwa 2007).

The distribution of breakdown for a given gap follows approximately normal or Gaussian
distribution. The distribution of over voltages developed on a power system also follows the
Gaussian distribution. For the coordination of the electrical stresses due to lightning or
switching surges with the electrical strength of the insulators, the overvoltage distribution is
represented in the form of probability density function and the insulation breakdown
probability by the cumulative distribution function (Figure 3.18) (Wadhwa 2007).

overvoltage
distribution insulation
breakdown
probability

Vk
Figure 3.18: Overvoltage Distribution and Insulation Breakdown Probability (Wadhwa 2007)
PhD Thesis
Page [74 of 211]
 CITY , UNIVERSITY of LONDON

3.7 Design of Lightning Protection System

Generally, the annual number of direct lightning strokes in a substation is given by the
equation:

( )

Where;
W and Lare the dimensions of the substation (width and length)
N is the number of flashes to earth per square kilometer per year [Equations (2.9) to (2.15)].
The design of lightning protection system is performed following appropriate methodologies.
The most widely used design methods to protect substation against lightning hits are:
 Fixed angles method
 Empirical curves method.
 Electrogeometrical method

3.7.1 The fixed Angles Method

The fixed-angles method uses vertical angles in order to adjust the configuration of the
external lighting protection system, i.e. the number, the position, and the height of shielding
wires or masts. Figure 3.19 depicts the method applied for shielding wires and masts,
correspondingly (McDonald et al. 2000, IEEE Std 998-1996).

(a)
PhD Thesis
Page [75 of 211]
 CITY , UNIVERSITY of LONDON

(b)
Figure 3.19: Fixed Angles for Shielding Wires (McDonald et al. 2000, IEEE Std 998-1996)

Critical factors that influence the calculated angles are:

 The degree of lightning exposure
 The importance of the substation being protected
 The physical area occupied by the substation.
The value of the angle alpha (α) that is commonly used is 45o. Both 30o and 45o are widely
used for angle beta (β) (McDonald et al. 2000, IEEE Std 998-1996).

3.7.2 Empirical Curves Method

Empirical curves, obtained by field studies and laboratory measurements, are used in order to
determine the characteristics of the design lightning protection system, i.e. the number the
position and the height of the shielding wires and the masts (McDonald et al. 2000, IEEE Std
998-1996). The curves have been developed considering shielding failure rates of 0.1%, 1%,
5%, 10% and15%. A failure rate of 0.1% is commonly used in design. Figures 3.20 has been
developed for a variety of protected object heights, d.

PhD Thesis
Page [76 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.20: Single lightning Mast Protecting Single Ring of Object—0.1% Exposure.
(y: Height of Mast above Protected Object, s: Horizontal Separation,
d: Height of Protected Object) (McDonald et al. 2000, IEEE Std 998-1996)

Figure 3.21 depicts the areas that can be protected by two or more shielding masts. In case
that two masts are used to protect an area, the empirical curves give shielding information
only for the point B and for points on the semicircles drawn about the masts, with radius x.
Any single point falling within the cross-hatched area should have <0.1% exposure. Points
outside the cross-hatched area will have >0.1% exposure (McDonald et al. 2000, IEEE Std
998-1996). The lightning performance of the substation can be upgraded by moving the masts
closer together (Figure 3.27). As illustrated in Figure 3.27b, the size of the areas with an
exposure greater than 0.1% has been significantly reduced (McDonald et al. 2000, IEEE Std
998-1996).

Figure 3.21: Areas Protected by Multiple Masts

(a) With Two Lightning Masts;
(b) with Four Lightning Masts (McDonald et al. 2000, IEEE Std 998-1996).
PhD Thesis
Page [77 of 211]
 CITY , UNIVERSITY of LONDON

3.7.3 The Electrogeometrical Model

The electrogeometrical model (EGM) is based on the striking distance, a factor that is
determined by the lightning peak current and the hit position. The striking distance r in m is
given by the equation:

(3.15)

Where;
A and b are constants dependent on the termination point
I is the prospective stroke current in kA

Several researchers have proposed various electrogeometrical models with different constants
A and b. However, all of them agree to the following aspects (Christodoulou et al. 2010):
 Strokes arrive vertically
 The lightning leader develops unaffected by the existence of grounded objects until it
arrives within striking distance from the grounded object
 The striking distance is related to the current of the return stroke.

The termination point of a lightning stroke to substation is of great importance, in order to

predict the effects of the lightning surges and secure the installation against the incoming
overvoltages. The most vulnerable part of a substation is the overhead lines. In case of
overhead transmission lines, lightning can hit either a ground wire, phase conductor, tower or
even ground. The fractions of lightning strikes hA(IP) and hB(IP) that will terminate on a phase
conductor or on an overhead ground wire, can be calculated by the equations (3.17) and (3.18)
(Christodoulou et al. 2010):

( ) (3.16)

( ) (3.17)

Where;
DC is the shielding failure exposure distance.
D is the vertical distance between points C and A (Figure 3.22) (Christodoulou et al. 2010).

PhD Thesis
Page [78 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.22: The Electrogeometrical Model: Representation of Shield Wires and Phase Conductors
(Christodoulou et al. 2010)

The above analysis indicates that an overhead ground wire cannot always offer adequate
protection for every lightning‟s current amplitude. In details, some lightning currents can
penetrate the lightning protection system and terminate on a phase conductor; so, the
insulation of the installation must be appropriately designed in order to withstand the
developed overvoltages without flashover (McDonald et al. 2000).Figure 3.23 presents the
application of the above method for a typical substation, by using the rolling sphere method.
The rolling sphere method is based on the assumption that the striking distances to the
ground, a mast or a wire are the same. The imaginary sphere of radius is equal to the striking
distance rolls over the surface of the substation (Lightning masts, shield wires, substation
fences and other grounded metallic objects that can provide lightning shielding). The parts of
the equipment that remain below the curved surface of the imaginary sphere are protected.
These parts that touch the sphere are not protected (McDonald et al. 2000, IEEE Std 998-
1996).

Figure 3.23: Principle of the Rolling Sphere (McDonald et al. 2000, IEEE Std 998-1996)
PhD Thesis
Page [79 of 211]
 CITY , UNIVERSITY of LONDON

A typical substation, however, is much more complex. It may contain several voltage levels
and may utilize a combination of shield wires and lightning masts in a three-dimensional
arrangement. Figure 3.24 shows the application of the electrogeometrical model considering
four shield masts in a multiple shield mast arrangement. The arc of protection for stroke
current Is is shown for each set of masts. The protective zone can again be visualized as the
surface of a sphere that is rolled toward a mast until touching the mast, and then rolled up and
over the mast such that it would be supported by the masts. Using the concept of rolling
sphere of the proper radius, the protected area of an entire substation can be determined. This
can be applied to any group of different height shield masts, shield wires, or a combination of
the two. Figure 3.25 shows an application to a combination of masts and shield wires
(McDonald et al. 2000, IEEE Std 998-1996).

Figure 3.24: Multiple Shield Mast Protection for Stroke Current Is

(McDonald et al. 2000, IEEE Std 998-1996)

Figure 3.25: Protection by Shield Wires and Masts (McDonald et al. 2000, IEEE Std 998-1996)

PhD Thesis
Page [80 of 211]
 CITY , UNIVERSITY of LONDON

3.8 Surge Arresters

The lightning performance of a substation can be improved by installing surge arresters. Surge
arresters are semiconductor devices that protect the equipment of an electrical installation
against incoming surges. Several different types of arresters are available and all perform in a
similar way: they present high resistance during the normal operation of the network and low
resistance during surge conditions. Nowadays, gapped surge arresters with varistors made of
silicon carbide have been installed by gapless metal oxide arresters, due their advanced
characteristics (Hinrichsen et al. 2001). In details, metal oxide arresters present more extreme
non-linear voltage-current characteristic compared to silicon carbide ones, rendering
unnecessary the disconnection of the resistors from the line through serial spark gaps
(Hinrichsen et al. 2001). Metal oxide varistors have to fulfill the following demands:
 Low resistance during surges so that overvoltages are limited.
 High resistance during normal operation so as to avoid negative effects on the power
system.
 Sufficient energy absorption capability for stable operation.

Main drawback of these varistors is the continuous flow of current to the ground (some μΑ),
since their resistance is not infinite. So, considering the applied operating voltage, there is
constantly a leakage current (Siemens 2008).

The basic parts of a metal oxide surge arresters are:

 The cylindrical metal-oxide resistor blocks (varistor).
 The insulating housing.
 The electrodes.

Between the varistor column and the polymeric housing there is a glass-fiber structure that
either completely encloses the resistor blocks or exerts sufficient force on the ends of the
stack to hold the metal oxide blocks firmly together (Figure 3.26).

PhD Thesis
Page [81 of 211]
 CITY , UNIVERSITY of LONDON

Figure 3.26: Metal Oxide Surge Arrester cut

(1: Electrodes, 2: Fiberglass, 3: Nonlinear Aesistor, 4: External Insulator) (Spanias et al. 2010)

The non-linear voltage-current characteristic is described by equation (Lundquist et al. 1990,

Christodoulou et al. 2009):

I =kVα, α>1 (3.18)

Where;
I is the current through the arrester,
V is the voltage across the arrester,
α is a non-linearity exponent (measure of non-linearity) and
k is a constant depended on the arresters type.

The value αcharacterizes the non-linear V–I characteristic; the greater the value of α, the
„„better” the varistor. Modern arresters have values of a between 25 and 60. Figure 3.27
depicts the voltage-current characteristic of a 420kV gapless surge arrester.

Figure 3.27: Voltage – Current Characteristic of a Metal Oxide surge Arrester (Hinrichsen et al. 2001)

PhD Thesis
Page [82 of 211]
 CITY , UNIVERSITY of LONDON

The main electrical characteristics of gapless metal oxide surge arresters according to the IEC
60099-4 are (Hinrichsen et al. 2001, IEC 60099-4 2004–2005):

 Continuous Operating Voltage: Designated rms value of power frequency voltage that
may be applied continuously between the terminals of the arrester. MCOV of the arrester
must be higher than the maximum continuous operating voltage of the system.
 Rated Voltage: Maximum permissible rms value of power frequency voltage between
arrester terminals at which is designed to operate correctly under temporary overvoltages.
 Discharge Current: Impulse current which flows through the arrester.
 Residual Voltage: Peak value of the voltage that appears between arrester terminals
when a discharge current is injected.
 Rated Discharge Current: Peak value of lightning current impulse, which is used to
classify an arrester.
 Lightning Impulse Protective Level: Voltage that drops across the arrester when the
rated discharge current flows through the arrester.
 Energy Absorption Capability: Maximum level of energy injected into the arrester at
which it can still cool back down to its normal operating temperature. Standards do not
define energy capability of an arrester. In IEC exists the term line discharge class, but since
this is not enough information, various manufacturers present thermal energy absorption
capability in kJ/kV (Uc), defined as the maximum permissible energy that an arrester may
be subjected to two impulses according to IEC clause 8.5.5 (IEC 60099-4 2004–2005),
without damage and without loss of thermal stability (Hinrichsen et al. 2001, IEC 60099-4
2004–2005).

The installation position of the arresters plays important role, due to the fact that overvoltages
behave as travelling waves. The voltage level at every instant and at every point on the line;
results from the sum of the different instantaneous values of each individual voltage wave
considering the refractions and reflections because of the changes of the surge impedance.
A connected transformer behaves as an un-terminated end, since its winding inductivity for
fast voltage waveforms presents much higher impedance compared with the impedance of the
line. Figure 3.28 depicts an overvoltage running towards a transformer, assuming a
propagation velocity equal to the speed of the light. The arrester presents an ideal behavior,
limiting the desired residual voltage. Figure 3.29 presents the overvoltages at the terminal ends
of the arrester and the transformer. Note that, a voltage wave is totally reflected when
PhD Thesis
Page [83 of 211]
 CITY , UNIVERSITY of LONDON

reaching an un-terminated end of a line. The voltage level at every instant and at every point
on the line; results from the sum of the different instantaneous values of each individual
voltage wave. Thus, at the terminated end this value will be doubled. A connected transformer
appears similar to an un-terminated end since its winding inductivity for rapid functions
exhibits a great impedance compared with the surge impedance of the line (Spanias et al.
2010).

Figure 3.28: Power Transformer Protected by a Surge Arrester (L is the distance between arrester and
transformer, c is the velocity of the light, S is the steepness of the lightning current)

Figure 3.29: Developed Overvoltages at the Terminals of the Arrester and the Transformer(L is the distance
between arrester and transformer, c is the velocity of the light, S is the steepness of the lightning current)

A significant parameter that determines the efficiency of the installed arrester is its protective
zone, i.e. the maximum separation distance for which the insulation coordination demands are
satisfied. Arresters should be installed as close as possible to the device to be protected, since
fast-front surges may result to overvoltages at the terminals of the equipment under
protection much higher than the residual voltage of the arrester.

PhD Thesis
Page [84 of 211]
 CITY , UNIVERSITY of LONDON

The protective zone of an arrester, for the simple arrangement of a transformer connected to
the end of a single feeder, can be estimated as following:

( ) (3.19)

Where;
c is the propagation velocity of the voltage wave
S is the front steepness of the lightning overvoltage ( kV/μs) (typical value:1000)
LIWV standard lightning impulse withstand voltage of the device to be protected in kV
Uπ lightning impulse protective level of the arrester (kV)

The normal operation of metal oxide arresters can be degraded by various factors. Ingress of
moisture in the internal of the arresters, poor inter-disc contact and consequent losses due to
electrical discharges, housing pollution, mechanical fractures because of thermal runaway after
high-current surge are common reasons for arresters failures (James & Su 2008, Lundquist et
al. 1990, Christodoulou et al. 2009). The above indicate the necessity of arresters‟ condition
testing and monitoring, in order to ensure their reliable operation. Laboratory tests and on-line
monitoring have been proposed (Hinrichsen 1997) in order to examine the situation of the
protective devices. However, there is always an increased risk to result to false conclusions and
lead to unnecessary repairs or replacements. In any case, the leakage current is the most used
indicator for the condition of installed surge arresters.
Before the presentation of the equivalent circuit models of metal oxide surge arrester, it is
important to emphasize that laboratory results about the characteristics of metal oxide surge
arresters indicate a dynamic behavior: the residual voltage of the arrester is strongly influenced
by the waveform of the lightning current and increases as the current front time decreases
(IEEE W. G. 3.4.11 1992, Zˇitnik et al. 2006, Bayadi et al. 2003).

So, metal oxide gapless surge arresters cannot be modeled by only a non-linear resistance,
since their response depends on the magnitude and the rate of rise of the surge current. To
these directions, several frequent dependent models have been developed, in order to
represent the actual dynamic behavior of the arrester. The proposed equivalent circuit
models(Martinez & Castro-Aranda 2005, Tarchini & Gimenez 2003, Montanes et al. 2002,
Garcia-Gracia et al. 1999, Nakada et al. 1998, Zanetta & Pereira 2003)present some
differences, concerning mainly the calculation of their parameters.

PhD Thesis
Page [85 of 211]
 CITY , UNIVERSITY of LONDON

3.8.1. The Physical Model

Figure 3.30 depicts the most simple varistor model, which is non-frequency dependent. L is
the inductance of conducting leads and C is the capacitance of the device package and zinc
oxide material. At low currents, the varistor behaves as a high value resistance RL, and at very
high currents the low value bulk resistance RB of the zinc oxide grains dominates the varistor
response (Pinceti & Giannettoni 1999, Zˇitnik et al. 2006, Suljanovic et al. 2006, Bayadi et al.
2003).

Figure 3.30: The Physical Model(Bayadi et al. 2003)

3.8.2 The IEEE Model

The IEEE Working Group 3.4.11 (IEEE W. G. 3.4.11 1992) proposes the model of Figure
3.31. The circuit is consisted of two nonlinear resistances A0 and A1, separated by R–L filter.
For slow front surges the filter impedance is low and the non-linear resistances are in parallel.
For fast front surges filter impedance becomes high, and the current flows through the non-
linear resistance A0. The inductance L1 and the resistance R1 comprise the filter between the
two varistors, since the inductance L0 is associated with magnetic fields in the vicinity of the
arrester. R0 stabilizes the numerical integration and C represents the terminal-to-terminal
capacitance. The equations for the above parameters and the per-unit V–I characteristic of the
varistors are given in (IEEE W. G. 3.4.11 1992):

Figure 3.31: The IEEE Model (IEEE W. G. 3.4.11 1992)

PhD Thesis
Page [86 of 211]
 CITY , UNIVERSITY of LONDON

L1  15 d / n μH (3.20)

R1  65 d / n Ω (3.21)

Lo  0.2 d / n μΗ (3.22)

Ro  100 d / n Ω (3.23)

C  100 n / d pF(3.24)

Where;
dis the length of arrester column in meters and
n is the number of parallel columns of metal oxide disks.

3.8.3 The Pinceti–Gianettoni Model

The Pinceti–Gianettoni model is based on the IEEE model, but there is no capacitance and
the resistances R0 and R1 are replaced by one resistance (approximately 1 MΩ) (Figure 3.32).

Figure 3.32: The Pianceti–Gianettoni Model (Pinceti & Giannettoni 1999).

The non-linear resistors are based on the curves of (IEEE W. G. 3.4.11 1992). The
inductances L0 and L1 are calculated using the equations (Pinceti & Giannettoni 1999):

1 Vr1 / T2  Vr 8 / 20
L1    Vn μH (3.25)
4 Vr 8 / 20

1 Vr1 / T2  Vr 8 / 20
Lo    Vn μH (3.26)
12 Vr 8 / 20

PhD Thesis
Page [87 of 211]
 CITY , UNIVERSITY of LONDON

Where;
Vnis the arrester‟s rated voltage,
Vr8/20 is the residual voltage for a (8/20) 10 kA lightning current and
Vr1/T2 is the residual voltage for a (1/T2) 10 kA lightning current.

3.8.4 The Fernandez–Diaz Model

Based also on IEEE model, A0 and A1 are separated only by L1 (Figure 3.33). C is added in
arrester terminals and represents terminal-to-terminal capacitance of the arrester. The
procedure for the computation of the parameters is given in (Zanetta & Pereira 2003). The V–
I characteristics for A0 and A1 are calculated using manufacturers‟ data, considering the ratio I0
to I1 equal to 0.02. The inductance L1 is given as:

Figure 3.33:The Fernandez–Diaz Model (Fernandez & Diaz 2001).

L1= n . L΄1 (3.27)

Where;
n is a scale factor and
L΄1 is given in (Fernandez & Diaz 2001), computing the percentage increase of the residual
voltage as:

( ⁄ ) ( ⁄ )
( ) (3.28)
( ⁄ )

Where;
Vr(8/20) is the residual voltage for a 8/20 lightning current.
Vr(1/T2) is the residual voltage for a 1/T2 lightning current with the nominal amplitude.

PhD Thesis
Page [88 of 211]
 CITY , UNIVERSITY of LONDON

3.9 Conclusions

This chapter presents the basic configuration schemes of HV/MV substations and analyzes
the basic characteristics of the basic equipment of the substations. Substation σ are complex
systems, consisting of various types of components and their appropriate operation is
fundamental for the reliable and uninterrupted power supply. It is obvious that a potential
lighting hit can influence a series of devices, resulting in the destabilization of the system and
consequent malfunction. So, the need for lightning protection is highlighted, in order to
safeguard the protection of the substation against internal and external overvoltages. In
addition, the main methods and procedures for the lightning protection of the substation are
depicted, providing also the circuit models that will be implemented in the next chapters.

