torelli za ucinke
TRANSCRIPT
-
8/10/2019 Torelli Za Ucinke
1/180
1/138
Transient Overvoltages and Distance Protection:Problems and Solutions
A thesis submitted in fulfillment of the requirements for the
degree of Master of Engineering by Research
Leonardo Torelli
B.Eng. (Hons)
School of Electrical and Computer Engineering
RMIT University
November 2010
-
8/10/2019 Torelli Za Ucinke
2/180
2/138
Table of Contents
The Author .......................................................................................................................... 5
Acknowledgement...............................................................................................................6
Declaration .......................................................................................................................... 7
Abbreviations ...................................................................................................................... 8
Executive Summary ..........................................................................................................10
1 Introduction .................................................................................................................. 11
2 Neutral Earth Resistor and Transient Overvoltages .....................................................14
2.1 Introduction ..........................................................................................................14
2.2 Method of Earthing............................................................................................... 15
2.3 Review of the Existing Body of Knowledge ........................................................16
2.4 Design of the Experiment .................................................................................... 25
2.4.1 The Software Package .................................................................................. 252.4.2 The Model .................................................................................................... 26
2.4.3 The experimental design............................................................................... 27
2.4.4 System Configuration ................................................................................... 29
2.4.5 Simulation Characteristics ............................................................................ 31
2.4.6 Model Components ...................................................................................... 32
2.5 Simulation Results................................................................................................40
2.5.1 Simulation 1 : Unloaded Line - Fault at the End of the Line ....................... 41
2.5.2 Simulation 2 : Unloaded Line - Fault at the Beginning of the Line ............. 47
2.5.3 Simulation 3 : Capacitor Bank. Fault at the End of the Line....................... 50
2.5.4 Simulation 4 : Capacitor Bank. Fault at the Beginning of the Line ............. 56
2.5.5 Simulation 5 : Light inductive Load. Fault at the End of the Line............... 59
2.5.6 Simulation 6 : Energized 66/22kV Transformer Fault at the 66kV Bus. .... 62
2.5.7 Overvoltages and Traveling Waves.............................................................. 63
2.5.8 Impact of Time of Fault with the Overvoltage ............................................. 68
2.5.9 Overvoltage Response on Phase a ,b and c .................................................. 72
2.5.10 Overvoltage Harmonic Response ................................................................. 76
2.6 Conclusion............................................................................................................78
3 CVT and Transient Overvoltages ................................................................................. 80
3.1 Overview .............................................................................................................. 80
-
8/10/2019 Torelli Za Ucinke
3/180
3/138
3.2 The CVT Model ..................................................................................................81
3.3 Simulation Settings............................................................................................... 843.4 Simulation Results................................................................................................86
3.5 CVT Conclusion .................................................................................................100
4 Distance Protection Scheme ....................................................................................... 101
4.1 Introduction ........................................................................................................101
4.2 MHO Characteristics Theory ............................................................................. 107
5 Transient Overvoltages and Distance Protection: Solutions ...................................... 111
5.1 Introduction ........................................................................................................111
5.2 Polarisation Techniques...................................................................................... 113
5.3 Polarising Techniques ........................................................................................ 117
5.4 CVT Techniques.................................................................................................119
5.5 Setting Advice ....................................................................................................121
6 Conclusion..................................................................................................................125
6.1 Research Findings ..............................................................................................125
6.2 Areas for Further Investigation........................................................................... 127
7 References .................................................................................................................. 129
Appendix A ..................................................................................................................... 133
10th International Conference on Developments in Power System Protection, March
2010, Manchester, UK -Abstract.....................................................................................133
Appendix B ..................................................................................................................... 134
Overvoltage StudyPSCAD Data .................................................................................. 134
Appendix C ..................................................................................................................... 135
Overvoltage studySteady State Voltage Data ............................................................. 135
Appendix D ..................................................................................................................... 136
Overvoltage Study- PSCAD Model ................................................................................ 136
Appendix E......................................................................................................................137
Voltage and Current as function of SIR.......................................................................... 137
-
8/10/2019 Torelli Za Ucinke
4/180
4/138
Appendix F......................................................................................................................138
CVT Study
PSCAD Data .............................................................................................138
-
8/10/2019 Torelli Za Ucinke
5/180
5/138
The Author
My name is Leonardo Torelli. I have been working for fourteen years in power
industry. I have worked in Italy in Utilities and in Australia in consulting firms. I am
currently working at Hydro Tasmania Consulting and involved in Protection and
Power System Analysis projects.
It is not a surprise that the topic of this research goes across these two areas of
electrical engineering. The aim of this study was, indeed, proving how important is
the knowledge of power system analysis for Protection Engineers.
As result of this study, I presented a paper at the 10th International Conference on
Developments in Power System Protection, March 2010, Manchester, UK
This research has given me the opportunity to expand my skills, challenging my
knowledge and improved my analytical skills. I am committed to use this
experience at RMIT University as starting point for further studies.
-
8/10/2019 Torelli Za Ucinke
6/180
6/138
Acknowledgement
I thank Dr Selva Moorthy for his contribution, support and expert advice during this
study. I have appreciated his sincere interest in my professional development and
support of my personal life. Dr Moorthys feedback helped me to stay in line with
my schedule and achieve the completion of this research.
I would like to thank my former and existing colleagues at Hydro TasmaniaConsulting for their encouragement during this journey.
Last, I would like to thank my three beautiful children, my family and friends for
their warm and enthusiastic appreciation of my work.
.
-
8/10/2019 Torelli Za Ucinke
7/180
7/138
Declaration
I certify that except where due acknowledgement has been made, the work is that
of the author alone; the work has not been submitted previously, in whole or in
part, to qualify for any other academic award; the content of the thesis is the result
of work which has been carried out since the official commencement date of the
approved research program; and, any editorial work, paid or unpaid, carried out by
a third party is acknowledged.
.
Leonardo Torelli
Date
-
8/10/2019 Torelli Za Ucinke
8/180
8/138
Abbreviations
The following terms are used in this thesis:
! HVDC - High Voltage Direct Current
! MHO relay - Distance relay with circular characteristic which passes through
the origin of the R-X plane
! NER - Neutral Earth Resistor
! NEX - Neutral Earth Reactance
! PSCAD-EMTDC - Power system software package
! SIR- Source Impedance Ratio
! TRV- Transient Recovery Voltage
-
8/10/2019 Torelli Za Ucinke
9/180
-
8/10/2019 Torelli Za Ucinke
10/180
-
8/10/2019 Torelli Za Ucinke
11/180
11/138
1 Introduction
Power systems are designed and operated to supply electrical energy to the
customers in a safe, reliable and economical way. Faults represent one of the main
challenges for the power system. Among all the types of faults, phase to ground
faults represent the majority of the events, with records between 80 to 90 % of all
faults. Ground faults are generated from insulation breakdown, atmospheric
conditions and accidental contacts of birds or branches of trees with power lines.
Therefore, these faults are transitory in nature.
During this post fault initial period, the voltage change dramatically from a pre fault
steady state value to a post fault steady state value. In addition, the voltage signal
is disturbed by higher and lower frequency components. Therefore, transient
overvoltages could affect the accuracy of the protection scheme. At the same time,
the protection scheme should be able to operate in a fast manner to reduce the risk
to personnel, equipment damage, and system stability.
It is evident that achieving a short operating time and high accuracy of the
protection scheme represents a major challenge for protection relay manufacturers
and protection engineers. This research investigated the performance of modern
digital distance relay during the transient overvoltages period which follows a
ground fault.
Initially, this study focused on power system analysis to have a clear understanding
of the transient overvoltage phenomena. The overvoltage study was conducted
using PSCAD-EMTC software package. The investigation was performed by using
the experimental method that involves the scientific manipulation of the variables
involved in the process and the systematic study of the behavior of the system.
The study was conducted on a radial system using a neutral earth resistor, NER,
as a method to earth the 66 kV network. However, the theory and the results
-
8/10/2019 Torelli Za Ucinke
12/180
12/138
obtained are largely applicable to higher transmission voltages and in power
systems earthed via a neutral earth reactor, NEX.
Following the transient overvoltages study, the research focused on the
performance of five modern digital distance relays.
In summary, the project objectives are as follow:
! Determine the characteristics of transient overvoltages generated by a
phase to ground fault. This work should highlight the main factors that
influence the voltage disturbances
! Analyze the characteristics of modern digital distance relays to determine if
the transient overvoltages could affect the accuracy ad operating time of
the protection relay
! Establish a method to predict the performance of the distance protection
relay following a phase to ground fault
! Elaborate a list of setting advice that could enhance the application of the
distance protection relay in relation to the transient overvoltages
disturbance
This study aimed to validate the existing body of knowledge, highlight new insight
and views and determine new findings.
The research is organised in six Sections:
! Section 2 -NER and Overvoltage
! Section 3 - NER and Capacitor Voltage Transformers
! Section 4- NER and Distance Protection
! Section 5- Distance Relay Applications
-
8/10/2019 Torelli Za Ucinke
13/180
13/138
In the last part of this study, Section 6, research conclusions and recommendations
for areas of further investigations are also provided.
-
8/10/2019 Torelli Za Ucinke
14/180
-
8/10/2019 Torelli Za Ucinke
15/180
15/138
2.2 Method of Earthing
The earthing of the power system involves the implementation of an electrical
connection between the neutral point and the ground.
