Documenti di Didattica
Documenti di Professioni
Documenti di Cultura
USING
TELECOMMUNICATIONS
Joint Working Group 34/35.11
December 2000
(Norway)
(Switzerland)
Alastair ADAMSON
(United Kingdom)
Ken BEHRENDT
(United States)
Michael CLAUS
(Germany)
Alouis W. H. GEERLING
(Netherlands)
(Spain)
Christopher HUNTLEY
(Canada)
(Spain)
Yoshizumi SERIZAWA
(Japan)
Kent WIKSTROM
(Sweden)
Corresponding members:
(Brazil)
Stephen HUGHES
(Australia)
David C. SMITH
(South Africa)
Comments and contributions received from:
Herv HOUKE
(France)
Trygve JORDAN
(Norway)
2 / 172
Contents
1
3.1.2
3.1.2.1
3.1.2.2
3.1.2.3
3.1.2.4
3.1.2.5
3.1.2.6
3.2.2
3.2.3
4.3.4
4.3.5
3 / 172
4.4.2
Digital Hierarchies................................................................................................................................72
4.5.3
4.5.3.1 Topology............................................................................................................................................................ 83
4.5.3.2 Media Contention and Protocols........................................................................................................................ 84
4.5.3.3 Advanced topologies ......................................................................................................................................... 85
6.1.2
6.1.3
4 / 172
6.3.1
6.3.2
CONCLUSIONS .............................................................................................................................................139
ANNEX A1
ANNEX A2
ANNEX A3
5 / 172
ABBREVIATIONS..................................................................................................................................................166
INDEX ......................................................................................................................................................................169
6 / 172
JWG 34/35.11 met for a kick-off meeting in Oslo in September 1997. The working group agreed
that a new version of the former Technical Brochure "Protection systems using
telecommunication" (Ref. No. 13, 1987) should be produced. The document should create
awareness for the requirements, opportunities and risks of protection systems using
telecommunications, and guide protection and telecommunication engineers towards a common
understanding for the design and operation of reliable teleprotection schemes that meet
performance requirements in the most economical way.
This Technical Brochure has the following content:
Chapter 2 describes power systems from a teleprotection point of view, with focus on power
system faults, their reasons and characteristics, and fault clearing requirements. It continues
with the definition of fault clearing systems, protection systems, protection schemes, and ends
up with explanations why teleprotection is needed, and how protection can use
telecommunication to meet fault clearing requirements.
Chapter 3 describes protection relaying principles and protection schemes using
telecommunications, and deals - from a power system point of view - with various aspects
around the need of teleprotection, its benefits and adverse implications if the teleprotection
service would fail.
Chapter 4 gives an overview of telecommunication systems, with focus on capabilities and
limitations related to protection signal transmission. Problems and risks that may arise with
different types of telecommunication technologies are addressed, and functional and reliability
aspects are dealt with, both under normal conditions and - most important - under power system
7 / 172
fault conditions.
Chapter 5 deals with interfaces. Requirements on interfaces between protection, teleprotection
and telecommunication devices are given.
Chapter 6 focuses on performance requirements on protection, teleprotection and
telecommunication functions.
Chapter 7 deals with protection system configuration and design. Compatibility issues between
protection schemes and telecom technologies are addressed to provide a guide for protection
and telecommunication specialists to design teleprotection systems that will meet fault clearing
requirements.
Chapter 8 gives an outlook on future trends and addresses some problems to be solved.
In Chapter 9 the document is summarized some conclusions are drawn.
Annexes A1 to A4 contain some related topics and additional information, which the JWG has
found valuable for the better understanding of the subject.
8 / 172
Node in network
= busbar
Breakers
Generator and
transformer unit 1
Overhead
power line
Overhead
power line
Generator and
transformer unit 2
Underground
or submarine
cable
Breakers
Figure 2.1-1:
Nodes at different voltage levels are connected by transformers. These connection points,
transformers and other units are made within a limited geographical area, called a station.
9 / 172
Configuration of a typical power generating station is shown on Figure 2.1-1, and Figure 2.1-2
shows a typical transformer station.
Node in network
= busbar
Breakers
Power
lines/cables
Transformer 1
Overhead
power line
Transformer 2
Overhead
power line
Underground
or submarine
cable
Load
Breakers
Figure 2.1-2:
Power flows through all healthy transmission lines in the electric power system network as it
moves from generation sources to consuming equipment owned by customers. Electric power
system networks operated by more than one electric power utility are often tied together to form
a large grid that supports the transmission of power over a very large area, sometimes spanning
several countries. Figure 2.1-3 shows the routes of major power lines connected in the
Scandinavian power grid.
10 / 172
TROMS
ALTA
ISLAND
NARVIK
REYKJAVIK
OFOTEN
TORNEHAMN
FINNLAND
PETJSKOSKI
SVARTISEN
PIRTTIKOSKI
LETSI
SCHWEDEN
AJAURE
PIKKARALA
GRUNDF.
TUNNSJDAL
LINNVASSELV.
VUOLIJOKI
STORNORRFORS
STORFINNF.
TRONDHEIM
ALAPITK
JRPSTRMMEN
NEA
ALAJR VI
HJLTA
HUUTOKOSKI
NORWEGEN
PETJVESI
KANGASALA
ULVILA
(132)
TOIVILA
OLKILUOT O
HIKI
HYVINK
BERGEN
LOVIISA
RJUKAN
HELSINKI
OSLO
INKOO
BORGVIK
ENKPING
HASLE
STOCKHOLM
STAVANGER
LISTA
NORRKPING
KRISTIANSAND
DC
DC
GTEBORG
NSSJ
OSKARSHAMN
RINGHALS
TJELE
HELSINGBORG
KARLSHAMN
[04/ \
MALM
KBENHAVN
KASS
50 100 150 km
FLENSBURG
HAMBURG
DEUTSCHLAND
Figure 2.1-3:
11 / 172
Insulator
Phase - Ground fault
Tower
Ground
Phase - Phase fault
Figure 2.2-1:
All power system components are exposed to faults due to insulation breakdowns. The
Scandinavian power system shown in Figure 2.1-3, for instance, typically experiences
approximately 3000 faults per year.
Voltage stresses caused by lightning and switching transients, and contamination due to
polluted air are major sources of insulation breakdowns. Mechanical stresses caused by wind,
vibration, ice, and snow-loading are major sources of insulator and supporting structure damage
that also leads to insulation breakdown.
For power lines, most insulation breakdowns are in air between phases and/or phases and
ground. Most frequently, insulation breakdowns are along the surface of insulators due to
excessive voltage stresses. An example of insulator flashover is shown in Figure 2.2-2
12 / 172
Figure 2.2-2:
Insulator flashover
Faults on both power lines and other power units can also
be caused by misoperation of earth switches and "forgotten" security ground connections.
Power system faults caused by weather, animals, high trees, humans, or equipment failure
disrupt normal power flow by diverting current through a short-circuited connection and
collapsing power system voltage. In addition to equipment damage, power system faults cause
transients that adversely affect sources of generation and customer loads. Consequently, faults
must be detected and isolated very quickly. Electric power system generators, transformers,
busbars, and power lines are therefore monitored by protective relays designed to detect power
system faults and operate isolating devices designed to interrupt damaging fault current.
13 / 172
accelerate, motors retard and severe voltage drops can easily result in tripping of complex
industrial plants. For the power system itself, severe disturbances can result in collapses and
blackout for regions, and, in severe cases, even for several countries. Todays society does not
accept frequent severe disturbances and blackouts because of their heavy reliance on electric
power consuming devices for business activities, safety, lighting, heating, cooking,
communication and many other conveniences.
Therefore, to avoid severe disturbances and blackouts, sound protection practices are used to
provide rapid fault clearing. In some cases, formal requirements for fault clearing are provided to
assure consistent levels of reliability levels throughout the power system. Formal requirements
may be grouped in external requirements and utility requirements as already done in [3].
External requirements may encompass:
- Customer power quality and interruption requirements.
- Requirements from insurance companies who underwrite equipment failures.
- Legal requirements to meet prudent utility practice and industry standards in case
primary equipment failures result in personal injury or property damage and legal actions
are taken against the utility by the parties incurring damage.
- International and national safety regulations, imposed by governmental and other
agencies.
- Requirements imposed by manufacturers of primary equipment in order to validate
equipment warranties.
- Requirements from occupational safety and hazard prevention.
Utility requirements
The power system must be designed and operated to avoid instability, loss of synchronism,
voltage collapse, undesired load shedding, and unacceptable frequency or voltage. Good
protection practices help meet these objectives by detecting and clearing faults rapidly. Rapid
fault clearing helps:
- Prevent severe power swings or system instability
- Minimize disruption of system power transfer capability
- Prevent unreliable services
- Limit or prevent equipment damage
It is very important to clear the fault within specified limits to ensure that the healthy remainder
of the power system can continue to serve its customers with acceptable quality and reliability.
Requirements on protection
Protection performance requirements are issued to satisfy external and utility requirements.
These requirements specify how protective schemes must perform on specific contingencies to
fulfill external and utility requirements. They typically provide a balance between the conflicting
goals of dependability and security. Dependability goals require maximum sensitivity and fast
response time to detect and clear all faults quickly with very low probability of a failure to trip.
Security goals require maximum selectivity and slow response time to minimize the probability
of an unwanted trip on an unfaulted circuit. Security is an issue during fault conditions as well as
during normal, unfaulted conditions.
Simply stated, the implementation of these protection requirements should result in dependable
operation of only those relays protecting the faulted unit, and secure non-operation of the relays
during non-fault conditions and when faults occur on adjacent power system units. This balance
is met only through proper protection scheme design, proper relay and equipment selection,
14 / 172
and proper connection and setting of these relays and equipment to achieve appropriate
sensitivity and coordination.
When protection schemes detect a fault on the equipment or line they protect, they signal
isolating devices, called circuit breakers, to open, isolating the faulty segment of the system,
and restoring normal system voltage and current flow in the power system.
Protection schemes command circuit breakers to isolate faults with no intentional time delay.
When the protection scheme and circuit breakers operate properly, the fault is isolated within
the required fault-clearing time. Protection applied on high voltage systems, where fault-clearing
times are most critical, typically detect faults and operate in about one to two cycles. Some
schemes operate in less than one cycle. Circuit breakers operate in one to three cycles. The
combination of high-speed protection schemes and fast circuit breakers can interrupt a fault in
about two cycles, although more common fault-clearing times range from three to six cycles.
Power units and lines are protected in zones to coordinate fault detection and clearing. A
15 / 172
protection zone is defined as all high voltage power system equipment (and all necessary
control, supervision and protection equipment) between two or more circuit breakers. Selective
fault clearing (selectivity) is to trip the breakers for the faulty zone, and not trip any additional
breakers for non-faulty zones. Five basic zones of protection are shown in Figure 2.4-1. These
zones of protection are identified as Generators (1), Transformers (2), Busses (3), Lines (4),
and Loads, such as Motors (5).
5
Station C
3
Station A
1
Station B
2
1
2
4
Station D
4
Figure 2.4-1:
The boundaries of each zone of protection, as it applies to protective relays, are determined by
the location of the current transformers that provide the representation of primary system
currents to the protective relays. Other parameters, such as voltage, are also used by some
protective relays to perform their protection function, but the current transformer location
determines the protection zone boundary. Overlapping zones of protection is an established
protection concept represented by Figure 2.4-2. As shown, the current transformers are typically
located on opposite sides of the circuit breaker, or on one side and as close as possible to the
circuit breaker that is tripped to clear faults in the respective protection zones.
Protection zone boundaries for power units such as generators, transformers, busses, and
motors are typically within the same substation, permitting one relay to monitor currents at the
boundary of its protection zone. Likewise, the same relay can easily be connected to issue trip
signals to all circuit breakers at the boundary of its protection zone.
The boundary for line protection, however, is typically located at two different stations that may
be separated by a considerable distance. This separation makes it impossible for one relay to
sense currents at both ends of the line, or control breakers at both ends of the line. It is
therefore common practice to install at least one relay and circuit breaker at each end of the
16 / 172
line. These relays may operate independently, or they may share information to improve their
operating speed, or they may require communication between them to operate.1
CT for Zone B
Zone B
Zone A
CT for Zone A
a. CTs on opposite sides of breaker
CT for Zone B
Zone B
Zone A
CT for Zone A
Protection schemes that share information to improve operating speed are sometimes referred to as non-unit protection
schemes. Protection schemes that require communication to operate are sometimes referred to as unit protection schemes.
17 / 172
circuit, permitting the exchange of information (analogue data and/or status) or transmission of
commands. In Figure 2.4-3, the telecommunication system may be dedicated point-to-point,
shared point-to-point, or a network.
High Voltage Equipment
Protection
Functions
Control
Circuit Breaker
Mechanism
Sensors
Teleprotection
Function
Protection
Scheme
Protection
Zone
Telecommunication
System
Teleprotection
Function
Sensors
Circuit Breaker
Protection
Functions
Control
Mechanism
Auxiliary power
Figure 2.4-3:
18 / 172
Protection functions in one station interchange information with protection functions in a remote
station via teleprotection and telecommunications. This sub-total functionality forms a
Protection Scheme.
