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 TU Delft: Dr. Ir. Marjan Popov Supervisor Alliander NV: Ir. Nico Steentjes &

Ing. John van Slogteren

- 1 -

Thesis committee: Prof. Ir. L. van der Sluis Delft University of Technology, Thesis supervisor Dr. Ir. M. Popov Delft University of Technology, Daily supervisor Ir. N. Steentjes Liander , Daily supervisor Dr. Ir. D. Djairam Delft University of Technology

- 2 -

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

- 3 -

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.

- 4 -

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.

- 5 -

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 -

- 6 -

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 -

- 7 -

List of symbols and abbreviations

NDCM: Not Delivered Consumer Minutes

PSCAD: Power System Computer Aid Design

I: Current (amps-A)

R: Resistance (ohms-Ω)

V: Voltage (volts-V)

HV: High Voltage (above 36 kV)

Rn: neutral resistor (ohms-Ω)

Xc0: capacitive reactance (ohms-Ω)

R0: zero sequence resistance (ohms-Ω)

positive reactance (ohms-Ω)

X0: zero sequence reactance (ohms-Ω)

T: time (seconds)

If: fault current (amps-A)

Ic: capacitance current (amps-A)

π: 3.1415927

C0: Single phase to earth capacitance (farads-F)

L: length of cable (meters-m)

Id: single phase fault current (amps-A)

In: neutral point current (amps-A)

Ir: nominal current of the neutral grounding resistor (amps-A)

Pr: nominal power of the grounding resistor (watt-W)

Pp: power of the Peterson coil (watt-W)

S: power reserve factor; 1.25-1.35

I0: zero sequence current (amps-A)

I positive sequence current (amps-A)

I2: negative sequence current (amps-A)

Zf: fault impedance (ohms-Ω)

Ia: current in phase a (amps-A)

Ib: current in phase b (amps-A)

Ic: current in phase c (amps-A)

Va: voltage in phase a (volts-V)

Vb: voltage in phase b (volts-V)

Vc: voltage in phase c (volts-V)

Z1: positive sequence impedance (ohms-Ω)

Z0: zero sequence impedance (ohms-Ω)

Z2: negative sequence impedance (ohms-Ω)

a: 1201

a²: 2401

: Electrical resistivity

A: cross section (m²)

P: active power (MW)

Q: reactive power (MVar)

S: apparent power = P + j Q

L-G: phase to ground

L-L: phase to phase

L-L-G: two phase to ground

- 8 -

L-L-L: three phase

L-L-L-G: three phase to ground

f: frequency (Hertz-Hz)

MV: medium voltage (1kV < U < 36kV)

- 9 -

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.

- 10 -

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

- 11 -

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.

- 12 -

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.

- 13 -

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: 3

XcoRn or RnRo 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.

- 14 -

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.

- 15 -

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.

- 16 -

- 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 over-

voltages 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.

- 17 -

- 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

- 18 -

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 ungrounded-

neutral 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:

3

100 0 LUCI l

c

LUC l03,181 (2.2)

Where lU is the rated line voltage; 0C is the single-phase earth capacitance, L is length of the

line.

To overhead line, the capacitance current can be derived simply by:

(2.3) 310)3.3~7.2( LUI lc

- 19 -

To cable line, the capacitance current can be derived simply by:

(2.4)

Capacitance current vs Cable lenght

0

50

100

150

200

250

300

1 2 4 5 10 20 30 40 50 60 70

L

Ic Ic

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 high-

voltage 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.

0.1c lI U L

- 20 -

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-phase-

to-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 high-

speed 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:

- 21 -

- 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]:

0I = zero sequence component

1I = positive sequence component

2I = 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

00 I (3.1)

0 IcIbIa (3.2)

Figure 3.3 of three phase fault

- 22 -

As the fault is symmetrical:

Ic

Ib

Ia

Zf

Zf

Zf

Vc

Vb

Va

00

00

00

(3.3)

The sequence voltages are given by

2

1

0

2

1

0

1

2

1

0

00

00

00

00

00

00

I

I

I

Zf

Zf

Zf

I

I

I

Ts

Zf

Zf

Zf

Ts

V

V

V

(3.4)

This gives the equivalent circuit of figure 3.7

ZfZ

VaIIa

1

1 (3.5)

1

2IaIb (3.6)

1aIIc (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.,

0Ia , therefore (3.8)