PhD Thesis
Page [89 of 211]
 CITY , UNIVERSITY of LONDON

References
ABB, Surge Arresters-Buyer‘s Guide, 5.1 ed., 2007
Abdel-Salam M., Anis H., A. El-Morshedy, Radwan R., ―High Voltage Engineering, Theory and
Practice‖, Marcel Dekkienrc, New York.
Ametani A. and Kawamura T., ―A method of a lightning surge analysis recommended in Japan using
EMTP,‖ IEEE Trans. Power Delivery, vol. 20, no. 2, pp. 867–875, Apr. 2005.
Ametani A., Kasai Y., Sawada J., Mochizuki Α., and Yamada Τ., ―Frequency–dependent impedance
of vertical conductors and a multiconductor tower model,‖ IEE Gener. Transm. Distrib., vol. 141, no. 4,
pp 339–345, Jul. 1994.
Baba Y. and Ishii M., ―Numerical electromagnetic field analysis on lightning surge response of tower with
shield wire,‖ IEEE Trans.Power Delivery, vol. 15, no. 3, pp 1010–1015, Jul. 2000.
Bayadi A., Harid N., Zehar K., Belkhiat S., Simulation of metal oxide surge arrester dynamic behavior
under fast transients, in: International Conference on Power Systems Transients, New Orleans, USA,
2003.
Chisholm W. A. and Janischewskyj W., ―Lightning surge response of ground electrodes,‖ IEEE Trans.
Power Delivery, vol. 4, no. 2, pp. 1329–1337, Apr. 1989.
Chisholm W. A., Chow Y. L., and Srivastava K. D. ―Travel time of transmission towers,‖ IEEE
Trans. Power App. Syst., vol. PAS–104, no. 10, pp. 2922–2928, Oct. 1985.
Chowdhuri P., ―Grounding for protection against lightning,‖ in Electromagnetic transients in power
systems. Research Studies Press Ltd., John Wiley &sonsinc., New York, 1996, pp. 104-113.
Christodoulou C.A., Avgerinos M.V., Ekonomou L., Gonos I.F., Stathopulos I.A., ―Measurement of
the resistive leakage current in surge arresters under artificial rain test and impulse voltage subjection‖,
IET Science, Measurement & Technology, Vol. 3, Issue 3, pp. 256-262, May 2009.
Christodoulou C.A., Ekonomou L., Fotis G.P., Gonos I.F., Stathopulos I.A., ―Assessment of surge
arrester failure rate and application studies in Hellenic high voltage transmission lines‖, Electric Power
Systems Research, Vol. 80, Issue 2, pp. 176-183, February 2010.
Christodoulou C.A., Perantzakis G., Spanakis G.E., Karampelas P., ―Evaluation of lightning
performance of transmission lines protected by metal oxide surge arresters using artificial intelligence
Techniques‖, Springer, Energy Systems, Vol. 3, Issue 4, pp. 379-399, December 2012.
CIGRE Working Group 33.01, ―Guide to procedures for estimating the lightning performance of
transmission lines,‖ Technical Bulletin 63, Oct. 1991.
Darveniza M., Popolansky F., and Whitehead E. R. ―Lightning protection of UHV lines,‖ Electra,
no. 41, pp. 39–69, Jul. 1975.
Fernandez F., Diaz R., Metal oxide surge arrester model for fast transient simulations, in: International
Conference on Power System Transients (IPAT‘ 01), Rio De Janeiro, Brazil, 2001, Paper 14.
Francisco de Le6n, Semlyen Adam, ―COMPLETE TRANSFORMER MODEL FOR
ELECTROMAGNETIC TRANSIENTS‖, IEEE Transactions on Power Delivery, Vol. 9, No.
1, January 1994.
Garcia-Gracia M., Baldovinos S., Sanz M., Montanes L., Evaluation of the failure probability for
gapless metal oxide arresters, in: IEEE Transmission and Distribution Conference, 1999.
Grigsby Leonard L., ―Electric Power Engineering Handbook‖, CRC Press, Taylor & Francis, 2007
Group
Gutierrez J. A., Moreno P., Guardado L., Naredo J. L., ―Comparison of transmission tower models for
evaluating lightning performance‖, in Proc. Power Tech, Bologna, Italy, 2003, pp. 1-6.
Hara T. and Yamamoto O., ―Modeling of a transmission tower for lightning–surge analysis,‖ IEE
Gener. Transm. Distrib., vol. 143, no. 3, pp. 283–289, May 1996.
PhD Thesis
Page [90 of 211]
 CITY , UNIVERSITY of LONDON

Hinrichsen V., Metal-Oxide Surge Arresters, Fundamentals, 1st ed., Siemens, 2001
Hinrichsen V.: ‗Monitoring of high voltage metal oxide surge arrester‘. VI Jornadas Internacionales de
Aislamiento Electrico, Bilbao, 22–23 October 1997 paper 6.4
IEC 60071-2, Insulation co-ordination – Part 2: Application guide, 1996
IEC 60099-4, Surge Arresters—Part 4: Metal-Oxide Surge Arresters WithoutGaps for a.c. Systems,
2nd ed., 2004–2005.
IEEE Std 1243-1997: IEEE guide for improving the lightning performance of transmission lines
(1997)
IEEE Std 80-2000, ―IEEE Guide for Safety in AC Substation Grounding‖, January 2000
IEEE Std 998-1996, ―IEEE Guide for Direct Lightning Stroke Shielding of Substations‖, 1996
IEEE Std 998-1996, ―IEEE Guide for Direct Lightning Stroke Shielding of Substations‖, 1996
IEEE Working Group 3.4.11, Modeling of metal oxide surge arresters, IEEE Trans. Power Delivery
7(1) (1992) 302–309.
IEEE Working Group on Estimating Lightning Performance of Transmission Lines, ―IEEE guide for
improving the lightning performance of transmission lines,‖ Proposed IEEE P1243 draft, 1996.
IEEE Working Group, ―A Simplified method for estimating lightning performance of transmission
lines,‖ IEEE Trans. Power App. Syst., vol. PAS-104, no. 4, pp. 919-932, Apr. 1985.
IEEE Working Group, ―Estimating lightning performance of transmission lines II – Updates to
analytical models,‖ IEEE Trans.Power Delivery, vol. 8, no. 3, pp. 1254-1267, Jul. 1993.
Ishii M., Kawamura T., Kouno T., Ohsaki E., Shiokawa K., Murotani K., and Higuchi T.,
―Multistory transmission tower model for lightning surge analysis,‖ IEEE Trans. Power Delivery, vol. 6,
no. 3, pp. 1327–1335, Jul. 1991.
James R.E., Su Q.: ‗Condition assessment of high voltage insulation in power system equipment‘ (IET
Power and Energy Series 53, 2008, 1st edn.)
Joint Working Group 33/23.12, ―Insulation co-ordination of GIS: Return of experience, on site tests
and diagnostic techniques,‖ Electra, no. 176, pp. 67-97, Feb. 1998.
Jordan C. A., ―Lightning computations for transmission lines with overhead ground wires,‖ Gen. Elec.
Rev., vol. 37, 1934.
Korsuntchev A. V., ―Application of the theory of similitude to the calculation of concentrated earth
electrodes,‖ Electrichestvo, no.5, pp. 31-35, May 1958.
Laurent, P. G., ―Les Bases Generales de la Technique des Mises a la Terre dans les Installations
Electriques,‖ Bulletin de la Societe Francaise des Electriciens, vol. 1, ser. 7, pp. 368–402, July 1951.
Lundquist J., Stenstrom L., Schei A., Hansen B.: ‗New method for measurement of the resistive leakage
currents of metal oxide surge arresters in service‘, IEEE Trans. Power Deliv., 1990, 5, (4), pp. 1811–
1822
Martinez J.A., Castro-Aranda F., Lightning performance analysis of overhead transmission lines using
the EMTP, IEEE Trans. PWRD 20 (3) (2005) 294–300.
Martinez-Velasco Juan A., ―Transient Analysis of Power Systems Solution Techniques, Tools and
Applications‖, 2015
McDonald, J. et al., ―Electric Power Engineering Handbook‖, CRC Press, Boca Raton, FL, 2000
McDonald, J., ―Substation integration and automation—fundamentals and best practices‖,
IEEEPower& Energy, 1, March 2003
Menemenlis C. and Chun Z. T., ―Wave propagation on no uniform lines,‖ IEEE Trans. Power App.
Syst., vol. PAS–101, no. 4, pp 833–839, Apr. 1982.
Montanes L., Garcia-Gracia M., Sanz M., Garcia M.A., An improvement for the selection of surge
arresters based on the evaluation of the failure probability, IEEE Trans. Power Delivery 17 (1) (2002)
123–128.
Motoyama H., Shinjo K., Matsumoto Y., and Itamoto N., ―Observation and analysis of multiphase
PhD Thesis
Page [91 of 211]
 CITY , UNIVERSITY of LONDON

back flashover on the Okushishiku test transmission line caused by winter lightning,‖ IEEE Trans.
PowerDelivery, vol. 13, no. 4, pp. 1391–1398, Oct. 1998.
Nakada K., Yokoyama S., Yokota T., Asakawa A., Kawabata T., Analytical study on prevention for
distribution arrester outages caused by winter lightning, IEEE Trans. Power Delivery 13 (4) (1998)
1399–1404.
Nieman, J., ―Unstellung von Hochstspannungs-Erdungsalagen Aufden Betrieb Mit Starr Geerdetem
Sternpunkt,‖ Electrotechnische Zeitschrift, vol. 73, no. 10, pp. 333–337, May 1952.
Nieman, J., ―Unstellung von Hochstspannungs-Erdungsalagen Aufden Betrieb Mit Starr Geerdetem
Sternpunkt,‖ Electrotechnische Zeitschrift, vol. 73, no. 10, pp. 333–337, May 1952.
Oettle E. E., ―A new general estimation curve for predicting the impulse impedance of concentrated earth
electrodes,‖ IEEE Trans.Power Delivery, vol. 3, no. 4, pp 2020–2029, Oct. 1988.
Picher, P.; Lapworth, J.; Noonan, T.; Christian, J. et al, ―Mechanical Condition Assessment of
Transformer Windings using Frequency Response Analysis,‖ Cigre Report 342, 2008.
Pigini A., Rizzi G., Garbagnati E., Porrino A., Baldo G., and Pesavento G., ―Performance of large
air gaps under lightning overvoltages: Experimental study and analysis of accuracy of predetermination
methods,‖ IEEE Trans. Power Delivery, vol. 4, no. 2, pp 1379–1392, Apr. 1989
Pinceti P., Giannettoni M., A simplified model for zinc oxide surge arresters, IEEE Trans. Power
Delivery 14 (2) (1999) 393–398.
Pinceti P., Giannettoni M., A simplified model for zinc oxide surge arresters, IEEE Trans. Power
Delivery 14 (2) (1999) 393–398.
Piotr ORAMUS, Marek FLORKOWSKI, Simulations of lightning overvoltages in HV electric power
system for various surge arresters and transmission lines models, ISSN 0033-2097, R. 90 NR
10/2014
Dommel, H., et al., Electromagnetic Transient Program Reference Manual (EMTP Theory Book),
1986
IEEE Fast Front Transients Task Force, Modeling Guidelines for Fast Front Transients, IEEE
Transactions on Power Delivery, Vol. 11, No. 1, January 1996
Marti J.R., ―Accurate modeling of frequency-dependent transmission lines in electromagnetic transient
simulations,‖ Power Engineering Review, IEEE, vol. PER-2, no. 1, pp. 29 –30, 1982
Rahimpour Ebrahim and Bigdeli Mehdi, ―Simplified transient model of transformer based on geometrical
dimensions used in power network analysis and fault detection studies‖, POWERENG 2009, Lisbon,
Portugal, March 18-20, 2009
Rondon D., Vargas M., Herrera J., Montana J., Torres H., Camargo M., Jimenez D., and Delgadillo
A., ―Influence of grounding system modelling on transient analysis of transmission lines due to direct
lightning strike,‖ in Proc. Power Tech, St. Petersburg, Russia, 2005, pp. 1-6.
Rüdenberg, R., ―Grounding principles and practices—Part 1, Fundamental considerations on grounding
currents,‖ Electrical Engineering, vol. 64, no. 1, pp. 1–13, Jan. 1945.
Sargent M. A. and Darveniza M., ―Tower surge impedance,‖ IEEE Trans. Power App. Syst., vol.
PAS–88, no. 5, pp. 680–687, May 1969.
Schwarz, S. J., ―Analytical expression for resistance of grounding systems,‖ AIEE Transactions on
Power Apparatus and Systems, vol. 73, no. 13, part III-B, pp. 1011–1016, Aug. 1954.
Siemens, Energy Sector, ―Power Engineering Guide: Answers for Energy‖, 5th Edition, 2008
Spanias C.A., Christodoulou C.A., Gonos I.F., Stathopulos I.A., ―Electric Field Measurements
around a Metal Oxide Surge Arrester‖, Proceedings of the 2010 International Conference on High-
Voltage Engineering and Application (ICHVE 2010), New Orleans, Luisiana, USA, October 11-
14, 2010, pp. 278-281
Suljanovic N., Mujcic A., Murko V., Practical issues of metal-oxide varistor modeling for numerical
simulations, in: 28th International Conference on Lightning Protection, Kazanaqa, Japan, 2006, pp.
1149–1154.
PhD Thesis
Page [92 of 211]
 CITY , UNIVERSITY of LONDON

Sunde, E. D., Earth conduction effects in transmission systems, New York: McMillan, 1968.
Sverak, J. G., ―Simplified analysis of electrical gradients above a ground grid; Part I—How good is the
present IEEE method?‖ IEEE Transactions on Power Apparatus and Systems, vol. PAS-103, no. 1,
pp.7–25, Jan. 1984.
Tarchini J.A., Gimenez W., Line surge arrester selection to improve lightning performance of transmission
lines, IEEE Bologna PowerTech Conference, Bologna, Italy, 2003.
Wadhwa C.L., ―High Voltage Engineering‖, New Age International Publishers, 2nd Edition, 2007
Wagner C. F. and Hileman A. R., ―A new approach to calculation of lightning performance of
transmission lines III – A simplified method: Stroke to tower,‖ AIEE Trans. Power App. Syst., vol.
79, pp. 589–603, Oct. 1960.
Weck K. H., ―Lightning performance of substations,‖ CIGRE SC 33, Rio de Janeiro, Brazil, 1981.
Weck K. H., ―The current dependence of tower footing resistance,‖ CIGRE 33-88 (WG01), 14 IWD,
1988.
Working Group 02 (Internal overvoltages) of Study Committee 33 (Overvoltages and Insulation
Coordination), ―Guidelines for Representation of Network Elements when Calculating Transients,‖
Cigré Report 39, 2000.
Yamada T., Mochizuki A., Sawada J., Zaima E., Kawamura T., Ametani A., Ishii M., and Kato
S., ―Experimental evaluation of a UHV tower model for lightning surge analysis,‖ IEEE Trans.
PowerDelivery, vol. 10, no. 1, pp 393–402, Jan. 1995.
Yasuda Y., Hirakawa Y., Shiraishi K., and Hara T., ―Sensitivity analysis on grounding models for
500kV transmission lines,‖ Trans.IEE Japan B, vol. 121, no. 10, pp. 1386–1393, 2001.
Zanetta L.C., Pereira C.E. de M., Application studies of line arresters in partial shielded 138-kV
transmission lines, IEEE Trans. Power Delivery 18 (1) (2003) 95–100.
Zitnik B., Zitnik M., Babuder M., The ability of different simulation models to describe the behavior of
metal oxide varistors, in: 28th International Conference on Lightning Protection, Kazanaqa, Japan,
2006, pp. 1111–1116.

PhD Thesis
Page [93 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 4
DESCRIPTION OF THE EXAMINED SYSTEMS

4.1 Introduction

Chapter 4 deals with the topology of the systems under examination. The basic equipment of
the substation is presented in details, providing the necessary electrical and geometrical
characteristic of each component. The depicted configurationσ correspond to typical
substations 150/20kV of the Hellenic electrical network. Note, that the substations are
interconnected, in order to increase the reliability of the system and ensure the uninterrubtible
operation of the system. Finally, the most used equivalent circuit models of each component
are presented.

4.2 Topology of the Examined Substations

In the current thesis a lightning performance study for substation will be carried out,
considering different configurations. Figures 4.1 – 4.3 depict the topology of the examined
cases, presenting the basic parts of the substations.

case a: arresters installed at position B and D

A case b: arresters installed at position D

transformer

B D C
cable

overhead transmission lines

Configuration (a)(Simple Topology of a 150/20kv Substation)

PhD Thesis
Page [94 of 211]
 CITY , UNIVERSITY of LONDON

Configuration (b)(Simulation Model)

Figure 4.1 Configuration1 (Simple Topology)and (Simulation Model) of a 150/20kv Substation

Configurations (a) and (b) in the Figure 4.1 are: (a) is a simple topology of a 150/20kV
substation and (b) is the Simulation Model. The basic components of the substation are:

 High voltage overhead transmission line

 High voltage cable
 Power transformer
 Busbars
 Surge arresters
 Grounding systems
 Circuit breakers.

A lightning flash current is injected in position A of the overhead transmission line, supposing
that substation outages are mainly caused by impinging overvoltage surges, incoming from the
connected overhead transmission lines, which exceeding the insulation level of the substation
equipment result in interruptions in power supply and consequently in economic losses

PhD Thesis
Page [95 of 211]
 CITY , UNIVERSITY of LONDON

(Mikropoulos et al. 2010). The substation is adequately protected against direct lightning
flashes by ground wires and masts, according to methodologies described in Chapter 3.
In details, a 150kV single-circuit overhead transmission line is connected with a HV cable, in
order to feed the power transformer. Each switchgear bay includes in practice the full
complement of disconnecting switches (to isolate equipment or to redirect current flow),
earthing switches (to protect the personnel), instrument transformers (voltage and current
measurement) and control and protection equipment (to protect the substation against short-
circuit currents). A lightning hits either a phase conductor or an overhead ground wire at
position A, considering that the majority of the substations failures happen due to shielding
failures or backflashover on the lines. The developed voltage surge travels through the
conductor and the cable to the substation‟s transformer. The developed overvoltages at the
joint position and at the terminals of the transformer are estimated. A sensitivity analysis is
performed, considering
 the installation position of the arresters (distance between the transformer and the
installed arrester CD=0m, 15m, 30m, 45m, 60m)
 the installation of arresters at the joint of overhead line and underground cable (case a and
case b)
 the length of the underground cable (BD=300m, 600m, 1000, 2000m)
 the tower footing resistance (R=1Ω, 5Ω, 10Ω, 15Ω, 20Ω, 25Ω)
Note, that an overhead transmission line of 15km is regarded, since the span between the
towers is considered equal to 300m.Moreover,a constant substation grounding resistance equal
to 1Ω has been supposed. The achievement of low grounding resistance values is a decisive
parameter, in order to ensure the adequate lightning protection of the substation.

Configuration 2 is a more complex topology, consisting of two interconnected transformers

(substation 1 is connected with substation 2 by using a high voltage cable). An overhead
transmission line is undergrounded at terminals I and II, which are connected by using a high
voltage cable. A lightning hits either the overhead transmission line 1 (case a) or the overhead
transmission line 2 (case b); in any case the developed overvoltages at various positions of the
examined system are estimated, performing a sensitivity analysis for the role of the grounding
resistance. Scope of the study is to indicate that the installation of external lighting protection
system does not guarantee the elimination of the lightning faults, since lightning strokes on the
connected transmission lines can create significant surges that stress the components of the
equipment (cables, transformers etc.) and can result to faults and damages.

PhD Thesis
Page [96 of 211]
 CITY , UNIVERSITY of LONDON

E
cable 1, 2km F

D B
CASE II . cable 4, 1km cable 3, 3km

CASE I
cable 2, 6km
C
overhead overhead
transmission line 1 transmission line 2

Figure 4.2 Configuration 2

4.3 Characteristics of the Substations’ Components

4.3.1 Transmission lines

The overhead transmission line has nominal voltage of 150kV (phase-phase, rms) and are
consisting of:
 The Metallic Towers.
 The Phase Conductors.
 The Overhead Ground Wires.
 The Insulators.
 The Grounding System.