19 / 172
Figure 2.5-1:
20 / 172
Step distance relays (without acceleration schemes) have limitations, for example:
Very short lines and cables due to their low impedance, which makes it difficult to
find an adequate setting to get a instantaneous trip for faults on the main part of the
line.
Multi-terminal lines, since the intermediate infeeds modify the impedance seen by
the distance relays, which depends not only on the distance to the fault, but also on
the infeed from the remote terminals, making impossible an accurate measure of the
impedance.
No potential transformers and only current transformers are installed at each end of the
line.
We can distinguish two types analog comparison protection systems: longitudinal current
differential protection and phase comparison protection. The current differential protection
compares the power frequency signals proportional to the primary power system currents
(amplitude and phase angle), while the phase comparison one is based on comparison of the
phase angle (or sign) between currents of each end of the protected line.
Since both of them use only current information, in comparison with the distance or other
system protections, analog comparison protections have the following advantages:
-
21 / 172
No mutual coupling problems from parallel lines. This may cause the line-to-ground fault
current reverses and flows into a weak source terminal, causing faulty directional
discrimination if other protection systems are used
Not subject to transient problems associated with coupling capacitor potential devices
With segregated current differential there are no problems of phase selection for single
pole auto-reclosing at simultaneous faults on different circuits and phases close to one
line end, because it operates only for faults between current transformers in each phase.
Some relaying problems in EHV transmission lines due to applying series capacitors are
also overcome, e.g. voltage reversal, current inversion or phase imbalance.
When phase selection is required for single phase tripping, especially at simultaneous faults on
different circuits and phases or in a faulty line when handling heavily loaded EHV lines, the
phase-segregated technique is used. The analogue information is transmitted separately for
each phase.
In cases where the complete information about the polyphase conditions is not essential and
single-phase tripping is not needed, the non-segregated technique is used. It reduces the threephase system of currents to a single-phase one by means of a mixing device. The
communication link needs therefore to only accommodate the transmission of this single phase
information. Some mixing techniques are described in [1].
3.1.1.1 Current differential protection
Operating principles
As mentioned above the current differential protection is an absolutely selective protection
system for transmission lines, tripping instantaneously for faults in the protected zone defined by
the current transformers of each end of the line.
It is based in the principle of current comparison. The Figure 3.1-1 shows a basic scheme of the
differential protection. In each terminal, an evaluation circuit compares the sum of the local and
remote current values, i.e. the differential current, with an operation threshold value Iop. In
normal operation conditions or external faults, the current entering at one end is practically the
same as one leaving at the other end, so the differential current value is practically zero and the
protection will remain stable. For a fault on the protected power line the differential current value
will exceed the operation value and the protection will trip.
When very large currents flow through the protected zone for a fault external to the zone a
differential current appears due to the different ratio error and saturation characteristic of the
current transformers, which could exceed the operation level. Such a maloperation of the
protection is prevented by the stabilizing. The stabilizing characteristic uses a bias current,
which is usually proportional to the sum of the absolute values of the currents at each terminal,
i.e. |iA| + |iB|, in order to make the protection less sensitive for higher through currents. This
technique is also called percentage restraint.
22 / 172
IA
IB
SA
TX
iA
Telecommunication
system
TX
SA
iB
DEL
DEL
iB
RX
Id
RX
TPF
Id> Iop
TPF
iA
+
Id
Id>Iop
Figure 3.1-1:
Figure 3.1-2 shows an example of percentage restraint characteristic with two slopes: the lower
slope ensures good sensitivity to resistive faults under heavy load conditions, while the higher
slope is used to improve relay stability against saturation of the current transformers and other
distortion effects under heavy through fault conditions.
The selection of the minimal operation current Is1 is based upon the magnitude of line
capacitance current and switching transients expected on the protected line. The capacitance of
the three conductors to earth and, except in single core cable, also between each other, makes
that under undisturbed conditions the current at both ends differs in angle and magnitude.
Particularly in cables, the capacitive charging current can attain significant values. Nevertheless,
usually the necessary rise of the Is1 does not involve an important loss of sensitivity.
Idiff
Idiff > k2xIbias - (k2-k1)Is2 + Is1
TRIP
slope k2
NO TRIP
slope k1
Is1
Is2
Ibias
Idiff = iA + iB
Ibias = |iA| + |iB|
Figure 3.1-2:
23 / 172
The differential principle may be applied to multi-terminal lines. The protection relies on the sum
of the inflowing currents, which are added geometrically. For this purpose, the measuring
circuits have to be so arranged that at each end of the line, the local current and the currents
from each of the others ends of the line are available for comparison. Generally, the most recent
designs allow up to three terminals applications.
For a multi-terminal system, the master/slave or centralized configuration is also used. In this
case, the current values are sent to a specific terminal for evaluation of the differential current.
This terminal will henceforth be noted as a master, while the terminal sending information about
currents will be denoted as a slave terminal. For a two-terminal system, the master/slave
configuration can, of course, also be used, but a master/master or distributed configuration,
where the current information is exchanged between both terminals and evaluated at both ends
is normally preferred, since this gives a shorter operating time than that in a master/slave
configuration. See Figure 3.1-4 and Figure 3.1-5 for more details about centralized and
distributed configurations.
The saturation of the current transformers for heavy through currents normally requires the
selection of a higher slope setting which involves a loss of sensitivity for internal faults. Recent
protections include some techniques to detect the saturation, so in only such conditions is the
protection desensitized increasing the restraint slope. To avoid the maloperation of the remote
protections, the terminal that detects the saturation includes a code in the message transmitted
to the other ends, so that all terminals increase the degree of stabilization.
Time delay compensation
As mentioned, the current values used in the differential protection must be taken at the same
instant at all ends of the power line for comparison, so a delay circuit is needed to compensate
the transmission time for the remote values. Classical designs incorporate an adjustable delay
for aligning the current values. However, when digital communication systems with automatic
route switch are used, the time delay can change and the protection must continuously adjust
the time alignment. For this purpose, digital devices incorporate different techniques in which
the messages of current values sent through the communication channel are tagged with the
sampling time. The principles of some synchronization techniques are described in more detail
in A4.1. An error in delay compensation results in a differential current that - according to Figure
3.1-2 - increases the risk of unwanted tripping. For more information see 6.1.2.2 and 6.3.1.1.
Additional functions
Generally, differential protections use intertrip functions, i.e. the sending of trip commands to the
remote ends. Intertrip commands are sent through the same communication channels used to
transmit the current values (switching the channel frequency to a specific intertrip frequency
when analogue links are used, or flagging the corresponding command bits in the out-going
data messages in digital links).
The intertrip function is activated either when the relay reaches a trip decision, or by closing an
external contact connected to an input of the relay.
The intertrip function can be used for:
- Breaker failure protection
- Stub protection: this is applied in switchyards with 1 circuit breaker configuration.
Operating an input by external contact when the line isolator opens allows to protect the
line between the circuit breakers and the line isolator.
24 / 172
ST
Pilot wires
MCT
MCT
TR
TR
ST
Evaluation circuit
Evaluation circuit
Figure 3.1-3:
In this case, the three-phase system is converted into a single AC current in the mixing
transformer MCT (non-segregated).
One differential system for each power phase (segregated) of the protected circuit can also be
provided. If high resistance faults are expected or faults on which the value of earth fault current
is relatively low, a fourth measuring system for the zero sequence component can be
introduced. This however, increases the number of pilot wires and therefore the communication
cost of the comparison information.
In both methods, a replica of the vector difference is formed at each line end by means of a
transformer ST for the stabilizing effect and a replica of the vector sum of the currents flowing at
each end by means of a further transformer TR for the tripping effect. These values are
evaluated separately at each line end in a measuring module and a tripping command is issued
to the circuit-breaker when the fault current has exceeded a permanently adjusted threshold
value.
Where the voltage induced into the pilot cables during earth faults may exceed the rated values,
the protective relays should be isolated from the pilot wires by isolating transformers, which can
also be used to subdivide the total length of the pilot wires into two or three sections. This
prevents the equipment from being subjected to excessive longitudinal voltage due to
25 / 172
Multi-terminal configuration
Transmission line protection based on a current differential scheme detects zone faults by using
each terminal current and transmits the detection results of the zone fault to the other terminals.
There are two types of multi-terminal current differential protection configurations; centralized
and distributed configurations. As these configurations are applied to a single zone protection,
they may be also applied to multi-zone and wide-area protections.
1
26 / 172
Centralized configuration
Figure 3.1-4 shows an example of line protection for a five-terminal EHV line [7]. Each terminal
has a terminal unit that detects the current and transmits the data to the main unit terminal via a
communication channel. This configuration simplifies the unit of each terminal and
communication channel. Since the main unit has current data of all terminals, the fault locator
function can be easily implemented by using these data.
Figure 3.1-4:
Centralized configuration
Distributed configuration
Figure 3.1-5 shows a distributed configuration of five-terminal current differential line protection
system. Each terminal has the current differential protection function as well as the signal
transmitting function that multiplexes current data at each terminal into one communication
signal. Master station A sends its own current data to slave station B. Slave stations B, C, D and
E multiplex their own current data over communication signal. Slave station E turns back this
signal toward slave station D. Now current data of all terminals are on the communication bus
and available for protection. In addition, this system contains sampling synchronization function
which enables the simultaneous sampling of current data at each terminal with high accuracy.
Many installations were conducted using a 1.544-Mbit/s fiber-optic communications channel for
HV double-circuit multi-terminal (up to ten terminals) or tapped lines [8]. In this network
configuration where current differential calculation is usually carried out at each terminal, a
centralized scheme where only master station conducts the calculation and sends the transfer
trip signal to all slave stations is also available.
Figure 3.1-5:
Distributed configuration
27 / 172
Comparison is made for each phase separately. A zero sequence circuit may also be
included => Segregated protection.
The currents of the three phases are mixed into one quantity for comparison => Nonsegregated protection.
The measurement is made twice every period => Full-wave phase comparison.
The measurement is made once every period => Half-wave phase comparison.
The phase angle signal is transmitted to the remote end only when a starter has picked
up.
Measuring is carried out continuously and the signals are permanently transmitted.
A phase comparison scheme can be designed for a blocking mode or for an unblocking
mode of operation, similar to a distance protection system using telecommunication.
The current which is used in the comparison is converted into a square wave signal, so that the
positive portion corresponds to the positive half-cycle and the zero portion corresponds to the
negative half cycle. The square wave from the remote terminal is compared with the local
square wave as shown in Figure 3.1-6.
SA
SQ
TX
iA
DEL
>
IB
IA
&
iB
Telecommunication
system
TX
iB
RX
RX
TPF
TPF
28 / 172
SA
SQ
DEL
&
iA
>
b) Internal fault
IA
IA
IB
IB
iA
iA
iB
iB
iA & iB
iA
& iB
>
<
Half-wave phase comparison
Figure 3.1-6:
In normal line conditions, there is a (small) phase current difference between line ends due to:
- The capacitance of the power line
- Errors due to the equipment, e.g. current transformers, sequencer, filters, squarer, etc.
- The time delay due to the signal propagation time between the terminals
To prevent false trips, a critical angle is defined, commonly called stabilizing angle, which limits
the maximum phase difference between currents, which would correspond to a boundary
between tripping and stabilizing.
In a non-segregated phase comparison protection, the three currents are mixed into one
quantity by means of a composite sequence network. The half-wave system use starters,
normally based on overcurrent detectors, to determine whether a fault has occurred, to initiate
signal transmission to the remote end and to permit local tripping. In the full-wave system, the
comparison is made for each semi-period and normally is therefore faster than the half-wave
type. Phase comparison information is transmitted all the time to the remote equipment, and no
starter is required.
The comparison in the segregated protection system is similar to the non-segregated protection
but, the comparison is made separately for each phase. It is very suitable for single pole
reclosing when handling heavily loaded EHV lines and parallel circuits on the same towers.
Segregated protection is more sensitive for earth faults than non-segregated protection, but it is
more costly and the requirements on communication are higher.
More details about non-segregated and segregated techniques can be found in [1].
29 / 172
Some recent designs use dynamic principles based on the variation of the instantaneous
current values in for example two periods, i.e. i = i(t) - i(t-2T), so that the signal to compare is
the sign of this variation value i. This principle, normally operating in combination with a
conventional phase comparison with starter, gives a higher sensitivity for high resistance ground
faults.
Time delay compensation
As described for differential protection system, time delay compensation must be also provided
in phase comparison protection in order that phase current values can be compared at the
same instant. Depending on the technology, the channel delay can be compensated either by
dynamic measuring techniques, or by a fixed delay setting in the protection relay. This later
case is only useful when there is no possibility for time delay variations.
Additional functions
A number of complementary functions may be included in the protection relay. Intertrip functions
are used to trip the remote breakers by means of sending a command through the same
communication channel used for comparison signals. An overcurrent criterion normally
supervises the remote trip to prevent tripping under normal conditions.
The same additional functions mentioned for current differential protection are also applicable
for phase comparison protection.
