The impact of grounding on the
Alliander power system
Master of Science thesis
Faculty of Electrical Engineering, Mathematics and Computer Science (EEMCS)
Delft University of Technology
Name:
Amrish Sookhlall BSc.
Student number:
1399489
Supervisor Alliander NV:
Ir. Nico Steentjes &
Supervisor TU Delft:
Dr. Ir. Marjan Popov
Ing. John van Slogteren
Thesis committee:
Prof. Ir. L. van der Sluis
Dr. Ir. M. Popov
Ir. N. Steentjes
Dr. Ir. D. Djairam
Delft University of Technology, Thesis supervisor
Delft University of Technology, Daily supervisor
Liander , Daily supervisor
Delft University of Technology
-1-
Preface
For completing my MSc study a thesis should be written about a certain subject in the field of
electrical power engineering (E.P.E). This project has been done for Liander. There are many
persons who helped me to achieve this goal. First I want to thank GOD for giving me the
strength and energy to finish my study. I also want to thank my parents Andre and Trude for
there support and motivation. From Delft University of Technology I want to thank dr.ir.
Marjan Popov for couching and supervising me through my project. From Alliander NV I
want to thank ir. Nico Steentjes, for being such a patient and helpful person, also ing. John
van Slogteren for always helping me and giving me that boost that I needed. I thank my
family in the Netherlands and my caring girlfriend, Vandana and her mother. At last my
friends especially my friend Vikaash and not to forget my sister Reena.
Amrish Sookhlall
Delft, Netherlands
-2-
Summary
Within the department Asset policy of Liander there is a current study into the way of
grounding of 10 and 20 kV networks. This research serves information as a context at the
drawing up of policy. The aim of the research is to examine the influence of nature errors at
different type of grounding methods on not delivered consumer minutes (NDCM) (in Dutch
SVBM = storingsverbruikersminuten). There has been an increase in the NDCM, but in the
last year a substantial reduction is achieved.
One of the main concerns of utilities, nowadays, is grounding the distribution system.
Distribution systems are usually three phase systems. Grounding the neutral wire will affect
the power quality and characteristics of distribution systems during unbalanced conditions,
specially phase to ground faults. This paper describes the impact of different grounding
practices in distribution systems on ground fault responses. Most companies use ground relays
to detect these fault current. These relays should act between certain boundaries, because
different fault location in a network shows different fault current values.
So in short the main problem is: In distribution networks, ground relays are used to
localize ground faults in the system. Thus, the time need to identify a fault location is
related to the sensitivity of the grounding method to the fault currents. This means that
different grounding method may result in different NDCM depending on the actual fault
current levels of the respective grounding method. What is the impact of grounding
types on the NDCM?
Chapter 1 gives a brief introduction about the thesis. It gives the background of the work, the
definition and scope. Some research questions are defined, and it is explained how these will
be done.
Chapter 2 describes the importance of grounding.
Different types of grounding that are currently used; their advantages and disadvantages are
described. The calculation of the grounding parameters in different grounding schemes is also
done, and the preferred grounding type in a network is given.
Chapter 3 describes the different types of faults that can occur in a network.
This chapter calculates the currents and voltages in different phases during the fault. Also
examples are given to clearly understand the fault situations.
Chapter 4 describes the different ways transformers can be coupled.
Chapter 5 describes causes and consequences of faults in the distribution network. This
chapter also gives an overview about the NDCM in different areas. It will be shown how these
NDCM are spread in different areas. In this chapter, diagrams and pictures are displayed to
see the differences.
Chapter 6 starts with a short summary about PSCAD, the way how PSCAD works and what
kind of calculation method is used in PSCAD. Different case studies are done with different
kinds of grounding and transformer couplings. The simulations results which are fault
currents and voltages are analyzed.
-3-
In the last chapter, chapter 7 the conclusions and recommendation are given. In this chapter
there is also a discussion on the types of grounding.
-4-
Contents
PREFACE ....................................................................................................................................................... - 2 SUMMARY ................................................................................................................................................... - 3 CONTENTS .................................................................................................................................................... - 5 LIST OF SYMBOLS AND ABBREVIATIONS ....................................................................................................... - 7 CHAPTER 1 INTRODUCTION .......................................................................................................................... - 9 1.1 BACKGROUND OF WORK ..................................................................................................................................- 9 1.2 DEFINITION AND SCOPE....................................................................................................................................- 9 1.3 RESEARCH QUESTIONS ...................................................................................................................................- 10 1.4 APPROACH ..................................................................................................................................................- 10 CHAPTER 2 IMPORTANCE OF GROUNDING................................................................................................. - 11 2.1 TYPES OF GROUNDING ...................................................................................................................................- 11 2.1.1 Resistive grounded system ............................................................................................................. - 11 2.1.2 Inductive grounded system (Peterson coil) ..................................................................................... - 13 2.1.3 Ungrounded system ........................................................................................................................ - 14 2.1.4 Solidly-grounded systems ............................................................................................................... - 16 2.2 GROUNDING SCHEMES SUITABLE FOR DIFFERENT NETWORK STRUCTURES .................................................................- 17 2.3 CALCULATION OF GROUNDING PARAMETERS AND CAPACITANCE CURRENT ................................................................- 18 2.4 CONCLUSIONS ..............................................................................................................................................- 19 CHAPTER 3 FAULT TYPES ............................................................................................................................ - 20 3.1 THREE PHASE ...............................................................................................................................................- 21 3.2 PHASE TO PHASE TO GROUND..........................................................................................................................- 22 3.3 PHASE TO PHASE...........................................................................................................................................- 23 3.4 SINGLE PHASE-TO-GROUND (GROUND FAULT) ....................................................................................................- 24 3.5 CONCLUSIONS ..............................................................................................................................................- 26 CHAPTER 4 TRANSFORMER CONNECTIONS ................................................................................................ - 27 4.1 TRANSFORMER CONNECTION ..........................................................................................................................- 27 4.2 CONCLUSIONS ..............................................................................................................................................- 30 CHAPTER 5 PROBLEMS CONCERNING EARTH FAULTS ................................................................................. - 31 5.1 TECHNICAL AND DAMAGED CAUSED BY DIGGING ACTIVITIES ..................................................................................- 32 5.1.1 Technical ......................................................................................................................................... - 32 5.1.2 Damaged caused by digging activities ........................................................................................... - 34 5.2 ANALYZING THE NDCM ................................................................................................................................- 35 5.3 CONCLUSIONS ..............................................................................................................................................- 45 CHAPTER 6 SIMULATIONS .......................................................................................................................... - 46 6.1 SIMULATIONS BY PSCAD ...............................................................................................................................- 46 6.2 NETWORK TYPE 1 .........................................................................................................................................- 47 6.2.1 Delta-wye connection of transformer ............................................................................................. - 49 6.2.2 Wye-wye connection of transformer .............................................................................................. - 54 6.2.3 Wye-delta and delta-delta connection of transformer ................................................................... - 56 6.2.4 Difference between delta-wye and wye-wye .................................................................................. - 57 6.2.5 Difference between wye-wye and delta-delta ................................................................................ - 62 6.3 NETWORK TYPE 2 .........................................................................................................................................- 64 6.4 CONCLUSIONS ..............................................................................................................................................- 68 CHAPTER 7 CONCLUSIONS AND RECOMMENDATIONS ............................................................................... - 69 -
-5-
LIST OF LITERATURE ................................................................................................................................... - 72 APPENDICES ............................................................................................................................................... - 74 APPENDIX A SIMULATION PROCEDURE FOR NETWORK TYPE 1 ......................................................................................- 74 Appendix A1 Delta-Wye connection of transformer ................................................................................ - 75 Appendix A2 Wye-Wye connection of transformer ................................................................................. - 83 Appendix A3 Wye-Delta connection of transformer ................................................................................ - 93 Appendix A4 Delta-Delta connection of transformer .............................................................................. - 96 APPENDIX B SIMULATION PROCEDURE FOR NETWORK TYPE 2 ......................................................................................- 99 Appendix B1 Delta-Wye vs Wye delta connection with different grounding types for ground faults. .. - 100 Appendix B2 Wye-Wye vs Wye-Wye connection with different grounding type for ground faults ....... - 103 Appendix B3 Wye-Wye vs Wye-delta connection with different grounding types for ground faults .... - 105 Appendix B4 Delta-Wye vs Wye-Delta connection for different grounding type on ground faults. ...... - 107 Appendix B5 Wye-Wye vs Wye-Wye connection for different grounding type on ground faults. ......... - 110 Appendix B6 Wye-Wye vs Wye-Wye connection for different grounding type on ground faults. ......... - 113 APPENDIX C REACTANCE GROUNDING...................................................................................................................- 115 APPENDIX D CALCULATION OF GROUNDING RESISTANCE ..........................................................................................- 124 -
-6-
List of symbols and abbreviations
NDCM:
PSCAD:
I:
R:
V:
HV:
Rn:
Xc0:
R0 :
X0:
T:
If:
Ic:
π:
C0 :
L:
Id:
In:
Ir:
Pr:
Pp:
S:
I0:
I
I2:
Zf:
Ia:
Ib:
Ic:
Va:
Vb:
Vc:
Z1:
Z0:
Z2:
a:
a²:
:
A:
P:
Q:
S:
L-G:
L-L:
L-L-G:
Not Delivered Consumer Minutes
Power System Computer Aid Design
Current (amps-A)
Resistance (ohms-Ω)
Voltage (volts-V)
High Voltage (above 36 kV)
neutral resistor (ohms-Ω)
capacitive reactance (ohms-Ω)
zero sequence resistance (ohms-Ω)
positive reactance (ohms-Ω)
zero sequence reactance (ohms-Ω)
time (seconds)
fault current (amps-A)
capacitance current (amps-A)
3.1415927
Single phase to earth capacitance (farads-F)
length of cable (meters-m)
single phase fault current (amps-A)
neutral point current (amps-A)
nominal current of the neutral grounding resistor (amps-A)
nominal power of the grounding resistor (watt-W)
power of the Peterson coil (watt-W)
power reserve factor; 1.25-1.35
zero sequence current (amps-A)
positive sequence current (amps-A)
negative sequence current (amps-A)
fault impedance (ohms-Ω)
current in phase a (amps-A)
current in phase b (amps-A)
current in phase c (amps-A)
voltage in phase a (volts-V)
voltage in phase b (volts-V)
voltage in phase c (volts-V)
positive sequence impedance (ohms-Ω)
zero sequence impedance (ohms-Ω)
negative sequence impedance (ohms-Ω)
1120
1240
Electrical resistivity
cross section (m²)
active power (MW)
reactive power (MVar)
apparent power = P + j Q
phase to ground
phase to phase
two phase to ground
-7-
L-L-L:
L-L-L-G:
f:
MV:
three phase
three phase to ground
frequency (Hertz-Hz)
medium voltage (1kV < U < 36kV)
-8-
Chapter 1 Introduction
1.1 Background of work
As it is know from an industrial point of view, fast detection and fault location is very
important. The longer it takes to find a fault the more money it costs. A fault in a system is
unavoidable, so if a fault occurs it should be within some limits and visible; with other words
it must be detected, and in this way the reliability can be improved. From theoretical
background it is known that grounding is the major factor to control fault currents and
voltages.