Connection of the neutral to the earth can be done in several methods:
! Solidly earthing
! Neutral earth resistor, NER
! Neutral earth reactance, NEX
! Compensated earthing
! Ungrounded
The type and size of the earthing system will affect the following system
parameters [1-4]:
! Fault current
! Personnel safety
! System Protection
! Steady state overvoltage during ground faults
! Transient overvoltage during ground faults
! Thermal stress on the equipment
! Effect on communication circuit
! Harmonic current on neutral connection
! System stability
-
8/10/2019 Torelli Za Ucinke
16/180
-
8/10/2019 Torelli Za Ucinke
17/180
17/138
Transient overvoltages are produced by sudden changes in the electrical system
as caused by lightning, faults or switching [6]. The method of earthing system doesnot have any effect on the system during normal operation [4]. However, during the
transient period the method of earthing has an impact on the system response and
magnitude of the transient overvoltage.
Power demand, power generated and system configuration vary continuously with
time. Therefore the power system always changes from one steady state to
another steady state. In this very short periods current and voltages may reach
dangerous values for the equipment and the insulation of the system.
Overvoltages are often classified according to their duration in two groups [7, 8]:
! Temporary Overvoltage above 200 ms
! Transient Overvoltage below 200 ms
Alternatively, overvoltages can be classified according to their origin:
! External overvoltage caused by atmospheric phenomena as lighting and
electrostatic charges
! Internal overvoltage generated within the power systems
This type of disturbances are also classified according to the frequency of the two
major overvoltage components [9]:
! Power frequency overvoltage
! Natural frequency overvoltage of a short duration superimposed on the
power frequency overvoltage
The sum of the power and natural frequency overvoltage is the voltage recorded in
the field. This voltage is commonly defined as a transient overvoltage.
-
8/10/2019 Torelli Za Ucinke
18/180
18/138
The overvoltage component with a power frequency does not theoretically have
any decaying component and it often termed as steady state overvoltage. Instead,the short duration component usually decays in less than 100 ms. Above all; the
magnitude of the overall overvoltage is of greatest interest for the impact on the
planning, design and financial cost of the power system.
A fault in the power system can be represented and analysed as the closing of a
switch in the electrical system. This change in the power system develops a new
redistribution of stored energy in the system [6]. Phase to ground faults represent
the majority of faults in the power system. These types of faults are produced by
atmospheric condition, mechanical breakdown of the insulation, objects such as
birds or branches or trees in contact with the overhead line and poles or structures.
Recent statistics conducted by protection relay manufacturers found that these
types of faults represent between 80 and 90 % of the faults in a given power
system [3, 4, 10]. A large majority of these faults involving overhead lines also
have transient characteristics and, therefore, self extinguish in a short period of
time.
When a phase to ground fault occurs, an overvoltage can be measured in the
system during the fault itself or after clearing [10]. The maximum transient
overvoltage is obtained by adding the peak of the temporary overvoltage at power
frequency overvoltage at 50Hz sinusoidal pattern to the transient overvoltage
which usually has a higher frequency. This relation is based on the worst case
scenario with the assumptions that the two components will have maximum peak
values at the same time. This is plausible because of the different time frames.Power frequency peak lasts long in the context of high frequency oscillations.
The magnitude of the power frequency overvoltage is strictly related to the
characteristic parameters of the system. This overvoltage can be calculated using
the symmetrical components theory. Considering some system simplification as
necessary, this engineering technique produces accurate results. The fundamental
frequency voltage, which is also referred to the steady state overvoltage, is absent
-
8/10/2019 Torelli Za Ucinke
19/180
19/138
in a solidly earthed system. In an ungrounded system the overvoltage reaches the
maximum value of !3=1.732. In this latter case, the phase to ground voltage isequal to the phase to phase voltage.
Application of the neutral earth resistor as a method of grounding increases the
fundamental frequency overvoltage from none to a maximum of 1.732 for the
insulated system.
Between these two boundaries of the overvoltage measured in the system, the
institution of Electrical and Electronic Engineers established an arbitrary limit of
the overvoltage measured, which is named effective grounding[11].
The effective grounding term indicates that, during a ground fault, the voltage on
the healthy phases will not exceed 80 % of the maximum line to line voltage. The
term effective grounding is commonly used in the power industry as a point of
reference for the design of the earthing system and sizing of the neutral
impedance. Therefore a power system can be effectively grounded or
ineffectively grounded. In other words, in a 66kV subtransmission system, thephase to ground voltage during fault condition will be:
-
8/10/2019 Torelli Za Ucinke
20/180
20/138
V Nominal
Vph-ph=66kV
Vph-g = 66kV/!3=38.1 kV
Solidly Grounded System
Vph-g fault= 66/ !3 = 38.1 kV
Effective grounding System
Vph-g fault= 66 * 0.8= 52.8 kV
Ungrounded System
Vph-g fault= 66kV
The main advantage of having an effective grounded system is limiting the
overvoltage magnitude which has a direct impact on the cost of the system. For
instance, the required insulation medium can be reduced with direct benefits to the
cost of the electrical equipment. This advantage is a key factor for the design of
subtransmission and transmission power systems.
The effective grounding system is nowadays considered the best compromise
between reducing the phase to ground fault and keeping the overvoltages at a
reasonable level. Therefore the effective grounding is the preferred solution for the
-
8/10/2019 Torelli Za Ucinke
21/180
21/138
higher voltages in countries where the power system is grounded via an
impedance[4] .
How to achieve an effective grounded system? IEEE Recommended Practice for
Grounding of Industrial and Commercial Power Systems states the required
system conditions:
! The system must be grounded through a sufficient low impedance such
that for all system conditions the ratio of zero sequence reactance to
positive sequence reactance is positive and equal or less than ( X0/X1"3)
and
! the ratio of zero sequence resistance to positive-sequence reactance is
positive and less than 1( R0/X1
-
8/10/2019 Torelli Za Ucinke
22/180
22/138
! 20
1
0"
X
X, which is inherently met in resistance grounding and with delta
star power transformers [12], unless the power system is earthed via a
large reactor
From the system conditions stated above, it can be derived that the size of neutral
earth resistor or neutral earth reactor is not critical by itself for the magnitude of the
overvoltage. The relation with other system parameters determines the magnitude
of the overvoltage. The above conditions also highlights that unplanned increase of
the size of the resistor will increase the value of the overvoltage due to the fault
clearing restrike.
Besides, the system conditions show that the reactance of the overall zero
sequence network shall be inductive and not capacitive.
These system conditions aim to limit transient overvoltages and, at the same time,
reduce the phase to ground fault current between 10 and 25 % of the steady state
fault current [9]. The lower level is dictated by the minimum current requirement for
relay operation. Nowadays with modern digital relays, this requirement could be
reconsidered. The 25 % upper level takes into consideration the resistor cost in
comparison to the reactive option.
Plots of power frequency overvoltage following faults as a function of different
system parameters or system parameters or ratios of system parameters have
been published and are available in the existing literature[9].
Existing studies on overvoltage and impact of the neutral earth resistor on the
measured overvoltages are based on field testing, software simulation and
theoretical analysis.
First researches on transient overvoltages during faults were conducted before the
Second World War. Overvoltages were analytically determined and tested using
-
8/10/2019 Torelli Za Ucinke
23/180
23/138
miniature models[9] . Maximum peak voltages were recorded mainly between 3
and 4 pu with 20 % above 4 pu. The researchers also considered current choppingproduced by the circuit breakers and restrike phenomena known as transient
recovery voltage, usually abbreviated as TRV. Overvoltage calculations and tests
using miniature system have also been conducted by other researchers[6, 8].
These papers provide a complete explanation of the maximum overvoltage in
relation to system parameters and formed the basis for future IEEE standards.
Measurements of overvoltages following a fault were performed in the distribution
network in the 10kV Croatian distribution network[13]. The system was earthed
using the NER. Maximum overvoltage factors are below 2 pu and minimum
variations are recorded for different values of the resistor. A transient software
program was used to validate the results.
It is possible to determine the maximum transient overvoltage by applying the RLC
circuit analysis and the use of symmetrical components. [14]. Generalized plots
which took into consideration the damping effect of the resistor were also
produced.
An overall indication of the effect of the resistor is defined by:
020 XR " circuit is oscillatory
020 XR $ critical damped
020 XR # over damped
Recent studies usually apply the traveling waves theory. In line with this theory,
modern transient software packages also avoid the use of lumped elements [15].
Lumped elements representation is adequate for steady state analysis but give
inaccurate results in transient studies. In the latter studies, where travel time of the
electromagnetic waves and the energy exchange between capacitance and
inductance must be taken into consideration, traveling waves theory is the most
-
8/10/2019 Torelli Za Ucinke
24/180
24/138
accurate method for the computation of the overvoltages along the transmission
line.
Fault generated traveling waves propagates not only on the overhead line to line.
but also line to earth with different propagation parameters [16]. A phase to ground
fault can be represented as disturbance that generates current and voltage
traveling waves on a subtransmission line. At each point of discontinuity the
traveling waves will be refracted and reflected according to the surge impedance
characteristic of the line [17].It is possible to represent the large number of forward
and backwards waves and determine the maximum overvoltage generated as a
function of time by using the Bewley Lattice diagram. However, there are also
some other considerations which must be dealt with which include mutual coupling
with other conductors and wave shape distortion experienced along the
transmission line.
It is important to point out that overvoltage studies and traveling wave application
mainly focus on switching and lightning transients [16]. However, overvoltages
produced by switching episode can be used as a reference for this type of study.
To conclude, what overvoltage magnitude can be developed under a ground fault?
From the existing literature it can be inferred generally that the maximum
overvoltages following a phase to ground fault are below 3 pu [18-20]. Maximum
overvoltage can instead reach 4.1 pu with trapped charges on the line as in the
case of switching a capacitor bank connected on the line [19].
