21-01-2010, 12:11 PM
[attachment=1262]
ABSTRACT
This is a new technique for the protection of transmission systems by
using the global positioning system (GPS) and fault generated
transients. In this scheme the relay contains a fault transient
detection system together with a communication unit, which is connected
to the power line through the high voltage coupling capacitors of the
CVT. Relays are installed at each bus bar in a transmission network.
These detect the fault generated high frequency voltage transient
signals and record the time instant corresponding to when the initial
traveling wave generated by the fault arrives at the busbar. The
decision to trip is based on the components as they propagate through
the system. extensive simulation studies of the technique were carried
out to examine the response to different power system and fault
condition. The communication unit is used to transmit and receive coded
digital signals of the local information to and from the associated
relays in the system. At each substation , the relay determine the
location of the fault by comparing the GPS time stay measured locally
with those received from the adjacent substations, extensive simulation
studies presented here demonstrate feasibility of the scheme .
1. INTRODUCTION
A century has passed since the application of the first electro
chemical over current relays in power system protection. The majority
of protection principles where developed with in the first three
decades of century .a rough guide to there development is shown in fig1
Distance protection has played an important role in power line
protection since it was first introduced in the early part of the
century. it has many advantages over the power line protection
techniques and can be adopted for fault location and back up
protection. however , like other power frequency based protection
techniques it suffers from limitation due to power system frequency
wave form , fault path resistance , line loading and source parameter
variations. In particular , the response speed of the relay cannot meet
the reqirements when very high speed fault clearance is required .
With the continuous development of modern technology, protection
relays have advanced with the development of electromechanical,
semiconductor, integrated circuits and microprocessor technologies. Al
tough decades of research have been put in to the continued development
and perfection of the relay technology , many of the basic relaying
principles of protection have not been changed and are still playing a
dominant role today. the introduction of computer technology have been
an important milestone in the history of power system protection .since
the concept was first raised in the late 60â„¢s relay technology has
gone through rapid development. digital techniques for transmission
line protection have been quickly developed and have included various
digital and numeric impedance algorithm for distance protection.
Modern development for power system network , the demand for fast
fault clearance to improve system stability and the need for
alternative protection principles have resulted in the search for
methods to increase the speed of relay response .in the late 70â„¢s this
led to the development of ultra high speed protection based on the
use of traveling waves and super imposed components these relays
offered the advantage of fast response , directionality , and where not
affected by power swing and CT saturation. However many distinct
advantages of the conventional protection techniques where not retained
for eg. Inherent back up protection.
In recent years, there is a growing interest in the use of fault
generated transients for protection purposes and extensive research
work has been conducted to develop new relaying principles and
techniques based on there detection.. this led to the new concept of
transient based protection (TBP). Among these the positional
protection offers attractive solutions for power line protection.
This technique is based on the detection of fault generated high
frequency transient signals and determine the actual portion of the
fault on the line by measuring the traveling time of the high
frequency transient voltage or current signals along the line . in
contrast to the conventional traveling wave based protection
techniques, this technique concentrates on the fault generated signals
during arcing and their associated high frequency signals. With this
approach not only the close in faults can be detected , but also the
problem of low fault inception angle , voltage zero faults is
effectively overcome since the faults arc signals vary little with the
inception angle.
The positional protection uses its associated GPS scheme to determine
the instant when it detects the fault generated high frequency
transient signals and uses the power line communication system to
communicate this information to the relays at the other substations. By
comparing the arrival time of the transient at different points in the
network , relay is able to identify where the fault is on the system
and pin point its location . the system can also respond to the high
frequency transient generated by switch gear operation, which provides
an immediate opportunity for comprehensive self testing and calibration
checking. Electro magnetic transient program(EMPT) software has been
used to simulate a model EHV transmission system in order to examine
the response of the protection scheme to a variety of different system
and fault condition. Results demonstrate that the proposed technique
offer a very fast relay response and high accuracy in fault location.
It has also been shown that the scheme is immune to power frequency
phenomena which can effect established types of relaying.
2. FAULT GENERATED TRANSIENTS AND ASSOSIATED PROTECTION TECHNIQUES
A comparison of different protection techniques in the frequency domain
is shown in fig . a power system fault indicates a variety of
additional transient components in additional components contain
extensive information about the fault and are spread through out the
spectrum ranging from Dc to may kilohertz and even mega hertz.
In conventional protection scheme , the high frequency signals are
considered to be noise and filtered out and as a result, considerable
research has been spent on the designing of the filters , protection
schemes based on detection of fault generated transient, such as the
ultra high speed protection schemes are generally limited by the band
width of transducers used.