Figure 4.3 depicts the geometrical configuration of a typical tower of a single circuit 150kV
overhead transmission line of the Greek electrical network . The insulators are consisting of
ten ceramic discs, offering an insulation level of 750kV. The overhead ground wires are
installed in a way to protect the phase conductors against external overvoltages, according to
the electrogeometrical model.

PhD Thesis
Page [97 of 211]
 CITY , UNIVERSITY of LONDON

Figure 4.3 Tower of 150kV Transmission Line

The cross-section of the Aluminium conductor steel-reinforced (ACSR) conductors is

636MCM, since the nominal current is 789A. Fig. 4.4 presents the configuration of the
grounding system of the line. The span between the towers is regarded equal to 300m.

Fig. 4.4 the Grounding System of the Tower (ABCD the Basis of the Tower, AE=DI=BF=CG=60m)

PhD Thesis
Page [98 of 211]
 CITY , UNIVERSITY of LONDON

4.3.2 Cables

Cables connect the overhead transmission lines to the substations; they also connect the
terminals of configuration 2. The conductors of the cables are made of aluminum (Al), their
cross section is 800mm2 and the nominal current is 830A. The insulation of the cable is made
of XLPE and the external layer is made of PVC. The cables are buried and the connection
between the different parts of the cables and the substations is performed by using
appropriate joints.

4.3.3 Power Transformers

The power transformers are the main part of the substation. Details about power transformers
are given in Chapter 3. Table 4.1 presents the electrical characteristics of the used power
transformers.

Table 4.1Electrical Characteristics of Power Transformer

Rated Power 50MVA

Rated Frequency 50Hz

Primary Rated Voltage 150kV

Secondary Rated Voltage 20kV

Vector Group Dyn1

4.3.4 Surge Arresters

Surge arresters are installed at the entrance of the substation, in order to protect the
equipment and keep the developed overvoltages below the Basic Insulation Level (BIL 750kV
for a system with nominal voltage 150kV). Table 4.2 presents the electrical characteristics of
the used surge arresters, which are installed at the high voltage side of the substations.

PhD Thesis
Page [99 of 211]
 CITY , UNIVERSITY of LONDON

Table 4.2 Electrical Characteristics of Surge Arresters

Maximum Continuous Operating Voltage 86kV

Rated Voltage 108kV

5kA 242kV
10kA 254kV
Residual Voltage
20kA 280kV
40kA 313kV
Discharge energy class Class 3
Energy Capability 8 kJ/kV

4.4 Simulation Models of the Substations’ Components

4.4.1 Circuit Representation of Overhead Transmission Lines and Cables

The circuit representation of overhead high voltage transmission lines is performed by using
the distributed parameter line model of Simulink, based on the Bergeron‟s theory. The
distributed parameter line block uses an N-phase distributed parameter line model with
lumped losses, based on the Bergeron's Traveling Wave Method (Dommel 1969). The lossless
distributed line is characterized by the surge impedance (ZC )and the phase velocity (v),
according to the equations:

√ (4.1)

(4.2)
√

Where;
l is the per unit length inductance of the line.
c is the per unit length capacitance of the line.

The model considers that the voltage at the start of the line should arrive at the end of the line
without changes, after a time given by the equation(Dommel 1969):

PhD Thesis
Page [100 of 211]
 CITY , UNIVERSITY of LONDON

(4.3)

Where;
τ is the transport delay of the voltage wave.
v is the propagation velocity of the wave.
d is the length of the line.

By lumping R/4 at both ends of the line and R/2 in the middle and using the current injection
method the circuit of Figure4.5 is extracted.

is ir

es Z Ish Irh Z er

Figure 4.5 Two-Port Equivalent Circuit Model of a Transmission Line

The above equivalent circuit is governed by the equations(Dommel 1969):

( ) ( ) ( ( ) ( )) ( ) ( ( ) ( ))(4.4)

( ) ( ) ( ( ) ( )) ( ) ( ( ) ( ))(4.5)

(4.6)

(4.7)

Where;
ZC is the surge impedance
τ is the transport delay of the voltage wave.
r is the per unit length resistance of the line.
his the height of the Transmission Line.
PhD Thesis
Page [101 of 211]
 CITY , UNIVERSITY of LONDON

4.4.2 Circuit Representation of Towers

The tower of the transmission lines were represented by the Chisholm et al. model, presented
in (Chisholm et al. 1985), according to the equation:

( ( ( ))) (4.8)

Where;
r is the tower base radius(m).

Six more tower models are also implemented, in order to examine the influence of the used
tower equivalent circuit model to the obtained simulation results. These are:

 Jordan (Jordan 1934).

 Wagner & Hileman (Wagner and Hileman 1960).
 Sargent & Darveniza (Sargent and Darveniza 1969).
 Ametani et al. (Ametani et al. 1994).
 Rondon et al. (Rondon et al. 2005).
 Yamada et al. (Yamada et al. 1995).

4.4.3 Circuit Representation of Surge Arresters

Metal oxide gapless surge arresters cannot be modeled only as non-linear resistances, since
their response, i.e. their residual voltage for a given current, is a function of the magnitude and
the slope of the injected pulse. The accuracy of the results is strongly depended on the
adjustment of the parameter values, for each model. In the current thesis, metal oxide gapless
surge arresters are represented by using the IEEE model (IEEE WG 3.4.11 1992), which
takes into consideration the dynamic behavior of the nonlinear resistors of the protective
device. The IEEE model is widely used for lightning performance studies and gives adequate
results. Moreover, two more circuit models are used for sensitivity analysis purposes, i.e. the
Pinceti-Giannettoni model and the Fernandez – Diaz model (Pinceti and Giannettoni 1999,
Fernandez and Diaz 2001). These models are simplified versions of the IEEE model and they
use different methodology for the computation of the circuit‟s parameters.

PhD Thesis
Page [102 of 211]
 CITY , UNIVERSITY of LONDON

4.4.4 Circuit Representation of Transformers, Busbars and Circuit Breakers

Substation apparatus and busbars also are modeled as frequency dependent
elements,according to (IEEE Task Force1996). In details, busbars are modeled as distributed
parameters line of impedance equal to 50Ω, since the power transformer is represented by the
circuit of Fig. 4.6 (CD=460pF, CE=2000pF, R1=148μΩ, L1=18mH). As far as the circuit
breakers is concerned, a capacitance equal to 50pF has been implemented. Fig. 4.6 depicts the

Figure 4.6: High frequency model of the HV transformer side (CD HV side widings capacitance, CE, R1, L1
capacitance, resistance and inductance of HV bushings)

4.4.5 Circuit Representation of Grounding System

The appropriate modeling of the grounding system of the tower and the substation is a great
importance, in order to obtain reliable results for the lightning performance of the system.
The achievement of low values of grounding resistance reduces the developed overvoltages
due to lightning hits and limits the probability of serious faults and damages. It is worth
mentioning that the grounding resistance depends on the soil resistivity, which is usually
known during the initial design stage. Furthermore, the use of the total fault current as the
current discharge by the grounding system is a contestable approach, since results in
overestimations and expensive solutions. The division of the fault current depends on (IEEE
St 80 2000):

 The number of transmission lines with shield wires tied to the ground grid
 The substation grounding resistance
 The number of power source and non-power source terminals
 The soil resistivity
 The self-impedance of shield wire
 The tower footing resistance.
 The mutual impedance between faulted phase conductor & shield wires.
PhD Thesis
Page [103 of 211]
 CITY , UNIVERSITY of LONDON

As far as the circuit modeling grounding resistance of the towers is concerned, the
corresponding Cigre model(CIGRE W.G. 33.01 1991)from Weck (Weck 1988)) is used,
considering the soil ionization phenomenon (ρ=200Ωm) (see Table 3.2).

4.4.6 Circuit Representation of Insulators

The simplest approach for the representation of the insulators is to model them as a switch,
which closes when the voltage exceeds a defined limit. Indeed, the developed overvoltages
stress the insulators of the line, resulting in a flashover in case of exceedance of the BIL
(Figure 4.7). However, the flashover depends on various factors, such as the shape of the
insulating disk, the pollution of the surface of the discs etc. In the current work The flashover
strength VFO(kV) is determined by the equation (3.4), which includes the insulator string
length and the elapsed time after a lightning hit. Furthermore, the leader progression models
of Table 3.3 are used, in order to perform a sensitivity analysis about the role of the used
models.

insulating flashover
disc

phase
conductor

Fig. 4.7 Insulation Flashover of a 150 kV Transmission Line

PhD Thesis
Page [104 of 211]
 CITY , UNIVERSITY of LONDON

4.5 Conclusions

Chapter 4 discussed the topology of the systems under examination. In the current thesis a
lightning performance study for substation will be carried out, considering different
configurations. The depicted configuration correspond to typical substations 150/20kV of the
Hellenic electrical network. The substations were interconnected, in order to increase the
reliability of the system and ensure the uninterruptible operation of the system. The basic
equipment of the substation was presented in details, providing the necessary electrical and
geometrical characteristic of each component. Finally, the most used equivalent circuit
models of each component were presented.

PhD Thesis
Page [105 of 211]
 CITY , UNIVERSITY of LONDON

References

Ametani A., Kasai Y., Sawada J., Mochizuki Α., Yamada Τ., ―Frequency–dependent impedance of
vertical conductors and a multiconductor tower model,‖ IEE Gener. Transm. Distrib., vol. 141, no. 4,
pp 339–345, Jul. 1994.
Chisholm W. A. and Janischewskyj W., ―Lightning surge response of ground electrodes,‖ IEEE Trans.
Power Delivery, vol. 4, no. 2, pp. 1329–1337, Apr. 1989.
Chisholm W. A., Chow Y. L., Srivastava K. D. ―Travel time of transmission towers,‖ IEEE Trans.
Power App. Syst., vol. PAS–104, no. 10, pp. 2922–2928, Oct. 1985.
CIGRE Working Group 33.01, ―Guide to procedures for estimating the lightning performance of
transmission lines,‖ Technical Bulletin 63, Oct. 1991.
Dommel, H., "Digital Computer Solution of Electromagnetic Transients in Single and Multiple
Networks," IEEE Transactions on Power Apparatus and Systems, Vol. PAS-88, No. 4, April,
1969
Fernandez F., Diaz R., Metal oxide surge arrester model for fast transient simulations, in: International
Conference on Power System Transients IPST‘01, Paper 14, Rio De Janeiro, Brazil, June 24–28,
2001.
IEEE Std 80-2000, ―IEEE Guide for Safety in AC Substation Grounding‖, January 2000
IEEE Task Force, ―Modeling guidelines for fast front transients,‖ IEEE Trans. Power Delivery, vol.
11, no. 1, pp. 493–506, Jan. 1996
IEEE Working Group 3.4.11, Modeling of metal oxide surge arresters, IEEE Transactions on Power
Delivery 7 (1) (1992) 302–309.
IEEE Working Group, ―A Simplified method for estimating lightning performance of transmission
lines,‖ IEEE Trans. Power App. Syst., vol. PAS-104, no. 4, pp. 919-932, Apr. 1985.
Jordan C. A., ―Lightning computations for transmission lines with overhead ground wires,‖ Gen. Elec.
Rev., vol. 37, 1934.
Mikropoulos P., Tsovilis T., Manousaridis I., Laloumis G., Dramis A., ―Lightning Risk Assessment
of a 170 kV GIS Substation connected to the Hellenic Transmission System through Underground
Cables‖, 7th Mediterranean Conference and Exhibition on Power Generation, Transmission,
Distribution and Energy Conversion (MedPower 2010), 7-10 Nov. 2010, Agia Napa.
Oettle E. E., ―A new general estimation curve for predicting the impulse impedance of concentrated earth
electrodes,‖ IEEE Trans.Power Delivery, vol. 3, no. 4, pp 2020–2029, Oct. 1988.
Pinceti P., Giannettoni M., A simplified model for zinc oxide surge arresters, IEEE Transactions on
Power Delivery 14 (2) (1999) 393–398.
Rondon D., Vargas M., Herrera J., Montana J., Torres H., Camargo M., Jimenez D., and Delgadillo
A., ―Influence of grounding system modelling on transient analysis of transmission lines due to direct
lightning strike,‖ in Proc. Power Tech, St. Petersburg, Russia, 2005, pp. 1-6.
Sargent M. A. and Darveniza M., ―Tower surge impedance,‖ IEEE Trans. Power App. Syst., vol.
PAS–88, no. 5, pp. 680–687, May 1969.
Wagner C. F. and A. R. Hileman, ―A new approach to calculation of lightning performance of
transmission lines III – A simplified method: Stroke to tower,‖ AIEE Trans. Power App. Syst., vol.
PhD Thesis
Page [106 of 211]
 CITY , UNIVERSITY of LONDON

79, pp. 589–603, Oct. 1960.

Yamada T., Mochizuki A., Sawada J., Zaima E., Kawamura T., Ametani A., Ishii M., and Kato
S., ―Experimental evaluation of a UHV tower model for lightning surge analysis,‖ IEEE Trans. Power
Delivery, vol. 10, no. 1, pp 393–402, Jan. 1995.
Yasuda Y., Hirakawa Y., Shiraishi K., and Hara T., ―Sensitivity analysis on grounding models for
500kV transmission lines,‖ Trans.IEE Japan B, vol. 121, no. 10, pp. 1386–1393, 2001.

CHAPTER 5
PhD Thesis
Page [107 of 211]
 CITY , UNIVERSITY of LONDON

CALCULATION OF THE DEVELOPED OVERVOLTAGES

5.1 Introduction

Scope of the presented analysis in the current Chapter is the calculation of the developed
overvoltages at different positions of the substations under study (presented in Chapter 4),
considering lightning hits on the phase conductors or the ground wires of the connected
overhead transmission lines. To this direction, the mechanisms of the development of the
incoming surges at the entrance of the substation due to shielding failure or backflashover are
analytically described. A sensitivity analysis is also performed, in order to examine the
influence of various parameters (grounding resistance, length of the cable, installation position
of the arresters, and number of the incoming lines) on the magnitude of the expected surges.
In addition, the impact of the used equivalent circuit modes of the equipment (transmission
lines, switches, arresters, cables, grounding system, insulators etc.) on the obtained results is
investigated.

5.2 Overvoltages in Case of Direct Lightning hit

In general, the substations are adequately protected against direct lightning hits, according to
the methodologies and the electrogeometrical models, presented in Chapter 4. Consequently,
lightning surges that impinge on the overhead transmission lines that are connected with the
substations are the main cause of overvoltage stresses of the insulation of the substations‟
equipment. In case that lightning hits directly the phase conductors of the transmission lines,
two travelling waves appear, the magnitude of which depends on the lightning peak current
and the surge impedance of the conductors (Figure 5.1). The overvoltage wave will arrive at
the entrance of the substation and can result in several serious damages of the components of
the substations. It is worth mentioning, that the developed overvoltage is not influenced by
the tower footing resistance and the unique protective measure is the installation of surge
arresters. Moreover, the position of the lightning hit, the length of the cable, the
characteristics of the implemented surge arresters, the position of the installation of the
arresters and the grounding resistance of the arresters are critical factors that determine the
efficiency of the lightning protection system and affect the lightning performance of the
substation.

PhD Thesis
Page [108 of 211]
 CITY , UNIVERSITY of LONDON

to the
to the substation
transmission line

phase conductor of the transmission

line
Figure 5.1: Lightning Hit on Phase Conductor of an Overhead Transmission Line

If surge arresters are installed between phase and earth of a transmission line, part of the
lightning current will be diverted to the grounding system (Figure 5.2), depending on the
achieved grounding resistance. Figure 5.3 depicts a surge arresters installed at the entrance of a
power transformer. The low values of the grounding resistance ensure that almost the total
current will pass through the arrester and the developed overvoltage will not exceed the
insulation level of the system.

metallic tower

insulator

surge arrester:
keeps U<BIL

to the to the substation

transmission line
phase conductor of the transmission
line
Figure 5.2: Overhead Transmission line Protected by Surge Arresters

PhD Thesis
Page [109 of 211]
 CITY , UNIVERSITY of LONDON

transformer

arrester

grounding
resistance

Figure 5.3: Power Transformer protected by Surge Arresters

The current analysis is performed for the configuration of Figure4.1. The amplitude of the
lightning surge which can occur is determined by using of the geometrical model, presented in
(A.R. Hileman, “Insulation coordination for Power Systems”, CRC Press, 1999). Time to half
values for the first and following strokes were selected according to (CIGRE, Working Group
01 (Lightning) of Study Committee 33 (Overvoltages and Insulation Coordination) “Guide to
Procedures for Estimating the Lightning Performance of Transmission Lines, brochure 63,
CIGRE, Paris, France, October,1991). The front time of the lightning surge is given by the
equation:

(5.1)
( )

where
I is the peak value of the lightning current,
P is the probability of a crest current exceeding the magnitude of I (H.M Ryan, “High voltage
engineering and testing”, Peter Peregrinus Ltd. 1st edition, 1994).
Applying the above procedure, a 35kA 7.5/200μs lightning current is applied for a lightning
hit on a phase conductor of the connected line. The developed voltage surge travels through
the conductor and the cable to the substation‟s transformer (150/20kV).

The developed overvoltages at the beginning (position B) and the end (position D) of the
cable are estimated, by using appropriate simulation tool(Simulink/MATLAB). Time-steps in
the range of 5-50ms are used in typical switching transients, ranging the simulation time from
PhD Thesis
Page [110 of 211]
 CITY , UNIVERSITY of LONDON

20-200ms. In lighting studies, the time-step depends upon the steepness of the surge, the
minimum length of travelling wave models, plus the use of flashovers gaps and surge arresters
with significant lead lengths. In the current work, it is in the range of 1-20 ns, with a
simulation length of 20-50 ms. A sensitivity analysis is performed, considering the length of
the cable, the installation position of the arresters and the tower footing. The towers of the
overhead lines and the installed metal oxide gapless surge arresters are modeled according to
the Chilsom model and the IEEE model, correspondingly. As far as the grounding resistance
is concerned the Cigre model is selected; finally, V-t curve is implemented for the circuit
representation of the insulators. Figures 5.4 – 5.5 depict the obtained results.

In details, the figures of results (Fig.5.4) presents the estimated overvoltages at the positions B
and D of the topology under study in function with the length of the cable (BD) in case that
the substation is not protected by surge arresters. The results indicate that long cables can be
adequately protected, since the expected overvoltages do not exceed the BIL (750kV for
150kV nominal voltage).

Fig. 5.4 The developed overvoltages at positions B (UB) and D (UD) in function with the length of the cable
(BD) in case of lightning hit on phase conductor of the connected line (no arresters)

Regarding the surge impedance of the cable and the transmission line equal to 40Ωand 350Ω,
respectively, a reflection factor equal to -0.8 is estimated; that means that the 20% of the initial
surge (U) comes into the cable. However, at position D (connection between cable and
transformer) the travelling wave is doubled and will be reflected. The reflected wave (0.4U) is
also reflected at position B (connection between cable and overhead transmission line)
considering a reflection factor 0.8. Thus, the developed surge (0.8 0.4U=0.32U) is
PhD Thesis
Page [111 of 211]
 CITY , UNIVERSITY of LONDON

superimposed to the initial surge 0.4U, resulting in the increase of the voltage at position D
(0.4U+0.32U=0.72U). The above procedure is repeated and the developed overvoltage may
exceed the withstand capability of the insulation.

The overvoltage at the positions B and D is given as (Siemens 2008, ABB 2007, Hinrichsen V
2001):

( )
( )( ) (5.2)
( )( )( ) )
{

Where;

( )

( )

( )

dis the length of the cable

C is the propagation speed.

[ ( )]

[ ( ) ( )( )] ( )
[ ( )( ) ( )( ( ) )]
{

Furthermore, the installation position of the arresters is of great importance for the
appropriate lightning performance of the substation. The increase of the distance between the
arrester and the transformer reduces the effectiveness of the installed arresters, in a way that
the overvoltage may exceed the BIL, depending on the other parameters, i.e. the magnitude of
the surge, the grounding resistance and the length of the cable. Especially, in case of cables of
length lower that 1000m the problem is more intense, since the initial surge is higher and
a non-appropriately installed arrester will not protect the equipment. For this reason, the
PhD Thesis
Page [112 of 211]
 CITY , UNIVERSITY of LONDON

protection distance of the arresters (see Chapter 3) has to be taken into consideration during
the initial design of the installation and the selection of the characteristics of the arresters.