-
8/10/2019 Torelli Za Ucinke
25/180
25/138
2.4 Design of the Experiment
2.4.1 The Software Package
Usually, software packages are used to facilitate transient overvoltage studies. In
fact, these investigations involve complex and time consuming computation that
can be achieved with sufficient accuracy with software packages.
PSCAD-EMTDC is a state of the art software package developed using the wave
propagation theory first published by DR Hernmann W Dommel in the transaction
of IEEE Power Apparatus and Systems in 1969. based on this paper another two
transient analysis software packages well known in the industry, EMTP and, later,
the ATP software package [21] were developed.
PSCAD is a very powerful and flexible graphical interface that uses the EMTDC
simulation program. The first version of PSCAD was produced by Denis Woodford
to model and simulate the high voltage DC system in Canada. The software is now
developed and maintained by the Manitoba HVDSC Research Centre and is
continually upgraded to incorporate the latest validated research.
The software allows one to model an electrical circuit, run a simulation, analyze the
results and manage the data information in an efficient way. The software package
is provided with an equipment library, but also allows the creation of new models
and building a personal equipment library. It is used by consulting engineers,equipment manufacturers and laboratories for planning and design of power
systems.
Typical projects involve transient stability, dynamic stability, relay coordination,
transformer saturation, insulation coordination studies, HVDC studies, harmonic
studies, power electronics studies and optimisation of controller parameters.
Further information is located on the website http://www.pscad.com
http://www.pscad.com/ -
8/10/2019 Torelli Za Ucinke
26/180
-
8/10/2019 Torelli Za Ucinke
27/180
27/138
Steady state overvoltage for a given value of the neutral earth resistor was
calculated by applying symmetrical components theory. Factor K2 and K0 wereused to determine fault currents and steady state overvoltages[4].
Z2/Z1K2 $
Z0/Z1K0 $
Where:
Z1= Positive system impedance
Z2= Negative system impedance
Z0= Zero system impedance
The factor K0 gives us a prompt understanding of the power frequency overvoltage
on the healthy phases during a ground fault.
The factor K2 is assumed to be 1. If the fault occurs close to the generator the
factor K2 may increase to 1.4 [4].
Knowing the steady state overvoltage and the total overvoltage, the transient
component can be calculated with:
Natural Frequency V = Measured Overvoltage - Power Frequency V
2.4.3 The experimental design
The most relevant factors that influence the overvoltage magnitude during phase to
ground faults are as follow:
-
8/10/2019 Torelli Za Ucinke
28/180
28/138
! Method of earthing
! Type of fault
! System configuration
! Number of earthing points and location in respect to the fault location
! Electrical parameters of the equipment
! Size of the neutral earth resistor
! Location of the fault
! Time of the fault applied
! Type of subtransmission network as underground cables and overhead
lines have different electrical characteristics
! Type of overhead line
! Length of the line
! Fault impedance
Each variable was analyzed in order to determine the most critical and relevant
factors to be used in this research. In the end, the study was based on thefollowing five factors:
! System configuration
! Type of faults
! Size of the neutral earth resistor
-
8/10/2019 Torelli Za Ucinke
29/180
29/138
! Location of the fault
! Time of the fault
2.4.4 System Configuration
Four system configurations were selected. These configurations were considered
the most likely configurations to produce the highest overvoltages during a ground
fault [5, 14]:
66kV Subtransmission Line
T
#1 #2RL
Resis
+
Figure 2.1: Overhead line with no load at the end of the line model
66kV Subtransmission Line
T
#1 #2RL
Resis
+30 [MVAR]
Figure 2.2: Overhead line with capacitor bank at the end of the line
-
8/10/2019 Torelli Za Ucinke
30/180
-
8/10/2019 Torelli Za Ucinke
31/180
31/138
Faults were simulated at the beginning and at the end of line.
Time of fault
Considering the symmetry of the alternating voltage cycle, the samples were
selected only on one half cycle. As one semi cycle at 50Hz takes 10 ms, 21 sample
spaced 0.5 ms apart were selected.
Line length
Line lengths were set to 100, 10 and 0.2 km.
These line lengths are based on the minimum, maximum and typical line length in
the Victorian Network. Typical subtransmission lines vary between 10-20 km in the
urban area around Melbourne and increase to 30-40 km in the country area. In
Victoria the maximum line length is about 80 Km. The line length of 100 km reflects
the need to investigate the system in the worst case scenario.
2.4.5 Simulation Characteristics
In this study no randomization factor is applied as the chance of biasing the results
in a software simulation is considered negligible. Therefore, simulation studies
were conducted in a logical order to maximize the efficiency of the work.
The simulations were conducted using PSCAD multiple run function. This software
tool allows completion of complex studies in an efficient manner by manipulating
variables sequentially and automatically. PSCAD Multiple Run tool can run and
modulate up to six variables.
-
8/10/2019 Torelli Za Ucinke
32/180
32/138
Maximum instantaneous voltage peak was recorded for each run and on each
phase. Only the maximum overvoltage among the three phases was selected forthe maximum overvoltage record.
Duration of the run time was set to 0.6 s with fault start at 0.4 s plus selected time
delay. The end time of the fault was 0.1 s with clearing fault at the first zero
crossing intersection of the fault current.
According to the Nyquist_Shannon Sampling Theorem, to avoid aliasing and
incorrect representation of the selected variable, sampling rate should be at least
double of the variable frequency to be represented. PSCAD recommends that 50s
for the time step of the process and computation would be sufficient. However,
time step resolution was reduced to 20 s to capture transient components up to
25 kHz [22].
Sampling rate=20 s
Frequency rate= 1/T=106/20= 50kHz
2.4.6 Model Components
Source
The source represents the power system which is upstream the 220/66 kV power
transformer in the substation. The voltage source selected from the PSCAD library
enabled to specify the positive, negative and zero-sequence impedance.
The impedance value is selected to produce a three fault current of 30 kA on the
220 kV bus at the substation. Positive impedance is equal to negative impedance.
Zero impedance was set to be equal to the positive sequence impedance to allow
for a phase to ground fault level equal to the three phase fault level.
The X/R ratio was set to 10.
-
8/10/2019 Torelli Za Ucinke
33/180
33/138
The control system for the output voltage was set using the Fixed Controlfunction
offered by PSCAD software package. Through several trial and error experimentsthis type of control produced a model behavior very similar to an infinite bus. This
setting enhances the robustness of the source model in respect to disturbances on
a lower voltage level. It is assumed that the impact of the overvoltage on the 66 kV
radial line is limited on the higher transmission network.
In summary:
Source Type = Three Phase Voltage-Source Model 3
MVA = 100 MVA
Vph- Vph = 220 kV
Frequency = 50 Hz
Time constant = 0.06 s
Z1=Z2=Z0= 4.23 Ohm
Z1 angle= 84.289 deg
Voltage Fixed Control
Es = 220 kV
Ph = 0.0 deg
Frequency = 50 Hz
Transformer
The two winding 220/66 kV power transformer connects the transmission system to
the subtransmission system.
-
8/10/2019 Torelli Za Ucinke
34/180
34/138
Often substation have more than one transformer operating in parallel with the bus
tie on the 220 kV closed and on the 66 kV bus open or closed according to systemrequirement, the fault level in the system and the rating of the equipment. To
simplify the study the model represents only one transformer.
Winding connection of the transformer is delta on the HV side and star on the LV
side. Tap changer operation is disabled
The transformer is represented in PSCAD according to the classical approach of
the transformer theory. Magnetising current is represented as well as core
saturation placed on the HV side. In PSCAD the HV side is the closest winding to
the core of the transformer.
Hysteresis losses in the transformer core are included. Copper losses are disabled
to reduce the number of nodes in the system. Considering that a typical X/R for a
power transformer of 100 MVA and 220/66 kV is above 30, this simplification was
considered acceptable.
TRANSFORMER DATA:
3 Phase
2 Winding Transformer
MVA Rating: 100.0 MVA
Freq= 50 Hz,
V ph-ph primary= 220 kV
Vph-ph secondary= 66kV
No Load Losses= 0.01 pu
-
8/10/2019 Torelli Za Ucinke
35/180
35/138
Copper Losses=0.0 pu.
Positive Sequence Reactance= 0.15 pu.
Saturation enabled simulated on primary winding
Air core reactance= 0.3 pu
In Rush Decay Time Constant= 0.03 pu.
Knee Point Voltage: 1.25 pu
Time to release Flux Clipping=0.1s
Magnetising Current= 0.02 pu.
Overhead line
In electromagnetic simulation studies there are two main methods to representtransmission lines[21]:.
! Simplified studies generally rely on the PI Section model
! Distributed transmission line, which is most suited for transient line response
modeling using a digital computer. This method was used for this study
PSCAD operates applying the traveling wave theory but also implementing mutual
coupling between conductors and wave shape attenuation produced by the
transmission line. Transmission lines under PSCAD transient software packages
are modeled using one of the three traveling wave models:
! Bergeron Model
! Frequency Dependent Mode Model
-
8/10/2019 Torelli Za Ucinke
36/180
36/138
! Frequency Dependent Phase Model
The Frequency Dependent Phase Model was selected for this study as is the most
accurate model. This model is a distributed RLC traveling wave model, which
incorporates the frequency dependence of all parameters in the model. This model
is recommended by the PSCAD Users Guide for this type of study.
The overhead line parameters are based on a typical 66 kV line as represented in
Victoria.
Conductor Data
400 mm2 aluminum
Conductor Resistance= 0.071 Ohm/km
Conductor Geometric Mean Radius 0.00982 m
Capacitor bank
The capacitor bank is implemented in Simulation 3 and 4. The capacitor bank is
connected at the end of the line to produce maximum peak value of the system
response during the transient period. The capacitance value was set under the
assumption that the load on the overhead line is 60 MVA with a power factor of
0.707.