Telecommunication systems used for phase comparison protection
Non-phase-segregated technique
In a half-wave comparison scheme it is very common to use power line carrier as
communication medium, with the same carrier frequency used for both directions. The carrier is
amplitude modulated i.e. switched on during positive half-cycles, and off during negative halfcycles, or vice-versa. This system operates as a blocking scheme. For an internal fault, if the
blocking signal from the other end is not received, the output of the comparator circuit sends a
trip command when the starters have picked up. This system might behave incorrectly in some
situations due to the noise generated during a fault, i.e. blocking the operation for internal faults
(=> delayed tripping) or deblocking for external faults (=> unwanted tripping).
In a full-wave comparison different frequencies for the two directions must be used. A FSK
(frequency shift keying) signal is used, which can be transferred over pilot wires, power line,
radio or fibre-optic link. The communication equipment continuously monitors itself and when a
fault occurs, the local signal is compared with the remote for both positive and negative halfcycle in the protection relay.
Phase-segregated technique
In this case, the values of each phase are transmitted separately via independent channels.
Most recent phase comparison systems usually operate in segregated mode and use digital
communication systems. The square signals to compare are sampled and converted to digital
data, which are transmitted serially to the opposite terminal by the telecommunication system.
Data rates and electrical or optic interfaces are the same as those mentioned for differential
protection.
When starters are used to initiate the comparison, a sequence of guard bits is transmitted in
30 / 172
normal state of operation, in order to monitor the channel availability and performance by the
receiver.
Some designs optionally include a modem to interconnect two terminals through a 4-wire audio
channel. In this case, a data rate of 9'600 or 19'200 bit/s may be used.
3.1.1.3 Charge comparison protection
Charge comparison is based on the principle of conservation of charge at a node. The charge
entering one line terminal must be approximately the same as the charge leaving the other line
terminal(s) of a healthy transmission line. This is also the principle from which Kirchoffs Current
Law (the theoretical basis of current differential relaying) is derived.
To perform charge comparison, the waveform of each line terminals phase and residual current
is sampled every millisecond. The half-cycle area under each wave is measured by
integrating current samples between zero-crossings. For each phase and ground, the resulting
ampere-second area (i.e., coulombs of charge) is stored in local memory, along with polarity
and start/finish time-tags. This storage operation occurs only if the magnitude exceeds 0.5
ampere r.m.s. equivalent and the half-cycle pulse width is equal to 6 ms or more.1
Every positive (negative 3Io) magnitude is also transmitted to the remote terminal, along with
phase identification and some timing information related to pulse width and queuing time (if any)
at the transmitting terminal. When the message is received at the remote terminal, it is
immediately assigned a received time-tag. A time interval is then subtracted from the received
time-tag. This interval represents the channel delay compensation (which does not have to be
precisely equal to the actual channel delay time) and the timing information contained in the
received message. The adjusted received time-tag (after subtraction) is then compared with the
local start and finish time-tags, looking for a nest, per Figure 3.1-7 (shown for an external
fault).
Remote current
Actual channel
delay time
Time adjusted in
received message
Channel delay
compensation
Time interval
subtracted
Local current
Start time-tag
Finish time-tag
Received time-tag
Adjusted received
time-tag
Figure 3.1-7:
1
Magnitude is actually measured in terms of ampere-seconds (i.e., coulombs). However, all values are converted to amperes rms
equivalent, based on a perfect 60 Hz (or 50 Hz) sine wave, without offset.
31 / 172
A nest is achieved when the adjusted received time-tag is greater than the local start time-tag
and smaller than the local finish time-tag, for a given half-cycle stored in memory.
When the nesting operation is successful, the local and remote current magnitudes (actually
charges converted to equivalent currents) are then added to create the scalar sum (sum of
absolute magnitudes). The scalar sum becomes the effective restraint quantity and the
arithmetic sum becomes the effective operate quantity, per the bias characteristic shown in
Figure 3.1-8.
BIAS LEVEL
ARITHMETIC
SUM
(TRIP)
(RESTRAINT)
SCALAR SUM
Figure 3.1-8:
The bias level is an operate threshold which provides security in the presence of spurious
operate current due to line charging current, current transformer mismatch, analog-to-digital
conversion quantizing errors, etc. As shown in Figure 3.1-8, the bias level rises sharply after the
scalar sum reaches a high value. This provides security for unequal CT saturation during high
current external faults. At lower currents, the bias level has a slight upward slope. This takes
care of the relatively minor non-communications-related errors that increase with current level,
such as CT ratio errors.
The operating characteristic of charge comparison, when plotted on a polar diagram, is the
ideal rainbow-shape of Figure 3.1-9. Referring to Figure 3.1-7, if the adjusted received timetag nests with a local negative half-cycle, this is equivalent to the upper half of Figure 3.1-9. If
the adjusted received time-tag nests with a local positive half-cycle, then the arithmetic sum and
scalar sum are equal to each other, which describes a 45 degree line on the bias characteristic
(well above the bias threshold for all except very small values of current). This is equivalent to
the lower half of Figure 3.1-9.
IR
IL
RESTRAINT
REGION OF IR
Protected
line
IL
Figure 3.1-9:
32 / 172
The bias level of charge comparison is significantly more sensitive than that of conventional
current differential relays for line protection. The conventional relay requires a gradually
increasing bias to take care of increasing spurious operate current for a given assumed error in
channel delay compensation (the biggest single source of spurious operate current). In contrast,
charge comparison introduces no additional communications-related error as the currents get
bigger, for a given error in channel delay compensation. Furthermore, for a given magnitude of
through current, no operate error current is introduced, at all, for increasing channel delay
compensation error (up to + 4 ms, at which point a total relay misoperation occurs typical of a
digital system). The + 4 ms misoperation threshold for charge comparison is almost three times
the + 1.5 ms (approximately + 30 degrees on 60 Hz systems) misoperation threshold which is
typical of conventional current differential schemes with circular polar diagram characteristics.
Lit: [38]
33 / 172
impedance variations. The primary relay elements used to detect line faults are therefore set to
see or reach either short of the remote line terminal (this is called under reaching), or to see or
reach past the remote line terminal (this is called over reaching).
Communication between line terminals at different electric power substations could be
accomplished by simply extending a number of wires between the substations. Connecting a
relay contact output from a relay scheme at one terminal to a relay scheme control input at the
other line terminal with a pair of copper wires provides the communication necessary for one
relay scheme to tell the other relay scheme that it has, or has not, seen a fault. Unfortunately,
connecting communication wires directly between substations is not that simple and can even
be hazardous. Voltage drop, induced voltages, and ground potential rise between substations
during a fault make direct metallic wire connection between relay schemes unreliable, insecure,
and hazardous.
Communication for state comparison protection schemes must therefore be designed to provide
safe, reliable, secure, and fast information transfer from one relay scheme to another. The
communication scheme must also be able to transmit information in both directions at the same
time. The amount of information required to transfer between relay schemes depends on the
relay scheme logic. The basic and most common state comparison protection schemes are
described in the following subsections. Their communication requirements are discussed within
these subsections. The order in which they are presented does not imply their priority, relative
importance, or usage. Other schemes and combinations of schemes may be designed to meet
specific protection needs, however, they are typically all based on the basic schemes described
in this document.
The terminology used to describe these state comparison protection schemes may differ from
utility to utility and country to country. State comparison schemes are basically defined
according to the impedance zone which sends the protection signal to the remote end of the
line. The following Table 3.1-1 shows the preferred CIGRE scheme names and alternate
scheme names used elsewhere. CIGRE scheme names will be used throughout this document.
CIGRE State Comparison Protection Scheme
Name
Zone acceleration
Table 3.1-1:
34 / 172
gap between the protection zones where faults would not be detected.
RU
RU
Bkr 1
Bkr 2
Protected Line
Teleprotection
Equipment
RU
OR
TX
TX
RX
RX
TRIP Bkr 2
TRIP Bkr 1
RU
OR
Protection
Equipment
RU - underreaching trip function, must be set to reach short of remote terminal
and must overlap in reach with RU at remote terminal
Figure 3.1-10:
For internal faults within the overlap zone, the underreaching functions at each end of the line
operate and trip their associated line breaker directly. At the same time, the RU function keys
its respective transmitter to send a direct transfer trip signal to the relay scheme at the remote
line terminal. Receipt of the trip signal from the remote line terminal also initiates line breaker
tripping.
This scheme provides high speed tripping at both line terminals for all faults within the protected
line section under most conditions. However, it will not provide tripping for faults beyond the
reach of one of the RU functions if the remote breaker is open or if the remote channel is
inoperative. If only one communications channel is used at each terminal, security may be
jeopardized because any erroneous output from the channel initiates an instantaneous breaker
trip. For this reason, this scheme is often applied with dual channels where both outputs must
provide a TRIP signal to initiate a breaker trip. Or a slight delay may be added to a single
channel output to ensure that the remote trip signal is valid before tripping the breaker.
Time-delayed overreaching back-up tripping functions that do not interface with the
communication scheme are usually added to trip the associated line breaker for faults beyond
the reach of the RU functions when the remote breaker is open, or when the communication
channel is inoperative.
This scheme may use virtually any communication media that is not adversely affected by
electrical interference from fault generated noise or by electrical phenomena, such as lightning,
that cause faults. Communication media that use a metallic path are particularly subject to this
type of interference, and must, therefore, be properly shielded, or otherwise designed to provide
an adequate communication signal during power system faults.
35 / 172
Bkr 1
Bkr 2
Protected Line
Teleprotection Equipment
RU
TX
RO
RX
&
OR
TRIP Bkr 1
TX
RU
RX
RO
TRIP Bkr 2
OR
&
Protection Equipment
RU - underreaching trip function, must be set to reach short of remote terminal
and must overlap in reach with RU at remote terminal
RO - overreaching trip function, must be set to reach beyond remote end of line
Figure 3.1-11:
When the underreaching relay elements detect a fault, they trip the local breaker directly and
key a TRIP signal to the remote line terminal. Unlike the Intertripping Underreach Distance
Protection Scheme, the Permissive Underreach Distance Protection Scheme supervises the
received trip signal with an overreaching relay element. Communication assisted tripping occurs
only if the overreaching relay element detects a fault during the time that a trip signal is received
from the remote line terminal via the communication channel.
Because the received communication signal is supervised by the output from an overreaching
relay element, there is less concern about a false signal causing an incorrect trip. This scheme
is therefore typically applied with a single duplex communication channel. This scheme may use
virtually any communication media that is not adversely affected by electrical interference from
36 / 172
fault generated noise or by electrical phenomena, such as lightning, that cause faults.
Communication media that use a metallic path are particularly subject to this type of
interference, and must, therefore, be properly shielded, or otherwise designed to provide an
adequate communication signal during power system faults.
The overreaching (RO) relay elements often start a zone 2 timer to provide time delayed tripping
for faults outside the reach of the underreaching (RU) relays elements if the communication
channel is inoperative.
3.1.2.3 Permissive Overreach Distance Protection
The Permissive Overreach Distance Protection scheme requires only overreaching relay
functions. Phase distance functions are used almost exclusively for detection of multi-phase
faults, whereas ground distance functions or directional ground overcurrent functions can be
used for the detection of ground faults. The scheme is usually applied with an active duplex
communication channel that transmits a GUARD signal during quiescent, or unfaulted,
conditions. The transmitter is keyed to a TRIP signal when the associated overreaching relay
element detects a fault within its reach. Basic logic for the Permissive Overreach Distance
Protection scheme is shown in Figure 3.1-12.
RO
RO
Bkr 1
Bkr 2
Protected Line
Teleprotection Equipment
RO
TX
RX
&
RO
TX
RX
TRIP Bkr 1
TRIP Bkr 2
&
Protection Equipment
RO - overreaching trip function, must be set to reach beyond remote end teminal
Figure 3.1-12:
For a fault anywhere on the protected line, both of the RO functions operate and assert one of
the inputs to the logic AND (&) gate. At the same time, RO also keys the transmitter TRIP
signal. Receipt of the TRIP signal at each terminal, and an output from the RO function,
satisfies the logic AND (&) gate to produce a TRIP output to the breaker. For external faults, the
RO functions at only one end of the line will operate, so communication assisted breaker
tripping is not initiated at either terminal.
The scheme is very secure in that it does not trip for any external fault if the channel is
inoperative. Conversely, the scheme is lacking in dependability because it will not trip for any
internal faults if the channel is inoperative. The scheme also will not trip for any fault if the fault
is not detected at all terminals of the line. The scheme may not trip at high speed for close-in
37 / 172
faults at the strong terminals because the fastest tripping time that can be expected is
dependent on the slowest function to operate for an internal fault. Some means must be used
to key the transmitter at an open breaker if tripping is to be initiated for faults seen at the other
terminals. Breaker auxiliary contact switch keying with echo logic is commonly used to provide
this requirement. Time-delayed back-up tripping can be provided because the scheme uses
overreaching functions. Because the GUARD signal is transmitted continuously, the channel
can be monitored on a continuous basis.
This scheme may use virtually any communication media that is not adversely affected by
electrical interference from fault generated noise or by electrical phenomena, such as lightning,
that cause faults. Communication media that use a metallic path are particularly subject to this
type of interference, and must, therefore, be properly shielded, or otherwise designed to provide
an adequate communication signal during power system faults.