It is accepted however that the fault generated high frequency
transient components contain a wealth of information about the fault
type, location , its directions and duration . the use of these high
frequency transient signals enables the realization of new protection
principles that could not be implemented using only power frequency
signals . this has led to the development of the transient based
protection and the transient identification shown in fig 2.
The transient based protection technique operate by extracting the
fault generated high frequency signals through specially designed
detection devices and their associated algorithms. The high frequency
current signal are directly extracted from the CT out puts . although
conventional iron cored CTâ„¢s alternate the high frequency signals,
their characteristics are such that sufficient signals can be detected
for relaying measurement and several researches are studying there use
for high frequency relaying. Following initial analogue filtering, fast
signal processing algorithms are then applied to the measured signals
for fault identification.
3. CHARECTERSTICS OF HIGH FREQUENCY CURRENT SIGNALS.
The theoretical aspects of the characteristics of the propagation of
high frequency signals on transmission lines have been well
documented.
3.1 detection of fault position and fault generated current
transients
when a fault occurs on a transmission line , wide band voltages and
current signals propagate away from the fault point along the power
conductors. In time, these signals reach discontinuities on the
transmission line and some of the signals is reflected back towards the
fault point. The characteristics of these waves are dependent on
several factors including , the fault position on the line , fault
path resistance and the characteristics impedance of the power
conductors .this propagation can be shown graphically.
Here relays are located at all of the substations in the power system
and independently monitor the power system. The frequency range of
interest for monitoring these fault generated high frequency signals is
between 40-80 KHz and the signal processing is designed as to determine
the arrival of a high frequency transient characteristics of those
generated by a fault. at these frequencies , bus bars are dominated by
their capacitive elements , and as a result, the incoming high
frequency current signal is both inverted and reflected . a resistive
fault in their frequency range will also reflect a current wave of the
opposite polarity.
3.2 Fault current transient detector
The proposed scheme uses a specially designed transient current
detector fed from the primary CTs . This extracts are high frequency
signals associated with the fault generated current transients. A
simplified block diagram of the detector arrangement is shown in the
figure . the circuit comprises of an analogue input circuit for signal
conditioning and a digital circuit for determining the transients.
Particular emphasis has been placed on the development of digital
circuit.
The detector is designed to interrogate signals in the range of
frequencies from 40-80KHz. analog circuit acts as a band pass filter
which extracts the band of fault generated transient current signal
from the line. as a result , the response of the scheme is not affected
by the power frequency short circuit band at the busbar or the presise
configration of the source side networks.
3.3 Signal Processing Unit
model transformation is employed to decouple the signal in to
their respective aerial modes. The signal mixing circuit receive the
signal from the 3 phase CTs and continue these to form mode2 and mode3
signals. There are filters to remove any spurious noise. The outputs of
the analog circuit are then passed to the digital circuit.
The sampling frequency of the analog to digital A/D converter is 1
Mz and the speed of propagation of the high frequency transient is
similar to the speed of light. The digital processing includes filters
sequence recording, amplitude comparison, counters and decision logic.
4. BASIC PRINCIPLES AND RELAY DESIGN
A short circuit fault on a power transmission line generates voltages
and current signals over a wide frequency range. These signals
propagate away from the fault point in both directions along the
transmission system with velocity close to the speed of light. It has
been long recognized that the actual faulted position could be
determined on line if the transient signals could time tagged at key
points on the power system network. The global positioning system ,
with its ability to provide synchronization with an accuracy of
microsecond over the wide area, provides an ideal tool for performing
this time tagging of the receipt of fault generated transients.
4.1 Basic Principle
The basic principle of the technique can be demonstrated by referring
to the 400Kv, EHV transmission network, shown in figure. Relays are
installed at the bus bars P,Q,R and S and are responsible for the
protection of the network . for this paper , the study has been
concentrated on the protection of the network PQR and tripping of the
breakers associated with that network, high frequency signals are
generated at the fault point and travel outward from that point along
the network conductors. In time they will reach the monitored bus bar
and be detected by the relays connected to them. each relay record the
arrival instant of the signal generated by the fault.
The relays then code this time information with details of their
identification.. and transmit this to their neighbouring relays. All
relays are continuously ready to receive the coded massages send by
other devices. Data protocols are used to avoid conflict between
information sent by different devices along the same line. following
an event , the relays compare the fault transient arrival time recorded
at its sight with those send by other relays .from this they determine
whether the fault is with in the protected zone. appropriate tripping
instructions are then send to the relevant local circuit breakers. The
actual location where the fault occurs can be clearly identified at
each relay location by this method.
4.2 Relay Design
A simplified block diagram of the relay unit is shown in fig.. the
transient detector uni5t is connected to the line using three phase
CVTs . these are able to detect the fault generated high frequency
voltage signals. The communication unit, containing a transmitter and a
receiver circuits , also uses the CVTs together with a hybrid unit to
separate the transmitted and received transmitted signal.