Grounding of Power systems has a large impact on short circuit currents and voltages during
faults. System grounding is used to control short circuit currents and voltages at the
occurrence of a fault. There are different kinds of grounding as it will be described later on.
As this research is done for Liander, a brief description about this company will be given.
Liander is a large distribution system operator of gas and electricity in the Netherlands. They
serve over almost 3 million customers in several Provinces of Netherland.
One of main concerns of the company is to keep the time duration of outages at a low
acceptable level. These outages are related to NDCM know as: not delivered consumer
minutes. NDCM are related to time duration of faults that occur in the network. The Liander
distribution network is spread all over The Netherlands. The Netherlands is divided into
several Provinces, and each of these Provinces has different network structures and also their
own NDCM.
To get a better view into the NDCM of these Provinces, they should be analyzed separately as
long as failure data is available. To get a better understanding of the relation between types of
grounding and fault currents/voltages simulations should be done. The outcome of the
simulation results can be used to give more detailed information about the NDCM in different
cities. PSCAD is used as simulation program to perform dynamic simulation. More
information about this program will be described in chapter 6.
1.2 Definition and scope
Grounding is the main aspect of this research. Different kinds of grounding together with
different types of fault are used to run simulations. Practically, it is known that about 70% to
90% of the faults are single phase to ground, so this type of fault is intensively used in the
simulations. In many cases the three phase fault is the most severe type of fault, focus will be
also on this fault type.
The title of this study is formulated as:
The impact of faults on the network with different grounding schemes
The research was performed by using a model that was used in Vision. All the data was used
from the vision program and put into PSCAD. Some modifications were needed especially
with the cable system, which is more advanced in PSCAD.
-9-
1.3 Research questions
Objective 1
Here some general knowledge is gathered about grounding, fault types and transformer
connection. The differences and other important parameters about the three subjects
mentioned are provided. Objective 1 is a theoretical research.
Objective 2
This is a more analytical research where available data is analyzed
This is done to know to main reasons for outages and on what conditions.
What are the main reasons for outages?
What are the network specifications?
Objective 3
Here simulations are done to understand the different behaviour of currents and voltages in
different grounding schemes.
What is the impact of faults on different grounding schemes?
Is there any difference between different grounding methods?
1.4 Approach
This project was done in the following sequence:
- first some basics were given about grounding techniques
- different types of fault that can occur in the power system
- variation of transformers couplings
- simulation were done using a network model ( different grounding and generation
scheme)
- finally, conclusions and recommendation are provided
- 10 -
Chapter 2 Importance of Grounding
The importance of grounding is due to the fact that malfunctioning of industrial equipment is
more costly compared to loss of production time. Therefore controlling the level of fault
current in the equipment was needed to minimize the damage.
In power system there are two types of grounding: system grounding and equipment
grounding.
System grounding refers to the intentional connection of a phase or neutral conductor to earth,
the purpose of which is to control the current to earth or to keep this within predictable limits.
It also provides a path for current to flow, which allows the detection of unwanted connection
between the system and ground.
Equipment grounding refers to the interconnection and the grounding of the non-electrical
metallic elements of a system. Examples of such equipment grounding system are motor
frames and equipment enclosures.
System grounding is one of the most important elements for evaluating power systems. It is
important to understand the basic circuit parameters behind system grounding in order to
determine the most preferable method used for grounding a power system.
Grounding of a system is very important due to the following:
- It limits the magnitude of fault current to ground
- Safety of the staff
- Limit insulation and mechanical stress on equipment
- Avoid loss of equipment
- Continuity of process and preventing immediate shutdowns
- Simplified ground fault location
- Improved system and equipment fault protection
- Reduced maintenance time and expense
- Improved lightning protection
- Reduction in fault frequency
2.1 Types of grounding
There are different ways to connect the system to ground. These depend on how the system is
connected to earth.
The following different types of grounding are distinguished:
- Resistance (low & high)
- Resonance
- Solidly
- Ungrounded
2.1.1 Resistive grounded system
Resistance grounding is achieved by inserting a resistor between the system neutral and the
ground.
- 11 -
In addition, limiting fault currents to predetermined maximum values permits the designer to
selectively coordinate the operation of protective devices, which minimizes system disruption
and allows quick location of the fault.
There are two broad categories of resistance-grounding: low-resistance and high-resistance. In
both types of grounding, the resistor is connected between the neutral of the transformer
secondary and the earth ground.
Low Resistive grounding
Low-resistance-grounding of the neutral limits the ground fault current to a relatively high
level (typically 50 amps or more), in order to operate protective fault-clearing relays and
current transformers. These devices are then able to quickly clear the fault, usually within a
few seconds. This fast response time is important, since it limits damage to equipment,
prevents additional faults from occurring, provides safety for personnel and localizes the fault.
The limited fault current and fast response time also prevents overheating and mechanical
stress on conductors. It must be noted that the circuit must be shut down after the first ground
fault. Low-resistance-grounding resistors, typically rated 400 amps for 10 seconds, are
commonly found on medium- and high-voltage systems.
Advantages of low-resistance grounding:
- localisation of the fault is easy
- safety for personnel
- limits damage to equipment
Disadvantages of low-resistance grounding:
- cost of the grounding neutral resistor
- not practical at HV system, as magnitude of fault current remains high
High Resistive grounding
To provide high-resistance grounding, the value of the neutral ground resistor would be high.
Implementation of high resistance-grounding allows continuity of operations and allows fault
current detection. A single-line-to-ground fault will result in an increased current flow in the
grounding resistor. In medium voltage system, high-resistance grounding uses a distribution
transformer with a secondary resistor.
High-resistance grounding has advantages similar like of an ungrounded system. Continuity
of process is maintained with a single-line-to-ground fault. Immediate shutdown of process is
minimum. Ground fault currents that often occur in ungrounded and high-resistance grounded
systems do not draw enough current to trigger protective devices, thus making them difficult
to locate.
High resistance grounding is applied where the system charging current is relatively low and
continuity of service is required during the first ground fault, thus providing a path to trigger
the detection and alarm system rather than the protective device for an orderly shutdown of
the system at the desired time.
- 12 -
Figure 2.1 High/ low resistance grounding
Advantages of high resistance grounding:
- Able to limit the transient over-voltage to safe level
- No immediate clearing of ground faults since the fault current is limited to a low level
- Provides all the benefits as the ungrounded systems
- Irregular arc voltages are eliminated
- Reduces flash hazard
Disadvantages of high resistance grounding:
- Overvoltage are high when higher resistances are applied
- During a ground fault, voltage to ground on the faulted phases increases to as high as
line-to-line voltage, thus imposing stresses on equipment
- Not to be used in HV power system as it could cause burning damage to the
equipment when the fault current gets too high
To be effective the size of the resistor must be carefully selected for each system. The relation
between neutral resistor Rn and the capacitive reactance should be: Rn  Xco or Ro  3Rn .
3
(2.1)
2.1.2 Inductive grounded system (Peterson coil)
Peterson coil grounding produces an inverse current to compensate the capacitance current
during single-phase faults, thus reducing the single-phase fault current, then the fault arc can
be extinguished easily. That’s why it’s called an arc-suppression coil or ground fault
neutralizer. The distribution system can be operated with faults in about 2 hours, the same as
the ungrounded system, which ensures the reliability and continuity of power supply.
- 13 -
Figure 2.2 Grounding through Peterson coil
Advantages of Peterson coil grounding:
- Net fault current is reduced to a low value by the parallel resonant circuitry
- Prevents the occurrence of transient over-voltages as result of irregular fault
conditions
- Provides all the benefits as the high-resistance grounding systems
Disadvantages of the Peterson coil grounding:
- cost is high as compared to the other grounding types
Note: this type of grounding is not used in cable-type grids.
2.1.3 Ungrounded system
An ungrounded system is one in which there is no intentional connection between the
conductors and earth. However, as in any system, a capacitive coupling exists between the
system conductors and the adjacent grounded surfaces. Consequently, the “ungrounded
system” is, in reality, a “capacitively grounded system” by virtue of the distributed
capacitance.
Under normal operating conditions, this distributed capacitance causes no problems. In fact, it
is beneficial because it establishes, in effect, a neutral point for the system. As a result, the
phase conductors are stressed at only line-to-neutral voltage above ground. However,
problems can arise when there is a ground fault condition. A ground fault on one line results
in full line-to-line voltage appearing throughout the two healthy lines. Thus, a voltage 1.73
times the normal voltage is present on all insulation in the two healthy lines, as shown in
figure 2.3.
- 14 -
Figure 2.3 Effect on ungrounded network voltage after ground fault
This situation can often cause failures in older motors and transformers, due to insulation
breakdown. The interaction between the faulted system and its distributed capacitance may
cause transient over-voltages (several times the nominal) to appear from line to ground during
normal switching of a circuit having a line-to-ground fault (short). These over-voltages may
cause insulation failures at points other than the original fault. In addition, a second fault on
another phase may occur, before the first fault can be cleared. This can result in very high
line-to-line fault currents, equipment damage and disruption of both circuits. In addition to the
cost of equipment damage, ungrounded systems complicate locating fault(s) involving a
tedious process of trial and error: first isolating the correct feeder, then the branch, and finally,
the equipment at fault. The result is unnecessarily lengthy and expensive downtime. An
ungrounded system, despite the drawbacks, does have one main advantage. After the first
ground fault, assuming it remains as a single fault, the circuit may continue operation
permitting continued production until a convenient shut down for maintenance can be
scheduled.
Figure 2.4 Ungrounded system
Advantages of ungrounded system:
- First fault between a line conductor and ground does not cause circuit interruption,
thus no loss of power that can disrupt operations. (First ground fault)
- Lower initial costs, as no expenditures are required for the grounding system
conductors.
- 15 -
-
The system can operate for a definite time with ground faults, and overvoltage control
can be successful with the use of surge arrestors.
Disadvantages of ungrounded system:
- Difficult in locating the first line-to-ground fault
- No control of fault current and overvoltage during fault conditions
- If a second fault occurs before the first one is cleared, then a major fault could occur
between the two phases and ground causing extensive damage involving other circuits
in the whole system.
- The capacitive coupling can cause the ungrounded system to have dangerous overvoltages for intermittent ground faults and resonant effects, thus ungrounded systems
are prone to insulation failures.
There are two methods used to detect ground faults in ungrounded systems:
One method is to monitor the voltages between the phases and ground. This is done in the
control room by an engineer, where voltages and currents are displayed. Each phase has an
indicator light on it. As a ground fault occurs, the faulted phase will collapse to ground
potential, causing an indicator light to dim. The indicator lights on the un-faulted phases
become brighter, because of higher voltage on the un-faulted phases.