The role of the grounding resistance is also crucial in order to limit the overvoltages at the
entrance of the substation. In general, the voltage that is implemented at the primary of the
primary transformer is given by the equation(ABB 2007):

( )

Where;

i is the lightning current.

L is the inductance of the connection wires of the arrester.

Rsa is the grounding resistance of the arrester.

Utrasformer is the voltage at the terminals of the transformer.

Uarrester is the residual voltage of the arrester.

Figures 5.5-5.8 present the results of the performed sensitivity analysis for cases a and
bregarding the following parameters:

 The Installation Position of the Arresters

 The Grounding Resistance of the Arresters
 The Length of the Cable.

In details :

- The Installation Position of the Arresters: The distance between the transformer and
the arrester varies between 0m and 60m (0m, 5 m, 15 m, 30 m, 45 m, 60m)
- The Grounding Resistance of the towers: The grounding resistance varies between
1Ωand 25Ω(1 Ω, 5 Ω, 10 Ω, 15 Ω, 20 Ω, 25Ω)

- The Length of the Cable: The length of the cable ranges between 300m and 2000m
(300 m, 600 m, 1000 m, 2000m).

PhD Thesis
Page [113 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.5: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the cable
length (BD) in case of a Lightning hit on the Phase Conductor of the Connected Transmission Line (case a)

Figure 5.6: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the cable
length (BD) in case of a Lightning hit on the Phase Conductor of the Connected Transmission Line (case a)

PhD Thesis
Page [114 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.7: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the cable
length (BD) in case of a Lightning hit on the Phase Conductor of the Connected Transmission Line (case b)

Figure 5.8: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the cable
length (BD) in case of a Lightning hit on the Phase Conductor of the Connected Transmission Line (case b)

The performed sensitivity analysis results in the following conclusions, about the behavior of
the substation in case of a lightning hit on the phase conductor of the connected overhead
transmission line:

 Τhe developed overvoltages at the end of the cable are greater than those ones at the
beginning, due to the reflection phenomena of the traveling waves.

 The developed overvoltages are strongly dependent on the length of the cable. Long
cables present a better lightning performance.
PhD Thesis
Page [115 of 211]
 CITY , UNIVERSITY of LONDON

 Tower footing resistance has no influence on the expected overvoltages.

 The arresters should be installed near the transformer; otherwise the voltage at the
equipment to be protected will be considerably higher than the residual voltage at the
terminals of the arrester.

 The installation or not of surge arrester at the joint between the incoming overhead
transmission line and the underground cable definitely influences the magnitude of the
expected overvoltages. Surge arresters at position B contribute significantly to the
reduction of the developed surges and the improvement of the lightning performance of
the system.

5.3 Overvoltages in Case of Lightning hit on the ground wire

When a lightning hits the ground wire or the tower of an overhead transmission line, the
magnitude of the developed overvoltage depends on the tower footing resistance, the
induction of the tower and the rise time of the injected lightning current (Figure 5.9). If the
overvoltage exceeds the insulation level of the line, a backflashover is occurred. The resulting
surge propagates to the connected substation and can lead to serious malfunctions of the
system. Backflashover at the connected transmission lines is a common reason of faults and
damages of the substations‟ equipment.

metallic tower

insulator

if U=K(Ri+Ldi/dt)>BIL 
backflashover

to the to the substation

transmission line
phase conductor of the transmission
line

Figure 5.9: Lightning hit on the Metallic Tower of an Overhead Transmission Line
PhD Thesis
Page [116 of 211]
 CITY , UNIVERSITY of LONDON

The lightning performance of the system can be improved by installing surge arresters in
parallel with the insulators and at the primary side of the transformer (Figure 5.10). Surge
arresters divert the current of the lightning strike to earth and restrain the voltage at the
terminals of the equipment below the BIL. The installation of metal oxide gapless surge
arresters is necessary; else the incoming voltage waves will stress the insulation of the
equipment, resulting in faults and interruption of the normal operation of the substation.

metallic tower

insulator

surge arrester:
keeps U<BIL

to the to the substation

transmission line
phase conductor of the transmission
line

Figure 5.10: Installation of Surge Arresters at an Overhead Transmission Line

Applying the procedure presented in the previous section (equation 5.1), a 200kA, 10.5/200μs
lightning current is implemented in case of hit on overhead ground wire of the incoming
150kV overhead transmission line. If the arising voltage exceeds the basic insulation level
(750kV) then a backflashover is occurred. The developed voltage surges travel through the
conductor and the cable to the substation‟s transformer (150/20kV). The developed
overvoltages at the beginning (position B) and the end (position D) of the cable are estimated,
by using appropriate simulation tool (Simulink/MATLAB). The towers of the overhead lines
and the installed metal oxide gapless surge arresters are modeled according to the Chilsom
model and the IEEE model, correspondingly. As far as the grounding resistance is concerned
the Cigre model is selected; finally, V-t curve is implemented for the circuit representation of
the insulators. A sensitivity analysis is performed, considering the length of the cable, the
installation position of the arresters and the tower footing resistance.

It is worth mentioning that the selected level of the injected lightning current is extremely
higher compared to the considered one in case of direct hit, since according to the applied

PhD Thesis
Page [117 of 211]
 CITY , UNIVERSITY of LONDON

electrogeometrical model, high currents strike the towers or the overhead ground wires and
low currents hit the phase conductors.

A lightning current of 200kA peak may create an overvoltage above the defined limit,
influenced by the grounding resistance, the insulators and the implementation (or not) of
arresters. Figures 5.11 and 5.12 present the estimated overvoltages at position B and D of the
configuration of Figure 4.1, taking into consideration the length of the cable (BD).The role of
the tower footing resistance is critical, since it determines the occurrence or not of the
backflashover and influences the magnitude of the voltage surge. Moreover, the length of the
cable also defines the lightning performance of the system, in a way that long cables are less
vulnerable to stresses due to lightning hits.

Figure 5.11: The Developed Overvoltage at Position D (UD) In function with Tower Footing Resistance
(Lightning hit on the Ground Wire of the Connected Transmission Line, No Arresters Installed)

PhD Thesis
Page [118 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.12: The Developed Overvoltage at Position B (UB) in function with the Tower Footing Resistance
(Lightning hit on the Ground Wire of the Connected Transmission Line, No Arresters Installed)

The following figures 5.13-5.20 for case a (see Fig. 4.1) depict the calculated voltages in case
that appropriate surge arresters have been installed. The achievement of low grounding
resistance values is a decisive parameter, in order to ensure the adequate lightning protection
of the substation. Note that a constant grounding resistance of 1Ωof the substation has been
supposed. Surge arresters cannot offer the demanded protection level, if the grounding system
is not appropriate. Consequently, the installation of surge arresters cannot always warrant the
reduction of the expected overvoltages.

The efficiency of the surge arresters is also reinforced by the length of the cable. The
calculations show that the estimated overvoltages are strongly dependent on the distance
(BD). In addition, the installation position of the arresters is of great importance, regarding
the protective distance, according to the analysis of Chapter 3.

PhD Thesis
Page [119 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.13: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=300m,
case a)

Figure 5.14: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the
Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line,
BD=600m, case a)

PhD Thesis
Page [120 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.15: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the
Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line,
BD=1000m, case a)

Figure 5.16: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=2000m,
case a)

PhD Thesis
Page [121 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.17: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=300m,
case a)

Figure 5.18: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=600m,
case a)

PhD Thesis
Page [122 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.19: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=1000m,
case a)

Figure 5.20: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=2000m,
case a)

PhD Thesis
Page [123 of 211]
 CITY , UNIVERSITY of LONDON

Figures 5.21 and 5.22 summarize the above results, indicating the impact of the length of the
cable, the tower footing resistance and the installation position of the surge arresters,
considering case a.

Figure 5.21: The Developed Overvoltage at Position D (UD) in function with the Distance CD, the length of
the cable (BD) and the Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case a)

Figure 5.22: The Developed Overvoltage at Position D (UD) in function with the Distance CD, the length of
the cable (BD) and the Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case a)

PhD Thesis
Page [124 of 211]
 CITY , UNIVERSITY of LONDON

Figures 5.23-5.30 depict the obtained results, concerning case b of Fig. 4.1. The extracted
outcomes indicate the impact of the installation of arresters at position B, compared with the
developed surges of case a.

Figure 5.23: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=300m,
case b)

Figure 5.24: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the
Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line,
BD=600m, case b)

PhD Thesis
Page [125 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.25: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the
Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line,
BD=1000m, case b)

Figure 5.26: The Developed Overvoltage at Position D (UD) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=2000m,
case b)

PhD Thesis
Page [126 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.27: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=300m,
case b)

Figure 5.28: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=600m,
case b)

PhD Thesis
Page [127 of 211]
 CITY , UNIVERSITY of LONDON

Figure 5.29: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=1000m,
case b)

Figure 5.30: The Developed Overvoltage at Position B (UB) in function with the Distance CD and the Tower
Footing Resistance (Lightning hit on the Ground Wire of the Connected Transmission Line, BD=2000m,
case b)

PhD Thesis
Page [128 of 211]
 CITY , UNIVERSITY of LONDON

Figures 5.31 and 5.32 summarize the above results (case b), indicating the impact of the length
of the cable, the tower footing resistance and the installation position of the surge arresters.

Figure 5.31: The Developed Overvoltage at Position B (UB) in function with the Distance CD, the length of
the cable (BD) and the Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case b)

Figure 5.32: The Developed Overvoltage at Position D (UD) in function with the Distance CD, the length of
the cable (BD) and the Tower Footing Resistance (Lightning hit on the Ground Wire of the Connected
Transmission Line, case b)
PhD Thesis
Page [129 of 211]
 CITY , UNIVERSITY of LONDON

In conclusion, lightning hits on the tower or the ground wires of the connected transmission
lines consist a serious threat for the safety and the normal operation of the substations. The
appropriate protection against backflashover phenomena requires the knowledge of the
expected surges, in order to design the lightning protection configuration and select the
electrical characteristics of the protective means. To these directions, the performed analysis
highlights various aspects that the designer should take into account during the initial study of
the lightning protection system of the substation:

 Low values of grounding resistance values, either for the tower or for the surge arresters,
always improve the lightning performance of the line. Low tower footing resistances do
not allow the potential of the tower to exceed the BIL and prevents backflashover. The
arresters‟ grounding resistance must be low keeps the voltage at the terminals of the
transformer below the desired limits.

 Long cable lengths contribute to the reduction of the expected overvoltages. Especially, in
case that the grounding resistance of the towers cannot be improved, the increase of the
cable length can balance the negative effects of the inadequate grounding system.

 The installation of the arresters away from the equipment to be protected reduces their
efficiency, resulting in the development of significant overvoltages higher than the
nominal residual voltage of the arresters. Furthermore, the installation of arresters at the
joint between the line and the cable reduces the expected surges and provides more
efficient protection against the incoming overvoltages.

5.4 Impact of Used Models of the Components to the Obtained Results

The current section deals with the examination of the impact of the implemented simulation
models of each device of the substation on the extracted outcomes. Considering for each
component a default model, an appropriate analysis is performed, in order to examine the
influence of the equivalent circuit models of the various components (i.e. tower, arresters,
grounding, and insulators) to the obtained results.

- Tower; table 5.1 gives the considered equivalent circuits models for tower.
-

PhD Thesis
Page [130 of 211]
 CITY , UNIVERSITY of LONDON

Table 5.1: Used Tower Models for the Sensitivity Analysis

No Tower Models
1 Jordan (Jordan 1934)
2 Wagner (Wagner and Hileman 1960)
3 Sargent (Sargent and Darveniza 1969)
4 Ametani (Ametani et al. 1994)
5 Rondon (Rondon et al. 2005)
6 Yamada & Chilsom (Yamada et al. 1995)
- Surge Arresters: table 5.2 gives the used surge arresters equivalent circuits models for the
sensitivity analysis

Table 5.2: Used Surge Arresters Models for the Sensitivity Analysis
No Surge Arrester Model
1 Physical(Bayadi et al. 2003)
2 Pinceti – Giannettoni(Pinceti & Giannettoni 1999)
3 Fernandez – Diaz(Fernandez & Diaz 2001)
4 IEEE(IEEE W. G. 3.4.11 1992)

- Grounding Resistance: table 5.3 gives the implemented models for the sensitivity
analysis.

Table 5.3 Used Grounding Resistance for the Sensitivity Analysis

No Grounding Resistance Models
1 Oettle (Oettle E. E. 1988)
2 Chisholm et al. (a) (Chisholm & Janischewskyj 1989)
3 Chisholm et al. (b) (Chisholm & Janischewskyj 1989)
4 IEEE (IEEE W. G. PAS-104 1985, IEEE Std 80 2000)
5 Yasuda et al. (Yasuda et al. 2001
6 Sverak (Sverak 1984)
7 CIGRE WG (CIGRE W.G. 33.01 1991)

PhD Thesis
Page [131 of 211]
 CITY , UNIVERSITY of LONDON

- Insulator Representation: Table 5.4: Used Insulator Models for the Sensitivity Analysis

Table 5.4: Used Insulator Models for the Sensitivity Analysis

No Insulator Models
1 Weck (Weck 1981)
2 Pigini (Pigini et al. 1989)
3 V-t curve (IEEE W.G no. 3 1993)

The default models; the considered default models are :

- Chilsom model for the Towers,

- IEEE model for the Arresters,

- Cigre model for the grounding resistance,

- V-t curve for the insulator.

The following figures depict the comparison of the calculated overvoltages by using the above
circuit models.

Tower: Figures 5.33-5.34 depict the developed overvoltages at positions B and D,

respectively, for various tower models (Table 5.1), considering:

- The lightning hit on the phase conductor of the connected transmission line.

- The tower footing resistance is equal to 10Ω.

- The length of the cable is 2km.

- The distance between the transformer and the arrester is 1m.

- Arresters have been installed at the entrance of the power transformer (case b).

PhD Thesis
Page [132 of 211]
 CITY , UNIVERSITY of LONDON

No surge arresters
500 With surge arresters
450

400

350

UB (kV) 300

250

200

150

100

50

0
1 2 3 4 5 6 7
Tower Model

Figure 5.33: The Developed Overvoltage at Position B (UB) for Various Tower Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of Connected Transmission Line)

No surge arresters
With surge arresters
600

500

400
UD (kV)

300

200

100

0
1 2 3 4 5 6 7
Tower Model

Figure 5.34: The Developed Overvoltage at Position D (UD) for Various Tower Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of Connected Transmission Line)

The obtained results indicate that the selected tower model influences the theoretical
estimations, but its impact is not critical.
PhD Thesis
Page [133 of 211]
 CITY , UNIVERSITY of LONDON

Surge Arrester : Figures 5.35 and 5.36presentthe Influence of the Used Surge Arresters
Models, according to Table 5.2.

All the simulated models are able to reproduce with good accuracy the residual voltage and the
absorbed energy. Although the Fernandez–Diaz model gives in general greater peak residual
voltage values, the IEEE model absorbs more energy; this is owned to the shape of the
waveform, since the area under the residual voltage curve is bigger for the IEEE model.

The advantage of the three dynamic models, apart from the fact that they decrease the relative
error, is that they can represent the frequency-dependent behavior of the metal oxide surge
arrester, which is really important when the time to crest value of the lightning current is not
constant.it has been proved that the decision on which model should be used depends on the
available data, the complexity of the system and the demanded speed of the simulation, since
the Pinceti–Gianetoni and the Fernandez–Diaz model are simpler than the IEEE model.

500
With surge arresters
450

400

350

300
UB (kV)

250

200

150

100

50

0
1 2 3 4
Surge arrester Model

Figure 5.35: The Developed Overvoltage at Position B (UB) for Various Surge Arrester Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of Connected Transmission Line)

PhD Thesis
Page [134 of 211]
 CITY , UNIVERSITY of LONDON

550
With surge arresters
500

450

400

350

300
UD (kV)

250

200

150

100

50

0
1 2 3 4
Surge arrester Model

Figure 5.36: The Developed Overvoltage at Position D (UD) for Various Surge Arrester Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of the Connected Transmission Line)

Grounding Resistance:Figures5.37 – 5.38depictthe Calculated Overvoltages for Various

Grounding Resistance Models according to Table 5.3

No surge arresters
500 With surge arresters
450

400

350

300
UB (kV)

250

200

150

100

50

0
1 2 3 4 5 6
Grounding resistance Model

Figure 5.37: The Developed Overvoltage at Position B (UB) for Various Grounding Resistance Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of the Connected Transmission Line)
PhD Thesis
Page [135 of 211]
 CITY , UNIVERSITY of LONDON

No surge arresters
With surge arresters
600
550
500
450
400
350
UD (kV)

300
250
200
150
100
50
0
1 2 3 4 5 6 7
Grounding resistance Model

Figure 5.38: The Developed Overvoltage at Position D (UD) for Various Grounding Resistance Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of the Connected Transmission Line)

From the previous figures (5.37 – 5.38)it can be concluded that the estimated values of the
overvoltages vary insignificantly. Model no 3[ Chisholm et al. (b)] gave a lower estimations of
the arising overvoltages in comparison to the other models. The use of model 2 [ Chisholm et
al. (a) ] resulted in increased values of overvoltages.

Insulator: Figures 5.39 – 5.40examinethe Influence of the Insulator Model According to

Table 5.4.

PhD Thesis
Page [136 of 211]
 CITY , UNIVERSITY of LONDON

No surge arresters
500 With surge arresters
450

400

350

UB (kV) 300

250

200

150

100

50

0
1 2 3
Insulator Model

Figure 5.39: The Developed Overvoltage at Position B (UB) for Various Insulator Models (R=10Ω,
BD=2000m, LCD=1m, Lightning hit on the Phase Conductor of the Connected Transmission Line)

No surge arresters
600 With surge arresters
550
500
450
400
350
UD (kV)

300
250
200
150
100
50
0
1 2 3
Insulator Model

Figure 5.40: The Developed Overvoltage at Position D (UD) for Various Insulator Models
(R=10Ω, BD=2000m, LCD=1m, Lightning hit on Phase Conductor of the Connected Transmission Line)

PhD Thesis
Page [137 of 211]
 CITY , UNIVERSITY of LONDON

The previous figures (5.33 – 5.40)show that the estimated surges do not vary significantly
among the implemented equivalent circuit models. The difference between the highest and the
lowest value of the estimated overvoltage at positions B and D does not exceed 4%, indicating
that the used models result in similar outcomes.

The above analysis was repeated for :

- The lightning hit on the ground wires of the connected transmission lines,

- The lighting peak current is 200kA,

- The tower footing resistance is 5Ω,

- The length of the cable is 1000m ,

- The distance between the arresters‟ installation position and the transformer to be
protected equal to 1m.

Figures 5.41 – 5.48 present the results of the performed sensitivity analysis in case of indirect
lightning hit.