0210 III (3.9)

ZfIcIbVcVb )( (3.10)

Figure 3.4 of phase to phase to ground fault

Thus,

VbaaVa

VbaaVa

VbVa

Vc

Vb

Va

aa

aa

V

V

V

)(

)(

2

3

1

1

1

111

3

1

2

2

2

2

2

1

0

(3.11)

- 23 -

Which gives 21 VV and

0

0 0 1 2

0 0 1 0

0 1 0

1( 2 )

3

1( ) 2( )

3

1( 2 ) 2(3 )

3

3

V Va Vb

V V V V Ib Ic Zf

V V V I Zf

V V ZfI

(3.12)

This gives the equivalent circuit of figure 3.7

The fault current is

1

1 2 0 3

VaI

Z Z Z Zf

(3.13)

12 0

1

2 0

( 3 )

3

VaI

Z Z ZfZ

Z Z Zf

(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,

0,Ia Ib Ic . (3.15)

The sequence components of the currents are:

Figure 3.5 of phase to phase fault

aa

aa

Ic

Ic

aa

aa

I

I

I

2

2

2

2

2

1

0 0

3

10

1

1

111

3

1 (3.16)

From equation 3.16

0 0I and 1 2I I (3.17)

- 24 -

2

1

0

22

2

1

0

2

2 0

1

1

111

110110

V

V

V

aaaa

V

V

V

aa

aa

Vc

Vb

Va

VcVb (3.18)

Therefore,

ZfIaa

ZfaIIa

VVaaVcVb

1

2

21

2

21

2

)(

)(

))((

(3.19)

This gives

ZfIVV 121 )( (3.20)

The equivalent circuit is shown in figure 3.7

Also

ZfIaaIb 1

2 )( (3.21)

And,

ZfZZ

VaI

21

1 (3.22)

The fault current is

1 2

3VaIb Ic

Z Z 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 0Zf , 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.

- 25 -

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

Fault conditions in phase domain 0 IcIb (3.24)

Single line-to-ground fault voltage IaZfVag * (3.25)

We now transform (3.22) and (3.23) to the sequence domain:

Ia

Ia

IaIa

aa

aa

I

I

I

3

1

0

0

1

1

111

3

1

2

2

2

1

0

(3.26)

Also,

IaIII3

1210 (3.27)

2211002100 )(3 VIZIVaZIVVVZfI (3.28)

which gives,

ZfZZZ

VaI

3210

0

(3.29)

The fault current Ia is

ZfZZZ

VaIIa

3

33

021

0

(3.30)

This shows that the equivalent fault circuit using sequence impedances can be constructed as

shown in figure 3.7 [2].

- 26 -

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.

Single phase to ground fault: ZfZZZ

VaIIa

3

33

021

0

, 0 IcIb

Two phase fault between phase b and c:ZfZZ

VajIcIb

21

3, 0Ia

Two phase to ground fault between phase b and c: 1

2 01

2 0

( 3 )

3

VaI

Z Z ZfZ

Z Z Zf

,

01

2 0

3

3

Z ZfIb I

Z Z Zf

, 2

1

2 0 3

ZIc I

Z Z Zf

Three phase fault:ZfZ

VaIIa

1

1 , 1

2IaIb , 1aIIc

- 27 -

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.

- 28 -

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

- 29 -

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

- 30 -

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.

- 31 -

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

- 32 -

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

Joints 55% Cable 19% Protection 5% Circuit breaker 8% Transformer 10% End enclosure 3%

- 33 -

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

- 34 -

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

- 35 -

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.

- 36 -

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.

- 37 -

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

- 38 -

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 mid-

voltage 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.

- 39 -

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.

- 40 -

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.

- 41 -

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

- 42 -

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.

- 43 -

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.

- 44 -

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]

- 45 -

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 Failure cause Grounding type Fault type

Nekaldiet cable

joint

Age/

manufacturing

Impedance/

ungrounded

Single phase

Cable lead-

paper

Digging Impedance/

ungrounded

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.