Load Current = S/ ( !3 * V ) = 60 *106 / !3 * 66000 )= 524.8 A
P = S * PF = 52.42 MW
If PF is 0.707
Power angle = 45 degrees
-
8/10/2019 Torelli Za Ucinke
37/180
-
8/10/2019 Torelli Za Ucinke
38/180
-
8/10/2019 Torelli Za Ucinke
39/180
39/138
Phase to ground and phase to phase to ground the fault were simulated. Current
interruption was simulated at 0 current crossing to avoid any overvoltage related tomagnetic field being trapped in the circuit. This type of overvoltage is usually
referred as current chopping[3].
Fault resistance is set to zero to evaluate the impact of the neutral earth resistor by
itself.
In summary:
Fault Type = Phase to ground and phase to phase to ground
Fault start =0.4 s to fault 0.410 s
Fault length =0.1 s
Fault ON resistance = 0.01 ohm
Fault OFF resistance = 106ohm
Current Chopping limit= 0 kA
-
8/10/2019 Torelli Za Ucinke
40/180
40/138
2.5 Simulation Results
Simulation results are organized according to the six system configuration and
related simulations.
Simulation results data is attached in Appendix B.
Each simulation result Section contains two graphs. The first plot summarizes the
main objective of this investigation which is the maximum overvoltage as a function
of the reduced fault current. The fault current was calculated by applying the
symmetrical components theory as illustrated in Appendix B.
The second plot highlights the specific weight of the transitory overvoltage
component in respect to the steady state overvoltage component. Power frequency
overvoltage was subtracted from the measured overvoltage only if the recorded
maximum overvoltage took place during the fault and not at fault clearing. After
fault clearing, the power frequency overvoltage can be considered absent as the
three phases are now connected to a symmetrical voltage source.
Additional investigations were also performed on the following topics:
! Overvoltages and traveling waves
! Influence of the fault time
! Overvoltage response on healthy and faulty phases
! Frequency spectrum analysis
-
8/10/2019 Torelli Za Ucinke
41/180
41/138
2.5.1 Simulation 1 : Unloaded Line - Fault at the End of the Line
TABLE 2-1SIMULATION 1 RESULTS
Line Length= 100km
NER(Ohm)
MaxMeasuredV from
PSCAD(kV)
Iph-g withNER/ Iph-g
solidlyearthed K0
V Meas /
NominalVoltage
Power
FrequencyV (pu)
Natural
Frequency V(pu)
0 126.216 1.000 1.0 2.342 1.002 2.3425 124.898 0.977 1.1 2.318 1.103 2.31810 126.607 0.946 1.3 2.349 1.204 2.34915 128.809 0.909 1.5 2.390 1.296 2.39020 128.021 0.867 1.8 2.376 1.378 2.37625 129.201 0.824 2.0 2.398 1.448 2.398
30 131.007 0.782 2.3 2.431 1.507 2.43135 131.669 0.740 2.6 2.443 1.555 2.44340 132.052 0.700 2.9 2.450 1.596 2.45045 132.457 0.663 3.2 2.458 1.629 2.45850 132.599 0.628 3.6 2.461 1.656 2.461100 103.588 0.394 6.8 1.922 1.763 1.922550 112.552 0.082 36.0 2.089 1.757 0.3321000 122.539 0.045 65.4 2.274 1.747 0.5271000000 140.632 0.000 65217.4 2.610 1.736 0.873
Line Length=10km0 90.447 1.000 1.0 1.678 1.006 1.6785 99.668 0.882 1.7 1.850 1.412 1.850
10 100.711 0.705 3.0 1.869 1.647 1.86915 99.279 0.561 4.3 1.842 1.745 1.84220 99.031 0.457 5.7 1.838 1.783 0.05425 96.724 0.383 7.1 1.795 1.797 -0.00330 96.071 0.328 8.5 1.783 1.802 -0.01935 95.465 0.286 9.8 1.772 1.802 -0.03040 95.437 0.253 11.2 1.771 1.800 -0.02945 95.512 0.227 12.6 1.772 1.797 -0.02550 95.526 0.206 14.0 1.773 1.794 -0.021100 94.901 0.105 27.9 1.761 1.771 -0.010
-
8/10/2019 Torelli Za Ucinke
42/180
42/138
NER(Ohm)
Max
MeasuredV fromPSCAD(kV)
Iph-g withNER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequency V(pu)
550 101.180 0.019 153.0 1.878 1.740 0.1371000 107.676 0.011 278.1 1.998 1.736 0.2621000000 120.050 0.000 278041.7 2.228 1.736 0.491
Line Length=0.2km0 76.644 1.000 0.9 1.422 1.011 1.4225 86.617 0.806 2.3 1.607 1.358 1.607
10 92.223 0.564 4.4 1.711 1.455 0.25715 95.204 0.415 6.5 1.767 1.478 0.28820 96.245 0.324 8.6 1.786 1.822 -0.03625 96.587 0.264 10.8 1.792 1.817 -0.02530 96.206 0.223 12.9 1.785 1.810 -0.02535 96.501 0.192 15.1 1.791 1.804 -0.013
40 96.337 0.169 17.2 1.788 1.798 -0.01045 96.163 0.151 19.3 1.784 1.792 -0.00850 95.991 0.136 21.5 1.781 1.788 -0.007100 94.486 0.068 42.9 1.753 1.763 -0.010
550 96.337 0.012 236.1 1.788 1.738 0.0501000 102.533 0.007 429.2 1.903 1.735 0.167
1000000 118.593 0.000 429200.4 2.201 1.736 0.464
Configuration 1- Fault at the of the line- No load- Measured Overvoltage
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu)
L=100km
L = 10km
L= 0.2km
Figure 2.5: Configuration 1 Measured overvoltage result
-
8/10/2019 Torelli Za Ucinke
43/180
-
8/10/2019 Torelli Za Ucinke
44/180
44/138
explained using the factor K0=Z0/Z1. The implementation of the resistor will
increase Z0. As the neutral impedance has a bigger impact on the factor K0, in theshortest line the expected steady state overvoltage has a bigger increase
The results show that the maximum overvoltages recorded in the simulation with a
line length of 10km are similar to the results obtained from the simulation with a
line length of 0.2km. This characteristic is recorded throughout the study. We can
explain this characteristic by referring to the overvoltage plots as function of system
parameters of IEEE standards[18]. The different line length does not have a
particular impact on the Xco/X1. For both line lengths, the ratio is larger than 100 .
Beyond this value the expected overvoltage is considered constant.[6].
Xsource= 0.0087 pu
X Tx= 0.15 pu
X line=0.0088 Ohm/km
C= 9.4 nF/Km
Line length 10km
Xc = 1/(2f C )/ 10 = 1/( 2* 50 * 9.4 10-9)/ 10 = 33879 Ohm
X1= Xsource+ Xtx + X line = 0.0087 + 0.15 + 0.088 ( 10km) = 0.2467 pu
Z base = 484 Ohm
Xc in pu = Xc value/ Z base= 33879/484= 70 pu
Ratio Xco/X1= 70/ 0.2467= 283.7
Line length 0.2km
-
8/10/2019 Torelli Za Ucinke
45/180
45/138
Xc = 1/(2f C )/0.2 = 1/( 2* 50 * 9.4 10-9)/ 0.2 = 1693996 Ohm
X1= Xsource+ Xtx + X line( 0.2km) = 0.0087 + 0.15 + 0.0018 = 0.1605 pu
Z base = 484 Ohm
Xc in pu = Xc value/ Z base= 636/484= 3500 pu
Ratio Xco/X1= 3500/ 0.1605= 21806
Line length 100 km
Xc = 1/(2f C )/0.2 = 1/( 2* 50 * 9.4 10-9)/ 100 = 3387.9 Ohm
X1= Xsource+ Xtx + X line( 100km) = 0.0087 + 0.15 + 0.88 = 0.1605 pu
Z base = 484 Ohm
Xc in pu = Xc value/ Z base= 3387.9/484= 7 pu
Ratio Xco/X1= 7/ 0.1605= 43.6
The Xco/X1 ratio for line length of 100km is also proved to be below 100.
Therefore, the line will produce a higher overvoltage.
-
8/10/2019 Torelli Za Ucinke
46/180
46/138
Configuration 1- Fault at the of the line- No load- Transient Overvoltage Contribution
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu)
L=100km
L = 10km
L= 0.2km
Figure 2.6: Configuration 1 Transient overvoltage component
Figure 2.6 shows the contribution of the transient overvoltage to the maximum
recorded overvoltage.