No surge arresters
With surge arresters
800

700

600

500
UB (kV)

400

300

200

100

0
1 2 3 4 5 6 7
Tower Model

Figure 5.41: The Developed Overvoltage at Position B (UB) for Various Tower Models
(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission Line)
PhD Thesis
Page [138 of 211]
 CITY , UNIVERSITY of LONDON

No surge arresters
With surge arresters
900

800

700

600

500
UD (kV)

400

300

200

100

0
1 2 3 4 5 6 7
Tower Model

Figure 5.42: The Developed Overvoltage At Position D (UD) for Various Tower Models
(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission Line)

500
With surge arresters
450

400

350

300
UB (kV)

250

200

150

100

50

0
1 2 3 4
Surge arrester Model

Figure 5.43: The Developed Overvoltage at Position B (UB) for Various Surge Arresters Models
(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission Line)

PhD Thesis
Page [139 of 211]
 CITY , UNIVERSITY of LONDON

550
With surge arresters
500

450

400

350

300
UD (kV)

250

200

150

100

50

0
1 2 3 4
Surge arrester Model

Figure 5.44: The Developed Overvoltage at Position B (UB) for Various Surge Arresters Models
(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission Line)

No surge arresters
900 With surge arresters

800

700

600

500
UB (kV)

400

300

200

100

0
1 2 3 4 5 6
Grounding resistance Model

Figure 5.45: The Developed Overvoltage at Position B (UB) for Various Grounding Resistance
Models(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission
Line)

PhD Thesis
Page [140 of 211]
 CITY , UNIVERSITY of LONDON

No surge arresters
With surge arresters
900

800

700

600

500
UD (kV)

400

300

200

100

0
1 2 3 4 5 6 7
Grounding resistance Model

Figure 5.46: The Developed Overvoltage at Position D (UD) for Various Grounding Resistance
Models(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission
Line)

No surge arresters
850 With surge arresters
800
750
700
650
600
550
500
UB (kV)

450
400
350
300
250
200
150
100
50
0
1 2 3
Insulator Model

Figure 5.47: The Developed Overvoltage at Position B (UB) for Various Insulator Models
(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission Line)

PhD Thesis
Page [141 of 211]
 CITY , UNIVERSITY of LONDON

No surge arresters
950 With surge arresters
900
850
800
750
700
650
600
550
500
UD (kV)

450
400
350
300
250
200
150
100
50
0
1 2 3
Insulator Model

Figure 5.48: The Developed Overvoltage at Position D (UD) for Various Insulator Models
(R=5Ω, BD=1000m, LCD=1m, Lightning hit on the Ground Wire of the Connected Transmission Line)

Summarizing the presented analysis in the current section, the developed surges at the
entrance of a 150/20kV substation due to lightning hit on the connected overhead
transmission line are computed, considering various equivalent circuit models for the main
devices of the substation‟s equipment. The effect of each model does not seem to be crucial
for the reliability of the simulation results, since the computed overvoltage do not vary
significantly among the implemented models. The performed analysis can be useful for
insulation coordination studies, in order to select the appropriate equivalent circuits.

5.5 Impact of the Number of the Incoming Transmission Lines

In this section the relation between the developed overvoltages and the number of the
incoming lines to the substation is examined. The arising overvoltages are strongly dependent
on the number of the lines and the lightning hit position (L is the distance between the
lightning hit position and the busbar, E is the overvoltage of the busbar). The peak value and
the rise time of the incoming voltage surges decrease with the increase of the number of the
transmission lines, since the overvoltage “faces‟ the parallel combination of the incoming
lines. Furthermore, the probability of lightning hit on a transmission lines is reduced as the
number of the incoming lines is increased. Considering that all the transmission lines present
PhD Thesis
Page [142 of 211]
 CITY , UNIVERSITY of LONDON

the same surge impedance, the overvoltage at the entrance of the substation is given by the
equation (McDonald, J. et al 2000, Siemens 2008):

( )

Where;

U is the peak value of the incoming voltage surge

E is the overvoltage at the entrance of the substation

n is the number of the incoming lines.

It is worth mentioning that U is not the overvoltage at the position of the lightning hit, but the
surge just before the entrance of the substation. The following equation is an empirical
approach to estimate the voltage U:

( )

Where;

Vmax is the overvoltage at the lightning hit position.

lis the distance between the lightning hit position and the substation.

kis an empirical coefficient between 0.0001 and 0.0004 (kV km)-1.

Figure 5.49 present the calculated results for the topology of Fig. 4.2. Τhe obtained outcomes
indicate that the expected overvoltage at the entrance of the substation is reduced for higher
number of incoming transmission lines. The impact of the lightning hit position is also
highlighted, indicating the influence of the distance between the lightning stroke position and
the substation to the incoming surge voltages.

PhD Thesis
Page [143 of 211]
 CITY , UNIVERSITY of LONDON

2000
E(n=1) (kV)
1900
1800 E(n=2) (kV)
1700 E(n=3) (kV)
1600 E(n=4) (kV)
1500
1400
E(n=5) (kV)
1300
1200
1100
E (kV)

1000
900
800
700
600
500
400
300
200
100
0
0 1 2 3 4 5 6 7 8 9 10 11
L (km)

Figure5.49: Voltage at the Entrance of the Substation in function with the Distance between the Lightning hit
Position and the Substation

5.6 Calculation of Developed Overvoltages for Interconnected Substations

Figures. 5.50 – 5.55 present the estimated overvoltages at various positions of the topology
depicted in Fig. 4.2. Τhe performed simulations consider the case of lightning hit on the phase
conductors of the ground wires and the implementation or not of metal oxide gapless surge
arresters. The behavior of the expected overvoltage for each position of the system is
examined in function with the tower footing resistance. The results highlight the need for
protection of the system by installing surge arresters and the achievement of low values of
grounding resistance. The analysis performed by using appropriate simulation tool
(Simulink/MATLAB). The towers of the overhead lines and the installed metal oxide gapless
surge arresters are modeled according to the Chilsom model and the IEEE model,
correspondingly. As far as the grounding resistance is concerned the IEEE model is selected;
finally, V-t curve is implemented for the circuit representation of the insulators.

PhD Thesis
Page [144 of 211]
 CITY , UNIVERSITY of LONDON

UA case (a) no surge arresters

1600 UA case (a) with surge arresters
1500 UA case (b) no surge arresters
1400 UA case (b) with surge arresters
1300
1200
1100
1000
900
UA (kV)

800
700
600
500
400
300
200
100
0
0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5
Rtower (Ohm)

Figure5.50: Overvoltage at Position A (UA) in function with Grounding Resistance for Direct (case (I)) or
Indirect (case (II)) Lightning hit for the Topology of Figure 4.2, Protected or Not by Surge Arresters

UB case (a) no surge arresters

1600 UB case (a) with surge arresters
1500 UB case (b) no surge arresters
1400 UB case (b) with surge arresters
1300
1200
1100
1000
900
UB(kV)

800
700
600
500
400
300
200
100
0
0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5
Rtower (Ohm)

Figure5.51 : Overvoltage at position A (UB) in function with Grounding Resistance for Direct (case (I)) or
Indirect (case (II)) Lightning hit for the Topology of Figure 4.2, Protected or Not by Surge Arresters

PhD Thesis
Page [145 of 211]
 CITY , UNIVERSITY of LONDON

1600 UC case (a) no surge arresters

1500 UC case (a) with surge arresters
1400 UC case (b) no surge arresters
1300 UC case (b) with surge arresters
1200
1100
1000
900
UC(kV)

800
700
600
500
400
300
200
100
0
0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5
Rtower (Ohm)

Figure5.52:Overvoltage at Position C (UC) in function with Grounding Resistance for Direct (case (I)) or
Indirect (case (II)) Lightning hit for the Topology of Figure 4.2, Protected or Not by Surge Arresters

UD case (a) no surge arresters

1600 UD case (a) with surge arresters
1500 UD case (b) no surge arresters
1400 UD case (b) with surge arresters
1300
1200
1100
1000
900
UD(kV)

800
700
600
500
400
300
200
100
0
0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5
Rtower (Ohm)

Figure5.53:Overvoltage at Position D (UD) in function with Grounding Resistance for Direct (case (I)) or
Indirect (case (II)) Lightning hit for the Topology of Figure 4.2, Protected or Not by Surge Arresters

PhD Thesis
Page [146 of 211]
 CITY , UNIVERSITY of LONDON

UE case (a) no surge arresters

1600 UE case (a) with surge arresters
1500 UE case (b) no surge arresters
1400 UE case (b) with surge arresters
1300
1200
1100
1000
900
UE(kV)

800
700
600
500
400
300
200
100
0
0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5
Rtower (Ohm)

Figure5.54: Overvoltage at Position E (UE) in function with Grounding Resistance for Direct (case (I)) or
Indirect (case (II)) Lightning hit for the Topology of Figure 4.2, Protected or Not by Surge Arresters

UF case (a) no surge arresters

1600 UF case (a) with surge arresters
1500 UF case (b) no surge arresters
1400 UF case (b) with surge arresters
1300
1200
1100
1000
900
UF(kV)

800
700
600
500
400
300
200
100
0
0,0 2,5 5,0 7,5 10,0 12,5 15,0 17,5 20,0 22,5
Rtower (Ohm)

Figure5.55:Overvoltage at Position E (UF) in function with Grounding Resistance for Direct (case (I)) or
Indirect (case (II)) Lightning hit for the Topology of Figure 4.2, Protected or Not by Surge Arresters

PhD Thesis
Page [147 of 211]
 CITY , UNIVERSITY of LONDON

5.7 Conclusions

The efficient protection of HV/MV substations against overvoltages is an issue of priority for
electrical power corporations, since lightning is a major cause of dielectric stress of the
equipment, ageing of the insulators and possible outages and damages of the substation
components. A substation fault results in important consequences for the normal operation of
the system, i.e. interruption of the power transmission and distribution and need for repair or
replacement of the destroyed parts of the installation. Common practice for the protection of
the HV/MV substations against lightning hits is the installation of external lightning
protection system combined with surge arresters at the primary and the secondary of the
power transformer. However, many factors affect the efficient protection that the
implemented surge arresters offer; the lightning hit position, the distance between the arresters
and the equipment to be protected, the grounding resistance are factors that influence the
lightning performance of the system and determine the range of the arising overvoltages.

In case of open substations, the achievement of low values of grounding resistance and the
installation of surge arresters at the entrance of the substation can improve the lightning
performance of the system. In case that the transformer and the overhead transmission lines
are connected with HV cables, the length of the cable is a critical factor for the restriction of
the developed surges. The performed simulations indicate the dependence of the arising
overvoltages at the entrance of HV/MV substations on the tower footing resistance, the
installation position of the arresters and the length of the cable that connects the incoming
overhead transmission lines with the transformer. The implementation of surge arrester
definitely reduces the range of the developed voltage surges, compared to those ones without
the use of surge arresters, but in cases of high grounding resistance values or high distance
between the protective measures and the transformer, the developed overvoltages exceed the
BIL. The significance of the length of the cable is also highlighted. As far as the influence of
the used equivalent circuit models is concerned, the results show that the applied models give
satisfactory results, presenting acceptable variations. Moreover, the relation between the
developed overvoltages and the number of the incoming lines to the substation and the
lighting hit position is examined, indicating that the range and the rise time of the incoming
voltage surges decrease with the increase of the number of the transmission lines.

Main novelty of the current work is that examines the lightning performance of a substation,
separately for a lightning hit on a phase conductor or a ground wire of the connected line,
considering not only the grounding resistance (which is the common practice in similar studies
of other researchers) but the cable length and the installation position of the arresters. Indeed,
PhD Thesis
Page [148 of 211]
 CITY , UNIVERSITY of LONDON

the literature review indicated that the majority of the relative studies mainly focus on the role
of the grounding resistance, ignoring or minimizing the importance of the other parameters.
Regarding that the external LPS of the substation protects adequately the substation against
direct lightning hits, the Thesis highlights that the lightning performance of the substations
depends on various factors, except from grounding resistance. Moreover, in case of the
Hellenic system, arresters are usually installed at the primary of the power transformer. In the
current work, different scenarios for the placement of the transformer are examined
(installation at the joint of the transmission line with the cable), revealing its impact on the
expected overvoltages. The efficient design of the substation has to take into account these
factors, in order to reduce the outage rate of the substation. It is worth mentioning that for
long cables, the developed surges do not exceed the basic insulation level even if no arresters
have been installed. Finally, the thesis reveals the role of the number of the incoming lines to
the arising overvoltages.

PhD Thesis
Page [149 of 211]
 CITY , UNIVERSITY of LONDON

References

Bayadi A., Harid N., Zehar K., Belkhiat S., Simulation of metal oxide surge arrester dynamic behavior
under fast transients, in: International Conference on Power Systems Transients, New Orleans, USA,
2003.2Pinceti P., Giannettoni M., A simplified model for zinc oxide surge arresters, IEEE Trans.
Power Delivery 14 (2) (1999) 393–398.
Chisholm W. A., Chow Y. L., and Srivastava K. D. ―Travel time of transmission towers,‖ IEEE
Trans. Power App. Syst., vol. PAS–104, no. 10, pp. 2922–2928, Oct. 1985.4
CIGRE Working Group 33.01, ―Guide to procedures for estimating the lightning performance of
transmission lines,‖ Technical Bulletin 63, Oct. 1991.
CIGRE, Working Group 01 (Lightning) of Study Committee 33 (Overvoltages and Insulation
Coordination) ―Guide to Procedures for Estimating the Lightning Performance of Transmission Lines,
brochure 63, CIGRE, Paris, France, October,1991
Fernandez F., Diaz R., Metal oxide surge arrester model for fast transient simulations, in: International
Conference on Power System Transients (IPAT‘ 01), Rio De Janeiro, Brazil, 2001, Paper 14.IEEE
Working Group 3.4.11, Modeling of metal oxide surge arresters, IEEE Trans. Power Delivery 7(1)
(1992) 302–309.-
Hileman A.R., ―Insulation coordination for Power Systems‖, CRC Press, 1999
IEEE Std 80-2000, ―IEEE Guide for Safety in AC Substation Grounding‖, January 2000Yasuda
Y., Hirakawa Y., Shiraishi K., and Hara T., ―Sensitivity analysis on grounding models for 500kV
transmission lines,‖ Trans.IEE Japan B, vol. 121, no. 10, pp. 1386–1393, 2001.6 Sverak (Sverak
1984)
IEEE Working Group, ―Estimating lightning performance of transmission lines II – Updates to
analytical models,‖ IEEE Trans.Power Delivery, vol. 8, no. 3, pp. 1254-1267, Jul. 1993.
Jordan C. A., ―Lightning computations for transmission lines with overhead ground wires,‖ Gen. Elec.
Rev., vol. 37, 1934Wagner (Wagner and Hileman 1960)
Oettle E. E., ―A new general estimation curve for predicting the impulse impedance of concentrated earth
electrodes,‖ IEEE Trans.Power Delivery, vol. 3, no. 4, pp 2020–2029, Oct. 1988.2Chisholm et al.
(a) (Chisholm & Janischewskyj 1989)
Pigini A., Rizzi G., Garbagnati E., Porrino A., Baldo G., and Pesavento G., ―Performance of large
air gaps under lightning overvoltages: Experimental study and analysis of accuracy of predetermination
methods,‖ IEEE Trans. Power Delivery, vol. 4, no. 2, pp 1379–1392, Apr. 1989
Rondon D., Vargas M., Herrera J., Montana J., Torres H., Camargo M., Jimenez D., and Delgadillo
A., ―Influence of grounding system modelling on transient analysis of transmission lines due to direct
lightning strike,‖ in Proc. Power Tech, St. Petersburg, Russia, 2005, pp. 1-6Yamada & Chilsom
(Yamada et al. 1995)
Ryan H.M, ―High voltage engineering and testing‖, Peter Peregrinus Ltd. 1st edition, 1994).

PhD Thesis
Page [150 of 211]
 CITY , UNIVERSITY of LONDON

Sargent M. A. and Darveniza M., ―Tower surge impedance,‖ IEEE Trans. Power App. Syst., vol.
PAS–88, no. 5, pp. 680–687, May 1969Ametani (Ametani et al. 1994)
Siemens, Energy Sector, ―Power Engineering Guide: Answers for Energy‖, 5th Edition, 2008
Weck K. H., ―Lightning performance of substations,‖ CIGRE SC 33, Rio de Janeiro, Brazil, 1981.

PhD Thesis
Page [151 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 6

LIGHTNING RISK ASSESSMENT OF SUBSTATIONS

6.1 Introduction

The current Chapter deals with the computation of the outage rate due to lightning
overvoltages, concerning the substation configurations of Chapter 4. A detailed methodology
for the computation of the outage rate of HV/MV substations is provided, considering
lighting hit on the phase conductors or the overhead ground wires of the connected
transmission lines. Assuming that a HV/MV substation is adequately protected against direct
lightning strikes, incoming surges from the connected transmission line are the main cause of
substation outages, resulting in interruption and repair/replacement costs. The above
consequences are more intense in case that cables are used for the connection of the incoming
line with the substation, since potential damages of the cable insulation results in higher repair
costs and longer term outages (Mikropoulos et al. IET). For these reasons, a lightning risk
assessment of HV/MV substations and the evaluation of their lightning performance is
critical, in order to adopt the necessary protective measures against lightning phenomena. In
the current work, lightning risk assessment of a 150/20kV substation of the Hellenic network
is carried out, considering the incoming lightning surges due to either backflashover or
shielding failure in the connected overhead transmission lines, based on the scenario and the
data of the previous chapters (configuration of the substation (Fig. 4.1.), sensitivity analysis
data, simulation models).

6.2 Calculation of Outage Rate due to Lightning Hits on Phase

Conductors

The outage rate of the substation due to shielding failures in the connected lines, i.e. lightning
flash to phase conductors, SORSF (outages/year), is given as(IEC 60071-2, 1996):

PhD Thesis
Page [152 of 211]
 CITY , UNIVERSITY of LONDON

[ ( ) ( )] ( )

Where:
SFRX (shielding failures/year) is the shielding failure rate of the connected line
XSF(km)is the distance between the substation entrance and the lightning hit position
The shielding failure rate of the connected line SFRX is given by the equation:

( )

Where;
SFR (shielding failures/100km/year) is the shielding failure rate of the incoming overhead line
(Mikropoulos & Tsovilis-IEEE, Mikropoulos et al., 2010).
F(I) is the probability of lightning crest current being greater than I, given by the equation
Anderson, 1982):

( ) ( )
(⁄ )

Where
ICSF (kA) is the minimum value of the lightning current hitting the phase conductor of the
connected transmission line at a distance XSF from the substation entrance that results in the
development of overvoltages greater than the basic insulation level(Hileman, 1999, IEEE T.
F., vol. 11, no. 1, 1996).
IMSF is the maximum shielding failure current of the connected line (Mikropoulos & Tsovilis,
IET).

6.3 Calculation of Outage Rate due to Lightning hits on ground wires

The substation outage rate due to incoming backflashover surges, that is, flashover of line
insulation caused by lightning flash to shield wire, SORBF (outages/year), can be estimated
as(IEC 60071-2, 1996):

( ) ( )

Where;
PhD Thesis
Page [153 of 211]
 CITY , UNIVERSITY of LONDON

NSX (strikes/year) is the rate of lightning strikes to shield wires of the connected line
XBF (km)is the limit distance between the substation entrance and the lightning hit position.
NSX is given as:

( )

Where
NS(strikes/100km/year) is the number of lightning hits on ground wires of the overhead
transmission line (Mikropoulos & Tsovilis, IEEE).
ICBF (kA) is the minimum value of the lightning current hitting the ground wire at a distance
XSF Distance from the substation entrance that results in the development of overvoltages
greater than the basic insulation level (Hileman, 1999, IEEE T. F., vol. 11, no. 1, 1996).