A better method to detect a ground fault is to measure the insulation resistance. As the
insulation deteriorates, a relay continuously monitoring the insulation resistance can alarm at
different levels for preventive maintenance. A visual indication or meter can also be used.
2.1.4 Solidly-grounded systems
In solidly grounded systems, the neutral points have been intentionally connected to earth
ground with a conductor having no intentional impedance. This reduces the problem of
transient over-voltages found on the ungrounded system and speeds the location of faults.
However, solidly grounded systems lack the current-limiting ability of resistance grounding
and the extra protection provides equipment damage and arcing ground faults.
The well-known and well-documented destructive nature of arcing ground faults in solidly
grounded systems is caused by the energy dissipated in the fault. A measure of this energy can
be estimated from the formula:
Kilowatt cycles = (V•I•T)/1000, where V and I are the voltage and current of the arc, and T is
the duration of the arc in cycles.
The faulted circuits need to be tripped, with conventional time current coordination. Often, it
is difficult to selectively trip the circuit with minimum time delay and high equipment damage
due to arcing ground fault is tolerated. Ground fault relays with zone-selective instantaneous
protection need to be used to provide maximum equipment protection, while retaining
selectivity and coordination.
To effectively build a solidly grounded system, there are some requirements to be met. These
are:
- Short circuit or fault current from a single-line to ground fault is at least 60 % of the
three phase balanced fault.
- 16 -
-
In terms of resistance and reactance, Ro ≤ X1 and Xo ≤ 3X1 where X1, Xo and Ro are
the positive-sequence, zero-sequence reactance and the zero-sequence resistance
respectively.
Figure 2.5 solidly grounded system
Advantages of a solidly grounded system:
- Overvoltage control is achieved, because the system neutral is solidly referenced to the
station ground.
- Sufficiently large current are able to activate or operate protective device when the
first phase-to-ground fault occurs
- The ease of locating the fault
Disadvantages of a solidly grounded system:
- The first phase-to-ground fault that opens the protective device shuts off power, lights
and control. Thus sudden loss of power can be severe as flash hazard do exists with a
phase to ground fault.
- It has the highest magnitude of fault current as compared to other grounding schemes.
- High fault current causes damages to the electrical equipment.
2.2 Grounding schemes suitable for different network structures
It is seen there are different types of grounding. But the question arises where and when
which specific type of grounding should be used. Some of the general rules used in practice.
Ungrounded network:
The fault should be excluded in a relatively short time (2h), in order to prevent the
development of inter-phase short-circuit and long-time over-voltage, causing damage to
equipment.
Power frequency over-voltage would rise during the single-phase faults, especially when the
capacitance current is large. Meanwhile, the electric arc grounding over-voltage risk also rises
- 17 -
and thus the entire distribution network requires a higher level of insulation, the investment in
equipment insulation will be increased.
Ungrounded-neutral mode is only used in the distribution with small capacitance current.
According by the electric power industrial standard, to the 20kV distribution, the ungroundedneutral mode should be used when the capacitance current is smaller than 10A.
Peterson coil grounding mode:
With the distribution running in Peterson coil grounding mode, Peterson coil grounding mode
can produce an inverse current to compensate the capacitance current during single-phase
faults, thus reducing the single-phase fault current, then the fault arc can be extinguished
easily. The distribution system can be operated with faults in about 2 hours, the same to the
ungrounded mode, which ensures the reliability and continuity of power supply.
In this grounding mode, over-voltage caused by single-phase faults also requires high level of
insulation. To the MV distribution (not in underground cable systems), the Peterson coil
grounding should be used when the capacitance current is above 10A, however, the current
should not be more than 150A.
In practical problem arises when a network should be extended. Because of extension the
network will require more inductive current to compensate the increased capacitive current.
Tuning the Peterson coil is option, but it should be kept in mind that the capacitive current
should not exceed the maximal tuning.
Resistance grounded system:
Resistance grounding mode can constrain power frequency over- voltage and electric arc
grounding over-voltage effectively. In a single-phase fault, for the single fault current is large,
protection equipment can act immediately to cut off the faults.
However, large single fault current in this mode would cause higher contact voltage and pace
voltage, threatening the safety of body and equipment.
Moreover, protection equipment acts immediately in single-phase faults, thus increasing
tripping times and inducing the reliability of power supply. To the MV distribution, the
resistance grounding mode is used when the capacitance is above 150A.
2.3 Calculation of grounding parameters and capacitance current
Calculation of capacitance current [12]
The capacitance current is one of the basic to choose proper grounding mode in MV
distribution. The capacitance current of lines can be derived by:
100C0U l L
 181,3C0U l L
(2.2)
3
Where U l is the rated line voltage; C0 is the single-phase earth capacitance, L is length of the
line.
Ic 
To overhead line, the capacitance current can be derived simply by:
I c  (2.7 ~ 3.3)U l L  10 3
(2.3)
- 18 -
To cable line, the capacitance current can be derived simply by:
I c  0.1U l L
(2.4)
Capacitance current vs Cable lenght
300
250
Ic
200
150
Ic
100
50
0
1
2
4
5
10
20
30
40
50
60
70
L
Figure 2.6 Capacitance current in amps, in relation to cable length in km
The capacitance current changed with the length of pure cable lines is indicated in figure 2.4
the capacitance current of system should be above 150A when the length of the cable line is
more than 45 km, then resistance grounding mode should be used. The Peterson coil
grounding mode can be when the length is less than 45 km. In the primary stage of
distribution construction, the length summation of the cable lines from the substation might be
less than 45 km, however, with the consideration of the development speed of urban
distribution, to pure cable lines, the resistance grounding mode should always be used.
2.4 Conclusions
Ungrounded delta systems have many operating disadvantages. High transient over-voltages
can occur that are not immediately evident. In addition, ground faults are difficult to locate.
Solidly grounded neutral systems provide greater safety for personnel, limit the system
potential to ground, and speed the detection and location of the ground fault. However, the
system must be shut down after the first ground fault.
Low-resistance-grounded neutral systems only limit the magnitude of the ground fault current
so that serious damage does not occur. The system must still be shut down after the first
ground fault. This level of resistance-grounding is generally used on medium and highvoltage systems.
High-resistance-grounded neutral systems limit the fault current to a tolerable level,
permitting continued production, until the fault can be located and corrected at a convenient
time. It also provides an economical method of upgrading older, ungrounded systems without
expensive addition of fault-clearing relays and breakers.
- 19 -
Chapter 3 Fault types
Fault current causes high dynamic and thermal stress for plants and materials. The percentage
or frequency of fault incidents on a power system varies with time and with many other
factors, such as climate, physical location, construction, and so on. In general terms 70% to
85% of the faults are phase-to-ground, 8% to 15% phase-to-phase, 4% to 10% double-phaseto-ground, and 3% to 5% three phase. Faults can also evolve from one type to another type,
especially where the protective equipment is slow in dissipating or isolating the fault. Thus a
phase-to-ground fault may develop into a double-phase-to-ground fault. Also a phase-to-phase
fault may become a double-phase-to-ground fault or three-phase fault. With moderate or highspeed relay protection, this generally does not happen. Phase faults are the most dangerous
kind of faults which can occur on electrical power system. Phase faults currents of ten times
rated current are not infrequent. In this thesis the focus is mainly on single-phase-to-ground
faults.
Short-circuit currents
Short-circuit currents have a dynamic behavior, this can be seen in the figure below.
Figure 3.1 Dynamic behaviour of short-circuit currents
During the first periods, the current is built up from the initial symmetrical short circuit
current Ik” and a decaying DC current. This results in a peak short-circuit current Ip, which
occurs after about a half period. The current Ik” contains the contribution of the generator and
has a contribution of the rest of the network. The duration of the contribution of the generator
is limited in time. After that, the short-circuit current reduces towards the steady state
symmetrical short-circuit current Ik. In the case of no generator being near the fault location,
the values of Ik” and Ik are more or less equal.
The peak short-circuit current Ip is decisive for the mechanical stresses. The initial and steady
state short-circuit currents are decisive for the thermal heating. Thermal stresses depend on
the amount of energy dissipated in the components and devices and are therefore related to the
duration of initial and steady-state short-circuit current. These values are specified by the
manufacturers. During a single-phase fault, a part of the fault current will return through the
sheath of the cable. This gives extra heating of the cable core and the cable sheath. The
amount of current and the maximum allowed time is therefore limited. Also these values are
specified by the manufacturers.
A distinction is made between the following main kinds of faults:
- 20 -
- three phase faults
- phase-to-phase faults
- phase-to-phase-to-ground faults
- ground faults
These are discussed in the following paragraphs.
It is known that currents can be represented in sequence components [6]:
I 0 = zero sequence component
I1 = positive sequence component
I 2 = negative sequence component
Zf = fault impedance
Figure 3.2 Representation of the sequence components
3.1 Three phase
When there is a three phase fault, the three phases are short-circuited through equal fault
impedances Zf, figure 3.3. The vector sum of fault currents is zero, because it is a
symmetrical fault and it is considered there is no path to ground.
The current are related as
I0  0
Ia  Ib  Ic  0
(3.1)
(3.2)
Figure 3.3 of three phase fault
- 21 -
As the fault is symmetrical:
Va
Zf
0
0 Ia
Vb  0
Zf
0 Ib
Vc
0
Zf Ic
0
(3.3)
The sequence voltages are given by
V0
V1  Ts
1
V2
Zf
0
0
Zf
0
0
0
I0
0 Ts I 1  0
0
0 I0
Zf
0 I1
Zf
0
Zf I 2
I2
Zf
0
(3.4)
This gives the equivalent circuit of figure 3.7
Ia  I 1 
Va
Z1  Zf
(3.5)
Ib  a 2 I1
Ic  aI1
(3.6)
(3.7)
3.2 Phase to phase to ground
A double-phase-to-ground fault is illustrated in figure 3.4. Phases b and c are connected to
ground through a fault impedance Zf. The current in the ungrounded phase is zero, i.e.,
Ia  0 , therefore
(3.8)
I 0  I1  I 2  0
(3.9)
(3.10)
Vb  Vc  ( Ib  Ic)Zf
Figure 3.4 of phase to phase to ground fault
Thus,
V0
1 1
1
V1  1 a
3
V2
1 a2
Va  2Vb
1
a Vb  Va  (a  a 2 )Vb
3
a Vc
Va  (a  a 2 )Vb
1 Va
2
(3.11)
- 22 -
Which gives V1  V2 and
1
V0  (Va  2Vb)
3
1
V0   (V0  V1  V2 )  2( Ib  Ic) Zf
3
1
V0   (V0  2V1 )  2(3I 0 ) Zf 
3
V0  V1  3ZfI 0

This gives the equivalent circuit of figure 3.7
The fault current is
Va
I1 
Z1   Z 2 Z 0  3Zf 
Va
I1 
Z ( Z  3Zf )
Z1  2 0
Z 2  Z 0  3Zf
(3.12)
(3.13)
(3.14)
3.3 Phase to phase
Figure 3.5 shows a phase-to-phase fault. A short circuit occurs between phase b and c,
through fault impedance Zf. The fault current circulates between phases b and c, flowing back
to source through phase b and returning through phase c,
(3.15)
Ia  0, Ib   Ic .