-
8/10/2019 Torelli Za Ucinke
47/180
47/138
2.5.2 Simulation 2 : Unloaded Line - Fault at the Beginning of the Line
TABLE 2-2SIMULATION 2 RESULTS
Line Length= 100km
NER(Ohm)
MaxMeasuredV from
PSCAD(kV)
Iph-g withNER/ Iph-g
solidlyearthed K0
V Meas /
NominalVoltage
Power
FrequencyV (pu)
Natural
FrequencyV (pu)
0 72.597 1.000 0.9 1.347 1.011 1.3475 94.147 0.804 2.4 1.747 1.361 1.747
10 99.371 0.561 4.4 1.844 1.456 1.84415 101.219 0.412 6.6 1.878 1.479 1.87820 102.529 0.321 8.7 1.903 1.488 1.90325 103.542 0.262 10.9 1.921 1.817 0.10430 104.549 0.220 13.1 1.940 1.810 0.13035 106.192 0.190 15.2 1.971 1.803 0.167
40 107.291 0.167 17.4 1.991 1.797 0.194
45 108.035 0.149 19.6 2.005 1.792 0.21350 108.486 0.134 21.7 2.013 1.788 0.226100 111.955 0.068 43.4 2.078 1.763 0.315550 110.022 0.012 238.7 2.042 1.738 0.3041000 111.633 0.007 434.0 2.072 1.735 0.3361000000 111.941 0.000 433986.8 2.077 1.736 0.341
Line Length=10km0 72.781 1.000 0.9 1.351 1.011 1.3515 91.485 0.804 2.4 1.698 1.361 1.698
10 97.134 0.561 4.4 1.802 1.456 1.802
15 97.917 0.412 6.6 1.817 1.479 0.33820 97.710 0.321 8.7 1.813 1.488 0.32525 97.291 0.262 10.9 1.805 1.817 -0.01230 96.880 0.220 13.1 1.798 1.810 -0.01235 96.520 0.190 15.2 1.791 1.803 -0.012
40 96.212 0.167 17.4 1.785 1.797 -0.01245 95.950 0.149 19.6 1.781 1.792 -0.01250 95.727 0.134 21.7 1.776 1.788 -0.011100 94.570 0.068 43.4 1.755 1.763 -0.008
-
8/10/2019 Torelli Za Ucinke
48/180
48/138
NER(Ohm)
Max
MeasuredV fromPSCAD(kV)
Iph-g withNER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequencyV (pu)
550 108.220 0.012 238.7 2.008 1.738 0.2701000 112.946 0.007 434.0 2.096 1.735 0.3611000000 119.533 0.000 433986.8 2.218 1.736 0.482
Line Length=0.2km0 67.090 1.000 0.9 1.245 1.011 1.2455 91.473 0.804 2.4 1.697 1.361 1.697
10 97.117 0.561 4.4 1.802 1.456 0.34615 97.898 0.412 6.6 1.817 1.479 0.33820 97.681 0.321 8.7 1.813 1.488 0.32525 97.272 0.262 10.9 1.805 1.817 -0.01230 96.861 0.220 13.1 1.797 1.810 -0.01335 96.501 0.190 15.2 1.791 1.803 -0.01340 96.193 0.167 17.4 1.785 1.797 -0.01245 95.931 0.149 19.6 1.780 1.792 -0.01250 95.708 0.134 21.7 1.776 1.788 -0.012100 94.570 0.068 43.4 1.755 1.763 -0.008550 108.314 0.012 238.7 2.010 1.738 0.272
1000 112.946 0.007 434.0 2.096 1.735 0.3611000000 113.404 0.000 433986.8 2.104 1.736 0.368
Configuration 2- Fault at the bus- No load- Measured Ov ervoltage
0.0
0.5
1.0
1.5
2.0
2.5
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(p
u)
L=100km
L = 10kmL= 0.2km
Figure 2.7- Configuration 2 Measured overvoltage result
-
8/10/2019 Torelli Za Ucinke
49/180
-
8/10/2019 Torelli Za Ucinke
50/180
50/138
2.5.3 Simulation 3 : Capacitor Bank. Fault at the End of the Line
TABLE 2-3SIMULATION 3 RESULTS
Line Length= 100km
NER(Ohm)
MaxMeasuredV fromPSCAD(kV)
Iph-g withNER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequencyV (pu)
0 187.403 1.000 1.0 3.478 1.002 2.4765 185.285 0.977 1.1 3.438 1.102 2.33610 184.351 0.946 1.3 3.421 1.204 2.21715 184.572 0.909 1.5 3.425 1.296 2.12920 185.656 0.867 1.8 3.445 1.378 2.06725 187.384 0.824 2.0 3.477 1.448 2.02930 189.670 0.782 2.3 3.520 1.507 2.01335 192.279 0.740 2.6 3.568 1.555 2.01340 195.097 0.700 2.9 3.620 1.596 2.025
45 199.190 0.663 3.2 3.696 1.629 2.06850 203.527 0.628 3.6 3.777 1.656 2.121100 245.786 0.394 6.8 4.561 1.763 2.798550 406.039 0.082 36.0 7.535 1.757 5.7781000 453.542 0.045 65.4 8.416 1.747 6.6691000000 526.382 0.000 65217.4 9.768 1.732 8.036
Line Length=10km0 110.323 1.000 1.0 2.047 1.006 1.0415 123.489 0.882 1.7 2.292 1.412 0.87910 127.826 0.705 3.0 2.372 1.647 0.72515 128.511 0.561 4.3 2.385 1.745 0.64020 124.144 0.457 5.7 2.304 1.783 0.52125 154.689 0.383 7.1 2.871 1.797 1.07330 160.269 0.328 8.5 2.974 1.802 1.17235 164.680 0.286 9.8 3.056 1.802 1.25440 168.268 0.253 11.2 3.123 1.800 1.32345 171.273 0.227 12.6 3.178 1.797 1.38150 173.731 0.206 14.0 3.224 1.794 1.430100 186.406 0.105 27.9 3.459 1.771 1.688
-
8/10/2019 Torelli Za Ucinke
51/180
51/138
NER(Ohm)
Max
MeasuredV fromPSCAD(kV)
Iph-g withNER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequencyV (pu)
550 198.797 0.019 153.0 3.689 1.740 1.9491000 200.148 0.011 278.1 3.714 1.736 1.9781000000 182.224 0.000 278041.7 3.381 1.736 1.645
Line Length=0.2km0 109.095 1.000 0.9 2.024 1.011 1.0145 135.208 0.806 2.3 2.509 1.358 1.151
10 135.715 0.564 4.4 2.518 1.455 1.06415 127.208 0.415 6.5 2.361 1.478 0.88220 148.710 0.324 8.6 2.760 1.822 0.93725 152.521 0.264 10.8 2.830 1.817 1.01330 162.300 0.223 12.9 3.012 1.810 1.20235 166.755 0.192 15.1 3.094 1.804 1.29140 170.288 0.169 17.2 3.160 1.798 1.36245 173.169 0.151 19.3 3.213 1.792 1.42150 175.588 0.136 21.5 3.258 1.788 1.471100 188.724 0.068 42.9 3.502 1.763 1.739550 200.208 0.012 236.1 3.715 1.738 1.977
1000 201.435 0.007 429.2 3.738 1.735 2.0031000000 203.044 0.000 429200.4 3.768 1.736 2.031
Configuration 3- Fault at the end of the line- Cap Bank connected at the end of the line- Measured
Overvoltage
0.0
2.0
4.0
6.0
8.0
10.0
12.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu)
L=100km
L = 10km
L= 0.2km
Figure 2.9: Configuration 3 Measured overvoltage result
-
8/10/2019 Torelli Za Ucinke
52/180
52/138
Configuration 3- Fault at the end of the line- Cap Bank connected at the end of the line-
Transient Overvoltage Contribution
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
8.0
9.0
0.00.10.20.30.40.50.60.70.80.91.0Ifault/I ph-g so lidly earthed (pu)
Overvoltage(pu)
L=100km
L = 10km
L= 0.2km
Figure 2.10: Configuration 3 Transient overvoltage component
The dangerous effects of capacitor switching are well known in the engineering
field [23]. The results obtained with this simulation show that, following a fault in the
network with no or small loads and a capacitor bank connected to the grid
represent a danger to the insulation of the power system.
The overvoltages measured in this simulation are the highest of the entire study
reported here. The results are in line with previous studies [5, 14] except for
results obtained with line length of 100 km . This is due to the combination of the
following system parameters:
!
Length of the line
! NER size
These parameters affect the3
00 XC
R " transient overvoltages condition. If this
condition is not met, dangerous fault restrikes are generated [9]. Is it worth noting
that field experience of arcing ground fault above 7 pu are quite rare as the
insulation of the power system usually breaks down between 6 to 7 pu [9]
-
8/10/2019 Torelli Za Ucinke
53/180
53/138
Line length impact is shown in Figure 2.9 and 2.10. Instead, to review the impact of
the NER size further simulations were performed. Four snapshots with a NER of10, 100, 1000 and 1000000 Ohm were taken as shown in Figures 2.11 to 2.14.
-80
-60
-40
-20
0
20
40
60
80
100
Voltage
(kV)
Vs
Figure 2.11: Configuration 3 Transient overvoltage Snapshot with NER of 10 Ohms
-300
-200
-100
0
100
200
300
VV
-
8/10/2019 Torelli Za Ucinke
54/180
54/138
Figure 2.14: Configuration 3 Transient overvoltage Snapshot with NER of 1000000 Ohms
The graphs show that the increase of the resistor will increase the transient
overvoltages.
The dangerous transient overvoltages are also due to the conservative approach in
selecting the size of the capacitor bank. A further simulation is conducted to
validate this theory. A smaller capacitor bank of 15 MVA was used for the analysis.
Configuration 3- Line Length 100km- Comparison between 30MVAR and 15MVAR Capacitor BankResults- Measured Overvoltage
0.0
2.0
4.0
6.0
8.0
10.0
12.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu)
15MVAR
30MVAR
Figure 2.15: Configuration 3 30 and 15 MVAR Cap bank overvoltage comparison
-
8/10/2019 Torelli Za Ucinke
55/180
55/138
The system response with a smaller capacitor gives a result which is acceptable
and in line with previous studies [5, 19]. Maximum overvoltage with the insulatedsystem is 4.1 pu.