3.1.2.4 Accelerated Underreach Distance Protection
Basic logic for the Accelerated Underreach Distance Protection scheme is shown in Figure
3.1-13. This scheme requires the use of underreaching relay element functions (RU) that can be
extended in reach by the receipt of a TRIP signal from the relay scheme at the remote line
terminal. The RU functions must be set to overlap in reach to avoid a gap in their fault
detection. This generally requires the use of ground distance functions for the detection of
ground faults, whereas phase distance functions are used for the detection of multi-phase
faults. The scheme is often applied with an active communication channel that transmits a
GUARD signal during quiescent, unfaulted conditions, and is keyed to a TRIP signal when the
associated RU function detects a fault within its reach.
Extended RU
RU
RU
Extended RU
Bkr 1
Bkr 2
Protected Line
Teleprotection Equipment
RU
TX
TX
RX
RU
RX
Extend RU
Extend RU
TRIP Bkr 2
TRIP Bkr 1
Protection Equipment
Figure 3.1-13:
For an internal fault within the overlap zone of the RU functions, breaker tripping is initiated
directly at both line terminals and each communication channel is keyed to the TRIP signal.
Receipt of the TRIP signal extends (accelerates) the reach of the RU functions to beyond the
remote line terminal. This reach extension has no further affect because breaker tripping has
38 / 172
already occurred at each line terminal. For an internal fault near one terminal, the RU function
at that terminal operates, tripping the breaker and keying its transmitter to the TRIP signal.
Receipt of the TRIP signal at the other terminal extends the reach of that terminals RU function,
which then detects the fault to initiate tripping. For external faults, none of the RU functions
operate, therefore tripping does not occur at either line terminal.
The scheme is more secure than the Direct Underreach Distance Protection scheme because it
does not trip directly on receipt of a trip signal. Conversely, it is slower than the Permissive
Underreach and Overreach Distance Protection schemes because it must wait for the extended
RU function to detect the fault before tripping. As mentioned before, it also requires a special
relay with zone extension capability.
This scheme may use virtually any communication media that is not adversely affected by
electrical interference from fault generated noise or by electrical phenomena, such as lightning,
that cause faults. Communication media that use a metallic path are particularly subject to this
type of interference, and must, therefore, be properly shielded, or otherwise designed to provide
an adequate communication signal during power system faults.
3.1.2.5 Blocking Overreach Distance Protection
Basic logic for a Blocking Overreach Distance Protection scheme is shown in Figure 3.1-14. The
scheme requires overreaching tripping functions (RO) and blocking functions (B) as shown.
Distance functions are used almost exclusively for multi-phase fault protection, but either ground
distance functions or ground directional overcurrent functions are used for ground fault
detection. A quiescent, or OFF/ON, communications channel is typically used with this type of
scheme. The power line itself is often used as the communications medium because the
communication channel is not required when the fault is on the protected line. The
communication channel is only used to transmit a block trip signal when the fault is external to
the protected line. Audio tone over leased phone lines, microwave radio, and fibre-optic media
are also used. The transmitter is normally in the OFF state for quiescent conditions and is
keyed to the ON state by operation of any one of the blocking functions. Receipt of a signal
from the remote terminal applies the NOT or inverted input to BLOCK the trip output.
The overreaching tripping functions (RO) must be set to reach beyond the remote terminal of
the transmission line with margin so they will be able to detect a fault anywhere on the
transmission line. The blocking functions (B) are used to detect any fault not on the protected
line that the remote tripping functions are capable of detecting; so they must be set to reach
further behind the terminal than the tripping function at the remote terminal.
For a fault external to the protected line, one or more of the blocking functions operate to key its
respective transmitter to send a blocking signal to the remote terminal. Receipt of the blocking
signal blocks tripping in the event one of the tripping functions has operated for the remote fault.
The coordinating timer, TL1, is required to allow time for a blocking signal to be received from
the remote terminal. It is set to compensate for channel time, signal propagation time and for
any difference in operating time that might result if the remote blocking function is slower than
the local tripping function.
For a fault anywhere on the transmission line, one or more of the tripping functions (RO) at each
terminal will operate and apply an input to its respective AND gate (&). The blocking functions
will not operate for an internal fault, therefore neither transmitter is keyed, so that there is no
output from either receiver. The logic at each terminal produces an output that starts the TL1
timer. When the TL1 timer expires, the scheme produces an output to trip the breaker.
39 / 172
RO
B
B
RO
Bkr 1
Bkr 2
Protected Line
Teleprotection Equipment
TX
RO
RX
TL1
C
0.0
&
Simplex or Duplex
Communication Link
TRIP Bkr 1
TX
RX
RO
TRIP Bkr 2
&
TL1
C
0.0
Protection Equipment
RO - overreaching trip function, must be set to reach beyond remote end of line
B - blocking function, must be set to reach beyond overreaching trip function at remote end of line
C - Coordinating time, required to allow time for blocking signal to be received
(set equal to channel time plus propogation time plus margin)
Figure 3.1-14:
The scheme is very dependable because it will operate for faults anywhere on the protected line
even if the communication channel is out of service. Conversely, it is less secure than
permissive schemes because it will trip for external faults within reach of the tripping functions
(RO) if the channel is out of service. This scheme does not require breaker auxiliary contact or
echo logic keying when the remote breaker is open to permit tripping for faults anywhere on the
line. It provides relatively fast tripping (dependent on coordinating time delay) for most source
and line conditions. However, it may not trip weak terminals of the transmission line, if fault
levels are below the sensitivity of the tripping relays.
If quiescent (OFF/ON) communication channels are used there is no way to monitor the channel
continuously because the channel is only keyed on during external faults. A communication
channel check-back scheme is often used to periodically key a momentary block signal to check
the channel status. Some check-back schemes echo a signal back to verify that the channel is
operational in both directions. Other schemes must receive a signal within a preset time period
to declare the channel in service.
The overreaching functions can be used to drive timers so that time-delayed back-up tripping
can be provided for faults within reach of the overreaching functions.
3.1.2.6 Deblocking Overreach Distance Protection
As mentioned in some previous sections, metallic communication paths adversely affected by
fault generated noise may not be suitable for some teleprotection schemes that rely on a signal
transmitted during a protected line fault. With power line carrier, for example, the communication
40 / 172
signal may be attenuated by the fault, especially when the fault is close to a line terminal,
thereby disabling the communication channel. Multi-phase power line carrier coupling schemes
can be used to minimize this problem.
The Deblocking Overreach Distance Protection scheme includes logic specifically designed to
accommodate a loss of communication signal during the protected line fault. The Deblocking
Overreach Distance Protection scheme, like the Permissive Overreach Distance Protection
scheme, uses overreaching phase distance functions almost exclusively for multi-phase fault
detection, and ground distance or directional ground overcurrent functions for ground fault
detection. The logic requires the use of an active communication channel that transmits a
GUARD signal during quiescent, or unfaulted, conditions, and is keyed to a TRIP signal when
the associated overreaching relay element detects a fault within its reach. To overcome the loss
of signal caused by the internal line fault, deblocking logic permits a TRIP output if the loss of
signal occurs at nearly the same time the overreaching relay function(s) detect a fault. A
tripping period is controlled by a timer that is typically set between 150 and 300 milliseconds.
Basic logic for the Deblocking Overreach Distance Protection scheme is shown in Figure 3.1-15.
RO
RO
Bkr 1
Bkr 2
Protected Line
Frequency Shift Power Line
Carrier Communication Link
GUARD
OR
TRIP
TX
RO
RX
TRIP
LOG
GUARD
OR
TRIP
&
OR
TRIP
Bkr 1
TRIP
Bkr 2
OR
&
T
0.0
&
&
TX
RO
RX
TRIP
LOG
0.0
RO - overreaching trip function, must be set to reach beyond remote end teminal
LOG - Loss of GUARD detection from receiver, RX
T - deblocking time delay, typically set for 150 to 300 milliseconds.
Figure 3.1-15:
If the signal loss is due to a fault on the protected line, at least one of the overreaching trip
functions (RO) will be picked up. Thus, tripping will be initiated when the deblocking output is
produced. If none of the permissive trip functions are picked up, the channel will lock itself out
150 - 300 milliseconds after the signal is lost and will stay locked out until the GUARD signal
returns for a pre-set amount of time. It is important to understand that this logic requires that the
loss of signal associated with the operation of an overreaching relay element must only be
caused by a fault on the protected line. Loss of signal due to external line faults will cause false
trips. Therefore, the Deblocking Overreach Distance Protection Scheme Logic is used almost
exclusively with power line carrier communication.
41 / 172
Bus A
BP-A
Id-A
CBFP
-A-L1
BP-B
CB-A-L1
CB-B-L1
CBFP
-B-L1
Id-B
Line 1
Notation :
Bus-A is section A of the bus.
CB-A-L2
CB-B-L2
CBFP
-B-L2
Line 2
b
CBFP
-A-T
CB-A-T
CB-B-T
CBFP
-B-T
Transf
c
Figure 3.2-1:
42 / 172
43 / 172
Bus A
Bus B
BP-A
Id-A
CBFP
-A-L1
CB-A-L1 e
Line 1
CBFP
-A-L2
CBFP
-A-T1
CBFP
-B-L1
CB-L3-L4
CB-A-L3
CB-B-L4
Line 4
CB-T1-T2
CB-B-T2
T2
d
Line 3
CB-A-T1
Id-B
T1 is transformer 1.
T1
Figure 3.2-2:
CB-B-L2
Line 2
c
Notation :
BP-B
CB-L1-L2
f
CBFP
-B-L2
CBFP
-B-T2
CBFP-A-F1 is circuit
breaker failure protection
for breaker A on feeder 1.
Feeders may be lines,
transformers or any other
feeder.
Id-A is current differential
protection for bus zone A.
44 / 172
45 / 172
Bus A
BP-A
BP-B
I-A-L1
Id-A
I-B-L1
CB-L1
Id-B
Line 1
Notation :
Bus-A is section A of the bus.
T1 is transformer 1.
CB-L1 is circuit breaker for line 1.
Coupler-A-B
CB-T1
Figure 3.2-3:
T1
46 / 172
Short or long
overhead
power line
Step-up
transformer
Stop
Generator
Figure 3.3-1:
Transformer protection
&
Generator protection
~
Generator protection
47 / 172
Lockout
Intertripping
Telecommunication
system
Power line
protection
Power line
protection
Transformer
protection
Trip
Trip
Figure 3.3-2:
Transformer protection
Intertrip
Telecommunication
system
Intertrip
Reactor
Reactor
Reactor
protection
Reactor
protection
Figure 3.3-3:
Reactor protection
48 / 172
beginning in power transmission lines, busbars, transformers, and so on. If faults cannot be
cleared by a main protection, however, successive operations are needed in the forms of
backup, multi-zone and/or system-wide protections.
Time after
fault occurrence
Local backup
protection
System stabilising
protection
Remote backup
protection
Main protection #2
(Redundant or backup)
Wide-area or
system-wide
protection
Multi-zone
protection
Main protection #1
Protected zones/area
Figure 3.4-1:
Relationship between protected area and operate time with respect to protection
schemes
49 / 172
Figure 3.4-2:
Zone-2 of the distance protection at A must cover the entire power line from A to B. Zone-2 of
the distance protection at A must not reach beyond zone-1 of the distance protection at B.
Zone-2 of the distance protection at A backs up the distance protection at B. However, this is
true for only one part of the power line from B to C. Zone-3 of the distance protection at A
provides back-up for the rest of the power line from B to C. We have to delay the tripping from
zone-3 of the distance protection at A more than the tripping from zone-2 of the distance
protection at B, direction C.
Splitting protection for busbar using communication for multi-circuit multi-terminal line
For the configuration of double busbar and double circuit transmission lines, if a fault persists
due to a CB failure or main protection failure, separation of the busbar by using splitting
protection before remote back-up operation is effective to prevent interruption. However, for
multi-terminal lines the splitting protection is done by sequential tripping and the operation time
may not be coordinated with remote back-up operation. Figure 3.4-3 gives a sample application
for three-terminal transmission lines.
Figure 3.4-3:
Splitting protection (BD) using telecommunications for multi-circuit and multiterminal line. Ry, CB and Td denote operating times of relay (30 ms) and CB (40
ms) and time delay for coordination, respectively.
50 / 172
Splitting protections operate sequentially from (1) to (2), and then to (3). Therefore, it cannot be
coordinated with the operation time (340 ms) of remote back-up protection (zone 2 of distance
relay) of substation A, and remote back-up tripping occurs at the substation A end on both lines,
which results in the isolation of substation C. To prevent such isolation, splitting protection at
substation C performs transfer tripping of busbar CBs at substations A and B using
communication channels. Busbars at substations A and B are separated within 260 ms which
allows coordinating of the remote back-up protection. Lit.: [4].
Coordination time control using communication
For the configuration where long distance transmission lines adjoin short distance transmission
lines, coordination between remote back-up protections may not be achieved. Figure 3.4-4
gives an example where there is a long distance transmission line between substations A and B
and a short distance transmission line between substations B and C. If fault F1 occurs at the
busbar in substation B, zone 2 of distance relay of substation A may operate. Zone 2 of
substation A cannot be coordinated in the standard zone-2 time setting of 270 ms. In this case,
the time setting needs to be changed from 270 ms to 370 ms, which is equivalent to the
operation time of zone 3. However, there is another problem that remote back-up operation
(zone 2) of the substation B is delayed for the busbar fault F2 at substation B. In order to
accelerate the operate time, the splitting protection operation signal is sent from substation B to
substation A by a communication channel, and the operation time of zone 2 in substation A is
shortened to 270 ms. Lit.: [4].