6.4 Failure Rate of Arresters

In case that the absorbed energy (by the arresters) exceeds their withstand capability, the
arresters are damaged (failure); the lightning energy E(in Joules) absorbed by a surge arrester is
computed by the relation (Christodoulou et al. 2010):

∫ ( ) ( ) ( )

Where;
u(t) is the residual voltage of the arrester in kV
i(t) is the value of the discharge current through the arrester in kA.
Considering the energy required causing damage to an arrester, the failure probability of an
arrester is calculated by (Christodoulou et al. 2010):

 
 
ArrFP1     f ( I p )  hA ( I P )dI p g (Tt )dTt
Tr I A (Tt ) 
(6.7)
  

ArrFP2     f ( I p )  hB ( I P )dI p g (Tt )dTt (6.8)
Tr 
I B (Tt ) 

Where;
PhD Thesis
Page [154 of 211]
 CITY , UNIVERSITY of LONDON

ArrFP1 is the probability that an arrester fails due to lightning stroke on a phase conductor,
ArrFP2 is the probability that an arrester fails due to lightning stroke on the overhead ground
wire,
IA(Tt) is the minimum stroke peak current in kA required to damage the arrester, when
lightning hits on a phase conductor, depending on each time-to-half value,
IB(Tt) is the minimum stroke peak current in kA required to damage the arrester, when
lightning hits on the overhead ground wire, depending on each time-to-half value,
f(IP) is the probability density function of the lightning current peak value,
g(Tt) is the probability density function of the time-to-half value of the lightning current,
Tr is the rise time of the incident waveform,
PT is the total failure probability of an arrester,
The total failure rate of the arrester is given by the equation(Christodoulou et al. 2010):
( ) (6.9)
AT is the arrester total failure rate in failures per year per line,
NL is the number of lightning flashes to a line per 100km per year, equal to (Christodoulou et
al. 2010):

( ) ( )

ht is the tower height in m, g is the horizontal spacing in m, between the ground wires,
Ng is the ground flash density in flashes per km2per year and l is the line length in km.

6.5 Calculation of the Total Substation Outage Rate

The total substation outage rate, SOR(outages/year), due to lightning hits on the connected
transmission lines is:

( )

Finally, in case that the arresters‟ failures are included to the substations damages, then the
total substation failure rate is given as:

( )

PhD Thesis
Page [155 of 211]
 CITY , UNIVERSITY of LONDON

6.6 Results
The above analysis indicates that the estimation of the substation outage rate is a complex
procedure that demands the knowledge of various parameters and factors that influence the
lightning performance of the system. In this section, a sensitivity analysis of the expected
failure rate of the substation of Fig.4.1 in function with the tower footing resistance, the
length of the cable and the arresters installation position is presented, based on the data
presented in previous Chapters.

The analysis performed by using the Simulink/MATLAB software tool. The towers of the
overhead lines and the installed metal oxide gapless surge arresters are modeled according to
the Chilsom model and the IEEE model, correspondingly. As far as the grounding resistance
is concerned the Cigre model is selected; finally, V-t curve is implemented for the circuit
representation of the insulators (see Chapters 3 and 4).

The following results highlight the great impact of the grounding resistance, the length of the
cable and the installation position of the implemented arresters to the expected failure rates.
In details, the grounding resistance is the dominant factor that determines the lightning
performance of the system, since it influences the occurrence or not of backflashover
phenomena and affect the effectiveness of the protection that surge arresters provide.
Furthermore, the length of the cable is of great importance, since influences the magnitude of
the expected overvoltages, as indicated in Chapter 5. The distance between the transformer
and the arresters affects the lightning performance of the system, imposing their installation
near the entrance of the substation. As far as the arresters failure rate is concerned, the
behavior of the arresters depends on the grounding resistance. In general, the achievement of
low values of grounding resistance contributes to the more efficient operation of the
protective devices, since ensures a low-resistance path in order to divert the lightning current
to earth.
In details Fig. 6.1 depict the substation outage rate in case of lightning hit oh phase conductor
of the connected line in function with the length of the cable and distance between the
arrester and the entrance of the power transformer, considering case A of Fig.4.1. It is worth
mentioning that the tower footing resistance has no impact on the expected outage rates. The
results indicate clearly the improvement of the lightning performance of the substation under
study after the installation of surge arresters and highlights the role of the length of the cable
and the installation position of the arresters, confirming the simulation results of Chapter 5.

PhD Thesis
Page [156 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.1 : Substation Outage Rate in case of lightning hit on the phase conductor of the connected line in
function with the length of the cable (BD) and the distance between the arrester and the transformer (CD)
(Case a)

Fig. 6.2 – 6.6 present the results in case of a lightning hit on the overhead ground wire of the
incoming line, considering case A of Fig. 4.1. The calculated outage rates reveal the dominant
role of the tower footing resistance, since low values of the grounding resistance reduce the
expected failures. Note that in case of long cables and installation of the arresters at the
entrance of the power transformers, the substation outage is almost eliminated.

Figure 6.2: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=300m (Case a)

PhD Thesis
Page [157 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.3: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=600m (Case a)

Figure 6.4: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=1000m (Case a)

PhD Thesis
Page [158 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.5: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=2000m (Case a)

Figure 6.6: Substation Outage Rate in function with the grounding resistance (R) and the distance between the
arrester and the transformer (CD) considering BD=300m (Case a)

PhD Thesis
Page [159 of 211]
 CITY , UNIVERSITY of LONDON

Figures 6.7 – 6.9 presents the total substation outage rate, according to the evaluations
presented in Figures 6.1 – 6.6. The installation of surge arresters combined with the
achievement of low values of tower footing resistance and long cables ensure the
improvement of the lightning performance of the substation.

Figure 6.7: Substation Outage Rate in function with the grounding resistance (R) and the distance between the
arrester and the transformer (CD) considering BD=600m (Case a)

Figure 6.8: Substation Outage Rate in function with the grounding resistance (R) and the distance between the
arrester and the transformer (CD) considering BD=1000m (Case a)

PhD Thesis
Page [160 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.9: Substation Outage Rate in function with the grounding resistance (R) and the distance between the
arrester and the transformer (CD) considering BD=2000m(Case a)

Figures 6.10 – 6.17 depict the expected failure rate for the arresters installed at positions B and
D. Critical factors that influence the energy passed through the arresters are the tower footing
reistance (that determines the incoming surge, due to backflashover) and the installation
position of the arresters (arresters at position B have higher failure rate compared with those
installed at position D).

Figure 6.10:Failure Rate of the arrester installed at position Bin function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=300m (Case a)

PhD Thesis
Page [161 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.11: Failure Rate of the arrester installed at position Bin function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=600m(Case a)

Figure 6.12: Failure Rate of the arrester installed at position Bin function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=1000m (Case a)

PhD Thesis
Page [162 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.13: Failure Rate of the arrester installed at position Bin function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=2000m(Case a)

Figure 6.14: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=300m (Case a)

PhD Thesis
Page [163 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.15: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=600m(Case a)

Figure 6.16: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=1000m (Case a)

PhD Thesis
Page [164 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.17: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=2000m (Case a)

Figures 6.18 – 6.21 present the total substation outage rate including the arresters failure rate.
Fig. 6.22 summarizes the obtained results for the case a of Fig. 4.1, comparing the substation
outage rate including or not the arresters failure rate.

Figure6.18: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=300m (Case a)

PhD Thesis
Page [165 of 211]
 CITY , UNIVERSITY of LONDON

Figure6.19: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=600m (Case a)

Figure6.20: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=1000m (Case a)

PhD Thesis
Page [166 of 211]
 CITY , UNIVERSITY of LONDON

Figure6.21: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=2000m (Case a)

Figure6.22: Comparison Substation Outage Rate including and not including Arresters Faults in function
with the Grounding Resistance and the distance and the length of the cable (BD) considering CD=0m (Case
a)

The same procedure is performed considering case b of Fig. 4.1 (installation of arresters only
at the entrance of the power transformer). The results indicate that the lightning performance
of the substation is downgraded, due the absence of arresters at the joint between
underground cable and the overhead transmission line. Figures 6.22 – 6.40 depict the
simulation results for case b, indicating the necessity of installation of arresters at position B,
in order to ensure the more efficient protection of the substation against lightning
overvoltages.
PhD Thesis
Page [167 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.23: Substation Outage Rate in case of lightning hit on the phase conductor of the connected line in
function with the length of the cable (BD) and the distance between the arrester and the transformer (CD)
(case b)

Figure 6.24: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=300m (case b)

PhD Thesis
Page [168 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.25: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=600m (case b)

Figure 6.26: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=1000m (case b)

PhD Thesis
Page [169 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.27: Substation Outage Rate in case of lightning hit on ground wire of the connected line in function
with the grounding resistance (R) and the distance between the arrester and the transformer (CD) considering
BD=2000m (case b)

Figure 6.28: Substation Outage Rate in function with the grounding resistance (R) and the distance between
the arrester and the transformer (CD) considering BD=300m (case b)

PhD Thesis
Page [170 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.29: Substation Outage Rate in function with the grounding resistance (R) and the distance between
the arrester and the transformer (CD) considering BD=600m (case b)

Figure 6.30: Substation Outage Rate in function with the grounding resistance (R) and the distance between
the arrester and the transformer (CD) considering BD=1000m (case b)

PhD Thesis
Page [171 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.31: Substation Outage Rate in function with the grounding resistance (R) and the distance between
the arrester and the transformer (CD) considering BD=2000m (case b)

Figure 6.32: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=300m (case b)

PhD Thesis
Page [172 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.33: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=600m (case b)

Figure 6.34: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=1000m (case b)

PhD Thesis
Page [173 of 211]
 CITY , UNIVERSITY of LONDON

Figure 6.35: Failure Rate of the arrester installed at position Din function with the Grounding Resistance and
the distance between the arrester and the transformer (CD) considering BD=2000m (case b)

Figure6.36: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=300m (case b)

PhD Thesis
Page [174 of 211]
 CITY , UNIVERSITY of LONDON

Figure6.37: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=600m (case b)

Figure6.38: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=1000m (case b)

PhD Thesis
Page [175 of 211]
 CITY , UNIVERSITY of LONDON

Figure6.39: Substation Outage Rate including Arresters Faults in function with the Grounding Resistance
and the distance between the arrester and the transformer (CD) considering BD=2000m (case b)

Figure6.40: Comparison Substation Outage Rate including and not including Arresters Faults in function
with the Grounding Resistance and the distance and the length of the cable (BD) considering CD=0m (case b)

PhD Thesis
Page [176 of 211]
 CITY , UNIVERSITY of LONDON

6.7 Conclusions

Lightning risk assessment of a HV/MV substation connected to the transmission system

through underground cables has been performed by considering both shielding failure and
backflashover surges incoming from the connected overhead lines. The obtained results by
the simulation procedures clearly indicate the improvement of the lightning performance of
the substation under study, after the installation of metal oxide gapless surge arresters,
highlighting, simultaneously, the dominant role of the grounding resistance, the length of the
cable and the installation position of the arresters. Low values of grounding resistance are
definitely contribute to the improvement of the lightning performance of the substation, but
the construction of adequate grounding systems is not always an attainable goal, due to
technical restrictions. The installation of the arresters near the transformer to be protected and
the use of long MV underground cables reduce the expected failures, even if the achieved
grounding resistance values are high. In any case, the design of the lightning protection system
and the demanded specification of the characteristics of the protection equipment have to
obey to technoeconomical criteria. Furthermore, the current work includes the arresters failure
rate to the total outage rate of the substation, denoting the fact that surge arresters are
vulnerable to damages, if high current pass through them, especially in case of low grounding
resistance values; low values of the grounding resistance contribute to the restrain of the
developed overvoltages and to the reduction of the outage rate, but increase the failure
probability of the installed arresters.

Innovation of the performed study consists the inclusion of the arresters failure rate to the
total substation failure rate, since arresters are part of the equipment of the substation and
their damage consequences repair of replacement costs and malfunction of the system..
Similar studies by other researchers do not examine the arresters behavior and do not regard
the arresters failure as a line/substation outage. Novelty of the Thesis is that highlights the
arresters failure rate, revealing how important is to avoid arresters damage; in that case,
arresters will not be operational and will not protect the substations, downgrading the future
lightning performance of the substation.. Note, that the appropriate selection of the electrical
characteristics and the installation position of the arresters is the first step for the reduction of
the expected lightning failures. A damaged arrester is not effective anymore and does not
protect the line or the substation; in addition, the replacement costs have to be taken into
consideration. For these reasons, arresters faults have to be included in the total substation
failure rate.

PhD Thesis
Page [177 of 211]
 CITY , UNIVERSITY of LONDON

References
Anderson G.J., Transmission Line Reference Book – 345 kV and above, 2nd ed. Palo Alto, CA:
Electric Power Research Institute,1982, ch. 12.
C.A. Christodoulou, L. Ekonomou, G.P. Fotis, I.F. Gonos, I.A. Stathopulos, ―Assessment of surge
arrester failure rate and application studies in Hellenic high voltage transmission lines‖, Electric Power
Systems Research, Vol. 80, Issue 2, pp. 176-183,February 2010.
CIGRE Working Group 33.01, ―Guide to procedures for estimatingthe lightning performance of
transmission lines,‖ Technical Bulletin63, Oct. 1991.
HilemanA. R., ―The incoming surge and open breaker protection,‖ inInsulation Coordination for Power
Systems. Boca Raton, FL: CRCPress, Taylor & Francis Group, New York, 1999.
IEC 60071-2, Insulation Coordination, 1996.
IEEE Task Force, ―Modeling guidelines for fast front transients,‖ IEEE Trans. Power Delivery, vol.
11, no. 1, pp. 493–506, Jan. 1996.
Mikropoulos P. N. and Tsovilis T. E., ―Estimation of lightning incidence to overhead transmission
lines,‖ IEEE Trans. Power Delivery, in press.
Mikropoulos P. N. and TsovilisT. E., ―Lightning attachment models and maximum shielding failure
current of overhead transmission lines: Implications in insulation coordination of substations,‖ IETGener.
Transm. Distrib., submitted.
Mikropoulos P. N. and Tsovilis T. E., ―Shielding performance of typical overhead transmission lines,‖
IEEE Trans. Power Delivery, submitted.
Mikropoulos P. N., Tsovilis T. E., and Zlitidis D. E., ―Software development for the evaluation of the
lightning performance of overhead transmission lines,‖ in Proc. 45th UPEC, Cardiff, Wales,UK, 2010,
accepted.

PhD Thesis
Page [178 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 7

SPECIAL ISSUES OF LIGHTNING PROTECTION OF

SUBSTATIONS

7.1 Introduction

This chapter deals with special issues of the lightning protection of HV/MV substations, i.e.
the arising induced overvoltages and the connection of surge arresters in parallel. The
magnetic field (due to the lightning discharge current) around the flash channel and the masts
or the overhead ground wires results in the development of induced overvoltages in the
formed wiring loops of the installation; these induction loops can include only active wires or
both live and protective conductors. The magnitude of the developed surges depends on the
lightning characteristics, the geometrical characteristics of the system and the position of the
stroke hit.

In details, the steepness of the lightning current determines the level of the induced
overvoltage, since the induced voltages are proportional to the self-inductance of the loop
multiplied by the steepness of the lightning current. Induced overvoltages increase also with
dimensions of the formed loops and decrease with attenuation of the magnetic field strength.
As far as the hit position concerns, the severity of the overvoltage is inversely proportional to
the distance of the point of impact. To this direction, a sensitivity analysis is performed, in
order to examine the way that various factors (peak current and steepness of the lightning
current, lightning hit position) affect the expected induced voltages. As far as the parallel
combination of the installed surge arresters is concerned, the decisive role of the voltage –
current characteristics of the non-linear resistors is demonstrated, in order to achieve the equal
sharing of the discharged impulse current and to reduce the arresters failure probability.

7.2 Induced Overvoltages

7.2.1 Literature review

The international literature deals mainly with the induced overvoltages on transmission and
distribution lines. The developed induced voltages have not been widely examine, since it is
regarded that cannot cause any damage to the substation equipment. Despite the fact that the

PhD Thesis
Page [179 of 211]
 CITY , UNIVERSITY of LONDON

induced overvoltages are expected to be low, they can influence already stressed insulation and
affect the electronic equipment of the substation. Various researchers have examined the
developed induced overvoltages in electrical networks, focusing mainly on the transmission
and distribution lines, since the field experience concerning HV/MV substations is limited.
The literature provides several methodologies and equivalent models for the evaluation of the
induced voltages, considering different systems configuration. An equivalent circuit for the
analysis of the lightning-induced voltage at a transition point is introduced in (Sekioka 2005).
The reliability of the proposed model is confirmed by comparing the calculated results of the
proposed equivalent circuit with results obtained by the application of the finite difference
method. A methodology for stochastic analysis of transient overvoltages caused by lightning
and switching in substations, is presented in (Kraulich L. et al. 2015), based on the ATP
simulation tool and the Monte Carlo Method for representing the randomness of these types
of events. In (Costea and Nicoara 2010) is analyzed how a lightning overvoltage occurred in a
superior voltage level (high or medium) could be transmitted to the inferior voltage network,
considering the influence of various factors. The study was performed on a real configuration
of network subjected of different shape and amplitude of lightning overvoltages. In (Ren et al.
2008), the lightning-generated electromagnetic fields over lossy ground produced by lightning
strikes either to flat ground or to a tall tower are estimated, applying the 2-D finite difference
time-domain method. The performed analysis reveals the dependence of the induced voltages
on ground conductivity somewhat influenced by return-stroke speed, and essentially
independent of return-stroke model. Furthermore, the induced overvoltages that may appear
on the SF6 substation enclosure during transient phenomena is examined in (Vukasović et al.
2010), highlighting the role of the peak current and the waveform of the injected lightning
surge, the dielectric strength of the SF6 gas and the geometry of the substation and its
grounding. In addition, lightning surges may also cause dangerous electromagnetic
interference problems to low voltage systems and especially to electronic devices [Psalidas M.
et al. 2004. Meliopoulos et al. 2001)].
Despite that the indirect hits result in the development of induced voltage surges with small
amplitude, comparing with those ones developed due to direct strikes, they influence the
nominal operation of electrical installations. Especially, in case of medium voltage level,
indirect strokes constitute main source of lightning faults (Viscaro 2005, Viscaro 2007, Sabiha
2009). For this reason the induced overvoltages developed at various positions of a substation
have to be studied, in order to prevent faults and damages due to nearby lightning hits.

The calculation of the induced-voltage flashover rate due to indirect lightning strokes,
demands data of the keraunik level of the region and the application of appropriate lightning
PhD Thesis
Page [180 of 211]
 CITY , UNIVERSITY of LONDON

attachment models, in order to estimate the interception radius of the conductors of the
overhead lines. An electrogeometrical model can be used for the estimation of the
interception radius (Mikropoulos and Tsovilis 2011, IEEE Std 1410-2004). However, the
interception radius can be also regarded as a statistical quantity, affected by the peak value of
the current, the geometrical characteristics of the line and the lightning interception
probability (Mikropoulos and Tsovilis 2008, Mikropoulos and Tsovilis 2009, Mikropoulos and
Tsovilis 2010).

The magnetic field around the flash channel of the lightning current may induce significant
overvoltages on overhead power lines situated in the vicinity. The return stroke phase of the
lightning discharge is considered to be the major responsible for the arising induced voltage
surges, since the intensity of the electromagnetic radiation during this phase is greater in
comparison to other phases of the phenomenon (Uman and McLain 1969). The estimation of
the induced overvoltages requires the following steps (Lambert et al. 2009, Nucci and Rachidi
1999):
a) Adoption of a return-stroke model: an appropriate return-stroke model that
determines the spatial and temporal distribution of the lightning current during the
return-stroke phase is selected. A certain number of return stroke models have been
proposed in the literature (Thottappillil and Uman 1993, Nucci 1995, Rakov and
Uman 1998, Nucci et al. 1988, Rakov and Dulzon 1991).

b) Calculation of the electromagnetic field radiated by the lightning return-stroke current:

for distances less than few kilometers, the perfect ground conductivity assumption
seems to be an adequate approach for the vertical component of the electric field and
for the horizontal component of the magnetic field (Zeddam and Degauque 1990,
Rubinstein 1996).Simplified equations are presented in (Rubinstein 1996, Cooray
1994, Wait 1997, Rachidi et al. 1996).

c) Evaluation of the induced overvoltages: overvoltages developed due to the interaction

between the electromagnetic field and the conductors of the electrical installation are
estimated by coupling models. The most used coupling models are those proposed by
Rusck (Rusck 1958), by Chowdhuri and Gross (Chowdhuri and Gross 1967), and by
Agrawal etal. (Agrawal et al. 1980).