- 46 -

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:

- 47 -

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

MVAKAKVIscUSsc 13640150*5.52*3**3

95.43*Isc

UZsc

Generator: V = 52.5 kV

Transformer: P = 30 MVA V1/V2 = 52.5/10.5

Cable: cm.10*8.2 6 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)

- 48 -

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:

6 3

6,7 26.6

* *cos

2 10.5 * *0.89

214

214*0.5 2 151

S P jQ MVA j MVAR

S

Pphase Uphase Iphase

MVA kV Iphase

Iphase A Ipeak

Iphase A

The network is now simulated the steady state values are given in figure 6.3

Figure 6.3 steady state values

- 49 -

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*8.4 14.55KV KV . 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.

- 50 -

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

- 51 -

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

- 52 -

Figure 6.9 gives the magnitude of the currents and voltages of different fault types

Delta-star connection of transformer Ungrounded

0

5

10

15

20

25

30

35

40

45

L-G L-L L-L-G L-L_L L-L-L-G

Types of fault

kA

/kV Current

Voltage

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 delta-

wye 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.

0

5

10

15

20

25

30

35

40

45

L-G L-L L-L-G L-L_L L-L-L-G

kA

/kV

Type of fault

Delta-star connection of transformer Star-grounded

Current

Voltage

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.

- 53 -

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

0

2

4

6

8

10

12

14

16

0,5 1 5 7 10 20

resistance

kA

/kV Current

Voltage

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).

- 54 -

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

0

10

20

30

40

50

L-G L-L L-L-G L-L-L L-L-L-G

types of fault

kV

/kA Current

Voltage

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

0

10

20

30

40

50

L-G L-L L-L-G L-L-L L-L-L-G

types of fault

kV

/kA Current

Voltage

Figure 6.2.2.2 types of fault in a directly grounded system with wye-wye connection of

transformer

- 55 -

Impedance grounded

Star-star connection of transformer impedance grounded

0

2

4

6

8

10

12

14

16

0,5 1 5 7 20

resistance

kV

/kA Current

Voltage

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

0

10

20

30

40

50

L-G L-L L-L-G L-L-L L-L-L-G

types of fault

kV

/kA Current

Voltage

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.

- 56 -

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

0

10

20

30

40

50

L-G L-L L-L-G L-L-L L-L-L-G

types of fault

kV

/kA Current

voltage

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

0

10

20

30

40

50

L-G L-L L-L-G L-L-L L-L-L-G

types of fault

kV

/kA Current

Voltage

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.

- 57 -

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.

- 58 -

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.

- 59 -

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.

- 60 -

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.

- 61 -

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.

- 62 -

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.

- 63 -

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.

- 64 -

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.

- 65 -

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 wye-

wye 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

- 66 -

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.

- 67 -

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-D-

IG

+ - + + + - + + + + + +

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-W-

IG

+ + + + + - + + + - + +

Tabel 6.3.3 comparison of voltages and currents at different locations. F1: fault location1;

F2: fault location 2; F3: fault location 3

- 68 -

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.

- 69 -

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 Failure cause Grounding type Fault type

Nikaldiet joint Age/

manufacturing

Impedance/

ungrounded

Single phase

Cable lead-

paper

Damaged

caused by

digging

activities

Impedance/

ungrounded

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?

- 70 -

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.

- 71 -

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.

- 72 -

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

- 73 -

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

- 74 -

Appendices

Appendix A Simulation procedure for network type 1

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

Simulations

Network type 1

Delta-Wye A

Wye-Wye B

Wye-Delta C

Delta-Delta D

U-G A1

D-G B1

I-G C1

L-G A2

L-L B2

L-L-G C2

L-L-L D2

L-L-L-G E2

Transformer connection

Grounding type

Fault type

- 75 -

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

- 76 -

A-A1-C2

A-A1-D2

- 77 -

A-A1-E2

A.1.2 Delta-Wye connection of transformer directly grounded system with different

fault types.

A-B1-A2

- 78 -

A-B1-B2

A-B1-C2

- 79 -

A-B1-D2

A-B1-E2

- 80 -

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

- 81 -

A-C1-A2 where C1 = 5 Ohm

A-C1-A2 where C1 = 7 Ohm

- 82 -

A-C1-A2 where C1 = 10 Ohm

A-C1-A2 where C1 = 20 Ohm

- 83 -

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

- 84 -

B-A1-C2

B-A1-D2

- 85 -

A-A1-E2

A.2.2 Wye-Wye connection of transformer directly grounded system with different fault