The transient detector is responsible for detecting the fault
generated fast transient signals and recording the time tag obtained
from the GPS clock.
The transmitter circuit sends this time tag corresponding to the
instant when the transient is captured, to the receivers of the other
relays installed involved in the network scheme.
Previous research has investigated the use of digital filters to
detect the high frequency signals generated by the fault and had shown
that the accuracy of fault location was a function of the sampling rate
used to digitalize the measured signal. the accuracy was directly
related to the sampling rate and higher the sampling rate , the more
accurate the measurement. in this system it was therefore proposed to
use continuous sampling. Ie an analog system and a pass band filter
tuned to operate between 40 and 80 KHz. The protection technique is
therefore divorced from the power system frequency.
The communication page link used in the scheme modeled and shown in fig.
Used power line carrier techniques. Although this has several
advantages , other communication system could be used, such as pilot
wire , optical fibre or microwave.
The decision to trip the local breaker depends on the comparison
between the times measured by the GPS system at that location and those
measured by other relays. Unlike the convectional protection scheme ,
where each relay associates with one circuit breaker on that line
section, the proposed relaying scheme will be responsible for
protection of several lines connected to the bus bar where it is
installed. For eg as shown in fig the relay at bus bar R responsible
for the protection of both line section , connected to the busbar, by
controlling both circuit breakers CB-RP and CB-RQ. Therefore the
technique offers a network protection scheme rather than than one which
concentrates on specific units of plant. This provides several
technical advantages over conventional relaying.
5. MODELLING AND SIMULATION
5.1 System Modeling
The response of the complete system was evaluated by modeling the
transmission line system together with the relays in the scheme using
the EMTP simulation program. EMTP is a general purpose computer program
for simulating high speed transient effects in electric power systems.
The EMTP program features an extremely wide variety of modeling
capabilities encompassing electro magnetic and electro mechanical
oscillations ranging in duration from micro seconds. Its main
application include switching and lightning surge analysis, insulation
co-ordinations, shaft torsional oscillations ferro resonance and HVDC
converter control and operations.
The EMTP simulation studies include
1. simulation of line and transformer energization , load
rejection and fault clearing which are done to help determine the
required transformer , circuit breaker and other equipment
charecterestics.
2. Additional simulations used to develop recommended procedure
for line and transformer energization.
3. Comparison of several recorded waveforms with the result of
EMTP simulation of same events.
5.2 Simulation Studies
The configuration of a transmission line network used in the studies is
shown in the fig. The line lengths , source capacities and fault
position studies are shown 9in fig. CB-PQ, CB-PR,CB-RQ are circuit
breakers responsible for isolating the different line sections.
Fig 3(a) shows the primary system voltage at the bus bars experienced
during a earth fault at the point F1 in the fig.as expected , the
seviarity of the fault depends on the impedance of the line connecting
the busbar and the fault point. It is evident that the high frequency
components are produced at the faulted and unfaulted phases.
Fig 3(b) shows the corresponding transient voltage signals captured by
the relays at P,Q and R. it can be seen that the magnitude of the
captured transient signal decreases with increasing distance between
the relay and the fault point. Upon detecting the arrival of the
transient signal , each relay time tag the signal and details of the
time are send to other relay locations.
The time taken for the communication will depend on the communication
system used. in this study a high speed communication system has been
modeled. the time taken for the communication is the system overload,
which will be added to the processing time required in the decision
making unit.
Determining which is the faulted section is reduced to a comparison of
time tags recorded when the fault transients where detected at the
relaying points through out the network .each relay compares the time
instant of the first wave to arrive at the location with those recorded
at the other location. .a time difference smaller than the time taken
to travel through the corresponding line length indicates that the
fault is with in the corresponding section .the actual fault location
can be determined with an accuracy of with in 300 meters using the
difference between time measurements taken at the end of the faulted
line.
The relation ship between the tag times and determining which feeder is
faulted and hence which breaker need to be tripped is given by
Tp-Tq < Lpq/V
Tp - arrival time of the transient wave as bus bar P
Tq - arrival time of the transient wave as bus bar Q
Lpq - length of the line between busbar P and Q
V - wave velocity on the line.
From the response shown in fig indicates that the fault occurs on the
line section PQ. Since this a TEED feeder , the trip decision will be
made up by the relays at locations P,Q and R respectively and
subsequently these relays trip their associated circuit breakers, as
shown in fig 5(b). for this fault , the relays at P and R are able to
discriminate between the TEED feeder PQR and line PQR and line PR by
considering their response and that from the relay at Q.