-
8/10/2019 Torelli Za Ucinke
56/180
56/138
2.5.4 Simulation 4 : Capacitor Bank. Fault at the Beginning of the Line
TABLE 2-4SIMULATION 4 RESULTS
Line Length= 100km
NER
(Ohm)
MaxMeasuredV fromPSCAD
(kV)
Iph-g withNER/ Iph-gsolidly
earthed K0
V Meas /Nominal
Voltage
PowerFrequency
V (pu)
NaturalFrequency
V (pu)0 73.660 1.000 0.9 1.367 1.011 1.3675 102.229 0.804 2.4 1.897 1.361 1.89710 110.677 0.561 4.4 2.054 1.456 2.05415 120.455 0.412 6.6 2.235 1.479 2.23520 139.620 0.321 8.7 2.591 1.488 2.59125 156.662 0.262 10.9 2.907 1.817 2.90730 171.647 0.220 13.1 3.185 1.810 3.18535 185.697 0.190 15.2 3.446 1.803 3.44640 198.103 0.167 17.4 3.676 1.797 3.67645 209.035 0.149 19.6 3.879 1.792 3.879
50 219.140 0.134 21.7 4.067 1.788 4.067100 282.383 0.068 43.4 5.240 1.763 5.240550 373.638 0.012 238.7 6.934 1.738 6.9341000 431.569 0.007 434.0 8.009 1.735 8.0091000000 553.568 0.000 433986.8 10.272 1.736 10.272
Line Length=10km0 73.885 1.000 0.9 1.371 1.011 0.3605 104.589 0.804 2.4 1.941 1.361 0.58010 115.953 0.561 4.4 2.152 1.456 0.69615 126.137 0.412 6.6 2.341 1.479 0.86220 129.761 0.321 8.7 2.408 1.488 0.92025 140.232 0.262 10.9 2.602 1.817 0.78530 143.196 0.220 13.1 2.657 1.810 0.84735 147.174 0.190 15.2 2.731 1.803 0.92840 151.560 0.167 17.4 2.812 1.797 1.01545 155.590 0.149 19.6 2.887 1.792 1.09550 158.744 0.134 21.7 2.946 1.788 1.158100 173.853 0.068 43.4 3.226 1.763 1.463550 185.633 0.012 238.7 3.445 1.738 1.707
-
8/10/2019 Torelli Za Ucinke
57/180
-
8/10/2019 Torelli Za Ucinke
58/180
-
8/10/2019 Torelli Za Ucinke
59/180
59/138
2.5.5 Simulation 5 : Light inductive Load. Fault at the End of the Line.
TABLE 2-5SIMULATION 5 RESULTS
Line Length= 100km
NER(Ohm)
MaxMeasured
V fromPSCAD(kV)
Iph-g with
NER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequency V (pu)
0 123.996 1.000 1.0 2.301 1.002 2.3015 121.497 0.977 1.1 2.255 1.102 2.25510 127.242 0.946 1.3 2.361 1.204 2.36115 124.414 0.909 1.5 2.309 1.296 2.30920 125.584 0.867 1.8 2.330 1.378 2.33025 126.472 0.824 2.0 2.347 1.448 2.34730 127.242 0.782 2.3 2.361 1.507 2.36135 127.753 0.740 2.6 2.371 1.555 2.371
40 128.210 0.700 2.9 2.379 1.596 2.37945 128.382 0.663 3.2 2.382 1.629 2.38250 128.563 0.628 3.6 2.386 1.656 2.386100 125.448 0.394 6.8 2.328 1.763 2.328550 106.256 0.082 36.0 1.972 1.757 0.2151000 114.514 0.045 65.4 2.125 1.747 0.378
1000000 131.252 0.000 65217.4 2.436 1.732 0.704
Line Length=10km0 89.031 1.000 1.0 1.652 1.006 1.6525 98.999 0.882 1.7 1.837 1.412 1.837
10 99.542 0.705 3.0 1.847 1.647 1.84715 98.172 0.561 4.3 1.822 1.745 1.82220 97.019 0.457 5.7 1.800 1.783 0.01725 95.755 0.383 7.1 1.777 1.797 -0.02130 95.531 0.328 8.5 1.773 1.802 -0.02935 95.024 0.286 9.8 1.763 1.802 -0.03840 94.367 0.253 11.2 1.751 1.800 -0.04945 94.358 0.227 12.6 1.751 1.797 -0.04650 94.180 0.206 14.0 1.748 1.794 -0.046100 93.855 0.105 27.9 1.742 1.771 -0.029
-
8/10/2019 Torelli Za Ucinke
60/180
60/138
550 99.236 0.019 153.0 1.841 1.740 0.101
NER(Ohm)
MaxMeasuredV fromPSCAD(kV)
Iph-g withNER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequency V (pu)
1000 109.857 0.011 278.1 2.039 1.736 0.3021000000 241.638 0.000 278041.7 4.484 1.736 2.748
Line Length=0.2km0 74.717 1.000 0.9 1.387 1.011 1.3875 85.939 0.806 2.3 1.595 1.358 1.595
10 91.418 0.564 4.4 1.696 1.455 0.24215 94.402 0.415 6.5 1.752 1.478 0.27320 95.404 0.324 8.6 1.770 1.822 -0.05225 95.743 0.264 10.8 1.777 1.817 -0.04030 95.531 0.223 12.9 1.773 1.810 -0.03735 95.640 0.192 15.1 1.775 1.804 -0.02940 95.477 0.169 17.2 1.772 1.798 -0.02645 95.301 0.151 19.3 1.768 1.792 -0.02450 95.131 0.136 21.5 1.765 1.788 -0.022100 94.017 0.068 42.9 1.745 1.763 -0.019550 97.021 0.012 236.1 1.800 1.738 0.062
1000 107.684 0.007 429.2 1.998 1.735 0.2631000000 126.718 0.000 429200.4 2.351 1.736 0.615
Configuration 5- Fault at the of end the line- Inductive Load connected at the end of the
line- Measured Overvoltage
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu) L=100km
L = 10km
L= 0.2km
Figure 2.18: Configuration 1 Measured overvoltage result
-
8/10/2019 Torelli Za Ucinke
61/180
61/138
Configuration 5- Fault at the end of the line- Inductive Load connected at the end of the line-
Transient Overvoltage Contribution
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
3.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu)
L=100km
L = 10km
L= 0.2km
Figure 2.19: Configuration 5 Transient overvoltage component
Maximum overvoltages recorded are similar to the results from Simulation 1. It is
likely that this inductive load was not large enough to produce any relevant impact
on the overvoltages. Therefore, not further analysis was conducted using the
results of this system configuration.
-
8/10/2019 Torelli Za Ucinke
62/180
62/138
2.5.6 Simulation 6 : Energized 66/22kV Transformer Fault at the 66kV
Bus.
TABLE 2-6SIMULATION 61 RESULTS
NER(Ohm)
MaxMeasuredV fromPSCAD(kV)
Iph-g withNER/ Iph-gsolidlyearthed K0
V Meas /NominalVoltage
PowerFrequencyV (pu)
NaturalFrequency V (pu)
0 54.929 1.000 0.9 1.017 1.011 0.0065 91.224 0.804 2.4 1.692 1.361 0.33110 96.846 0.561 4.4 1.792 1.456 0.33615 97.661 0.412 6.6 1.812 1.479 0.33320 97.464 0.321 8.7 1.806 1.488 0.31825 97.053 0.262 10.9 1.800 1.817 -0.01730 96.646 0.220 13.1 1.793 1.810 -0.01735 96.287 0.190 15.2 1.788 1.804 -0.01640 95.982 0.167 17.4 1.786 1.798 -0.01245 95.721 0.149 19.6 1.776 1.792 -0.01650 95.499 0.134 21.7 1.772 1.788 -0.016
100 94.343 0.068 43.4 1.763 1.763 0.000550 93.211 0.012 238.7 1.729 1.738 -0.0091000 93.089 0.007 434.0 1.727 1.735 -0.0081000000 92.838 0.000 433986.8 1.723 1.732 -0.009
Configuration 6- Fault at the bus- Tx 30 MVA delta/star at the bus- Measured Overvoltage
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
2.0
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthed (pu)
Overvoltage(pu)
Figure 2.20: Configuration 6 Measured overvoltage result
-
8/10/2019 Torelli Za Ucinke
63/180
63/138
Configuration 6- Fault at the bus- Tx 30 MVA at the bus-
Transient Overvoltage Contribution
-0.1
0.0
0.1
0.1
0.2
0.2
0.3
0.3
0.4
0.4
0.00.10.20.30.40.50.60.70.80.91.0
Ifault/I ph-g solidly earthe d (pu)
Overvoltage(pu)
Figure 2.21: Configuration 6- Transient overvoltage result
Overvoltages recorded in this study are relatively lower than in the other
simulations. Maximum overvoltage reaches a maximum of 1.8pu as phase to
ground fault decreases below 0.6 pu.
Please note that the transient overvoltage component disappears as the phase to
ground fault decreases below 0. 25 pu. Therefore, for high resistor values the
overvoltages are only produced by the power frequency voltage.
No further analysis was conducted using the data from this simulation
2.5.7 Overvoltages and Traveling Waves
This section documents the propagation of the voltage traveling waves along the
transmission line. A series of snapshots are presented to describe the behavior of
the overvoltages using the data from Simulation 1 No load at the end of the line-
and with the line length of 100 km
-
8/10/2019 Torelli Za Ucinke
64/180
64/138
Bus Current,Bus Voltage : Graphs
Time ... 0.3850 0.3900 0.3950 0.4000 0.4050 0.4100 0.4150 0.4200 0.4250 0.4300
-1.50
-1.00
-0.50
0.00
0.50
1.00
1.50
Current(kA)
Is
-100
-80
-60
-40
-20
0
20
40
60
80
Voltage
(kV)
Vs
Figure 2.22: Configuration 1 - Maximum Overvoltage - Phase to Ground Fault NER 0 Ohm
Snapshots were recorded at fault inception and fault clearing time. The graph also
shows that the transient overvoltage at higher frequency is superimposed on a
lower frequency overvoltage. The overvoltage has a period of 0.7 ms which is
equivalent to a frequency of 1400 Hz. The high frequency component decays inapproximately 10ms.