types

B-B1-A2

- 86 -

B-B1-B2

B-B1-C2

- 87 -

B-B1-D2

B-B1-E2

- 88 -

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

- 89 -

B-C1-A2 where C1 = 5 Ohm

B-C1-A2 where C1 = 7 Ohm

- 90 -

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

- 91 -

B-C1-C2 where C1 = 0.5 Ohm

B-C1-C2 where C1 = 7 Ohm

- 92 -

B-C1-D2 where C1 = 7 Ohm

B-C1-E2 where C1 = 0.5 Ohm

- 93 -

B-C1-E2 where C1 = 7 Ohm

Appendix A3 Wye-Delta connection of transformer

C-A1-A2

- 94 -

C-A1-B2

C-A1-C2

- 95 -

C-A1-D2

C-A1-E2

- 96 -

Appendix A4 Delta-Delta connection of transformer

D-A1-A2

D-A1-B2

- 97 -

D-A1-C2

D-A1-D2

- 98 -

D-A1-E2

- 99 -

Appendix B Simulation procedure for network type 2

Network type 2

Starting with ground faults at location 1

Network type 2

Wye-Delta C

Wye-Wye B

Delta-Wye A

Delta-Delta D

Transformer 1 T1

Transformer 2 T2

I-G C1

D-G B1

U-G A1

L-G A2

L-L B2

L-L-G C2

L-L-L D2

L-L-L-G E2

Transformer connection

Grounding type

Fault type

- 100 -

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

- 101 -

T1 A-C1-A2 vs T2 C-A1-A2

T1 A-B1-A2 vs T2 C-B1-A2

- 102 -

T1 A-C1-A2 vs T2 C-C1-A2

- 103 -

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

- 104 -

T1 B-B1-A2 vs T2 B-B1-A2

T1 B-C1-A2 vs T2 B-C1-A2

- 105 -

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

- 106 -

T1 B-C1-A2 vs T2 C-A1-A2

- 107 -

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

- 108 -

T1 A-C1-A2 vs T2 C-A1-A2

T1 A-B1-A2 vs T2 C-B1-A2

- 109 -

T1 A-C1-A2 vs T2 C-C1-A2

- 110 -

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

- 111 -

T1 B-B1-A2 vs T2 B-B1-A2

T1 B-C1-A2 vs T2 B-A1-A2

- 112 -

T1 B-C1-A2 vs T2 B-C1-A2

- 113 -

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

- 114 -

T1 B-A1-A2 vs T2 B-B1-A2

T1 B-B1-A2 vs B-B1-A2

- 115 -

Appendix C reactance grounding

Delta-Wye connection of transformer

X = 7 Ohm

L = 22.3 mH

Fault location 1

Wye-wye

- 116 -

Fault location 2

Delta-wye

Wye-wye

- 117 -

Fault location 3

Delta-wye

Wye-wye

- 118 -

Changing the reactance

0.01H wye-wye

0.02H wye-wye

- 119 -

0.03H wye-wye

Changing reactance

0.01H delta-wye

- 120 -

0.02H delta-wye

0.03H delta-wye

- 121 -

With second generator

Fault location 1

Delta-wye L = 22.3 mH vs wye-delta

Delta-wye vs wye-delta L = 22.3 mH

- 122 -

Delta-wye L = 22.3 mH vs wye-delta L = 22.3 mH

Wye-wye L = 22.3 mH vs wye-wye

- 123 -

Wye-wye vs wye-wye L = 22.3 mH

Wye-wye L = 22.3 mH vs Wye-wye L = 22.3 mH

- 124 -

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 single-

phase fault current dI should be first fixed, and then the current of the neutral point can be

derived by: 22

Ncd III [12] (D.1)

Where dI is the single-phase fault current; cI is the capacitance current; and NI is the neutral

point current. The resistance value can be derived by:

N

l

NI

U

R 3 (D.2)

Where lU is the nominal voltage; NI is the neutral point current; and NR 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 15

1 times

the neutral point current namely: 15

Nr

II (D.3)

Where NI is the neutral point current; rI is the nominal current of the neutral grounding

resistor.

Then the nominal capacity of the grounding resistor can be derived by:

Nrr RIP 2 (D.4)

Where rI is the nominal current of the neutral grounding resistor; rP 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:

3

lc

USIQ (D.5)

Where Q is the capacity of the Peterson coil; S is the capacity reserve factor; 1.25~1.35; cI is

the capacitance current; lU is the nominal voltage.