The distance we to the fault is calculated at both terminals line
section between bus bar P and Q. the time tag data and the measured
fault location are given in the table1. The tripping signals shown
assume a high speed communication system.
Fig 4 shows the corresponding responses for an Ëœaâ„¢ phase to ground
fault occurring at the point F2 in fig. In this case , the time
difference between Tp and Tq , Tp and Tr correspond to the wave travel
time from P to Q and P to R respectively, and therefore the fault is
onside the area considered in the study.
Tp - Tq = Lpq/V
Tp - Tr = Lpr/V
The relays therefore restrain the circuit breakers associated with
sections PQR from tripping.
The relay installed at bus bar ËœSâ„¢ will detect the time tag the fault
generated transient wave together with the time tag data received from
the relay at bus bar ËœPâ„¢ section is SP and trip the corresponding
breaker. The relay located on the bus bar P will respond in a similar
manner.
Fig 5 shows the relay response for a Ëœbâ„¢ phase to ground fault
occurring near a voltage zero at point F3. as expected although the
magnitude of the transient signal has been reduced as compared to
those shown in fig 3 & 4 , relays able to make correct decision based
on the signals detected . details of the relays response is shown in
table 1.
Again although the magnitude of the signal s captured are relatively
lower due to increase in fault path resistance , the result clearly
shows that the scheme is still able to operate.
Fig 6 shows an Ëœaâ„¢ to Ëœbâ„¢ fault at point F5 on the TEED feeder PQR
. the high frequency transients are readily detected at the relay
location and summary of results are shown in table 1.
Switching operations at any substation will also generate high
frequency transients, which will be detected by the relays. However
the time difference between the time tags will correspond to the
transient time along the feeders and the protection will diagnose that
the disturbance is not on the protected feeders . the response of the
system to those of the system to these switching operations offers the
opportunity for a comprehensive self-testing of the fault detection
GPS and the communication system.
The GPS clock has an accuracy of 1 microseconds roughly corresponds to
an accuracy in fault location of 300 meters. This assumes that the
transients travel at the speed of light in vacuum, where as their speed
will be less along the power conductors.
Error in the time tagging to introduce an uncertainity for faults
occurring close to a bus bar . to for a complete protection scheme
which covers, faults on any part of the transmission line system, the
technique need to be complemented by a high speed directional relaying
technique . such relays could be provided by measuring either current
or voltage, transient signals. In the complete scheme , the direction a
fault as determined at a bus bar also be transmitted to the adjacent
relays using the communication link.
6. CONCLUSION
A new technique for the protection of a transmission line network is
presented in this paper. this uses a dedicated fault detector to
extract the fault generated high frequency voltage transient signal and
GPS system to time tag these signals. The traveling time of the
transient high frequency signal from the point of fault to the adjacent
substation is used to determine the fault positions.
Simulations studies have been carried out the operation of the system
when applied to an EHV transmission network containing both plain and
TEED feeders. Results show that the proposed scheme is able to identify
the faulted section of a transmission network and issue the trip
command to the circuit breaker associated with the faulted section. The
protection is inherently high speed but is dictated by the data
communication system used.
Studies show that the proposed technique is able to offer a high
accuracy in fault location. Since the accuracy of fault location is
proportional to digital sampling was chosen, ie an analog fault
detector.
Unlike traditional protection schemes , this technique offers a new
concept in network protection. The protection inherently monitors the
network to which it is connected and is not limited to individual units
of plant.
7. REFERENCES
o Zhiqian Q Bo Weller, Tom Lomas and Miles A, Redfern
Positional Protection of Transmission system Using global Positioning
System IEEE Trans. On Power delivery, vol 15 no 4 oct 2000
o Z Q Bo G Weller F.T Dai and M A redfern Positional technique
for power transmission lines in IPEC 99 proceedings of the
international power engg conference
o Protective relays application guide: ALSTOM T & D protection
and control ltd
ACKNOWLEDGEMENT
I express my sincere gratitude to Dr. P.M.S. Nambisan, Prof.
and Head, Department of Electrical and Electronics Engineering, MES
College of Engineering, Kuttippuram, for his cooperation and
encouragement.
I would also like to thank my seminar guide and Staff in-charge
Mrs. Sunitha. M (Department of EEE) for his invaluable advice and
wholehearted cooperation without which this seminar would not have seen
the light of day.
Gracious gratitude to all the faculty of the department of EEE
and friends for their valuable advice and encouragement.
CONTENTS
1 Introduction
1
2 Fault generated transients and associated protection techniques
4
3 Characteristics of high frequency current signals
6
4 Basic principles and relay design
9
5 Modeling and simulation
13
6 Conclusion
23
7 References
24