The sequence components of the currents are:
Figure 3.5 of phase to phase fault
I0
1 1
1
I1  1 a
3
I2
1 a2
1
0
1
a 2  Ic   a  a 2
3
a Ic
 a2  a
From equation 3.16
I 0  0 and I1   I 2
0
(3.16)
(3.17)
- 23 -
Va
1
1
Vb  Vc  0 1  1 Vb  0 1  1 1 a 2
Vc
1
a
1 V0
V0
a V1  0 a 2  a a  a 2 V1
2
a V2
(3.18)
V2
Therefore,
Vb  Vc  (a 2  a)(V1  V2 )
 (a 2 I 1  aI 2 ) Zf
(3.19)
 (a 2  a) I 1 Zf
This gives
(V1  V2 )  I1 Zf
(3.20)
The equivalent circuit is shown in figure 3.7
Also
Ib  (a 2  a) I1 Zf
(3.21)
And,
I1 
Va
Z1  Z 2  Zf
(3.22)
The fault current is
Ib   Ic 
 3Va
Z1  Z 2  Zf
(3.23)
3.4 Single phase-to-ground (ground fault)
Definition of Ground Fault
A ground fault is an inadvertent contact between an energized conductor and ground or
equipment frame. The return path of the fault current is through the grounding system, any
personnel or equipment that becomes part of that system. Ground faults are frequently the
result of insulation breakdown. It’s important to note that damp, wet, and dusty environments
require extra diligence in design and maintenance. Since water is conductive, it exposes
degradation of insulation and increases the potential for hazards to develop.
Consider a single line-to-ground fault from phase a to ground at the general three-phase bus
shown in figure 3.6. For generality, we include fault impedance Zf. In the case of a bolted
fault Zf  0 , whereas for an arcing fault Zf is the arc impedance. In the case of a
transmission-line insulator flashover, Zf includes the total fault impedance between the line
and ground, including the impedances of the arc and the transmission tower, as well as the
tower flooring if there are no neutral wires.
- 24 -
The relations to be derived here apply only to a single line-to-ground fault on a phase a.
However, since any of the three phases can be arbitrarily labelled phase a, consideration of
the other single line-to-ground faults on other phases is not needed.
Figure 3.6 of single phase to ground fault
Ib  Ic  0
Fault conditions in phase domain
Single line-to-ground fault voltage Vag  Zf * Ia
(3.24)
(3.25)
We now transform (3.22) and (3.23) to the sequence domain:
I0
1 1
1
I1  1 a
3
I2
1 a2
1 Ia
Ia
1
a 2 0  Ia
3
a 0
Ia
(3.26)
Also,
I 0  I1  I 2 
1
Ia
3
3I 0 Zf  V0  V1  V2   I 0 Z 0  (Va  I1 Z1 )  I 2V2
(3.27)
(3.28)
which gives,
I0 
Va
Z 0  Z1  Z 2  3Zf
(3.29)
The fault current Ia is
Ia  3I 0 
3Va
Z1  Z 2  Z 0  3Zf
(3.30)
This shows that the equivalent fault circuit using sequence impedances can be constructed as
shown in figure 3.7 [2].
- 25 -
Figure 3.7 gives an insight of the sequence connection during different types of fault
Figure 3.7 Sequence connections of different types of fault (Prov et all)
3.5 Conclusions
Summarizing this chapter some concluding remarks can be made. Calculating the short-circuit
current, it can be noticed from the formulas, that Zf plays a major role in determining the fault
current. To keep the fault current in predictable limits, Zf should be varied. For bolted fault Zf
is equal to zero.
3Va
Single phase to ground fault: Ia  3I 0 
, Ib  Ic  0
Z1  Z 2  Z 0  3Zf
Two phase fault between phase b and c: Ib   Ic 
 j 3Va
, Ia  0
Z1  Z 2  Zf
Two phase to ground fault between phase b and c: I1 
Va
,
Z ( Z  3Zf )
Z1  2 0
Z 2  Z 0  3Zf
 Z 0  3Zf 


Z2
Ib   I1 
 , Ic   I1 

 Z 2  Z 0  3Zf 
 Z 2  Z 0  3Zf 
Three phase fault: Ia  I 1 
Va
, Ib  a 2 I1 , Ic  aI1
Z1  Zf
- 26 -
Chapter 4 Transformer connections
4.1 Transformer connection
Transformer connection is important in power systems. Grounding of the power system is in
many cases through a star connected transformer. It’s essential to know how your transformer
is coupled, because the flow of zero sequence current depends on it.
Transformer connections can be done in four main ways [19] :
-
DELTA-WYE; industrial application (∆-Y)
DELTA-DELTA; commercial and industrial (∆-∆)
WYE-DELTA; high voltage transmission (Y-∆)
WYE-WYE; cable systems (Y-Y)
DELTA-WYE:
The primary winding acts as a harmonic filter by circulating currents through the phases. The
secondary winding with a properly grounded neutral, operates single phase and three phase
loads, provides dual voltages, is ground fault protected and the phase voltages are solidly
ground referenced.
Figure 4.1a detla-wye connection
The DELTA-WYE can have two schemes one with neutral ungrounded and second neutral
grounded.
Figure 4.1b sequence representation with secondary grounded
Figure 4.1c sequence representation with secondary ungrounded
The DELTA-WYE is very popular in office buildings but has serious limitations in most
industrial facilities as the secondary is very susceptible to harmonics.
- 27 -
DELTA-DELTA:
The primary as well as the secondary winding can dissipate harmonic noise, can run "open
delta" with only two transformers either by design or in an emergency. The primary and
secondary coils of the bad phase can simply be disconnected and run in most (58 percent) of
the original 3-phase transformer bank capacity. This has been the workhorse of the industry as
production activities can continue even with one phase solidly grounded and required
maintenance can locate and repair during non-production times.
Figure 4.2a delta-delta connection
Figure 4.2b sequence representation
WYE-DELTA:
The primary winding is solidly ground referenced for phase voltage stability and the
secondary winding dissipates harmonics. This design can operate with one phase solidly
grounded. The WYE-DELTA transformer can operate "open-WYE/open DELTA" with only
two transformers either by design or in an emergency. The primary and secondary coils of the
bad phase can simply be disconnected and run in most (58 percent) of the original 3-Phase
transformer bank capacity. The Primary-Secondary phase shift is 210 degrees.
Figure 4.3a wye-delta connection
The WYE-DELTA can have two schemes one with neutral ungrounded and second neutral
grounded.
Figure 4.3b sequence representation with primary ungrounded
- 28 -
Figure 4.3c sequence representation with primary grounded
WYE-WYE:
The primary and secondary coils are normally both grounded and some are supplied with an
integral ground common to the primary and secondary.
Figure 4.4a wye-wye connection
The WYE-WYE transformer connection can have different structures depending on were the
connection to ground is added.
Figure 4.4b sequence representation with primary grounded and secondary ungrounded
Figure 4.4c sequence representation with primary and secondary grounded
- 29 -
4.2 Conclusions
The transformer connection together with the type of grounding determines how the zero
sequence current will flow in the network during a fault.
Most common transformer connection types are:
- Delta-delta connection
- Delta-wye connection
- Wye-delta connection
- Wye-wye connection
In most of the networks are grounded through the neutral of a transformer. In some cases a
transformer is used mainly as grounding equipment.
- 30 -
Chapter 5 Problems concerning earth faults
There are different reasons to interruption of electricity to customers. Interruptions are
responsible for the rising NDCM, these NDCM can not be decreased to zero but it can be
minimized to a certain value. This can be done by using better components, redundancy,
maintenance, replacement and repair strategy.
The two main reasons for interruption are:
1) technical problem
2) damaged caused by digging activities
Figure 5.1 Causes for interruption in 2010
The technical problem can be divided into smaller parts. Within the technical area, there are
several components that effect the interruption of delivery; these are most of the time:
- cable
- cable joints
- protection equipment
- transformers
- circuit breakers
- cable end-closure
- 31 -
Joints 55%
Cable 19%
Protection 5%
Circuit breaker 8%
Transformer 10%
End enclosure 3%
Figure 5.2 Technical interruption reasons
5.1 Technical and Damaged caused by digging activities
5.1.1 Technical
In this paragraph the technical problems are described with their effects. The possible
solutions and the impact on the not-delivered consumer minutes will be discussed.
As was mentioned earlier in this chapter, 50% of technical kind comes from the cable joints.
One of the cable joints that causes the most outages are the nekaldiet cable joints. Studies
have been done before on these joints to show there voltage life, weakness, failure rate etc. It
can be seen from the figure 5.3, that from 2007 the nekaldiet cable joints are showing a
decreasing numbers of failures.
Figure 5.3 Number of nikaldiet cable-joint failure per year
- 32 -
This can be explained as follow: as known, these joints are being replaced by new types
which are more reliable and secondly they are also being tested in the field to replace the
weak joints. In the coming years the nekaldiet failures will decrease further until all the joints
are replaced. So it’s essential to know the best option to decrease the effect of these joints.
Options for reducing the impact of nekaldiet failures on the system:
Option 1 replacing all joints
There are tens of thousands of nekaldiet joints in the network of Alliander. Replacing all these
joints will have a great positive impact on the system. The failures due to nekaldiet joints will
be decreased. Replacing tens of thousands of joints will cost a lot of money, so there should
be another way to find a balance between cost and system reliability.
Option 2 using replacement strategy
Option 2 is a more practical way of replacing the nekaldiet joints.
First the most critical areas should be studied, critical in the sense of that they are responsible
for a large part in the NDCM. If this is found out than the next step should be:
- analyzing the network
- finding the costumers attached to a string
- data about the fault history of string
Option 2 is further elaborated here.
The mid-voltage (10-20 kV) network has a meshed structure and sometimes radial structure.
Because of net openings in the meshed network faulted section could be taken out of order to
make repairs. Due to net openings a feeder could be energize from more than one place, in
this way the number of costumers attached to the faulted feeder could be minimized. In the
case of a radial structure the impact could not be minimized.
In case that a feeder or total substation cannot be energized from another nearby substation or
feeder, for this particular case an external generator will be used.
If the network structure is known, then the total number of costumers attached to the outgoing
feeders should be known.
Figure 5.4 Meshed network
- 33 -
Data needed for meshed network:
- total number of costumers on feeders
- can the impact of fault on feeder be reduced by net openings?
- historical fault data on that feeder
- the location of the feeder in the network
- cost of replacement related to reliability
Advantage:
Essentially meets the requirements of two alternative feeds to give 100% continuity of supply.
Disadvantage:
For faults at T1 fault current is fed into fault via two paths in parallel, hence the fault current
is much higher.