Figure 3.4-4:
51 / 172
connected to that busbar. The data are transmitted by the terminal equipment to the central
equipment through data communication channels. As this protection scheme has to be installed
for every busbar, one central equipment may cover two or more protected areas or busbars. For
the double-busbar and double-circuit configuration shown in the figure, the conventional backup protection firstly performs busbar splitting protection to prevent the interruption of a sound
circuit or isolation of transmission lines, which leads to longer operate time, wider outage area
or isolated transmission lines. When a fault occurs at F in the busbar in substation B, for
example, and if the busbar protection fails to operate, the conventional remote back-up
protections or distance relays of substation A and C operate in zone-2 after the bus-tie splitting
protection operates to prevent disruption of the sound circuits. The wide-area back-up
protection operates to minimize the outage area, which is the same as the main protection in
this case. Therefore, the operate time is 140 ms shorter than the conventional protection and
the outage area is smaller. Lit.: [6].
Figure 3.4-5:
The wide-area current differential protection system requires wide-area timing synchronization
for simultaneous current sampling. As some current differential multi-terminal line protections
employ centralized timing synchronization scheme in their telecommunication circuits, a similar
scheme may be applied to such wide-area protections. More terminals, however, lead to the
complexity of achieving total synchronization among the terminals using telecommunication
circuits. Satellite-based wide-area timing synchronization such as GPS may be an alternative
solution. Since back-up protections are initiated after a main protection operated, delays for
transmitting current data and tripping signals are not necessarily crucial, while timing
synchronization and data integrity and reliability are still important.
52 / 172
Since this kind of wide-area protection system employing telecommunications can collect
various kinds of power system data simultaneously sampled throughout the area, further
sophisticated power system monitoring, control, protection, and restoration could be achieved.
53 / 172
Figure 3.4-6:
Another system stabilizing protection, which consists of central processing unit, fault detecting
unit, and transfer trip unit, operates for stabilizing power frequency or transient instability [11].
The central processing unit collects data on generated energy and load as controllable
quantities from power stations and substations, automatically recognizes electric power system
connections, performs calculations in advance to prepare for faults, and automatically
determines control quantity and objects for each pattern of separation. If a fault occurs, the
central processing unit sends a trip signal to the transfer trip unit based on the calculation
results.
The fault detecting unit detects a route disconnection fault in any of the EHV lines, and
calculates power flow through the main lines, frequency, and voltage drops, and transmits these
data to the central equipment. If faults occur, the transfer trip unit receives a transfer trip signal
from the central processing unit, and sheds the generators and/or loads as controllable
quantities based on the received information to stabilize the frequency of each separated part.
A means of high-speed multiplexed data transmission of large volume of information is essential
to a power stabilizing system that provides adaptive approach at high speed. A dedicated
transmission unit is used for the important information such as fault or route-off detection and
transfer trip signals to ensure high-speed and reliable transmissions, while relatively large
volumes of information that do not necessarily require high speed are transmitted by an
EMS/SCADA communication network. The overall operate time and transmission time
requirements are similar to the previous system stabilizing protection.
54 / 172
The telecommunication link should have a high degree of availability and should transmit the
protection signal as fast as necessary to the remote station with the highest possible reliability.
The actual requirements on transmission speed, dependability for wanted operation and
security against unwanted operation may vary for different protection schemes and line
configurations. Practical and economical reasons may define which type of transmission
medium has to be used.
All communication systems are subject in varying degrees to interference and noise of various
forms. These can corrupt the information arriving at the receiver, either by simulating a signal
when no such signal has been transmitted, or by delaying or blocking a true signal. In analog
systems, there are many ways in which transmission can be degraded. For example, the signalto-noise ratio may be poor, or the signal may suffer distortion or crosstalk from one user to
another, or the system may clip the input signal. In comparison, a digital system has the
parameters: bit rate, error rate, delay, and delay variation. Channel impairments may result in
bad messages, no messages, excessive message delay, excessive message delay variation
and/or excessive delay difference in the transmit and receive direction.
The quantity of information per unit time (bits per second) which a communication channel can
transfer depends on its bandwidth and on the received signal quality1 (normally expressed as
Signal-to-Noise Ratio, SNR).
The signal transfer delay introduced by the medium is normally low for terrestrial links, since in
most media the signals propagate typically at speeds between 60% and almost 100% of the
speed of light in vacuum. The propagation delay is for example about 3.3s/km for open-wire
(e.g. Power Line Carrier) and microwave radio links, about 5s/km for optical fibres and 5 ....
10s/km for pilot wires. The significant part of the overall operating time of a teleprotection
system is normally introduced by the terminal equipment including their interfaces to the
protection, by intermediate repeater stations and network node devices with channel routing
functions. See also Figure A1-5 in ANNEX A1.
Transmission time delay, bandwidth and signal quality are important parameters when
considering the design of a telecommunication system used for protection. The criteria apply
equally for both analogue and digital communication systems. For digital systems it is however
1
The maximum information flow that a communication channel can transfer without errors is called its capacity. According to
Shannon's law the channel capacity is given by the formula C = B x ld(1 + SNR), with C = Capacity in bits/s, B = Channel
bandwidth in Hertz, ld = logarithm to the base of 2, and SNR = Signal-to-Noise Ratio of the received signal. The channel capacity
is a theoretical value that can only be approximated at the cost of excessive signal transfer delay due to infinite coding- and
decoding efforts. The formula also indicates that a bandwidth related data rate increase is compromized by the bandwidth related
SNR deterioration (=> the wider the bandwidth is, the more noise is captured).
55 / 172
more convenient to use the data rate instead of the bandwidth and to express the transmission
quality in terms of bit errors (e.g. bit error rate, errored seconds etc.) rather than signal-to-noise
ratio.
Circuit or service providers however may offer circuits or services with guaranteed performance,
which seem to be applicable to protection.
56 / 172
Important parameters for analogue circuits are the bandwidth, the frequency response ( =
attenuation and group delay) and the signal-to-noise ratio at the receiver input. Analogue
circuits are characterized by their graceful degradation under disturbed channel conditions, i.e.
the quality of the received signal deteriorates gradually with increasing disturbance and noise.
Important parameters for digital circuits are the bit rate, the data transfer delay and delay
variation (timing jitter) and the error rate of the received data. The impairments of the
communication medium, which may be quite severe, are hidden from the user. Error rates can
be bounded to very low values by placing regenerative repeaters at periodic intervals
(intermediate stations) along the physical medium.
Digital circuits are characterized by their threshold behavior under disturbed channel
conditions. Simply speaking, they are either very good or not available.
The protection of power systems normally imposes very stringent demands on the
communication system regarding its real-time properties. The signal transfer time and transfer
time variation is for example much more critical for protection signal transmission than for
general data or voice communication.
Voice frequency circuits
The term voice frequency (VF) circuit is used for analogue circuits that pass frequencies
between approximately 300 Hz and 3400 Hz and block frequencies outside this range.
Historically this frequency range has been defined for the transmission of speech signals.
Today, analogue voice frequency circuits are provided by both analogue and digital
telecommunication systems and may be used by voiceband modems for data transmission up
to approximately 33 kbit/s.
Voice frequency circuits may further be characterized according to the number of wires that are
required: 2-wire circuits employ the same wire pair for transmitting and receiving, whilst with 4wire circuits one wire pair is used for transmitting and the other wire pair is used for receiving.
57 / 172
shares an aggregate with other services. Switching, routing and multiplexing techniques bear a
certain risk that a transmitted signal is directed to a receiver which it was not originally intended
for (channel cross-over or signal loop-back).
The consequences of signal misdirections on protection are different for analog comparison and
state comparison protection schemes. For example, signal misdirection is typically less critical
for state comparison protection systems that are normally in the guard (non-operate) state, than
for analog comparison protection systems, which depend on continuous exchange of
information between line ends and may immediately trip the line if signals from the wrong
transmitter are received.1
In addition to inadvertent signal misdirection there exist some other network-related risks that
are new or have a different impact when compared to traditional hard-wired point-to-point links,
for example:
-
Automatic re-route to some non-defined alternate path with inadequate performance for
protection
Automatic re-route to a pre-defined non-preemptible (dedicated) path
Excessive outage time until re-route completed
Different propagation time delays between the various paths selected
Possibility of different go and return propagation time delays
Protection circuits may be bumped at the expense of others when re-routing after a link
failure, unless prevented by adequate circuit priority rating mechanisms
Channel may not revert back to its original path unless manually optimized, eg on a least
cost basis
Unacceptable signal transfer delays due to queuing mechanisms in networks with
dynamic bandwidth allocation
Availability may be less than expected due to the particular definition of "Available Time"
for telecom ISDN circuits according to ITU-T G.821
58 / 172
tendency is to replace pilot wires by optical fibres which are free from electromagnetic
interference, their use may still be justified for economical reasons.
Pilot wire circuits may be utility owned or rented from telecommunication companies. Utility
owned pilot wires often follow the same route as power cables. Since pilot wires may be subject
to dangerous induced voltages during faults in the primary system, appropriate installation
precautions must be taken in order to prevent maloperation, hazards to personnel and damage
to equipment.
The electrical parameters of pilot wires such as signal attenuation and signal delay per unit
length depend on their mechanical parameters like wire diameter, insulation material and cable
construction, as well as on the signal frequency. Values for the attenuation coefficient between
0.5 dB/km and 3 dB/km in the audio frequency range are typical.
The signals, which are transmitted over pilot wires are historically sometimes DC signals or
signals at power frequency (50 or 60 Hz) e.g. from pilot wire differential relays. Transmission of
DC or AC signals at power frequency is hardly used any more due to the pronounced
susceptibility to interference from the primary system. Normally the information is modulated
onto a carrier which shifts the information from the power frequency range into the audio
frequency range for transmission. At the receiving end, the information signal can be separated
from the power frequency by means of filtering. This function is usually performed by means of
a teleprotection equipment operating over a 2- or 4-wire circuit.
2-wire circuits use the same pair of wires for transmitting and receiving. Transmit and receive
signals are normally separated by their respective frequencies. With 4-wire circuits, a pair of
wires is allocated to the transmitter and a pair of wires is allocated to the receiver. The same
frequency is normally used for transmitting and receiving.
A typical application for pilot wires is the transmission of binary on/off protection commands
using dedicated teleprotection equipment in conjunction with distance or directional comparison
relays. The protection command is modulated onto an audio frequency carrier somewhere in the
0.3 kHz to 3.4 kHz range, which makes the transmission less susceptible to power frequency
interference and high frequency noise. The teleprotection equipment may also multiplex several
commands from different relays onto the same wires.
Internet access and multimedia services had a tremendous impact on the development of new
high speed transmission principles for copper wires. High-speed modems would allow the use
of pilot wires for higher data rates, ranging from several tens of kbit/s for voiceband modems to
up to 10 Mbit/s over short distances for wideband (xDSL) modems. However, due to their
inherent high signal transfer delay (latency) and their sensitivity to channel disturbances, the
use of high-speed modems is not recommended for the transmission of protection signals.
Electromagnetic interference, power frequency harmonics and wideband noise produced by
faults in the power system are likely to block the modem receiver just in that moment when the
communication is truly needed.
59 / 172
Summary
Advantages and disadvantages of Pilot Wires as related to protection signal transmission are:
Advantages
Disadvantages
Wide deployment
Table 4.3-1:
60 / 172
the transmission of protection commands the speech and data signals may be switched off such
that the maximum transmit power is available to the protection signal. This "boosting" of the
protection signals compensates the additional signal attenuation, which is introduced by the
fault on the protected power line when signalling over faulty lines. As voice and data are
interrupted during protection signalling, boosting is not recommended for the transmission of
persistent commands, as might be the case for reactor protection for example.
Emerging Digital PLC systems translate serial digital data into one or several 4 kHz slots at
carrier frequency range using bandwidth efficient digital modulation techniques, such as
quadrature amplitude modulation (QAM) or multicarrier (MCM) modulation. The serial aggregate
data may accommodate digitized speech and digital data by time-division-multiplexing (TDM,
see 4.4.1.2). Protection command transmission is usually accomplished by means of a
dedicated subsystem in order to achieve the required dependability and minimum signal
transfer delays under faulty line conditions.
For both analogue and digital PLC systems the signal at carrier frequency is amplified to
typically 5 to 100 Watts output power (PEP, peak envelope power) and coupled to the power
line via an impedance matching device and a high voltage coupling capacitor. For optimum
transmission performance under faulty line conditions coupling onto two phases in push-pull
mode is normally preferred. Line traps in series to the power line prevent the carrier signals from
being shunted by the local busbar and prevent signal leakage to adjacent lines. Several PLC
terminals may share a common coupling equipment.
The propagation of the signal along a multi-conductor power line may be explained by the
combined transmission of independent modes whose number is equal to the number of nonearthed conductors above ground. Each mode propagates with its specific attenuation and
velocity. The signal attenuation depends on the construction of the power line, the line condition
and on the carrier frequency used. It is typically in the range of 0.02 dB/km to 0.2 dB/km,
increasing with frequency.