Traveling waves are triggered at the same time on both healthy phases and travel
on the conductors. The steps pattern of the transient overvoltage is the effect of the
traveling waves that reflect back and forward between the two ends of the line.
-
8/10/2019 Torelli Za Ucinke
65/180
65/138
Bus Current,Bus Voltage : Graphs
Time ... 0.5050 0.5075 0.5100 0.5125 0.5150 0.5175 0.5200 0.5225 0.5250 0.5275
-1.50
-1.00
-0.50
0.00
0.50
1.001.50
Current(kA)
Is
-100
-75
-50
-25
0
25
50
75
100
125
Voltage
(kV)
Vs
Figure 2.23: Configuration 1- Snapshot of after clearing overvoltage- Phase to Ground Fault NER
0 Ohm
Transient overvoltage superimposed on the power frequency voltage has a
frequency of 590 Hz. Maximum overvoltage appears on the phase which has the
highest voltage at the clearing time. The overvoltage is produced by the energy
stored in the electric field of the capacitance that will discharge at fault clearing
time.
-
8/10/2019 Torelli Za Ucinke
66/180
66/138
Bus Current,Bus Voltage : Graphs
Time ... 0.490 0.500 0.510 0.520 0.530 0.540 0.550
-1.50
-1.00
-0.50
0.00
0.50
1.001.50
Current(kA)
Is
-100
-50
0
50
100
150
Voltage
(kV)
Vs
Figure 2.24: Configuration 1- Snapshot of Maximum Overvoltage- Phase to Phase to Ground Fault
NER 0 Ohm
In the phase to phase ground fault the maximum overvoltage is recorded at the
fault clearing time. Frequency of the superimposed transient overvoltage is 590 Hz
as in the phase to ground faults. The overvoltages on each phase have the same
frequency. This transient overvoltage decays after 30ms.
-
8/10/2019 Torelli Za Ucinke
67/180
-
8/10/2019 Torelli Za Ucinke
68/180
68/138
As the NER approaches 50 Ohm, the pattern of the plots remains similar to plots
with NER equal to 0 Ohm. Overvoltage frequencies are 590 Hz in all three phases.
Bus Current,Bus Voltage : Graphs
Time ... 0.4000 0.4050 0.4100 0.4150 0.4200 0.4250
-0.100
-0.050
0.000
0.050
0.100
0.150
0.200
Current(kA)
Is
-125
-100
-75
-50
-25
0
25
50
75
100
Voltage
(kV)
Vs
Figure 2.27: Configuration 1- Snapshot of Maximum Overvoltage- Phase to Ground Fault NER
1000 Ohm
The large size of the resistor now makes the phase to ground fault as the most
prominent overvoltage. Traveling waves are visible in the current and voltage plots.
2.5.8 Impact of Time of Fault with the Overvoltage
This investigation highlights the relation between the time of fault and the
amplitude of the measured overvoltage. Plots are divided into phase to ground fault
and phase to phase to ground fault.
-
8/10/2019 Torelli Za Ucinke
69/180
69/138
Simulation 1- Measured Overvoltage
Ph-g fault- All Data Plot
0
0.5
1
1.5
2
2.5
3
0 50 100 150 200 250 300 350
Sample
Overvoltag L=100km
L=10km
L=0.2km
Solidly Earthed
SystemInsulated
System
Figure 2.28: Configuration 1- All Data Plot
Simulation 2- Measured Overvoltage
Ph-g Fault- All Data Plot
0
0.5
1
1.5
2
2.5
0 50 100 150 200 250 300 350
Sample
Overvoltag L=100km
L=10km
L=0.2km
Solidly Earthed
SystemInsulated
System
Figure 2.29: Configuration 2 All Data Plot
Simulation 3- Measured Overvoltage
Ph-g Fault- All Data Plot
0
2
4
6
8
10
12
0 50 100 150 200 250 300 350
Sample
Overvoltag L=100km
L=10km
L=0.2km
Solidly Earthed
SystemInsulated
System
Figure 2.30: Configuration 3 All data plot
-
8/10/2019 Torelli Za Ucinke
70/180
70/138
Simulation 4- Measured Overvoltage
Ph-g Fault- All Data Plot
0
2
4
6
8
10
12
0 50 100 150 200 250 300 350
Sample
Overvoltag L=100km
L=10km
L=0.2km
Solidly Earthed
SystemInsulated
System
Figure 2.31: Configuration 4- All data plot
Simulation 1- Measured OvervoltagePh-Ph-G Fault- All Data Plot
0
0.5
1
1.5
2
2.5
3
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
Series1
Series2
Series3Solidly EarthedSystem
InsulatedSystem
Figure 2.32: Configuration 1- All data plot
Simulation 2- Measured OvervoltagePh-Ph-G Fault- All Data Plot
0
0.5
1
1.5
2
2.5
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
L=100km
L=10km
L=0.2kmSolidly Earthed
System
InsulatedSystem
Figure 2.33: Configuration 2 All data plot
-
8/10/2019 Torelli Za Ucinke
71/180
71/138
Simulation 3- Measured OvervoltagePh-Ph-G Fault- All Data Plot
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
L=100km
L=10km
L=0.2kmSolidly E arthedSystem
InsulatedSystem
Figure 2.34: Configuration 3- All data plot
Simulation 4- Measured Overvoltage
Ph-Ph-G Fault- All Data Plot
0
1
2
3
4
5
6
7
8
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
L=100km
L=10km
L=0.2kmSolidly EarthedSystem
InsulatedSystem
Figure 2.35: Configuration 3 All data plot
These graphs are produced by using the maximum overvoltage recorded on each
simulation. It is reminded that 11 simulations were performed for each NER sizeconfiguration. Besides, 1 ms additional delay time was applied on each simulation.
Points were plotted sequentially, starting on the simulation with a NER of 0 Ohm to
the maximum value of the NER, which corresponds to the insulated system.
In these eight plots, Figure 2.28 to Figure 2.35, it is possible to recognize the 15
peaks related to the 15 resistor sizes used in the simulations.
-
8/10/2019 Torelli Za Ucinke
72/180
72/138
Phase to ground faults for transmission line of 10 and 0.1km showed a smoother
pattern with lower peak values. Only in Simulation 2 there is an irregularity in themaximum overvoltage. Overall, the maximum overvoltage in the phase to ground
fault is therefore less dependent on the time of fault.
Phase to ground fault maximum overvoltages show a similar pattern across the
three different subtransmission line lengths. Maximum overvoltages appear only for
a brief period of time and at the same time in all the three line lengths.
The 8 plots also show similar results for simulations conducted with
subtransmission line of 10km and 0.1km.
These graphs confirm that time of fault is relevant in determining the maximum
overvoltage during phase to phase to ground faults.
2.5.9 Overvoltage Response on Phase a ,b and c
The section investigates the relationship between faults and different overvoltage
responses on each phase.
This study was conducted only for subtransmission line of 100km in the simulation
1, 2, 3 and 4. Graphs are divided in phase to ground faults and phase to phase to
ground faults.
On PSCAD software package phase to ground fault takes place on phase a and
phase to phase to ground fault takes place on phases a and b.
-
8/10/2019 Torelli Za Ucinke
73/180
73/138
Simulation 1-Measured Overvoltage
Phase Comparison-Phase to Ground Fault- L=100km
0
0.5
1
1.5
2
2.5
3
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
Phase a
Phase b
Phase c
Solidly Earthed
System
Insulated
System
Figure 2.36: Configuration 1 Phase comparison
Simulation 2-Measured Overvoltage
Phase Comparison-Phase to Ground Fault- L=100km
0
0.5
1
1.5
2
2.5
0 50 100 150 200 250 300 350
Simulation
Overvoltage(p)
Phase a
Phase b
Phase c
Solidly Earthed
System
Insulated
System
Figure 2.37: Configuration 2 Phase comparison
Simulation 3-Measured Overvoltage
Phase Comparison-Phase to Ground Fault- L=100km
0
2
4
6
8
10
12
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
Phase a
Phase b
Phase c
Solidly Earthed
System
Insulated
System
Figure 2.38: Configuration 3 Phase comparison
-
8/10/2019 Torelli Za Ucinke
74/180
74/138
Simulation 4-Measured OvervoltagePhase Comparison-Phase to Ground Fault- L=100km
0
2
4
6
8
10
12
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
Phase a
Phase b
Phase c
Solidly EarthedSystem
InsulatedSystem
Figure 2.39: Configuration 4 Phase comparison
Simulation 1-Measured Overvoltage
Phase Comparison-Phase to Phase to Ground Fault- L=100km
0
0.5
1
1.5
2
2.5
3
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
Phase a
Phase b
Phase c
Solidly Earthed
System
InsulatedSystem
Figure 2.40: Configuration 1 Phase comparison
Simulation 2-Measured Overvoltage
Phase Comparison-Phase to Phase to Ground Fault- L=100km
0
0.5
1
1.5
2
2.5
0 50 100 150 200 250 300 350
Simulation
Overvoltage(pu)
Phase a
Phase b
Phase c
Solidly Earthed
System
Insulated
System
Figure 2.41: Configuration 2 Phase comparison
-
8/10/2019 Torelli Za Ucinke
75/180
-
8/10/2019 Torelli Za Ucinke
76/180
76/138
As noted in Section 2.4.9, the plots for the phase to phase to ground faults show
that the time of fault is a key element to determine in which phase the maximumovervoltage is recorded.