Protection must therefore be fast and discriminate correctly so that other consumers are not
inconvenienced
Figure 5.5 Radial system
Data needed for radial network:
- total number of costumers on feeders
- historical fault data on that feeder
- cost of replacement related to reliability
Advantage:
If a fault occurs at T2 then only the protection on one leg is called into operation.
The other consumers are not affected.
Disadvantage:
If the conductor to T2 fails, then supply to this particular consumer is lost completely.
Having these data can be useful to make decisions about replacing the nekaldiet joints.
5.1.2 Damaged caused by digging activities
One third of outages are caused by digging. This is a big concern for Liander. The cost for
such a fault is paid by an insurance company, so most of the time costs related to digging are
not of concern. The problem is what about the NDCM? Will digging have any effect on the
- 34 -
NDCM? The answer is simple YES! They will not be foreseen from electricity because of
single phase to ground faults or even multiphase faults. How can these third party damages be
reduced?
To reduce the NDCM caused by digging is one of the main items going on these days.
Possible solutions that have been mentioned are:
- use material above the cables to protect them
- software with cables routes in it
- cable tracking system
- use of ground radar on excavator as shown in figure 5.6
Figure 5.6 ground radar of excavator
5.2 Analyzing the NDCM
Electrical companies are always trying to keep the NDCM at low acceptable levels. These
NDCM has a great impact on the company in many ways. The most important reasons are
economical losses because of penalties and image of the company. In this paragraph the trend
of the NDCM are discussed and analyzed in two steps.
- 35 -
Step one
In this step there are more analytical solutions in decreasing the NCDM. This has been done
by Alliander N.V. in the middle of 2010. Figure 5.7 shows when the program started the
NDCM started to decrease. First a summary about the analytical solutions will be given.
The graph below shows the development of the NDCM over the last few years.
Figure 5.7 Development of the NDCM from 2008 till end 2010
It can be seen that the NDCM are increasing until 2010. Alliander has started a program in
mid 2010 to decrease these minutes. This program seems to be effective showing a slight
decrease in the NDCM, see figure 5.7, from mid 2010 forward. The next picture will give
better understanding of how these NDCM are built up.
Figure 5.8 time interval for fault detection till electricity delivery is recovered
Explanation of the picture above:
S_BEGIN is the time at which a fault is detected, the beginning of an outage.
A mechanic is called and the location of the outage is given to him.
S_PLAATS is the time at which the mechanic is arrived at location of the outage. Now he
should search for the fault.
If the fault is found then he will start doing the reparation.
DEF_EINDE is the time at which the reparation is finished and the delivery is recovered.
- 36 -
There is a certain time gap from outage to recovery. The whole time gap multiplied by the
total of consumers without electricity divided by the total consumers of the whole electricity
network gives the NDCM for one outage.
The Alliander network has approximately 3 million customers. The distributed area is divided
into smaller parts. Every part has a different network structure. These parts are also analyzed
separately and they are combined together to give a total of the NDCM. The table beneath
describes the different parts.
Figure 5.9 Overall performance of different regions with their average NDCM
Data from figure 5.9 shows that Gelderland East has the highest NDCM and Veluwe has the
lowest. The average NDCM value from all the division is 27.5 minutes.
Explanation of figure 5.9:
TOT: the total yearly NDCM of the different parts
MS: TOT for the mid-voltage network 10/20kV
LS: TOT for the low-voltage network 440/380/220V
SVBM is the same as NDCM
Target is the goal Alliander has set for this year to achieve.
As mentioned earlier, a certain time gap is needed from outage to recovery.
What are of interest are the average times for the different stage in the time gap. The time of
the different stages are a part of the NDCM. This is explained as follows:
The total time from outage to recovery multiplied by the number of costumers affected by the
outage, this divided by the number of (connected) costumers. This gives the NDCM value for
that outage. Every time stage that could be minimized will be a great achievement to reduce
the NDCM. That’s why it is important to know the time of the different stages.
The next table gives a complete view of these stages.
Figure 5.10 the average time of the 3 stages for the whole network
- 37 -
The coordination time is about 11 minutes. The time for the mechanic to reach at the fault
location is about 25 minutes and reparation takes about 42 minutes. This is data for the midvoltage network. The values in table 5.10 are approximated values of all the regions. The next
table has data for the different regions.
Figure 5.11 the stages of the time gap for different regions
Until now all the data that has been gathered, has been used to analyze the composition of the
NDCM. Looking for possible solutions to decrease the different stages of the time gap is the
main goal.
In short: step 1 is about decreasing the coordination, driving and solution time.
Some possible solutions to decrease the time gap:
Coordination: better communication and fast response to problems
Driving: more mechanics
Solution time: experienced mechanics, fast fault detection methods.
Step two
The analysis of failure data in step 2 is more or less the same as in step 1. The difference lies
in the fact that the type of grounding is included in step 2. The impact of grounding on these
NDCM will now be analyzed.
For research of step 2 data is needed regarding:
- Kind of grounding
- NDCM
- Type of fault
- Causes for faults
- Failed component
In this section different tables/figures are given about failure data.
To begin with it is important to know which components cause most failures.
Figure 5.12a gives all the causes for failures in the Alliander network.
- 38 -
Figure 5.12a causes for failure
Figure 5.12b is a zoomed in picture of figure 5.12a, the last three years are zoomed in.
From figure 5.12 a. and 5.12 b. there are 3 causes which need most attention. They are:
- Unknown cause (dark blue)
- Cable paper-lead (reflector)
- Cable joint in paper-lead cable (red)
Figure 5.12 b zoomed in picture of causes for failures from 2008-2010
The first cause is a contribution of unknown factors it will be removed from the list for
analysis. The second and third (cable lead paper and cable joints in cable lead paper) are now
analyzed in more detail. As grounding method was the main research problem, all the analysis
will include grounding in it.
- 39 -
Cable paper-lead and type of grounding
The figure beneath gives the total number of failures on paper-lead cable in unknown,
ungrounded and impedance earthed networks. It’s important to known if grounding has any
effect on the total number of failures.
Figure 5.13 cable paper-lead with different grounding schemes
From the figure 5.13 it can be noticed that the total number of failures do not differ so much
between ungrounded and impedance earthed network. A failure is unavoidable we should try
to minimize the damage caused by such a failure.
Analyzing the cable paper-lead
The figure below gives the reason for fault in paper-lead cable.
Figure 5.14 causes for faults in cable paper-lead
As can be seen damages caused by digging activities (graafschades) are for 67% responsible
for outages. This is only the case for cable lead-paper.
- 40 -
Cable-joint paper-lead and type of grounding
The next graphs give a better view of the total number cable-joints in paper-lead together with
different grounding methods.
Figure 5.15 cable joints in different grounding schemes (cable paper-lead)
From figure 5.15 there is again no big difference between grounding types. A very important
thing that should be noticed is that, there is still a big amount of cable joint failures in
unknown grounding type network.
Cable joints
In figure 5.16 the different cable-joints that are responsible for outages are given.
Figure 5.16 different cable joints responsible for faults
- 41 -
As can be seen from figure 5.16 nekaldiet joints and lovink oil joints are the dominant types.
These two joints will be analyzed in more detail.
Nekaldiet cable-joints and their cause for outages.
Figure 5.17 type of fault for nekaldiet joint
From figure 5.17 the types of fault responsible for nekaldiet joint failures are:
- otherwise explain
- earth fault
- protection
- installation error
- testing
- manufacturing defect
- aging
- wear out
- excavation old and new
- old damage
- weather influence
- effect of soil
From all the types of fault otherwise explain, manufacturing defect, aging and wear out are in
majority.
- 42 -
Lovink-oil cable joints and the cause for outages
Figure 5.18 type of fault for lovink oil joints
From figure 5.18 the types of faults responsible for lovink oil joints failure are:
- otherwise explain
- not examined
- wear out
- earth fault
- excavation old and new
- aging
- wear out
- weather influence
- effect of soil
- overloading
- latent fault
From all the types of fault otherwise explain, not examined, aging and wear out are in
majority.
Both types of cable-joints nekaldiet and lovink oil have more or less the same kind faults. The
only difference is that nekaldiet joints have a manufacturing defect. As nekaldiet joints are in
majority in comparison to lovink oil joints, they will be analyzed is more detail.
- 43 -
Components responsible for NDCM
In the beginning of this graph results have shown that cable joints and paper-lead cable are the
main components which are the responsible for most failures. Are these component failures
responsible for the increasing NDCM? The next figure will answer this question.
Figure 5.18 Total NDCM vs cause
It can be seen that the components cable paper-lead and cable-joints are in majority
responsible for the NDCM. In figure 5.19 and 5.20 the impact of nekaldiet cable joints and
paper-lead cables on NDCM will be analyzed.
Figure 5.19 nekaldiet joints [total NDCM vs cause/ grounding type/ fault type]
- 44 -
Figure 5.20 cable paper-lead [total NDCM vs cause/ grounding type/ fault type]
5.3 Conclusions
From the data analyzed in this chapter some concluding remarks can be made:
-
Nekaldiet cable joints and cable lead-paper have most defects
Nekaldiet cable joints lifetime and manufacturing defect are the main reason for them
to fail
Impedance grounded and ungrounded system have major failures for nekaldiet joints
Cable lead-paper digging are the main cause for them to fail
Impedance grounded and ungrounded system have major failures for cable lead-paper
Single phase to ground fault are in majority for both nekaldiet and cable lead-paper
Two phase and three phase faults also shows to have some part in failures
Component
Nekaldiet cable
joint
Cable leadpaper
Failure cause
Age/
manufacturing
Digging
Grounding type
Impedance/
ungrounded
Impedance/
ungrounded
Fault type
Single phase
Single phase
Table 5.4 summarized table of faults
Overall: The relation of grounding type to failures is still not very clear due to the analysis
made in this chapter.
- 45 -
Chapter 6 Simulations
In this chapter the current and voltage behaviour will be analyzed through simulations.
The network in figure 6.1 is used to model the network in PSCAD for simulations.
Figure 6.1 model used for simulations (network type I)
The basic components are: generator, transformer, cable/line, loads and protection devices.
In this thesis protection is not modelled, because it was not necessary for the problem
analysis.
6.1 Simulations by PSCAD
PSCAD (Power Systems Computer Aid Design) is a powerful and flexible graphical user
interface to the world-renowned, EMTDC solution engine. PSCAD enables the user to
schematically construct a circuit, run a simulation, analyze the result, and manage the data in
a completely integrated, graphical environment. Online plotting functions, controls and meters
are also included, so that the user can alter system parameters during a simulation run, and
view the results directly.
PSCAD has a complete library of pre-programmed and tested models, ranging from simple
passive elements and control functions, to more complex models such as electric machines,
FACTS devices, transmission lines and cables. If a particular model does not exist, PSCAD
provides the flexibility of building custom models.
With PSCAD studies can be done beginning from contingency analysis to more complex
analysis such as impulse testing of transformers, design of control parameters etc.
For the simulation an educational version of PSCAD is used. The specific dimensions are:
- 46 -
200 nodes, 64 page Modules and 32.768 components.
Why simulations?