The signal quality may be impaired by various noise sources.
Corona noise results from electric impulse discharges along the surface of the phase
conductors. Its spectrum extends well into the carrier frequency range. Corona noise is always
present on an energised line and is perceived as background noise in a PLC receiver. Its level
depends on the power system voltage and design, the climatic conditions and the altitude above
sea level. It normally does not constitute a problem to protection signal transmission since its
level is less than other channel impairments caused by line faults for example.
Isolator operation creates high frequency noise of high amplitudes which cause poor signal-tonoise ratios in the PLC receiver. Its duration may last some seconds depending on the isolator
design. The signal quality degradation depends largely on the method of coupling and on the
characteristics of the equipment. The interference produced by isolator noise is most severe in
comparison with other noise sources. Because it occurs under healthy line conditions it may
cause unwanted operation of the protection system.
Operation of breakers produces disturbance similar to isolator operation. Its duration is
however limited to the operating period of the circuit breaker which is typically less than 20 ms.
During line faults the PLC channel is subject to strong transient noise at the onset of the line
fault until the arc has established, followed by an immediate increase in signal attenuation due
61 / 172
to the short circuit of the faulty phase(s). During the interruption of the fault current, noise is
produced again by the operation of the circuit breakers.
Interference produced by power system faults occurs during the time when the protection is in
active operation. It may therefore prevent operation of the protection system.
Channel impairments during line faults are the primary reason why PLC links have so far been
restricted for two applications for protection signal transmission, where they have however
proven to perform most effectively:
- Transmission of binary status information in conjunction with distance protection or
directional comparison relays in state comparison protection schemes
- Transmission of phase comparison signals in conjunction with phase comparison relays
in analog comparison protection schemes.
Summary
Advantages and disadvantages of PLC links as related to protection signal transmission are:
Advantages
Disadvantages
Channel
is
subject
to
increased
disturbance during faults in the primary
system
Table 4.3-2:
62 / 172
From a legal point of view terrestrial microwave radio systems may be broadly categorized into
licensed radio and unlicensed radio systems:
Licensed radio systems operate in a "protected" frequency range that has been licensed to the
utility by national authorities. Licensed radio systems typically constitute point-to-point multichannel links that are used in the backbone of the communication network.
Unlicensed radio systems operate in an "unprotected" frequency range that is open to the
public. Unlicensed radio systems usually share a common frequency band and support point-tomultipoint communication. Access to the shared medium (i.e. the common frequency band) is
accomplished either through TDMA (Time Division Multiplex Access) or CDMA (Code Division
Multiple Access), or a combination of the two, to prevent mutual interference between users.
See also chapters 4.4.1.2 and 4.4.1.3.1
4.3.3.1 Multichannel radio
Although the microwave equipment may be owned and operated by the power utility, the
frequency bands for its use have to be licensed from national authorities. The frequency bands
for microwave radio systems are typically between 400 MHz and 40 GHz.
In legacy analog microwave systems a number of voice baseband channels with 4 kHz
bandwidth each are combined onto a single aggregate signal by frequency division multiplexing
(FDM, see 4.4.1.1). One or several of the 4 kHz baseband channels may be used individually or
collectively for the transmission of protection signals.
Earlier analogue microwave systems used frequency modulation (FM) where the analogue
aggregate FDM signal varies the frequency of the emitted carrier. Analogue microwave systems
are mainly of historical interest since they have been gradually replaced by digital systems.
In digital microwave systems a number of digital data channels of typically 64 kbit/s each are
combined onto an aggregate data stream using time division multiplexing (TDM, see 4.4.1.2).
Analogue signals such as speech are converted into digital data prior to multiplexing. One or
several of the digital 64 kbit/s channels may be used either individually or collectively for the
transmission of protection signals.
In digital microwave systems frequency modulation has been replaced by phase shift keying
(PSK) modulation or combined phase-amplitude shift keying, which is also called quadrature
amplitude modulation (QAM), with 16-QAM being widely used today. Higher level QAM like 64QAM or 128-QAM provide a higher bandwidth efficiency, i.e. they allow to transmit more bits per
second in a given bandwidth, however at the expense of an increasing susceptibility against
interference and noise.
Licensed microwave radio links are normally point-to-point with maximum distances between 40
and 100 km. The distances that can be covered depend on the transmitter output power, on the
frequency band used, on atmospheric conditions, on the topography and on obstacles, which
may impede signal propagation or cause signal reflections. Signal reflections may lead to
multipath propagation which causes a certain additional attenuation or signal extinction when
the direct wave and the reflected wave are opposite in phase at the receiver.
Waves reflected by the ionized part of the atmosphere or by a changing refractive index due to
temperature or humidity variations have the same effect, but since the degree of reflection is
subject to a random process, the received signal varies statistically with time. This phenomenon
is called fading as the received signal can fade or disappear on a statistical basis.
1
TDMA and CDMA are general media access technologies that are for example typically used in point-to-multipoint radio systems
where many users share a common frequency band.
63 / 172
These drawbacks can be overcome by careful link planning, by the positioning of intermediate
repeaters, by the selection of the transmitted power and by the antenna design.
Diversity is commonly used to improve the availability of microwave links by adding some
degree of redundancy. Space diversity is obtained if different antennas are located in different
positions on the antenna tower. Frequency diversity is used when the same signal is transmitted
using different frequencies. At the receiving side the signals coming from different antennas are
combined to achieve the best possible signal at the output, permitting to limit the outage time of
the link during the worst period of the year.
4.3.3.2 Single channel radio
Unlicensed radio systems normally constitute single-channel point-to-multipoint short-haul links
that can be set up on the fly at moderate cost. Single-channel point-to-multipoint microwave
radio links have been used at MV level with less stringent demand on signal transfer times.
Access to the common medium - i.e. the shared radio frequency band - is accomplished by
means of TDMA (Time Division Multiple Access) or CDMA (Code Division Multiple Access) to
prevent mutual interference between transmitters operating simultaneously, see also 4.4.1.3.
Successful operation of such systems has been reported from South Africa. Both intertripping
as well as differential protection signals are transferred over point-to-point TDMA-based singlechannel radio links between outstations, with typical signal propagation delays (outstation to
outstation) in the range of 14ms to 22ms1 at data rates of 19.2kbit/s and 64kbit/s.
The use of unlicensed spread spectrum radio for the transmission of state indication in a state
comparison scheme for a 138 kV line has been reported from the USA.
Summary
Advantages and disadvantages of microwave radio links as related to protection signal
transmission are:
Advantages
Disadvantages
Caution: It is most important to mention that the upstream and downstream time delays between masterstation and outstation
are different with certain types of TDMA equipment this can lead to difficulties with differential protection systems: An additional
dummy outstation may have to be installed at the master-station site for the sole purpose of equalizing the go and return time
delays!
64 / 172
Table 4.3-3:
65 / 172
66 / 172
Summary
Advantages and disadvantages of optical fibre links as related to protection signal transmission
are:
Advantages
Disadvantages
Table 4.3-4:
4.3.5 Satellites
The race for satellite communication has been on ever since the announcement of pocket-sized
ground terminals to provide a truly global mobile telephone service. At present, there are many
different satellite systems that have been proposed to complement terrestrial communication
networks, all at varying developmental stages. Narrowband satellite systems which carry many
voice or low speed data channels - up to 9'600 bits per second - are more advanced in terms of
development than wideband systems supporting SDH and ATM (see Chapters 4.4.2.2, 4.5.1
and 4.6.2.2 on SDH and ATM). The reason is mainly due to new or more acute issues related to
creating broadband satellite links with QoS (Quality of Service) guarantees (ANNEX A3).
Projects have been launched worldwide to investigate the integration of terrestrial wideband
networks with satellite networks.
Satellites are usually classified according to the type of orbit they are in.
4.3.5.1 GEO - Geosynchronous Earth Orbit Satellites
GEO satellites are placed in the orbit such that their period of rotation exactly matches the
Earths rotation, i.e. they appear stationary from earth. Earth station antennas do therefore not
need to move once they have been properly aimed at a target satellite in the sky.
Today, the majority of satellites in orbit around the earth are positioned in GEO at 36000 km
orbital height. It is at the precise distance of 36000 km that a satellite can maintain an orbit with
a period of rotation exactly equal to 24 hours.
Due to the long distance of 36000 km GEO satellites experience long up-down signal
propagation delays of about 250 ... 280 ms which normally excludes them from being used as a
communication medium for protection signal transmission, with perhaps few exceptions for
wide-area protection applications.
67 / 172
68 / 172
Summary
Advantages and disadvantages of satellite links as related to protection signal transmission are:
Advantages
Disadvantages
Table 4.3-5:
Terminal addressing will also protect against protection maloperation when signals are (inadvertently) looped back for testing or
maintenance reasons.
69 / 172
information signal may be either analog or digital, the combined FDM signal is inherently an
analog waveform. FDM is therefore primarily used with analogue transmission systems.
Wavelength Division Multiplexing (WDM)
With optical fibre systems, a special form of FDM called WDM (Wavelength Division
Multiplexing) is increasingly being introduced to further exploit the huge capacity of optical
fibres. Several transmission systems, each using a different wavelength or 'colour', may be
stacked onto the same fibre using WDM. In its simplest form, WDM uses different optical
windows for the multiplexing, e.g. the windows centred around 1300 nm and 1550 nm
wavelength. More sophisticated systems multiplex a number of optical channels (e.g. 4, 16, 32
or 64) within the same optical window centred around 1550 nm wavelength. As the spacing
between the different wavelengths becomes very narrow in this case, the technology is called
Dense Wavelength Division Multiplexing (DWDM). An in-depth treatment of WDM technology is
found in [12].
As WDM actually creates 'virtual fibres' it may also be employed for the de-coupling of
transmission systems from each other. Dedicated teleprotection links that operate quasi-isolated
from other telecom services could be realized using WDM for example: In Figure 4.4-1, system
1 consists of a protection relay with internal or external teleprotection function plus a fibre-optic
transmitter/receiver operating at wavelength 1. System 2 could be any other fibre-optic
communication system operating at wavelength 2 and carrying other services such as data and
voice. A failure or maloperation of System 2 should not adversely affect System 1, as the only
common parts of the two systems are the optical fibre and the passive optical wave-division
multiplexer / demultiplexer.
Although the isolation of the teleprotection from other services by means of WDM appears
attractive from an operational point of view, it may not be easily justified for cost reasons.
System 1
1
System 1
1+2
System 2
Figure 4.4-1:
Optical fibre
WDM
WDM
System 2
Figure 4.4-1 shows a simplex (i.e. unidirectional) communication for simplicity reasons. Full-duplex (i.e. bi-directional) operation
would require either a second fibre, or a 3rd and a 4th wavelength (3 and 4 respectively) on the same fibre.
70 / 172
Most modern telecommunication systems employ some form of TDM for transmission over longdistance routes. The multiplexed signal may be sent 'directly' (called 'baseband' transmission)
over optical fibres, or it may be modulated onto a carrier signal for transmission over analogue
media, such as microwave radio or coaxial cables for example.
TDM can be split into various subclasses. The most important are Fixed TDM and Statistical
TDM.
Fixed TDM
In fixed TDM - sometimes also called synchronous TDM - each channel has its assigned
timeslot which sustains a fixed data rate and uses aggregate bandwidth irrespective of actual
user data being transmitted or not. The number of channels is normally equal to the number of
timeslots in a frame. Due to the fixed allocation of channels and timeslots, data can always be
transmitted. Buffering and flow control are not required. Continuous data flow at a fixed bit rate
without delay variations is ensured, a condition which is a prerequisite for protection signal
transmission.
Statistical TDM
Statistical - sometimes also called asynchronous TDM - multiplexers rely on the bursty traffic
characteristics of certain information sources. Data may be transmitted in any timeslot as long
as there are free slots available. Relying on the statistics of the data, the number of channels or
the peak data rate which is supported by the statistical multiplexer may be larger than the total
number of timeslots or the aggregate data rate in a frame. Data buffering and flow control is
employed to store and withhold data until a free timeslot or free cells become available.
Buffering and flow control introduce extra delay as well as delay variations, and data may be
discarded in case of overload. Loss of information is normally not acceptable for protection
signal transmission. Statistical multiplexing has therefore to be avoided unless the required
quality of service is explicitly guaranteed.
A multiplexing technology which was originally intimately bound up with the emerging SDH
(Synchronous Digital Hierarchy, see 4.4.2.2) standards is ATM (Asynchronous Transfer Mode,
see also 4.5.2.3 and 4.6.2.2) which was conceived as a way in which arbitrary-bandwidth
communication channels could be provided within a multiplexing hierarchy consisting of a
defined set of fixed bandwidth channels. ATM multiplexers support both constant bit rate (CBR)
and variable bit rate (VBR) traffic, where CBR which basically emulates fixed TDM is a
prerequisite for todays protection systems using telecommunication.