The magnitude of the expected induced overvoltages depends on various parameters, such as
the distance between the lighting hit position and the developed voltage at a given point of the
system, the characteristics of the lightning current and the grounding resistance. The steepness
PhD Thesis
Page [181 of 211]
 CITY , UNIVERSITY of LONDON

of the lightning current determines the level of the induced overvoltage, since the induced
voltages are proportional to the self-inductance of the loop multiplied by the steepness of the
lightning current. Induced overvoltages increase also with dimensions of the loop and
decrease with attenuation of the magnetic field strength. As far as the hit position concerns,
the severity of the overvoltage is inversely proportional to the distance of the point of impact.

In details, the first step includes the calculation of the electromagnetic field radiated by the
lightning current. The lightning channel is assumed to be a vertical antenna with a current
flowing from the channel base to the cloud at a constant speed chosen equal to 1.108 m/s.
The variation of the current amplitude is supposed to be given by the following double-
exponential function (Lambert et al. 2009, Uman & McLain 1969):

( ) ( ) (7.1)

The current is given by the equation:

( ) ( ) (7.2)

( ) (7.3)

Where;
z is the height of the considered point in the channel
v is the velocity of the current waveform.
The vertical and horizontal components of the electric field radiated by the current channel
are then calculated by considering the channel as a set of vertical elementary dipoles and by
integrating over the channel height the contribution of each dipole and image. Those two
calculated field components are required as input source terms for the adopted field -to-
overhead line coupling model used for evaluating the induced voltages.

For the computation of the induced voltages, the Agrawal field-to-transmission line model is
implemented (Agrawal et al. 1980).

PhD Thesis
Page [182 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.1 Geometry for the Calculation of Overvoltages Induced on an Overhead Power Line
by an Indirect Lightning Return Stroke (Mimouni et al. 2007).

The coupling equations for a single wire horizontal line at height H above the ground
(Figure7.1 ) are given as:

( ) ( )
( ) ( ) (7.4)

( ) ( )
( ) (7.5)

Where;
R is the line resistance per unit length
L is the line inductance per unit length
G is the line conductance per unit length
C is the line capacitance per unit length
( )is the horizontal component of the exciting electric field along the x-axis at the
conductor‟s height H,
s
u (x,t) is the so-called scattered voltage, defined by:

( ) ∫ ( ) (7.6)

Where;
( ) z is the vertical component of the scattered electric field.

PhD Thesis
Page [183 of 211]
 CITY , UNIVERSITY of LONDON

The exciting electric field is the sum of the incident field radiated by the lightning stroke and
of its ground reflected field, both considered in absence of overhead wire. The scattered
electric field represents the reaction of the overhead wire to the exciting field. The boundary
conditions for the scattered voltage are:

( ) ( ) ( ) (7.7)

( ) ( ) ( )(7.8)

Where;
e e
u (0,t) and u (L,t) are the voltage sources at the two extremities of the line, obtained from the
e
exciting voltage u (x,t) given by:

( ) ∫ ( ) (7.9)

s
The solution of equation (5.12) and (5.13) gives the scattered voltage u (x, t ) at a given point
along the line, from which one can obtain the total voltage according to:

( ) ( ) ( ) (7.10)

Figure 7.2 depicts the waveforms of the induced overvoltage at a given position of an
electrical system as a function of the distance of the lightning hit point, presented in(Nucci
and Rachidi 2009).
.

PhD Thesis
Page [184 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.2Induced Overvoltage for Various Distances between the Measurement Point and the Lightning Hit
Position (Nucci and Rachidi 2009)

The line height and the ground conductivity influence substantially the number of induced
voltages greater than a given BIL (Borghetti and Nucci 1998). In Figure 7.3, the dependence
of the number of events as a function of the BIL for a 2-km long line is presented, taking into
account different values for the ground conductivity (Nucci and Rachidi 2009).

Figure 7.3 Number of Induced Voltages with Magnitude exceeding the BIL
in function with the BIL for a MV line (Nucci and Rachidi 1999, Borghetti and Nucci 1998)

7.2.2 The LIOV code

Based on the presented theoretical analysis, an appropriate simulation tool has been
developed(Nucci and Rachidi 2003) for the calculation of the induced overvoltages, which can
be combined with other tools. The LIOV (Lightning-Induced Over Voltage) code, is a
computer code which allows for the calculation of lightning-induced voltages on multi
conductor lines above a lossy soil as a function of the line geometry, lightning current wave
shape, return-stroke velocity, soil electrical parameters, etc. (Nucci and Rachidi 2003).
PhD Thesis
Page [185 of 211]
 CITY , UNIVERSITY of LONDON

The evaluation of lightning induced overvoltages is generally performed in the following way:
• The lightning return-stroke electromagnetic field change is calculated at a number of
points along the line employing a lightning return-stroke model, namely a model that
describes the spatial and temporal distribution of the return stroke current along the
channel. To this end, the return stroke channel is generally considered as a straight,
vertical antenna.
• the evaluated electromagnetic fields are used to calculate the induced overvoltages making
use of a field-to-transmission line coupling model which describes the interaction between
the LEMP and the line conductors.
• the Modified Transmission Line model with Exponential Decay (MTLE) (Nucci et al.
1988, Rachidi and Nucci 1990) is used for the specification of the spatial-temporal
distribution of current along the lightning return stroke channel.
For the calculation of the vertical component of the electric field, the assumption of a
perfectly conducting ground was considered (this assumption has been shown to be
reasonable for distances not exceeding a few kilometers or so). On the other hand, the
horizontal component of the electric field is appreciably affected by the finite conductivity of
the ground. This component is computed using the Cooray-Rubinstein formula (Rubinstein
1996, Cooray 2002). Note that in the freely downloadable version, the line and ground are
supposed to be perfect conductors. For the evaluation of the electromagnetic coupling, the
implemented model in LIOV is based on the transmission line theory and on the field-to-
transmission line coupling model of Agrawal et al. (Agrawal et al. 1980). The coupling
equations are solved using the finite difference-time domain-technique (FDTD).The LIOV
code has been experimentally validated using a number of experimental data related to natural
and triggered lightning, Nuclear Electromagnetic Pulse Simulators and reduced scale models
(Paolone et al. 2009, Rachidi 2009).The full version of the LIOV code takes into account a
certain number of features such as frequency dependence of line parameters (Rachidi et al.
2009), multi conductor lines (Rachidi et al. 1997), leader induction effects (Rachidi et al. 1997),
corona phenomenon (Nucci et al 2000), nonlinear components (surge protective devices) and
shielding wires (Paolone et al. 2004), etc.
In (Paolone M et al. 2013) interface between the LIOV code and the EMTP-RV has been
developed in order to properly simulate the response of realistic distribution networks against
external electromagnetic fields produced by nearby lightning; the developed interface is
proved to be appropriate for the distribution networks response against nearby lightning,
insulation coordination studies, lightning-to-fault correlation, design and optimal placement of
protection devices. In (Paolone et al. 2005) the LIV-ATP and LIOV-EMTP software tools for

PhD Thesis
Page [186 of 211]
 CITY , UNIVERSITY of LONDON

the estimation of induced overvoltages due to lightning hits are compared. The theoretical
approach of the programs is discussed and simulation procedure are carried out, concerning
various topologies and scenarios.

7.2.3 Configuration of the system under examination

In the current work the arising induced overvoltages at the entrance of the substation of the
configuration of Figure7.4 are calculated by using the LIOV (Lightning-Induced Over
Voltage)tool, taking into consideration various factors that affect the expected magnitudes of
the developed surges, i.e. the grounding resistance, the steepness of the waveform of the
lightning discharge current, the crest value of the lightning current and the lightning position
hit.
30m
K
l=10, 20, 30, 40m

overhead transmission lines L

MV lines
20m

A Β D
transformer
high voltage circuit breaker

main medium voltage circuit breaker

60m
medium voltage circuit breaker mast

switching device ground wire

Figure 7.4Configuration of the Model of the Arising Induced Overvoltages at the Entrance of the Substation

Three different lightning stroke positions are considered according to Fig. 7.4 (A, K, L). The
grounding resistance of the towers and the substation are equal to 5Ω and 1Ω, respectively.
For each lightning stroke position, the developed overvoltages at the beginning and the end of
the cable (BD) are evaluated for various lightning current waveforms, considering the cases a
and b of Fig. 4.1.

7.2.4 Results
Figure 7.5-7.16 depict the obtained results, highlighting the impact of the magnitude and the
rise time of the lightning current, the lightning hit position and the installation or not of surge
arresters at the joint between the incoming overhead transmission line and the underground
cable.

PhD Thesis
Page [187 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.5Induced Overvoltages at the position B for lightning hit on position A for various current waveforms
(case a)

Figure 7.6 Induced Overvoltages at the position D for lightning hit on position A for various current
waveforms (case a)

PhD Thesis
Page [188 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.7 Induced Overvoltages at the position B for lightning hit on position K for various current waveforms
(case a)

Figure 7.8 Induced Overvoltages at the position D for lightning hit on position K for various current waveforms
(case a)

PhD Thesis
Page [189 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.9 Induced Overvoltages at the position B for lightning hit on position L for various current waveforms
(case a)

Figure 7.10 Induced Overvoltages at the position D for lightning hit on position L for various current
waveforms (case a)

PhD Thesis
Page [190 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.11 Induced Overvoltages at the position B for lightning hit on position A for various current
waveforms (case b)

Figure 7.12 Induced Overvoltages at the position D for lightning hit on position A for various current
waveforms (case b)

PhD Thesis
Page [191 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.13 Induced Overvoltages at the position B for lightning hit on position K for various current
waveforms (case b)

Figure 7.14 Induced Overvoltages at the position D for lightning hit on position K for various current
waveforms (case b)

PhD Thesis
Page [192 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.15 Induced Overvoltages at the position B for lightning hit on position L for various current
waveforms (case b)

Figure 7.16 Induced Overvoltages at the position D for lightning hit on position L for various current
waveforms (case b)

The results indicate that higher steepness of the lightning current waveform results in
increased induced overvoltages. Moreover, the crest value of the impulse current is a
dominant factor that determines the magnitude of the developed induced voltages. Indeed,
higher peak values of the discharge current result in the increase of the expected voltage
surges due to induction phenomena. The dependence of the developed induced overvoltages
on the distance between the lightning hit position and the substation is clearly showed in

PhD Thesis
Page [193 of 211]
 CITY , UNIVERSITY of LONDON

Figures7.7-7.8 and 7.13-7.14. The attenuation of the magnetic field reduces the induced surges
as the distance increases. As far as the installation position of the arresters is concerned, the
outcomes reveal the role of the implementation of arresters at position B (case b).In
conclusion, the results indicate that the arising induced overvoltages are not a significant threat
for the normal operation of the substation and their influence can be adequately restrained by
adopting appropriate protective measures. However, the impact of the computed overvoltages
on the data/communication cables of the substation is an issue of further research.

7.3 Installation of Parallel Arresters

Metal oxide surge arresters constitute a basic protective measure against external and internal
transient phenomena, installed at the entrance of the substations, ensuring that the incoming
voltage surges will not exceed the insulation withstand capability of the system. However,
surge arresters are stressed by the discharge currents flowing through them and their
effectiveness and normal operation may be degraded. In case that the absorbed energy
exceeds their energy absorption capability, arresters cannot cool back-down to their normal
operating temperature and, consequently, they fail. A damaged arrester cannot protect
anymore the installation against lightning and switching overvoltages and has to be repaired or
replaced. If not, subsequent lightning hits can cause serious damages to the substation. The
international literature does not provide extended research results, concerning the installation
of metal oxide gapless surge arresters in parallel at various positions of HV/MV substations.
Studies about distribution transformers and low voltage circuits have been presented,
examining the current sharing between the parallel non-linear resistors (Chatterton 2002,
Putrus et al. 2001).

The installation of parallel surge arresters contributes to the reduction of the absorbed energy
by them, since the lightning current is shared, diverted to earth by two paths. In this way, the
energy rating of the arresters is doubled and the arresters failure probability due to lightning
overvoltages is reduced. Indeed, manufacturers of surge arresters recommend their parallel
combination, taking into consideration their voltage – current characteristics. In case that
voltage – current characteristics are not carefully matched, the sharing of the discharge
currents will be not adequate to reduce the failure probability of the arresters.

Figure 7.17 depicts the voltage – current characteristics of two different metal oxide gapless
surge arresters. Arrester 1 present lower resistance (Red line) compared with arrester 2 (Green

PhD Thesis
Page [194 of 211]
 CITY , UNIVERSITY of LONDON

line), so the lightning current will not be equally shared, but the major part will pass through
arrester 1.

voltage
Idischarge current

Figure7.17Voltages – Current Characteristics of Two Different Metal Oxide Gapless Surge Arresters

The above remark highlights the need for matching of the voltage –current characteristics of
the combined arresters, otherwise arresters that present the more intense conducting behavior
will be stressed more.

In the current work, two arresters for each phase at the entrance of a substation are
paralleled(Figures 7.18) and the expected current that will pass through each arrester is
computed, considering three cases, depending on the difference between the voltage – current
characteristic of the arresters. Details about the system characteristics are given in Chapter 4.

Figures 7.18 Two Arresters at the Entrance of a Substation in Paralleled for each Phase

Figures 7.19 – 7.21 depict the characteristics of the arresters for each case (case 1, case 2 and
case 3, depending on the difference between the voltage – current characteristic of the
arresters1.).

PhD Thesis
Page [195 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.19Voltages – Current Characteristic of the Two Arresters in Parallel (Case 1)

Figure 7.20Voltages – Current Characteristic of the Two Arresters in Parallel (Case 2)

PhD Thesis
Page [196 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.21Voltages – Current Characteristic of the Two Arresters in Parallel (Case 3)

Figures 7.22 – 7.24 present the results of the performed analysis. The current sharing is
strongly dependent on the differences between the voltages – current characteristics of the
combined arresters. Even slight differences between the characteristics result in uneven
sharing of the lightning discharge current. The arresters which present more intense non
linearity of its characteristic are stressed more and present greater probability to fail. So, the
careful choice of the electrical characteristics of the arresters to be connected is critical, in
order to achieve an adequate sharing of the lightning current and reduce the annual failure rate
of the arresters.

Figure 7.22 Sharing of the Lightning Current (Cases 1, 2, 3)

PhD Thesis
Page [197 of 211]
 CITY , UNIVERSITY of LONDON

Figure 7.23The Developed Overvoltages at Position B and D of the Examined Substation (Cases 1, 2, 3)

Figure 7.24Substation Outage Rate and Arresters Failure Rate (Cases 1, 2, 3)

7.4 Conclusions

The current Chapter examines the range of the developed induced overvoltages at the
entrance of a HV/MV substations, revealing the role of the lightning current waveforms, the
lightning hit position and the installation position of the arresters. The results indicate that the
arising induced overvoltages are not a significant threat for the normal operation of the
substation and their influence can be adequately restrained by adopting appropriate protective
measures. Main contribution of the current work is the examination of the role of the arresters

PhD Thesis
Page [198 of 211]
 CITY , UNIVERSITY of LONDON

to the magnitude of the developed induced voltages. Further work would include the impact
of the induced overvoltages of the data/communication cables, that can influence the normal
operation of the system.
As far as the parallel combination of the arresters, three cases are examined, considering
arresters with different voltage – current characteristic. The obtained results indicate that
slight differences between the characteristics can lead to unequal sharing of the injected
impulse current. The arresters which present more intense non linearity of their characteristic
are stressed more and present greater probability to fail. In the current work, two arresters for
each phase at the entrance of a substation are combined in parallel and the expected current
that will pass through each arrester is computed, considering three cases, depending on the
difference between the voltage – current characteristic of the arresters. Main contribution of
the performed analysis is the clarification that the parallel installation of surge arresters has no
significant impact on the reduction of the arising surges and reduces the energy stress of the
arresters, only if specific requirements are satisfied.

PhD Thesis
Page [199 of 211]
 CITY , UNIVERSITY of LONDON

References
Agrawal A.K., Price H.J., Gurbaxani S.H., "Transient response of a multiconductor transmission line
excited by a nonuniform electromagnetic field", IEEE Trans. on EMC, Vol. EMC-22, No. 2, pp. 119-
129, May 1980.
Borghetti A., Nucci C.A., "Estimation of the frequency distribution of lightning-induced voltages on an
overhead line above a lossy ground: a sensitivity analysis", Int. Conf. On Lightning Protection, Birmingham,
Sept. 1998.
Chatterton Baden George, ―The use of surge arresters in parallel for the lightning protection of pole mounted
distribution transformers in Eskom‖, School of Electrical and Electronic Engineering University of Natal,
December 2002, Master of Science Thesis
Chowdhuri P., Gross E.T.B., "Voltage surges induced on overhead lines by lightning strokes" Proc. IEE,
Vol. 114, No. 12, pp. 1899-1907, Dec. 1967.
Cooray V., "Horizontal fields generated by return strokes," Radio Science, vol. 27, pp. 529-37, 1992; _
―Some considerations on the "Cooray-Rubinstein" formulation used in deriving the horizontal electric field of
lightning return strokes over finitely conducting ground‖, IEEE Trans on EMC, Vol. 44, No. 4, November
2002.
Costea M. and Nicoara B., ―Influence factors on the transmitted overvoltages from high voltage to low voltage
networks‖, U.P.B. Sci. Bull., Series C, Vol. 72, Iss. 1, 2010
Franc B., Filipović-Grčić B., Milardić V., ―Lightning overvoltage performance of 110 kV air-insulated
substation‖, Electric Power Systems Research Vol. 138, pp. 78-84, Sept. 2010
IEEE Guide for improving the Lightning Performance of Electric Power Overhead Distribution Lines,
IEEE Std. 1410-2004, Jul. 2004
Iorio R., Nucci C.A., Porrino A., Rachidi F., "Lightning-induced overvoltages on distribution overhead lines
in presence of distribution transformers or short cables ", International CIGRÉ Symposium on Power System
Electromagnetic Compatibility, Lausanne, Oct. 1993.
Kraulich L., Marchesan A. C., Mariotto L., Loose F., Camargo M. C., Marchesan G., Cardoso G.,
―Stochastic Analysis of Transient Overvoltages at Substations by Applying the Monte Carlo Method
Integrated with ATP‖, International Conference on Power Systems Transients (IPST2015), Cavtat, Croatia
June 15-18, 2015
Lambert R., Xémard A., Fleury G., Tarafi R., Zeddam A., ―Estimating the expected failure rate of
distribution type equipment due to lightning induced overvoltages‖, IPST 2009
Mikropoulos P. N and Tsovilis T. E.: ―Estimation of lightning incidence to overhead transmission lines‖,
IEEE Trans. Power Delivery, Vol. 25, No. 3, pp. 1855-1865, Jul. 2010
Mikropoulos P. N. and Tsovilis T. E. ―Interception probability and shielding against lightning‖, IEEE
Trans. Power Delivery, Vol. 24, No. 2, pp. 863-873, Apr. 2009