In chapter 5 the results of analyzing data did not show great differences between grounding
types in relation to the NDCM. So the next step is chapter 6, where simulations will be
performed to see differences in grounding types. The network model in chapter 6.1 is used to
do these simulations. Faults are applied to the system, where the grounding type and
transformer coupling will be varied, to see the current and voltage behavior. These currents
and voltages are important because of their influence on components in the system. For
example: A voltage increase cause stresses on your component, which in turn has impact on
the remaining life of the component.
Faults in the system can have the following consequences:
- Dip (when the voltage drops in the system)
- Swell (when the voltage is increased to phase to phase voltage)
- transient voltage
- transient current
- over-current
Before the simulations are done some data is needed as input.
Isc is 150 KA this used at Alliander NV
Ssc  3 *U * Isc  3 * 52.5KV *150KA  13640MVA
U
Zsc 
 4.95
Isc * 3
Generator: V = 52.5 kV
Transformer: P = 30 MVA V1/V2 = 52.5/10.5
Cable:   2.8 *106 .cm for aluminium
Surface of the cable is 240 mm²
Load: all loads are equal and are 1 MVA + 0.5 MVAR per phase
The sequence components of the cable are automatically calculated by PSCAD.
Some notes:
Three types of grounding are used in the simulations, which are:
- Ungrounded
- Directly grounded
- Resistively grounded
These are the usual types used at Alliander NV, for resistive/ impedance grounding they use 7
Ohm.
During the simulations the resistance/ impedance is varied.
6.2 Network type 1
To see the behavior of the voltages and currents during faults simulations are done in PSCAD.
During the simulations two types of networks are simulated.
Network type 1:
Infinite bus
30 MVA transformer
Two cables in series with lengths of 10 km
Loads of total 6 MW and 3 MVAR (loads are coupled between cable and at the end)
- 47 -
Fault location 1, 2 and 3 seen in figure 6.1
Figure 6.2 Network type 1
The following will be varied during simulations:
- grounding type (ungrounded, star grounded and impedance grounded)
- fault type (L-G, L-L, L-L-G, L-L-L, L-L-L-G)
- fault location 1,2 and 3 (also in combination for cross country faults)
- transformer coupling (delta-star, star-delta, delta-delta and star-star)
First the steady state values are calculated:
S  P  jQ  6MVA  j 3MVAR
S  6, 726.6
Pphase  Uphase * Iphase *cos 
2MVA  10.5kV * Iphase *0.89
Iphase  214 A  Ipeak
Iphase  214*0.5 2  151A
The network is now simulated the steady state values are given in figure 6.3
Figure 6.3 steady state values
- 48 -
In the first graph, the first figure on the right gives voltage at location 1 and on left the
current. Second figures are the voltages and current at location 2 and the third at location 3.
As can be seen from the figure the current is 150 Amps and the voltage 8.4 KV.
The first simulations have the following specification:
Delta-star connection of transformer
Variation will be made on: grounding type and types of fault
6.2.1 Delta-wye connection of transformer
Fault location 1
Ungrounded
Single phase to ground fault (phase a to ground)
Figure 6.4 single phase to ground fault
The problem that arises in ungrounded system is that a single phase fault causes 3 times the
voltage in the other phases. This can be clearly seen in figure 6.3, the first graph the voltage
is 3 *8.4KV  14.55KV . The problem that occurs in ungrounded system, with weak
capacitive coupling between earth and phase, is that the fault current is difficult to detect. In
this case, figure 6.3, the fault current is large enough to be detected, because the capacitive
coupling between phase and earth is strong. Otherwise the 3 time nominal voltage stays for a
longer time on the two phases. The stresses on these phases are increased due to the higher
voltage. Small defects in the cable can now grow very fast in to major defects, which can
result in breakdown in the cable. This will result in phase to ground fault in other phases or
even multiphase faults. The level of the current is then increased very fast.
This type of fault is called cross country fault. At the end of this graph some cross country
faults are simulated to see the effect.
- 49 -
Ungrounded
Phase to phase fault (phase a to b)
The network is ungrounded and a phase to phase fault is applied to phase a and phase b.
Figure 6.5 phase to phase to fault
The voltage at phase a and phase b are decreased, blue and green lines in figure 6.4. The
current reaches very high values about 40 kA. As we move further in the network at position
2 and 3 we see no difference in voltage profile. The currents at position 2 and 3 in phase a and
phase b decrease.
Ungrounded
Phase to phase to ground (phase a to b to ground)
A phase to phase to ground fault is applied to network type 1. The network is again
ungrounded and the faulted phases are a and b.
Figure 6.6 two phase to ground fault
- 50 -
Ungrounded
Three phase
A three phase fault is applied to the network type 1.
Figure 6.7 three phase fault
Ungrounded
Three phase to ground
Figure 6.8 three phase to ground fault
- 51 -
Figure 6.9 gives the magnitude of the currents and voltages of different fault types
Delta-star connection of transformer Ungrounded
45
40
35
kA/kV
30
25
Current
20
Voltage
15
10
5
0
L-G
L-L
L-L-G
L-L_L
L-L-L-G
Types of fault
Figure 6.9 magnitudes of currents and voltages for different fault types of an ungrounded
system.
Figure 6.4 through 6.8 are simulations results for different types of fault in ungrounded
network.
Figure 6.9 gives the maximum fault current and voltages for different fault types. A single
line to ground fault gives a very small fault current. Applying multiple faults there is an
incredible increase in fault current. The voltages for ground fault or multiple faults to ground
have the same magnitude. The voltages for multiple faults have the same magnitude, which is
lower than incase of a fault to ground.
Directly grounded
The network type 1 is now directly or solidly grounded. The transformer coupling is deltawye as is the case in this paragraph. The same faults are applied, from single phase to three
phase faults. Appendix A.1.2 gives the simulation results. The first figure starts with single
phase to ground fault.
Delta-star connection of transformer Star-grounded
45
40
35
kA/kV
30
25
Current
20
Voltage
15
10
5
0
L-G
L-L
L-L-G
L-L_L
L-L-L-G
Type of fault
Figure 6.10 magnitudes of currents and voltages for different fault types of a star grounded
system.
Figure 6.10 are the magnitudes of the currents and voltages during different fault types. The
magnitudes are the same for all types of faults, current about 40 kA and voltage 8.5 kV.
- 52 -
Resistively grounded
The resistively grounded system has a resistance inserted in the wye of the transformer
connection. The resistance will be varied during the simulations. The transformer coupling is
again delta-wye. In appendix A.1.3 the simulations result for single phase to ground fault for
different resistance values are given (varying from 0.5 ohm to 20 ohm).
Delta-star connection of transformer Impedance grounded
16
14
12
kA/kV
10
Current
8
Voltage
6
4
2
0
0,5
1
5
7
10
20
resistance
Figure 6.11 magnitudes of currents and voltages for different fault types of an ungrounded
system
In figure 6.11 the maximum magnitudes of the current and voltages are given for different
resistance values. The voltage is more or less the same. The current is affected by the change
of the resistance. The current decreases fast till 5 Ohm, from 5 Ohm till 20 Ohm the decrease
is very slow. The resistance added to the grounding is responsible for this.
In this graph Delta-star connection of the transformer was used as reference. During the
simulations the grounding type and fault type were varied. Coming to some concluding
remarks according to the changes made:
-
-
Ungrounded: The phase to ground voltage is around 15 kV for all types of fault. The
current is small for single phase to ground fault. As multiple fault are applied to the
system the current increases incredibly to about 40 kA.
Star grounded: Voltage and current magnitudes are more or less the same for all types
of fault. The phase to ground voltage is around 8.5 kV and current around 40 kA.
Resistively grounded system: Varying the resistance from 0.5 Ohm to 5 Ohm gives a
fast decrease of the current magnitude. From 7 Ohm to 20 Ohm we see a small
decrease in the current. The phase to ground voltage is around 15 kV for the varied
resistance (figure 6.20).
- 53 -
6.2.2 Wye-wye connection of transformer
In this section the transformer coupling is wye-wye connected. Here simulations are exactly
the same as in section 6.2.1 only difference is the transformer coupling. The results of these
simulations are in appendix A2. Only the maximum magnitudes of the current and voltages
during fault are put into figures.
Different fault types and grounding types are simulated using wye-wye connection of
transformer.
Fault location 1
Ungrounded
Star-star connection of transformer Ungrounded
50
kV/kA
40
30
Current
20
Voltage
10
0
L-G
L-L
L-L-G
L-L-L
L-L-L-G
types of fault
Figure 6.2.2.1 types of fault in an ungrounded system with wye-wye connection of transformer
Directly grounded
Star-star connection of transformer Star-ground
50
kV/kA
40
30
Current
20
Voltage
10
0
L-G
L-L
L-L-G
L-L-L
L-L-L-G
types of fault
Figure 6.2.2.2 types of fault in a directly grounded system with wye-wye connection of
transformer
- 54 -
Impedance grounded
kV/kA
Star-star connection of transformer impedance grounded
16
14
12
10
8
6
4
2
0
Current
Voltage
0,5
1
5
7
20
resistance
Figure 6.2.2.3 variation of the resistance in an impedance grounded system with star-star
connection of transformer
Star-star connection of transformer Impedance grounded
7 ohm
50
kV/kA
40
30
Current
20
Voltage
10
0
L-G
L-L
L-L-G
L-L-L
L-L-L-G
types of fault
Figure 6.2.2.4 types of fault in an impedance grounded system (7 Ohm) with star-star
connection of transformer.
Concluding remarks:
- Ungrounded: the phase to ground voltage varies between 15 kV and 8 kV and the
current is small for a single phase to ground fault. At multiple faults the current
starting to increase.
- Star grounded: The same as in ungrounded system happens with star grounded.
- Impedance grounded: the voltage has a constant value of 14.6 kV and the current is
also constant around 0.63 kA for all types of fault.
- 55 -
6.2.3 Wye-delta and delta-delta connection of transformer
As these two types of transformer connection are not used intensively in medium voltage
network they will not be analyzed extensively. The simulation results are in appendix A3 and
A4.
Wye-delta connection of transformer
Star-delta connection of transformer Ungrounded
network
50
kV/kA
40
30
Current
20
voltage
10
0
L-G
L-L
L-L-G
L-L-L
L-L-L-G
types of fault
Figure 6.2.3.1 types fault in an ungrounded network with star-delta connection of transformer
Delta-delta connection of transformer
Delta-delta connection of transformer Ungrounded
system
50
kV/kA
40
30
Current
20
Voltage
10
0
L-G
L-L
L-L-G
L-L-L
L-L-L-G
types of fault
Figure 6.2.3.2 types fault in an ungrounded network with delta-delta connection of
transformer
Both figures 6.2.3.1 and 6.2.3.2 have the same voltage and current behavior. Only difference
is that the current at two phase and two phase to ground fault star-delta connection are a little
bit smaller than in delta-delta connection.
- 56 -
6.2.4 Difference between delta-wye and wye-wye
In this section the differences between delta-wye and wye-wye connection will be analyzed.