2.5.10 Overvoltage Harmonic Response
A further analysis was conducted on the frequency spectrum of the transient
overvoltage. The study was done using the Fast Fourier Transform (FFT) function,
which determine the harmonic magnitude and phase of the input signal as a
function of time[21].
Base frequency was set to 12.5 Hz to detect noise elements lower than power
frequency voltage.
Here below few snapshots of the healthy and faulty phases for a phase to ground
fault following a fault are provided. Snapshots were captured after 10 ms fault
injection.
Referring to the graphs left hand side spectrum refers to 12.5 Hz component and
right hand side spectrum refers to 322.5 HZ noise element. The number in the plot
frame refers to the voltage magnitude in the first frequency band, 12.5 Hz.
Ea60.0
0.0
[1] 1.79434
Figure 2.44: Configuration 1 Harmonic analysis NER 0 ohmsafter 10ms
-
8/10/2019 Torelli Za Ucinke
77/180
77/138
Ea60.0
0.0
[1] 3.50508
Figure 2.45: Configuration 1 Harmonic analysis NER 5 ohmsafter 10ms
Ea60.0
0.0
[1] 6.13753
Figure 2.46: Configuration 1 Harmonic analysis NER 50 ohmsafter 10ms
From the graphs we can observe that voltage noise appears at frequencies higher
and lower values in respect of the power frequency voltage The major components
are in the band close to 50 Hz. Components with lower magnitude, less than 3 %
of nominal voltage, are visible on high frequency up to 300 Hz.
The harmonics were recorded on the healthy phase. The graphs show that thenoise is not limited to odd harmonics. Therefore, filtering this natural frequency
component could be problematic and slow down the information process of the
digital relays.
-
8/10/2019 Torelli Za Ucinke
78/180
78/138
2.6 Conclusion
1. Results obtained from this study are in line with the existing literature and
previous studies in this field. Maximum overvoltages are below 2.5 pu for a
ground fault in an unloaded subtransmission line and below 4 pu in an
unloaded line with a capacitor bank connected at the end of the line
2. Maximum overvoltage is influenced by the power system parameters. The
size of the neutral earth resistor itself does not give a clear indication of the
system response and expected maximum overvoltage during a ground fault
3. Increasing the size of the neutral earth resistor controls the maximum
overvoltage on unloaded line as the resistor has a positive impact in limiting
the natural frequency overvoltage until the fault current is equal to 10 % of
the fault current with a solidly earthed system
4. The limitation of the transitory overvoltage component is strictly connected
to the increase of the ratio R0/X0 until the R0
-
8/10/2019 Torelli Za Ucinke
79/180
79/138
analysis and related transient studies should be carefully considered for
long subtransmission lines as is usually done for transmission lines
8. There is also a consistent similarity of results for line length of 10km and
0.2km. This characteristic should be taken into consideration to optimize the
work during planning and design activities for short and medium length
subtransmission lines
9. Transient overvoltages appear in the faulty and healthy phases. There is a
relation between time of fault and location where the maximum overvoltages
takes place
10. There is not a strong relation between the time of fault and the magnitude of
the maximum overvoltage during phase to ground faults. Relation between
the magnitude of the maximum overvoltage and the time of fault is only
relevant on the phase to phase to ground faults
11. Considering the overvoltage component at higher frequency, this study
shows that beyond the debate of effective or non effective grounding, design
of the neutral earth resistors could safely aim to limit the ground fault current
to 0.1pu of the fault current in a solidly earthed system without
compromising the insulation level of the system
-
8/10/2019 Torelli Za Ucinke
80/180
80/138
3 CVT and Transient Overvoltages
3.1 Overview
Capacitive voltage transformer, CVT, represents the most common method of
voltage source for the impedance protection relays installed for transmission and
subtransmission power systems.
Until few years ago the use of the CVT was limited to extra high voltage system. In
recent years the use of the CVT has become more popular even for applicationwith nominal voltage down to 33kV (as applied by Energex and Ergon in
Queensland) due the lower capital cost compared to the traditional voltage
transformer, VT. CVTs also offer the advantage to couple power line carrier to
inject teleprotection signals between substations.
During a fault, ideally, the low voltage source provided by the CVT should be an
exact copy of the primary voltage. Unfortunately this is not the reality. CVT
contains a large number of stored energy components which must charge and
discharge during voltage changes [24-29]. The transfer of energy between
capacitances and inductances within the CVT requires some time to take place.
Therefore, the CVT introduces some transients that affect the magnitude and
shape of the voltage signal.
CVT creates some problems to the distance relays especially during faults that
severely depress the voltage at the relay location. Faults located at the beginning
of the protected line can bring the voltage down to few percents of the nominal
voltage and this could affect accuracy and operating time of the relay.
This section aims to investigate the CVT transient response during phase to
ground fault and the impact that this disturbance can have on the distance relay.
-
8/10/2019 Torelli Za Ucinke
81/180
-
8/10/2019 Torelli Za Ucinke
82/180
82/138
The tuning reactor is required to cancel out the value of the capacitive divider at
the system frequency [24] . Therefore, the reactor prevents the phase shiftbetween the voltage signal at the transmission line and the voltage fed to the relay.
The size of the reactor is tuned for the system frequency of 50 Hz. In this study the
reactor is 111 H.
Voltage transformer
The voltage transformer reduces the voltage from 10-15 kV to a voltage suitable for
the relay which is usually 100 or 110 V. Copper and iron losses were taken into
consideration by adding in the CVT model a resistance of 2000 Ohm.
Ferroresonance Suppression Circuit
The ferroresonance suppression circuit is installed in the low voltage side of the
voltage transformer to prevent dangerous overvoltages due to the saturation of the
iron core of the step down transformer.
Two types of ferroresonance suppression circuits are normally used by CVT
manufacturers [24, 25, 27]:
! Active ferroresonance suppression circuit
! Passive ferroresonance suppression circuit
The active ferroresonance suppression circuit consists of a parallel tuning circuit in
series with a damping resistor. The LC parallel circuit behaves as an open circuit at
the nominal power frequency to prevent shunt current flowing through the circuit.
During transient condition the tuning circuit will decrease the impedance to the
value of the damping resistor. The resistor will attenuate the energy of the transient
surge [24]
-
8/10/2019 Torelli Za Ucinke
83/180
83/138
The passive ferroresonance suppression circuit has a resistor, a saturable inductor
and air gap loading resistor. In normal condition the voltage is not high enough toflash the air gap. During transient conditions, the air gap will flash over through the
resistor and will attenuate the transient energy.
Existing literature and previous study demonstrated that the passive
ferroresonance circuit has better transient performance than the active device [24,
25, 27, 28, 31, 32]. However, as the study aims to test the CVT under the worst
case scenario, the active suppression circuit was implemented.
Device Data
R= 77379 Ohms
C=29 nF
L=111 H
CVT Burden
The burden connected at the end of the circuit represents the static or numerical
digital relays. Modern relays have a small burden. Typically the burden is below 5
VA .
The value of the burden is 1000 Ohms
-
8/10/2019 Torelli Za Ucinke
84/180
84/138
3.3 Simulation Settings
Fault time
System faults can happen at any time. Considering the two voltage magnitude end
range, the fault could take place at 0 voltage or at maximum voltage. Fault at zero
voltage are less common than faults at the maximum voltage. In fact, this type of
fault could only happen during lightning or close of the circuit breaker on fault
condition. Faults at the maximum voltage usually take place when insulation failure
occurs and are statistically more likely to happen.
Simulations were conducted with fault at zero crossing of the primary voltage. At
this time the energy store in the capacitor is at the maximum and this condition will
generate the worst voltage transient conditions. The sudden change in the system
will release the energy stored in the active elements of the CVT creating the worst
transient overvoltage condition. These voltage oscillations are composed of lowerand higher frequency components.
Network and Source Settings
66 kV subtransmission line model is as per the model used in Section 1
(Overvoltage). The line model was set to 2 km and 10 km. Short lines were
selected to increase the source impedance ratio, SIR, and testing the performance
of the CVT with a higher SIR. It is reminded that SIR is the ration between the
source impedance and the line impedance at the relay point.
-
8/10/2019 Torelli Za Ucinke
85/180
85/138
The source impedance was set to determine a starting SIR of 1 and 10 without the
implementation of the NER. Then, the neutral earth resistor was increased toobtain the desired SIR.
To conclude, four network models were investigated:
! 10km line length- SIR of 1( No NER), 5, 10, 20, 30 and 60
! 10km line length- SIR of 10( No NER), 20, 30 and 60
! 2 km line length- SIR of 1( No NER), 5, 10, 20, 30 and 60
! 10km line length- SIR of 10( No NER), 20, 30 and 60
-
8/10/2019 Torelli Za Ucinke
86/180
86/138
3.4 Simulation Results
Simulations results are provided in Appendix D and summarised in the graphs here
below:
Transient Overvoltage- 10km line- Initial SIR 1
0.00
5.00
10.00
15.00
20.0025.00
30.00
35.00
40.00
45.00
50.00
10 30 50 70
Time (ms)
Overvoltage(pu) SIR 1
SIR 5
SIR 10
SIR 20
SIR 30
SIR 60
Figure 3.1: Peak overvoltage after fault- Initial SIR 1- Line length 10 km
Transient Overvoltage- 10km line- Initial SIR 10
0.00
5.00
10.00
15.00
20.00
25.00
30.00
35.00
40.00
45.00
50.00
10 30 50 70
Time (ms)
Overvoltage(pu)
SIR 10
SIR 20
SIR 30
SIR 60
Figure 3.2: Peak overvoltage after fault- Initial SIR 10- Line length 10 km
-
8/10/2019 Torelli Za Ucinke
87/180