4.4.1.3 Code Division Multiplexing (CDM)
In CDM, several signals share a common medium (copper wires or radiowaves) using the same
frequency band simultaneously. Multiplexing of different channels is achieved by utilizing
different pseudorandom binary sequence codes that modulate a carrier. The process of
modulating the signal by the code sequence causes the power of the transmitted signal to be
spread over a larger bandwidth. Systems based on CDM are therefore sometimes also referred
to as 'Spread Spectrum' (SS) systems. The spreading of the spectrum enhances the noise
immunity of such systems.
CDM and in particular CDMA (code division multiple access) is mainly used with unlicensed
spread spectrum radio where many simultaneous users have to share the same frequency
band. CDM/SS techniques may also be used with wire-based systems to enhance the
transmission capacity and noise immunity. Its application for inter-substation communication
would however need to be further examined with respect to cost efficiency and transmission
performance.
71 / 172
The use of unlicensed single channel CDMA/SS radio for the transmission of protection
commands for a 138 kV line has been reported from the USA. However, no practical
installations of CDM/SS technology for the transmission of protection signals over copper wires
have been reported. A final conclusion on CDM is not possible at the time of writing.
Europe
North America
Japan
64 kbit/s
64 kbit/s1
64 kbit/s1
2048 kbit/s
1544 kbit/s
1544 kbit/s
8448 kbit/s
6312 kbit/s
6312 kbit/s
34368 kbit/s
44736 kbit/s
32064 kbit/s
139264 kbit/s
139264 kbit/s
97728 kbit/s
Table 4.4-1:
When multiplexing a number of digital signals with the same nominal bitrate they are likely to
have been created by different pieces of equipment each generating a slightly different bitrate
due to their independent internal clocks. A technique called bit stuffing is used for bringing the
individual signals up to the same rate prior to multiplexing. Dummy bits or justification bits are
inserted at the transmit side and discarded by the demultiplexer at the receiving end, leaving the
original signal. The same problem with rate alignment occurs at every level of the multiplexing
hierarchy. The process of multiplexing almost synchronous signals is called plesiochronous,
from Greek. The use of plesiochronous operation throughout the hierarchy has led to the
adoption of the term Plesiochronous Digital Hierarchy.
Plesiochronous operation does not allow extracting and inserting individual channels from the
aggregate without prior demultiplexing and subsequent re-multiplexing, leaving towers of
multiplexers. With the exception of vendor specific solutions, network management and
performance monitoring throughout the hierarchy is not adequately supported with PDH
systems either, as PDH systems have developed over time with insufficient provision for
standardized management. These disadvantages have - amongst others - finally led to the
definition of a new digital transmission hierarchy: the Synchronous Digital Hierarchy.
1
72 / 172
STM-1
STM-4
SONET
Aggregate Rate
Optical
Carrier level
Synchronous
Transport Signal
level
Aggregate Rate
Max. number of
simultaneous voice
channels
(informative)
OC-1
STS-1
51.840 Mbit/s
783
155.520 Mbit/s
OC-3
STS-3
155.520 Mbit/s
2'349
622.080 Mbit/s
OC-12
STS-12
622.080 Mbit/s
9'396
STM-16
2488.320 Mbit/s
OC-48
STS-48
2488.320 Mbit/s
37'584
STM-64
9953.280 Mbit/s
OC-192
STS-192
9953.280 Mbit/s
150'336
Table 4.4-2:
73 / 172
Although SONET and SDH were conceived originally for optical fibre transmission, SDH radio
systems exist at rates compatible with both SONET and SDH.
SDH/SONET network topologies and network resilience
A synchronous network will be more reliable than PDH due to both the increased reliability of
individual elements, and the more resilient structure of the whole network. SDH will allow
development of network topologies which will be able to achieve 'network protection', that is to
survive failures in the network by reconfiguring and maintaining service by alternate means.
Network protection can be accomplished by the use of cross-connect functionality to achieve
restoration, or through the use of self-healing ring architectures.
Two main types of synchronous ring architectures have been defined:
- The Dedicated Protection Ring - This is a dedicated path switched ring which sends
traffic both ways around the ring, and uses a protection switch mechanism to select the
alternate signal at the receive end upon failure detection.
- The Shared Protection Ring - This is a shared switched ring which is able to provide
'shared' protection capacity which is reserved all the way around the ring. In the event of
a failure, protection switches operate on both sides of the failure to route traffic through
the reserved spare capacity.
The ability to share protection capacity in shared protection rings can in many instances offer a
significant capacity advantage over dedicated protection rings. This means, in economic terms,
less equipment, lower cost and less operation efforts. However, this is at the expense of a
slower restoration time than a path switched ring.
Protection switching in a ring topology can be either "uni-directional" or "bi-directional". Unidirectional means that only the faulted path is reverted along the ring by selecting the healthy
fibre at the receiving end, whilst the non-faulted path follows the original route.
With bi-directional protection both the go and return path are switched to follow the opposite
direction along the ring.
It is noted that only bi-directional protection will maintain equal signal propagation delays for the
go and return path, whilst uni-directional protection may introduce unequal propagation delays
that may cause severe difficulties for current differential protection relays!
The synchronous ring structure, with its inherent resilience, is a powerful building block from
which survivable networks can be built:
A typical power system control network has a radial (star) topology, with point-to-point links
connecting a central control station with associated substations. SDH/SONET network
implementations may connect the substations in rings. The logical star connection is achieved
by configuring the channels within the SDH/SONET network in order to provide the required
logical point-to-point links. In case of a path interruption, signal flow may be reversed along the
ring such that communication is sustained.
More about SDH network design and -operation is found in chapter 4.6.2.1 of this document.
SDH/SONET for power system protection
Since SDH/SONET networks provide a set of fixed bandwidth channels with a deterministic
transmission characteristic, they are well suited for applications that rely on the transmission of
a sustained fixed data rate and short signal transfer times, as needed by differential current
protection for example. As SDH/SONET signals follow a fixed physical path through the
network, SDH/SONET channels will exhibit a fixed transmission delay with low delay variations
74 / 172
or "jitter" unless paths are re-routed automatically or manually due to network failures. Transmit
and receive directions may however still experience different signal propagation times when the
physical route does not follow the same path. Provisions to accommodate the non-equal signal
transfer times have to be built into the protection relay in this case, see also Chapter 6.3.1.1.
In conclusion, transport networks based on SDH/SONET technology can be designed to meet
the stringent requirements of legacy and future protection systems regarding signal transfer
times and error characteristics. Propagation velocity of the light pulses in optical fibres is around
200 km/ms, signal transfer delays between ports of an SDH/SONET node are typically well
below 1ms, and networks are designed to produce very low error rates (<< 10-6) under normal
operating conditions. Issues that are more critical to the operation of the protection scheme are
related to network management and network security, e.g., the impact of path re-routing on
transmission time variation and on circuit availability. These are however primarily matters of
network planning and network operation.
More about the subject on wideband transport technologies and networks can be found in [13]
and [2].
75 / 172
S E R V IC E
ACCESS
P O IN T S
S E R V IC E
LAY ER
TRA NS PO RT
LAY ER
Figure 4.5-1:
Network Architecture
The transport layer takes care of the communication between geographically separated sites
providing the transport service for the telecommunication service network and so connecting the
different parts of the whole service network.
Depending on the geographical distribution, the transport layer could form a mesh of channels
connecting different sites, or be reduced to the actual links that connect the switching nodes of
the service network.
The service layer performs all the necessary functions to offer the final service to the user. It is
formed by switching nodes connected by means of dedicated links or throug channels of the
transport network.
Service networks can be classified according to the technology on which they are implemented.
They can also be classified by the service they provide; nevertheless, this classification has
became obsolete due to fact that modern networks are designed to integrate a wide variety of
services.
Service networks could be structured, depending on the size and geographical distribution, in
access and core networks. The access network connects end-users to the closest core node
whereas the core network performs transit functions in order to set-up a connection between
access nodes thus establishing the final end-to-end connection.
Figure 4.5-2 shows the main components of a complete network. It can be seen that transport
functions are placed in the bottom layer as they provide the basic interconnection functionality
to the upper functions, the networking layer. This layer is formed by a set of functions that are
responsible for the delivering of the final service to the end user.
76 / 172
NETWORK SERVICES
SIGNALLING
ADDRESSING
NETWORKING
SWITCHING
ROUTING
PDH/SDH
TRANSPORT
PHYSICAL MEDIA
Figure 4.5-2:
Network components
It is important to comment that these components should interact among themselves to offer the
final networking service so that every node of the service network has to be able to support
those functions and work in a collaborating way so that a global performance could be
achieved.
77 / 172
channel in equal shares among its users by assigning a part of the time to every user.
Synchronous Digital Hierarchy SDH is the newest transport technology. It is used for high
capacity transport applications, being the transport technology in which the Broadband ISDN (BISDN) is based. See also chapter 4.4.2.
Application Programs
Application Programs
Application Layer
Application Layer
Presentation Layer
Presentation Layer 6
Session Layer
Session Layer
Transport Layer
Transport Layer
Network Layer
Network Layer
Physical Layer
Physical Layer
Medium
Figure 4.5-3:
Medium
78 / 172
Every layer has a well-defined functionality and provides a service to the upper layer in the
model. The functions carried out by every one of the OSI model layers are:
1. Physical layer. The responsibility of the physical layer is to transmit unstructured bits of
information across a link. It deals with the physical aspects such as the shape of connector,
pin assignment, etc.
2. Data link layer. The responsibility of the data link layer is to transmit the information across a
link. It deals with error detection and correction, information alignment and addressing when
several system are reachable as in LANs or multipoint links.
3. Network Layer. The responsibility of the network layer is to enable the communication
between any pair of end system in the network. The network layer deals with the route
calculation function, congestion control, etc.
4. Transport layer. The responsibility of the transport layer is to establish a reliable
communication stream between a pair of End systems. It deals with the detection and
correction of the errors introduced by the network layer, such as packet lost or duplicated,
reordering of out-of-order information, etc.
5. Session layer. The responsibility of the session layer is to co-ordinate the way data are
transferred throughout the communication provided by the transport layer.
6. Presentation layer. The responsibility of the presentation layer is to adapt, when necessary,
the different internal data representation format used by the End system that are transferring
information.
7. Application layer. The responsibility of the application layer is to deliver the communication
service to the application that is using the service provided by the network.
Service networks can be implemented by using different technologies. Every technology in
general could be more suitable to offer some services. Since the design of modern network is
focused towards service integration, only those technologies that can be able to offer service
integration are being considered for future designs. Nevertheless, we are going to mention not
only future trends but also existing technologies owing to their capabilities to support the related
applications.
4.5.2.1 Circuit Switched Networks (POTS, ISDN)
Circuit switched networks are connection oriented networks. The establishment of a connection
requires a call set-up that chooses a path in the network in which the necessary resources to
support the connection are reserved.
Resources are allocated to a connection whilst this connection is maintained, even though they
were not used. Only when the connection is released will the resources be liberated.
Circuit switched networks can be based on analogue transport technology, on digital transport,
whether PDH or SDH, or on a mixed configuration. Due to the fact that a circuit of constant bitrate, usually 64 kbit/s, is used to support the connection, a deterministic delay performance is
achieved in the final service offered. On the other hand, since every connection established in
the network is based on the use of 64 kbit/s channels, when a connection is used for the
transmission of information with a lower bit-rate poor resource efficiency is obtained, unless
sub-multiplexing techniques are applied.
Narrow-Band ISDN is the ITU-T standard for Digital Circuit Switched Networks. Switched and
permanent connections can be established. The integration of services is limited to connectionoriented constant bit-rate service types.
79 / 172
80 / 172
CBR
DTE
IWF
IWF
DTE
ATM
Figure 4.5-4:
Figure 4.5-4 shows the reference model of the CES. We can distinguish three main
components: The Data Terminal Equipment (DTE), the Internet-Working Function (IWF) and the
ATM network. The DTEs are the actual users of the service, a protection relay or a
teleprotection equipment providing 64 kbit/s or N times 64 kbit/s for instance. The IWF provides
the conversion of the bit-stream generated by the users into cells and the reconstruction of the
original bit-stream at the reception side including timing recovery and jitter removal. This
function is usually embedded in the ATM access device. Finally, the ATM network should
provide a Constant Bit-Rate (CBR) virtual channel that should have been dimensioned with the
fixed bandwidth required to carry the bit-stream provided by the users.
When Nx64 kbit/s working mode is selected, cross-connect (DXC, Digital Cross-Connect)
functionality can also be provided, thereby being possible to deliver every single 64 kbit/s
channel to different locations in a similar way as PDH cross-connect devices do.
Since the CES has to offer quality performance similar to a PDH/SDH connection, it has to fulfil
the requirement of the related standards. Therefore, a CES has to comply with ANSI T1.403
and ITU-T G.824 for jitter and wander performance of digital networks that are based on the
1544kbit/s hierarchy, and ITU-T G.823 for networks that are based on the 2048 kbit/s digital
hierarchy, see also Table 4.4-1. Other facilities related with the data format and structure such
as framing, alarm transmission, loops, etc, should comply with the relevant standards already
applied to PDH/SDH connections.
The Bit Error Rate (BER) of the emulated channel should comply with the ITU-T G.826
recommendation for E1 (2048 kbit/s) and the ANSI T1.510-1994 for DS1 (1544 kbit/s), or ITU-T
G.821 for lower bit rates, e.g. 64 kbit/s.