Also the differences during variation of the grounding type will be shown.
Fault location 1
Delta-wye connection
Ungrounded
The difference in voltage profile during single phase to ground fault in ungrounded system.
Figure 6.2.4.1 ungrounded system with delta-wye connection of transformer
Wye-wye
Ungrounded
Figure 6.2.4.2 ungrounded system with wye-wye connection of transformer
As can be seen from figure 6.2.4.1 and 6.2.4.2 the voltage profile of wye-wye connection is
sinusoidal and has no distortion in it. This is marked with red circles in figure 6.2.4.1. This
causes less stress on your component during fault occasion.
- 57 -
Delta-wye connection of transformer
Ungrounded system
Fault location 2
The difference in voltage profile during single phase to ground fault in ungrounded system.
Figure 6.2.4.3
Wye-wye
Ungrounded
Figure 6.2.4.4
At fault location 2 we see the same difference as at fault location 1. See figure 6.2.4.3 red
circles.
- 58 -
Delta-wye connection of transformer
Ungrounded system
Fault location 3
The difference in voltage profile during single phase to ground fault in ungrounded system.
Figure 6.2.4.5
Wye-wye
Ungrounded
Figure 6.2.4.6
At fault location 3 we see the same difference as at fault location 1. See figure 6.2.4.5 red
circles.
- 59 -
Fault location 1
Difference between delta-wye and wye-wye
Star/ effectively grounded
The difference in voltage profile during single phase to ground fault in ungrounded system.
Figure 6.2.4.7
Wye-wye connection
Star/ effectively grounded
Figure 6.2.4.8
As can be seen from figure 6.2.4.7 and 6.2.4.8 the voltage and current profile of star-star
connection is sinusoidal and has no distortion in it. This is marked with red circles in figure
6.2.4.7. This causes less stress on your component during fault occasion.
- 60 -
Fault location 1
Difference between delta-wye and wye-wye
Impedance grounded 7 ohm
The difference in voltage profile during single phase to ground fault in ungrounded system
Figure 6.2.4.9
Wye-wye connected
Impedance grounded 7 ohm
Figure 6.2.4.10
The figure 6.2.4.9 and 6.2.4.10 gives the same results as the figures before. Delta-wye
connection has a fault current of 1675 Amps and wye-wye connection 628 Amps.
- 61 -
6.2.5 Difference between wye-wye and delta-delta
Fault location 1
Ungrounded
Wye-wye connection single phase to ground fault
Figure 6.2.5.1
Delta-delta connection
Fault location 1
Figure 6.2.5.2
There is no difference in voltage and current profile between both figure 6.2.5.1 and 6.2.5.2.
- 62 -
We move further to see what happens if we have two phase to ground fault.
Wye-wye connection of transformer
Ungrounded
Fault location 1
Figure 6.2.5.3
Delta-delta connection
Ungrounded
Fault location 1
Figure 6.2.5.4
There is again no difference in voltage and current profile. If we look at the magnitudes we
see that the magnitudes are also the same.
- 63 -
6.3 Network type 2
Model specification:
Infinite bus
30 MVA transformer and a 10 MVA transformer
A 1MW generator
Two cables in series with lengths of 10 km
Loads of total 6 MW and 3 MVAR (loads are coupled between cable and at the end)
Fault location 1, 2 and 3 seen in figure 6.3.1
Figure 6.3.1
Steady state
The 2nd Generator delivers 1 MW
Figure 6.3.2 steady state
The voltage values are the same as in network 1, the current value is increased to 228 A.
- 64 -
In the previous case network type 1 there was only one transformer, the variation during
simulations were as follow:
- delta-wye ungrounded
- delta-wye directly grounded
- delta-wye impedance grounded
- wye-wye ungrounded
- wye-wye directly grounded
- wye-wye impedance grounded
- delta-delta ungrounded
- wye-delta ungrounded
The two last types will not be included in network type 2.
Network type 2 has two transformers, so the variation of transformer coupling could be on
both transformers. The variations are:
- delta-wye vs wye-delta
- delta-wye directly grounded vs wye-delta
- delta-wye impedance grounded vs wye-delta
- delta-wye directly grounded vs wye-delta directly grounded
- delta-wye impedance grounded vs wye-delta impedance grounded
The same variations were applied for wye-wye vs wye-wye, delta-wye vs wye-wye and wyewye vs wye-delta connection of transformers.
The first simulations have the following specification:
Delta-wye and wye-delta connection of transformer
Variation will be made on: grounding type
The focus will be on ground faults or single phase to ground fault.
Starting with faults at location 1
Delta-wye ungrounded vs wye-delta ungrounded
Figure 6.3.3 D-W U vs W-D U
D: Delta, W: Wye, U: ungrounded
- 65 -
Table 6.3.1 gives the maximum values of the fault current and voltages.
T1: transformer 1, T2: transformer 2
D: delta, W: wye, G: directly grounded, IG: impedance grounded,
V1 and I1 voltage and current at location 1
V2 and I2 voltage and current at location 2
V3 and I3 voltage and current at location 3
Table 6.3.1 Voltage and current for different transformer couplings and fault types.
So far these were results for fault location 1.
The simulation results of other types of grounding and transformer coupling are in appendix
B1, B2 and B3.
Fault location 2
Here simulations we are done at fault location 2. Different kind of grounding, transformer
coupling and fault type will be simulated.
Delta-wye ungrounded vs wye-delta ungrounded
Figure 6.3.4 D-W U vs W-D U
Table 6.3.2 gives the maximum values of the current and voltages at different locations in the
network. The fault was applied at location 2. The simulation results are in appendix B4 and
B5.
- 66 -
Table 6.3.2 Voltage and current for different transformer coupling during ground faults
To search for the best option for network type 2 one should consider the following:
- smooth profiles for voltages and currents
- not very high fault currents
- detectible fault currents
- 3 points mentioned above should be valid for all locations in the network
The simulation results for location 3 are in appendix B6
Comparison
First plus stand for no distortion and second plus for low magnitude
First minus stand for distorted signal and second minus for high magnitude
T1 vs T2
F1
F2
F3
V
I
V
I
V
I
D-W vs
+++
-+++W-D
D-W-G vs + ++++
++
++
W-D
D-W-IG
+++
+++
++
++
vs W-D
D-W-G vs + +
++++
++
++
W-D-G
D-W-IG
+++
+++
++
++
vs W-DIG
W-W vs
++
++
+++
+++
W-W
W-W-G vs + +
++
+++
+++
W-W
W-W-IG
++
++
+++
+++
vs W-W
W-W-G vs + +
++
+++
+++
W-W-G
W-W-IG
++
++
+++
+++
vs W-WIG
Tabel 6.3.3 comparison of voltages and currents at different locations. F1: fault location1;
F2: fault location 2; F3: fault location 3
- 67 -
6.4 Conclusions
Network type 1
Wye-wye connection is best solution for a network with a single transformer. The voltage
profile has no distortion during a fault. The best grounding type is impedance grounding,
because the fault current can be controlled.
Network type 2
Wye-wye connection for transformer 1 and wye-delta connection for transformer 2 is the best
solution. Again this has to do with the voltage profile which has no distortion during a fault.
The grounding type that is suitable is again impedance grounding.
- 68 -
Chapter 7 Conclusions and Recommendations
At the beginning of this project some questions were formulated. To find answers to these
questions, research was performed.
Objective 1
Here some general knowledge is gathered about grounding, fault types and transformer
connection. The differences and other important parameters about the three subjects
mentioned are done. Objective 1 is a theoretical research.
As objective 1 was a theoretical research, the following conclusions can be drawn:
- Low resistance grounding should be used in medium and low voltage networks
because the fault current is still high to cause damages in high voltage networks.
- High resistance ground is good for high voltage networks to limit the large fault
currents. It should not be used in medium and low voltage network because the fault
current becomes too small to be detected.
- Overall resistance grounding should be used for systems with weak capacitive
coupling to earth.
- Star or effectively grounded system causes very high short circuit current which is
very unfavourable for the components in the system.
- Ungrounded system causes difficulties for detecting a single phase to ground fault,
because the fault current is small. The system operates with voltages 1.73 times the
phase voltage on the two healthy phases. This causes stress on your components and if
not cleared will causes multiple faults.
Objective 2
This a more analytical research where available data is analyzed
This is done to know the main reasons for outages and on what conditions.
What are the main reasons for outages?
What are the network specifications?
The main reasons for outages are nikaldiet cable-joints and damaged caused by digging
activities.
The table gives an indication on what conditions these outages happen.
Component
Nikaldiet joint
Cable leadpaper
Failure cause
Age/
manufacturing
Damaged
caused by
digging
activities
Grounding type
Impedance/
ungrounded
Impedance/
ungrounded
Fault type
Single phase
Single phase
Objective 3
Here simulations are done to understand the different behaviour of currents and voltages in
different grounding schemes.
What is the impact of faults on different grounding schemes?
Is there any difference between different grounding methods?
- 69 -
Objective 3 simulations are done to see the best option for network to operate. Here
grounding types and transformer connection were varied.
The impact of faults is that the components are stressed by the voltages and current during a
fault. Grounding types and transformer connection gives different behaviour of the current
and voltage.
The following scenario’s where simulated:
Network 1 and network 2
Network 1 the following was used to do the simulations:
Delta-wye connection of transformer
Delta-delta connection of transformer
Wye-delta connection of transformer
Wye-wye connection of transformer
Grounding types: ungrounded, star grounded and impedance grounded
Fault types: L-G, L-L, L-L-G, L-L-L and L-L-L-G
In paragraph 6 section 6.2.4 and 6.2.5 differences between the above mentioned simulations
were shown. According to these simulations the following came out:
-
Wye-wye connection of transformer is the best option, because the voltage profile has
no distortion as the other connection types.
-
The grounding types are only suitable for single phase to ground faults. For the other
type of fault the grounding type does not have any influence.
-
Depending on your system specification and your protection device the grounding type
should be chosen, because the fault current is different for different network structures.
The fault current should be visible and detectible.
For network 2 the same applies as for network 1 only difference is that the current magnitudes
are increased.
Best option is: Wye-Wye connection of transformer and resistive/
impedance grounding
Overall: A component that fails in the system is not dependent on the transformer connection
or grounding type. A component can only be protected by these changes in the network, so
that the stresses on the component will be reduced.
- 70 -
Recommendation
-
Using professional edition of PSCAD to simulated more complicated networks.
The software I used for my simulations was a student version. There were some
limitations for expanding the network.
-
Using different kind of loads
The loads used in these simulations were static. The impact of dynamic loads such as
motors can have different impact. Also the upcoming trend of electric vehicles is one
of the scenario’s which will influence the behaviour of current and voltages during
fault.
-
Using other kind of generations (solar, wind etc)
Wind generation, solar and other types of generation are nowadays increasing. So it’s
interesting to use this kind of generation in the simulations.