PhD Thesis
Page [200 of 211]
 CITY , UNIVERSITY of LONDON

Mikropoulos P. N. and Tsovilis T. E., ―A statistical method for the estimation of induced-voltage flashover
rate of unshielded overhead distribution lines‖, XVII International Symposium on High Voltage Engineering,
Hannover, Germany, August 22-26, 2011
Mikropoulos P. N. and Tsovilis T. E.: ―Striking distance and interception probability‖, IEEE Trans.
Power Delivery, Vol. 23, No. 3, pp. 1571-1580, Jul. 2008
Mimouni A., Azzouz Z., Ghemri B., ―Lightning induced overvoltages on overhead lines: modelling and
experimental validation‖, Journal of Electrical Engineering, Vol. 58, no. 3, 2007, pp. 146-151
Nucci C. A., Mazzetti C., Rachidi F., Ianoz M., "On lightning return stroke models for LEMP
calculations", 19th International Conference on Lightning Protection, Graz, April 1988.
Nucci C.A and Rachidi F., "Interaction of electromagnetic fields generated by lightning with overhead electrical
networks," in The Lightning Flash, V. Cooray, Ed. London: IEE, 2003, pp. 425-478.
Nucci C.A., "Lightning-induced voltages on overhead power lines. Part I: return stroke current models with
specified channel-base current for the evaluation of the return stroke electromagnetic fields", Electra, No. 161,
pp. 74-102, August 1995.
Nucci C.A., Guerrieri S., Correia de Barros M.T., Rachidi F., "Influence of corona on the voltages induced
by nearby lightning on overhead distribution lines", IEEE Trans. on Power Delivery, Vol. 15, No. 4, pp.
1265-1273, October 2000.
Nucci C.A., Mazzetti C., Rachidi F., Ianoz M., "On lightning return stroke models for LEMP
calculations", Proc. 19th Int. Conf. on Lightning protection, Graz, April, 1988.
Nucci C.A., Rachidi F., ―Lightning-Induced Overvoltages‖, IEEE Transmission and Distribution
Conference, Panel Session "Distribution Lightning Protection", New Orleans, April 14, 1999
Paolone M, Rachidi F., Borghetti A., Napolitano F., Nucci C.A., ―An advanced interface between the
Lightning Induced Overvoltage Code (LIOV) and EMTP-RV, EMTP-RV European User Group
Meeting, April 3, 2012 – Paris, La Defense, France
Paolone M., Nucci C.A., Petrache E. and Rachidi F., "Mitigation of Lightning-Induced Overvoltages in
Medium Voltage Distribution Lines by Means of Periodical Grounding of Shielding Wires and of Surge
Arresters: Modelling and Experimental Validation", IEEE Trans. on Power Delivery, Vol. 19, No. 1, pp.
423-431, Jan. 2004
Paolone M., Perez E., Borghetti A., Nucci C.A., Rachidi F. and Torres H., ―Comparison of Two
Computational Programs for the Calculation of Lightning-Induced Voltages on Distribution
Paolone M., Rachidi F., Borghetti A., Nucci C.A., Rakov V.A., Uman M.A., "Lightning Electromagnetic
Field Coupling to Overhead Lines: Theory, Numerical Simulations and Experimental Validation", IEEE
Transactions on Electromagnetic Compatibility, Vol. 51, No. 3, August2009, pp. 532-547.
Psalidas M., Agoris D., Pyrgioti E., Karagiannopoulos C., ―Modeling for the Calculation of Overvoltages
Stressing the Electronic Equipment of High Voltage Substations due to Lightning‖, 8th WSEAS
International Conference on Circuits, July 2004
Putrus G.,. Ran L,. Ahmed M., ―Improving current sharing between parallel varistors‖, ISIE 2001, Pusan,
PhD Thesis
Page [201 of 211]
 CITY , UNIVERSITY of LONDON

KOREA
Rachidi F., "A Review of Field-to-Transmission Line Coupling Models with Special Emphasis to Lightning-
Induced Voltages", IEEE Transactions on Electromagnetic Compatibility, Vol 54, No. 4, pp. 898 - 911,
2012.
Rachidi F., Nucci C.A., "On the Master, Lin, Uman, Standler and the Modified TransmissionSystems‖,
International Conference on Power Systems Transients (IPST‘05) in Montreal, Canada on June 19-23, 2005
Paper No. IPST05 - 250
Rachidi F., Nucci C.A., Ianoz M., Mazzetti C., "Influence of a lossy ground on lightning-induced voltages on
overhead lines", IEEE Trans. On Electromagnetic Compatibility, Vol. 38, No. 3, pp. 250-264, Aug.
1996.
Rachidi F., Nucci C.A., Ianoz M., Mazzetti C., "Influence of a lossy ground on lightning-induced voltages on
overhead lines", IEEE Trans. on Electromagnetic Compatibility, Vol. 38, No. 3, August 1996.
Rachidi F., Nucci C.A., Ianoz M., Mazzetti C., "Response of multiconductor power lines to nearby lightning
return stroke electromagnetic fields", IEEE Trans. on Power Delivery, Vol. 12, pp. 1404-1411, July 1997.
Rachidi F., Rubinstein M., Nucci C.A., Guerrieri S., "Voltages induced on overhead lines by dart leaders
and subsequent return strokes in natural and rocket-triggered lightning", IEEE Trans. on Electromagnetic
Compatibility, Vol. 39, No. 2, May 1997.
Rakov V., Dulzon A.A., "A modified transmission line model for lightning return stroke field calculations",
Proc. 9th Int. Symposium on EMC, Zurich, Switzerland, pp. 229-235, March 1991.
Rakov V., Uman M.A., "Review and evaluation of lightning return stroke models including some aspects of
their application", IEEE Trans. On EMC, Vol. 40, No. 4, pp. 403-426, Nov. 1998.
Ren He-Ming, Zhou Bi-Hua, Rakov Vladimir A., Shi Li-Hua, Gao Cheng, and Yang Jian-Hua,
―Analysis of Lightning-Induced Voltages on Overhead Lines Using a 2-D FDTD Methodand Agrawal
Coupling Model‖, IEEE Transaction on Electromagnetic Compatibility, Vol. 5, no. 3, August 2008,
pp.651-659
Rubinstein M., "An approximate fomula for the calculation of the horizontal electric field from lightning at
close, intermediate, and long range," IEEE Trans. on Electromagnetic Compatibility, vol. 38, pp. 531-535,
1996 1996.
Rubinstein M., "An approximate formula for the calculation of the horizontal electric field from lightning at
close, intermediate, and long range", IEEE Trans. on EMC, Vol. 38, No. 3, Aug. 1996.
Rusck S., "Induced lightning overvoltages on power transmission lines with special reference to the overvoltage
protection of low voltage networks", Transactions of the Royal Institute of Technology, Stockholm, No. 120,
1958.
SabihaNehmdoh A., ―Lighting-induced overvoltages in medium voltage distribution systems and customer
experienced voltage spikes‖, Doctoral Dissertation Aalto University School of Science and Technology Faculty
of Electronics, Communications and Automation Department of Electrical Engineering 2009
Sekioka S., ―An Equivalent Circuit for Analysis of Lightning-Induced Voltages on Multiconductor System
PhD Thesis
Page [202 of 211]
 CITY , UNIVERSITY of LONDON

Using an Analytical Expression‖, International Conference on Power Systems Transients (IPST‘05) in

Montreal, Canada on June 19-23, 2005 Paper No. IPST05 – 029
Thottappillil R, Uman M.A., "Comparison of lightning return-stroke models", J. Geophys. Res., vol. 98, pp.
22,903-22,914, Dec. 1993.
Uman M.A., McLain D.K., "Magnetic field of lightning return stroke", J. Geophys. Res., Vol. 74, 1969.
Viscaro S., ―Direct Strokes to Transmission Lines: Considerations on the mechanisms of Overvoltage
Formation and their Influence on the Lightning Performance of Lines‖, Journal of Lightning Research, Vol. 1,
pp. 60-68, 2007
Viscaro S., ―Lightning: an Engineering Approach‖ (in Portuguese), ArtLiber Edit-Sao, Wait J.R.,
"Concerning horizontal electric field of lightning", IEEE Trans. on EMC, Vol. 39, No. 2, May 1997.
Vukasović V., Baus Z., Barukčić M., ―The Influence of Atmospheric Overvoltages on High-Voltage SF6
Substations‖, International journal of electrical and computer engineering systems, Vol.2 No.1 May 2011.
Zeddam A., Degauque P., "Current and voltage induced on telecommunications cable by a lightning return
stroke", Lightning Electromagnetics, Edited by R. L. Gardner, Hemisph. publ. Corp., pp.377-400, 1990.
Α.P. Meliopoulos, G. Cokkinides, A. Arora, J.C. Braz, ―Analysis and Mitigation of Induced Voltages on
Substation Control Cables‖, International Conference on Power System Transients, June 24-28, 2001, IPST
2001, Rio de Janeiro, Brazil, pp. 529-534

PhD Thesis
Page [203 of 211]
 CITY , UNIVERSITY of LONDON

CHAPTER 8

CONCLUSIONS, CONTRIBUTION AND FUTURE

WORK

8.1 Introduction

The current thesis dealt with the lightning performance of HV/MV substations and the
examination of the appropriate protective measures that should be adopted, in order to
restrain the developed overvoltages and reduce the expected failure rate. To this direction, two
different topologies of substations are examined and a sensitivity analysis is performed,
considering the impact of various factors on the expected voltage surges at different position
of the system and substation outage rate.

8.2 CONCLUSIONS

8.2.1 Estimation of the Expected Overvoltages at Different Positions of

the Substation

As far as the calculation of the arising overvoltages at the entrance of the substations is
connected, the performed sensitivity analysis results in the following conclusions, about the
behavior of the substation in case of a lightning hit on the phase conductor of the connected
overhead transmission line:
 Τhe developed overvoltages at the end of the cable are greater than those ones at the
beginning, due to the reflection phenomena of the traveling waves.
 The developed overvoltages are strongly dependent on the length of the cable. Long cable
lengths present a better lightning performance and contribute to the reduction of the
expected overvoltages. Especially, in case that the grounding resistance cannot be
improved, the increase of the cable length can balance the negative effects of the
inadequate grounding system. The results indicate that long cables can be adequately
protected, since the expected overvoltages do not exceed the BIL (750kV for 150kV
nominal voltage).

PhD Thesis
Page [204 of 211]
 CITY , UNIVERSITY of LONDON

 Low grounding resistance values contribute to the decrease of the developed overvoltages,
diverting significant part of the lightning current to the grounding system. Low values of
grounding resistance values, either for the tower or for the surge arresters, always improve
the lightning performance of the line. Low tower footing resistances do not allow the
potential of the tower to exceed the BIL and prevents backflashover. The arresters‟
grounding resistance keeps the voltage at the terminals of the transformer below the
desired limits.
 The arresters should be installed near the transformer; otherwise the voltage at the
equipment to be protected will be considerably higher than the residual voltage at the
terminals of the arrester. The protective distance of the installed arresters depends on the
steepness of the incoming surge and the nominal protection level of the device. The
installation of the arresters away from the equipment to be protected reduces their
efficiency, resulting in the development of significant overvoltages higher than the
nominal residual voltage of the arresters.

Furthermore, the impact of the number of the connected overhead transmission lines is
examined. The arising overvoltages are strongly dependent on the number of the lines and the
lightning hit position. The peak value and the rise time of the incoming voltage surges
decrease with the increase of the number of the transmission lines, since the overvoltage
“faces‟ the parallel combination of the incoming lines. The probability of lightning hit on a
transmission lines is reduced as the number of the incoming lines is increased.

8.2.2 The influence of the used Models of the Substations Components

A sensitivity analysis is performed, in order to examine the influence of the equivalent circuit
model of the various components (i.e. tower, arresters, grounding, and insulators) to the
obtained results. As far as the tower models concerns the Jordan, Wagner, Sargent, Ametani,
Rondon, Yamada and Chilsom circuits are considered. Physical model, Pinceti-Giannettoni,

Fernadez-Diaz and IEEE model are used for the representation of the surge arresters.
The implemented models for the grounding resistance include Oettle, Chilsom, Weck, Yasuda,
Sverak and IEEE models. Finally, the Weck, Pigini and V-t curve models are used for the
insulator representation. As default models are considered the Chilsom model for the towers,
the IEEE model for the arresters, and the Cigre model for the grounding resistance and the
V-t curve for the insulator. Two cases are examined, i.e. surge arresters have been installed or

PhD Thesis
Page [205 of 211]
 CITY , UNIVERSITY of LONDON

not. The obtained results indicate that the selected tower model influences the theoretical
estimations, but its impact is not critical.

8.2.3 Outage Rate of the Substation due to Lightning Strokes

A sensitivity analysis of the expected failure rate in function with the tower footing resistance,
the length of the cable and the arresters installation position is presented. Τhe obtained
outcomes indicate that the expected overvoltage at the entrance of the substation is reduced
for higher number of incoming transmission lines. The impact of the lightning hit position is
also highlighted, indicating the influence of the distance between the lightning stroke position
and the substation to the incoming surge voltages.

The results emphasize also the great impact of the grounding resistance, the length of the
cable and the installation position of the implemented arresters to the expected failure rates.
In details, the grounding resistance is the dominant factor that determines the lightning
performance of the system, since it influences the occurrence or not of backflashover
phenomena and affect the effectiveness of the protection that surge arresters provide.
Furthermore, the length of the cable is of great importance, since influences the magnitude of
the expected overvoltages. The distance between the transformer and the arresters affects the
lightning performance of the system, imposing their installation near the entrance of the
substation.

8.2.4 Induced Overvoltages

The magnitude of the expected induced overvoltages depends on various parameters, such as
the distance between the lighting hit position and the developed voltage at a given point of the
system, the characteristics of the lightning current and the grounding resistance. The steepness
of the lightning current determines the level of the induced overvoltage, since the induced
voltages are proportional to the self-inductance of the loop multiplied by the steepness of the
lightning current. Induced overvoltages increase also with dimensions of the loop and
decrease with attenuation of the magnetic field strength. As far as the hit position is
concerned, the severity of the overvoltage is inversely proportional to the distance of the point
of impact.
The presented study calculated the expected induced overvoltages at the entrance of the
substation under study, highlighting the impact of various factors and parameters on the
magnitude of the computed surges. However, the results indicated that the arising induced
PhD Thesis
Page [206 of 211]
 CITY , UNIVERSITY of LONDON

overvoltages are not a significant threat for the normal operation of the substation and their
influence can be adequately restrained by adopting appropriate protective measures.

8.2.5 Parallel Arresters

In the current thesis, two same arresters for each phase at the entrance of a150/20kV
substation were combined in parallel and the expected current that will pass through each
arrester was computed, considering three cases, depending on the difference between the
voltages – current characteristic of the arresters. The current sharing was strongly dependent
on the differences between the voltages – current characteristics of the combined arresters.
The performed analysis indicated that even slight differences between the characteristics result
in uneven sharing of the lightning discharge current. The arresters which present more intense
non linearity of its characteristic were stressed more and present greater probability to fail. So,
the careful choice of the electrical characteristics of the arresters to be connected was critical,
in order to achieve an adequate sharing of the lightning current and reduce the annual failure
rate of the arresters.

PhD Thesis
Page [207 of 211]
 CITY , UNIVERSITY of LONDON

8.3 CONTRIBUTION

The contribution of the PhD project is described as following:

 The current work examines the lightning performance of substations, considering the
influence of various parameters (not only the grounding resistance, which is the most
common factor presented in the international research bibliography). The performed
study highlights the role of the installation position of the arresters, the length of the cable
that connects the incoming overhead transmission line and the transformer and the
number of the connected lines. Compared with other research outcomes, main
contribution of the thesis is that does not focus only on the role of the grounding
resistance, but examines the dominant influence of other parameters, revealing that the
improvement of the lightning performance of the substation can be achieved by the
appropriate adjustment of various factors of the system.
 The obtained simulation results show that the impact of the used models equivalent circuit
for each component of the substation is not significant, since cannot change dramatically
the estimation of the arising voltage surges. Each equivalent circuit model has advantages
and drawbacks, but in general all the proposed models are appropriate for transient
analysis studies, extracting adequate results. The current work highlights the advantages
and drawbacks of each equivalent circuit model, providing a guide to other researchers in
order to select the appropriate models.

 The estimation of the substation outage rate due to atmospheric overvoltages is carried
out, considering the role of the tower footing resistance, the protection distance of the
arresters and the configuration of the substations. Innovation of the performed study
constitutes the inclusion of the arresters failure rate to the total substation failure rate,
since arresters are part of the equipment of the substation and their possible failure results
in malfunction of the nominal operation of the system.

• The induced overvoltages developed at the entrance of the system under examination are
estimated by using appropriate simulation tool, considering various lightning current
waveforms and lightning hit positions. The current thesis contributes to the study of the
induced surges on substations, by examining the role of the installed arresters for the
restrain of the expected induced voltages.

PhD Thesis
Page [208 of 211]
 CITY , UNIVERSITY of LONDON

 The parallel installation of surge arresters is a common practice in order to reduce the
absorbed energy by them and, consequently, their failure probability. However, the current
work emphasizes the need for good matching of the voltage – current characteristics of
the arresters; otherwise, the expected equal sharing of the injected lightning current will
not be achieved. The outcomes of the analysis reveal that the combination of the arresters
in parallel does not strongly influence the expected overvoltages, but has to do mainly
with the absorbed energy by the arresters.

PhD Thesis
Page [209 of 211]
 CITY , UNIVERSITY of LONDON

8.4 FUTURE WORK

Future work includes:

 Examination of other topologies (substations without cable connections, interconnection
between the power transformers, different types of metallic towers): the performed
analysis presented in the current Thesis should be carried for different topologies, in order
to gain a general perspective about the impact of the examined factors on the developed
overvoltages and the expected outage rate of the substations.
 Study of the lightning performance of MV/LV substations, in order to protect the LV
consumers against lightning surges: MV/LV transformers are vulnerable to atmospheric
overvoltages, that affect the normal power supply and may cause serious damages to the
consumers. Surge arresters installed at the MV side of the transformer offer protection
against direct or indirect lightning strokes to the MV distribution line. However, the
efficiency of the protection level that MV surge arresters provide is strongly dependent on
the achieved grounding resistance of the transformers and the installation position of the
arresters. Considering that the demanded grounding resistance (1Ω) is rarely achieved, the
impact of the grounding resistance on the developed overvoltages has to be studied,
considering also the installation position of the arresters. Furthermore, surges arising at
the LV side of the power transformer due to lightning strokes to the LV distribution line
have to be studied, concerning that LV side is not usually protected by surge arresters.
 Implementation of arresters with different characteristics (V-I characteristics, Maximum
Continuous Operating Voltage, Energy Class etc.): the current work examines the parallel
connection of surge arresters, presenting different voltage – current characteristics. The
study of parallel connection of arresters that present different maximum operating voltage
or energy class would contribute to reveal the role of the nominal electrical characteristics
of the arresters, in an effort to reduce their failure rate.
 Combination of arresters with other protective devices (e.g. spark gaps): it is common
practice to use spark gaps, instead of metal oxide gapless surge arresters, for the protection
of the HV side of the power transformer, in an effort to reduce the installation and the
potential repair or replacement cost. The examination of the effectiveness of the spark
gaps and the study of their behavior in parallel connection with surge arresters is an issue
for further investigation. Moreover, the arresters‟ failure probability in case that they
operate in combination with spark gaps should be evaluated.

PhD Thesis
Page [210 of 211]
 CITY , UNIVERSITY of LONDON

 Laboratory or field measurements in order to confirm the theoretical approaches: the

installation of monitoring systems in order to record the lightning hits and the arising
overvoltages at different positions of a substation, laboratory experiments for the
measurement of the discharge current through the arresters in case of installation of
parallel protective measures should be carried out, in order to verify the appropriateness
and the reliability of the performed theoretical analysis.
 Methodologies for the improvement of the tower footing resistance in order to reduce the
lightning failure rate of the substation: the current thesis indicated the critical role of the
tower footing resistance, in an effort to improve the lightning performance of the
substations and reduce the expected outage rate. The improvement of the grounding
resistance of the towers of the connected lines, according to technoeconomical criteria, is
an issue of priority for the power utilities. Despite the fact that several methods for the
improvement of the tower footing resistance are provided in the international literature,
the selection of the appropriate procedure (type and dimensions of grounding system, soil
etc) and the estimation of the critical number of towers, the grounding resistance of which
should be improved, are subject to be investigated.
 Development of appropriate computer tool for the optimum design of the Lightning
Protection System (LPS) of the substation and the selection of the electrical characteristics
of its components: the method has to take into consideration the geometrical
characteristics and the type of the substation, the resistivity of the ground, the grounding
system and the achieved grounding resistance, the characteristics and the installation
position of the arresters, the keraunik level of the regions and other critical parameters.
The proposed software will be useful for the engineers to design efficient LPSs,
considering the common practice and technoeconomical criteria.

PhD Thesis
Page [211 of 211]