- 71 -
List of literature
1
Nekaldiet reductie in risicostrengen, “onderzoek naar de reductie van VBM voor de
component Nekaldiet in risicostrengen” Koen Verstappen & Peter Dankmeijer
Alliander N.V.
2
“Power system grounding”
Dr.ir M. Popov
Lecture notes
3
Protection of electricity distribution networks (2nd edition)
Juan M.Gers and Edward J.Holmes © 2004 Institution of Engineering and Technology
4
PSCAD Electromagnetic Transients
Manitoba HVDC Research Centre
5
“Computer-Aided Power System Analysis”, volume 15
Ramasamy Natarajan 2002
6
“Power system analysis” 1 edition (1 January, 1994)
Juan J. Grainger, William D. Stevenson, Jr.
7
“Grounding of Industrial Power Systems”
Electrical Business “Grounding of industrial power system”
“High resistance is a far most effective method”
By Andrew Cochran (may 2002)
8
“Transient Overvoltage on Ungrounded System from Intermittent Ground Faults”
Eaton Power Business Worldwide
9
“Earth Fault in Extensive Cable Networks”
Electrical distribution system
Anna Gulbrand
10
“Short circuit Calculations in Networks with distributed distribution”
Thekla N. Boutsika, Stravos A. Papathanassiou
Electrical power system research 78 (2008)
www.sciencedirect.com
11
“Ground Fault protection on ungrounded and high resistance grounded system”
www.postglover.com
12
“Approach on the Neutral Grounding Modes of 20kV distribution Networks”
JI Yafei, Hou Yi Ming, IEEE
China Electric Power Research Institute
13
“Analysis and Selection of Neutral Grounding Modes in Cable distribution Network”
Yundong Song, Shun Yaun, Chunfang Zhao, Yangfen Jia, IEEE
Department of Electrical Engineering, Sheyang University of Technology
- 72 -
14
“Distribution System Grounding Impacts of Fault Responses”
S. A. Arefifar, Student Member, IEEE
15
“Distribution System Neutral Grounding”
Jim Burke and Mike Marshall
ABB Power T&D
16
“Study of Neutral Grounding for 22 KV Distribution System”
K.Oka, J.Yoshinaga, S.Koizumi, S.Uemura and Y.Ariga, IEEE
17
Www.littlefuse.com
18
“Intelligent Distribution Network Design”
Frans Provoost 2009
19
Frost & Sullivan Power systems
www.frost.com
- 73 -
Appendices
Appendix A Simulation procedure for network type 1
Simulations
Network type 1
Delta-Wye
A
U-G
A1
L-G
A2
Wye-Wye
B
D-G
B1
L-L
B2
Wye-Delta
C
I-G
C1
L-L-G
C2
Delta-Delta
D
Transformer
connection
Grounding
type
L-L-L
D2
L-L-L-G
E2
Fault
type
U-G : Ungrounded
D-G : Directly grounded
I-G : Impedance grounded
L-G : phase to ground
L-L : phase to phase
L-L-G : phase to phase to ground
L-L-L : three phase
L-L-L-G : three phase to ground
- 74 -
Network type 1
Appendix A1 Delta-Wye connection of transformer
A.1.1 Delta-Wye connection of transformer ungrounded system with different fault
types.
A-A1-A2
A-A1-B2
- 75 -
A-A1-C2
A-A1-D2
- 76 -
A-A1-E2
A.1.2 Delta-Wye connection of transformer directly grounded system with different
fault types.
A-B1-A2
- 77 -
A-B1-B2
A-B1-C2
- 78 -
A-B1-D2
A-B1-E2
- 79 -
A.1.3 Delta-Wye connection of transformer, variation of the impedance with ground
fault
A-C1-A2 where C1 = 0.5 Ohm
A-C1-A2 where C1 = 1 Ohm
- 80 -
A-C1-A2 where C1 = 5 Ohm
A-C1-A2 where C1 = 7 Ohm
- 81 -
A-C1-A2 where C1 = 10 Ohm
A-C1-A2 where C1 = 20 Ohm
- 82 -
Appendix A2 Wye-Wye connection of transformer
A.2.1 Wye-Wye connection of transformer ungrounded system with different fault
types.
B-A1-A2
B-A1-B2
- 83 -
B-A1-C2
B-A1-D2
- 84 -
A-A1-E2
A.2.2 Wye-Wye connection of transformer directly grounded system with different fault
types
B-B1-A2
- 85 -
B-B1-B2
B-B1-C2
- 86 -
B-B1-D2
B-B1-E2
- 87 -
A.2.3 Wye-Wye connection of transformer, variation of the impedance with ground fault
B-C1-A2 where C1 = 0.5 Ohm
B-C1-A2 where C1 = 1 Ohm
- 88 -
B-C1-A2 where C1 = 5 Ohm
B-C1-A2 where C1 = 7 Ohm
- 89 -
B-C1-A2 where C1 = 10 Ohm
A.2.4 Wye-wye connection of transformer, variation of impedance and fault types
randomly.
B-C1-B2 where C1 = 7 Ohm
- 90 -
B-C1-C2 where C1 = 0.5 Ohm
B-C1-C2 where C1 = 7 Ohm
- 91 -
B-C1-D2 where C1 = 7 Ohm
B-C1-E2 where C1 = 0.5 Ohm
- 92 -
B-C1-E2 where C1 = 7 Ohm
Appendix A3 Wye-Delta connection of transformer
C-A1-A2
- 93 -
C-A1-B2
C-A1-C2
- 94 -
C-A1-D2
C-A1-E2
- 95 -
Appendix A4 Delta-Delta connection of transformer
D-A1-A2
D-A1-B2
- 96 -
D-A1-C2
D-A1-D2
- 97 -
D-A1-E2
- 98 -
Appendix B Simulation procedure for network type 2
Network type 2
Transformer 1
T1
Delta-Wye
A
Wye-Wye
B
U-G
A1
L-G
A2
Transformer 2
T2
D-G
B1
L-L
B2
Wye-Delta
C
I-G
C1
L-L-G
C2
Delta-Delta
D
Transformer
connection
Grounding
type
L-L-L
D2
L-L-L-G
E2
Fault
type
Network type 2
Starting with ground faults at location 1
- 99 -
Appendix B1 Delta-Wye vs Wye delta connection with different grounding
types for ground faults.
Fault location 1
T1 A-A1-A2 vs T2 C-A1-A2
T1 A-B1-A1 vs T2 C-A1-A
- 100 -
T1 A-C1-A2 vs T2 C-A1-A2
T1 A-B1-A2 vs T2 C-B1-A2
- 101 -
T1 A-C1-A2 vs T2 C-C1-A2
- 102 -
Appendix B2 Wye-Wye vs Wye-Wye connection with different grounding type
for ground faults. Fault location 1
T1 B-A1-A2 vs T2 B-A1-A2
T1 B-B1-A2 vs T2 B-A1-A2
- 103 -
T1 B-B1-A2 vs T2 B-B1-A2
T1 B-C1-A2 vs T2 B-C1-A2
- 104 -
Appendix B3 Wye-Wye vs Wye-delta connection with different grounding types
for ground faults. Fault location 1
T1 B-A1-A2 vs T2 C-A1-A2
T1 B-B1-A2 vs T2 C-A1-A2
- 105 -
T1 B-C1-A2 vs T2 C-A1-A2
- 106 -
Appendix B4 Delta-Wye vs Wye-Delta connection for different grounding
type on ground faults.
Fault location 2
T1 A-A1-A2 vs C-A1-A2
T1 A-B1-A2 vs C-A1-A2
- 107 -
T1 A-C1-A2 vs T2 C-A1-A2
T1 A-B1-A2 vs T2 C-B1-A2
- 108 -
T1 A-C1-A2 vs T2 C-C1-A2
- 109 -
Appendix B5 Wye-Wye vs Wye-Wye connection for different grounding
type on ground faults.
Fault location 2
T1 B-A1-A2 vs T2 B-A1-A2
T1 B-B1-A2 vs T2 B-A1-A2
- 110 -
T1 B-B1-A2 vs T2 B-B1-A2
T1 B-C1-A2 vs T2 B-A1-A2
- 111 -
T1 B-C1-A2 vs T2 B-C1-A2
- 112 -
Appendix B6 Wye-Wye vs Wye-Wye connection for different grounding
type on ground faults.
Fault location 3
T1 B-A1-A2 vs T2 B-A1-A2
T1 B-B1-A2 vs T2 B-A1-A2
- 113 -
T1 B-A1-A2 vs T2 B-B1-A2
T1 B-B1-A2 vs B-B1-A2
- 114 -
Appendix C reactance grounding
Delta-Wye connection of transformer
X = 7 Ohm
L = 22.3 mH
Fault location 1
Wye-wye
- 115 -
Fault location 2
Delta-wye
Wye-wye
- 116 -
Fault location 3
Delta-wye
Wye-wye
- 117 -
Changing the reactance
0.01H wye-wye
0.02H wye-wye
- 118 -
0.03H wye-wye
Changing reactance
0.01H delta-wye
- 119 -
0.02H delta-wye
0.03H delta-wye
- 120 -
With second generator
Fault location 1
Delta-wye L = 22.3 mH vs wye-delta
Delta-wye vs wye-delta L = 22.3 mH
- 121 -
Delta-wye L = 22.3 mH vs wye-delta L = 22.3 mH
Wye-wye L = 22.3 mH vs wye-wye
- 122 -
Wye-wye vs wye-wye L = 22.3 mH
Wye-wye L = 22.3 mH vs Wye-wye L = 22.3 mH
- 123 -
Appendix D Calculation of grounding resistance
The experience and analysis have certified that the grounding arc can burn steadily when the
single-phase fault current is above 500~600A, thus the over-voltage caused by the intermittent
arc can be avoided, which is convenient for the correct acts of relay protection. The singlephase fault current I d should be first fixed, and then the current of the neutral point can be
derived by: I d  I 2 c  I N
2
[12]
(D.1)
Where I d is the single-phase fault current; I c is the capacitance current; and I N is the neutral
point current. The resistance value can be derived by:
Ul
3
(D.2)
RN 
IN
Where U l is the nominal voltage; I N is the neutral point current; and RN is the neutral
grounding resistance. It is usually know that the relay could remove the problem in less than 1
second. According to the principle that the heat-stable current reaches 15 times the nominal
current of the electric facilities, the nominal current of neutral grounding resistor is 1 times
15
I
the neutral point current namely: I r  N
(D.3)
15
Where I N is the neutral point current; I r is the nominal current of the neutral grounding
resistor.
Then the nominal capacity of the grounding resistor can be derived by:
(D.4)
Pr  I r2 RN
Where I r is the nominal current of the neutral grounding resistor; Pr is the nominal capacity of
the grounding resistor.
Calculation of the Peterson coil parameters
In the condition of the Peterson coil grounding mode, the capacity of the Peterson coil can de
derived by:
U
(D.5)
Q  SI c l
3
Where Q is the capacity of the Peterson coil; S is the capacity reserve factor; 1.25~1.35; I c is
the capacitance current; U l is the nominal voltage.
- 124 -