CES could find a direct application to connect existing protection relays or digital teleprotection
equipment to ATM networks without the need of a specific implementation or external
adaptation devices. Nevertheless, no practical experience of using this approach has been
reported until the moment of writing this document.
4.5.2.4 Datagram Networks (IP)
The traditional network concept we have discussed so far is based on the circuit-switched
approach. Each connection is associated to a circuit that has resources allocated for its
exclusive use along a path. There is no uncertainty about the bandwidth or delay along this path
so the Quality of Service in terms of bandwidth and delay can be guaranteed.
Datagram networks have introduced a very different mode of operation. Network resources
(bandwidth, buffers, etc.) are statistically shared among their users. This presents many
advantages for computer communication applications, since data traffic tends to be bursty so
that resource reservation would lead to low utilisation levels. In datagram networks, data
81 / 172
packets are delivered to the Network without any resource allocation, and the network exerts its
best effort to serve the packets.
Two different working modes can be distinguished in data networks, Hop-by-Hop control, used
in Virtual Circuit Networks and End-to-End control used in Datagram Network.
In the first approach, a connection is set up in the network, so that every intermediate system
involved in this connection change its internal state. Every node takes care of every packet and
guarantees its transmission towards its destination. This scheme suffers from a side effect
known as fate-sharing: The End-to-End connection depends on the state of all the
intermediate systems involved in this connection. If any of those systems fails, the connection
will be lost.
In the datagram approach, the End-to-End connection does not depend on the state of any one
of the intermediate systems of the network. If one of the intermediate systems fails, the
information will be routed using another path so that the final users will not be aware of this
change. This scheme increases the overall availability of the network with this effect being more
important for bigger networks.
In the End-to-End scheme, the responsibilities are shared between the network and its users.
The Network is responsible for the routing whereas the users are responsible for the control of
the communication. Thanks to this approach, the Datagram networks present an unmatchable
resilience level as well as the best resource optimisation. These characteristics make them
suitable for mission-critical applications such as most of the applications that can be found in
the Power Utility Control Network environment.
Datagram networks using IP (Internet Protocol) cannot assure the QoS as the network presents
a non-deterministic transmission delay. Then it cannot be applied to delay sensitive applications
such as teleprotection, unless some specific Quality of Service mechanisms were added in
order to guarantee bandwidth and/or delay. Refer to chapter A3.3 in ANNEX A3.
The great flexibility of this type of networks makes them suitable for service integration.
Although they cannot intrinsically offer a constant bit-rate service, thanks to a new application
protocol that has been defined (Real-time protocol) it is possible to eliminate the delay variation
at the application level. However this comes at the expense of an additional delay that may not
be acceptable for protection.
82 / 172
Topology: The way network devices are connected. Straight-line bus, ring, and star are
the most common arrangements.
Protocols and media contention: The rules and encoding specifications for data
interchange and for the administration of the shared medium.
Physical media: Devices can be connected by twisted-pair cable, coaxial cable, fibre or
wireless.
4.5.3.1 Topology
As shown in Figure 4.5-5, there are three basic topologies used in LANs:
-
Bus topology. All devices are connected to a central cable, called the bus or backbone.
Bus networks are relatively inexpensive and easy to install for small networks.
Ring topology. All devices are connected to one another in the shape of a closed loop,
so that each device is connected directly to two other devices, one on either side of it.
Ring topologies are relatively expensive and difficult to install and maintain, but they
offer higher bandwidth than bus topology and can span large distances.
Star Topology. All devices are connected to a central hub. Star networks are relatively
easy to install and manage, but bottlenecks can occur because all data must pass
through the hub.
S ta r
R in g
B us
Figure 4.5-5:
LAN Topologies
The market trend in terms of topology is going towards the star topology since both hubs and
switches have enough capacity to cope with all the traffic that can be generated in a LAN. Star
topology has been adopted to support internal substation communication in the new UCA
architecture.
Bus topology is a cost-effective solution for small LANs but due to the fact that is less reliable
than star topology, it is not recommended for substation applications.
Although ring topology is conceptually the best approach, its lack of flexibility and scalability has
put it aside of the main innovative applications though still maintains a considerable market
83 / 172
share.
4.5.3.2 Media Contention and Protocols
LANs are based on the use of a shared medium to connect all the users. In this environment,
only one user can transmit at a time. Some mechanism must therefore exist to allocate
bandwidth among users in such a way that:
- Each user gets a fair share of the bandwidth
- Each user gains access to the medium within a reasonable amount of time
- The bandwidth used for arbitration be minimized.
The two most popular bandwidth arbitration mechanisms used on LANs are token schemes and
contention schemes.
In a token scheme, a user can send a piece of information when it has the token. The token is
circulated from user to user.
In a contention scheme, every user can send at will when it sees the channel in idle condition.
When two or more users transmit at the same time a collision occurs. This situation is resolved
by means of contention mechanisms that have a probabilistically fair behavior.
The CSMA/CD (Carrier Sense Multiple Access with Collision Detection) working principle which
is used by the Ethernet LANs has proved to be the most efficient and flexible contention
scheme able to adapt to different speeds and physical media.
Every different aspect of LAN networks, from physical specifications to protocols, has been
developed by the IEEE 802 committee and later adopted by international standardization
bodies. Protocols involved in the data interchange in LANs are confined to the OSI data link
layer. Figure 4.5-6 shows the protocol stack of the Ethernet LAN and its relation with the OSI 7
layers reference model. The 100 Mbit/s Ethernet protocol stack has been shown since this is the
standard adopted for the internal substation communication.
OSI
7- Layer
reference
Model
IEEE
CSMA/CD
Model
Application
Higher Layers
Presentation
Session
Transport
Network
Media Independent
Interface MII
Data Link
Physical
Media
Physical
100 Mbit/s
Figure 4.5-6:
84 / 172
100BASE-TX that uses a 2 pairs UTP (Unshielded Twisted Pairs) category 5 cable in a
point-to-point arrangement (Star topology)
100BASE-FX that uses an optical fibre (2 strands) in a point-to-point arrangement (Star
topology)
85 / 172
that is to say, a group of users connected in any physical LAN segment can be logically
associated so that they work as if they were connected in a single LAN segment with
broadcast functionality. Between different Virtual LAN segments attached to the same
device, bridging functionality is provided.
Most of the off-the-shelf LAN access/control devices include Hub, Bridging and Switching
functions. These devices allow the network manager to define several virtual LANs that share
the same physical infrastructure and set when hub-, bridging- or switching functions have to be
used in order to optimize the LAN performance. When the LAN serves a site of a Wide Area
Network, routing functionality can also be included in most of these devices, being then possible
to forward packets to any other site of the same Wide Area Network.
86 / 172
that the digital path supporting the protection communication complies with the ITU-T standards
for the Quality of Service of digital connections the dependability will depend on the
performance of the protocol used by the protection communication device (i.e. teleprotection
function).
Availability of a communication path can be defined as the ability of this path to perform its
required function at any instant of time within a given time interval, assuming that the external
resources, if required, are provided. It can be expressed as:
A=
MUT
MDT + MUT
Where MDT is the Mean Down Time and MUT is the Mean Up Time.
In general, it is widely accepted that the availability objective for a Telecommunication Service
for protection should reach the 99,99%. This availability level, as can be seen from the abovementioned expression, is quite difficult to achieve with a single link so that a back-up link should
be considered.
When a network implements the communication path, other protection measures can be
included in order to improve the availability of the service without devoting several links for
every protection device. Among these protection measures such as recovery or self-healing
mechanisms are the more relevant issues to be considered. When these protection procedures
are implemented in a network, other functional components such as routing and addressing
should also be analyzed in order to prevent collateral effects that could affect the protection
service performance.
Out of the above mentioned concepts directly related with the protection service, there are other
requirements at the network level that should also be studied as they can influence the
protection service operation.
Service isolation and service prioritization facility is another important issue to be considered
in the design of a network for protection communication:
Modern digital networks are used to integrate different types of services. This is a common
practice in transport networks as the aggregation of different traffic services allows a more
efficient use of high capacity links such as fibre-optic cables or microwave radio links.
Integration of services implies that different users are sharing the network resources. Under
these conditions, the network should provide mechanisms in order to guarantee the isolation
among different users. By providing this, every user has a guarantee of use of their assigned
resources and so it can expect a guaranteed level of availability.
On the other hand, networks integrating mission-critical services should provide a priority
mechanism that allows the implementation of different availability levels. The priority mechanism
should guarantee that in case of lack of resources due to an outage situation or network
congestion, the service with a higher priority will always be able to use the remaining network
resources and so maintain the expected service availability.
Network Security is a very important aspect to be considered in the design of a network that
offers communication service to protection applications:
Network Security includes different aspects that all together provide the necessary means for
87 / 172
secure and reliable operation of the network. The goal of the security functions is to assure that
any other user of the network cannot interfere with, by any means, the proper operation of the
protection system. That is to say, the network should withstand any security attack coming from
any other user of the network whether on purpose or by mistake.
The security aspects of the Network Management Centre are standardized by the ITU-T and the
ISO. The operational details are discussed later in this chapter.
Among others, it should be mentioned the implementation of checking procedures to assure
that the parties connected by a path belong to the same type of user such as protection
devices of the same electrical subsystem, etc. This procedure could prevent a wrong
establishment of connections. The control of misconnection or misinsertion of information due to
a failure in any network component is also a relevant functionality.
88 / 172
89 / 172
The BER in an ATM link depends only on the characteristics of the physical media used as no
error detection or correction techniques are applied to the user data ( = payload). Nevertheless,
the BER could be considered as negligible due to the wide use of highly reliable links, mainly
based on fibre-optics.
However, the BER experienced by the protection service will be affected by the Cell Loss Rate
CLR (a QoS parameter, see A3.2 in ANNEX A1), which depends on the scheduling, and Call
Admission Control (CAC) algorithms used in the network.
The use of ATM networks to communicate protection devices requires a careful analysis of the
End-to-End delay of the service. The performance of an ATM connection in terms of delay
depends on the cell scheduling policy in the switches as well as on the Call Admission Control
algorithms implemented in the network. It is important to find out if the scheduling algorithms of
the switches are able to isolate protection information from the rest of the traffic of the network,
so that the possible interference among different traffic flows that could drive to an uncontrolled
End-to-End delay can be prevented.
Network topology should be designed taking into account the availability of low-delay
communication paths between protection devices. Routing algorithms should also be analyzed
in order to find out if they implement QoS routing. This facility is a necessary piece to guarantee
an End-to-End bounded delay.
The availability of the protection communication service can be improved if the network
implements some recovery mechanism that can use the spare capacity of the network.
Recovery mechanisms are under standardization process in the ATM-Forum. The great
advantage of ATM networks is that they can choose the back-up route in function of the QoS
requested by the service user. Thanks to this, the back-up path will always comply with the
requested End-to-End delay. The design of the network should take into account that in order to
protect the communication path used by the protection service the network must have some
spare capacity, and the topology design must include some physically disjointed routes with low
transmission delay.
Synchronization is not a critical issue in ATM networks. As they work in asynchronous mode,
the synchronization is carried out at link level and no synchronization plan is needed for the
whole ATM network. Nevertheless, the transport of Constant Bit Rate (CBR) signals and the
interconnection of PDH/SDH network throughout ATM network using the Circuit Emulation
Service (CES) requires a clock synchronization plan to be deployed.
Due to the fact that ATM networks are based on statistical multiplexing, the emulation of a
transmission circuit will require the following aspects to be taken into account:
-
The two former requirements are achieved by the proper ATM service specification whereas the
last one is not under the control of the ATM network.
Jitter and wander reduction as well as clocking facilities will have to be provided by the InterNetworking Function (IWF), see Figure 4.5-4.
90 / 172
Since the ATM network cannot transmit timing information suitable for network synchronization,
the IWF has to include the clock recovery functions. There are two methods to recover clock
information: the Synchronous Residual Time Stamp (SRTP) and the Adaptive Clock.
SRTP requires a common reference clock to be available at every access point of the ATM
network providing a transport service, whereas the adaptive method does not require any
external clock. Due to this fact, some wander may be introduced by the adaptive method
thereby not being recommended to use it in emulation of circuits that support the
communication of analog comparison protection devices.
The ATM Forum and the ITU-T have standardized this service. In the network design phase, the
synchronization plan should be carefully analyzed; taken into account that ATM network should
not be used to transport timing references and therefore, a network-wide reference clock will
have to be provided.
ATM networks present intrinsic service isolation. Every different service integrated in an ATM
network is based on a Virtual network; this network is formed by a set of Virtual Paths (VP).
Cells belonging to a VP cannot be delivered by any means to another VP.
Flow isolation in the same virtual network is also an intrinsic characteristic of ATM networks.
Cells belonging to a due flow are identified by a particular Virtual Connection Identifier. These
cells cannot be delivered by any means to another connection.
Security aspects in ATM networks should be considered in the design phase in order to
guaranteed the integrity of the Protection information transported by the network. Despite the
tremendous complexity involved, the signalling network can be used to perform a malicious
attack disturbing the proper operation of any service if no counterpart measures are
implemented. The ATM Forum is defining a common framework of security in ATM networks.
This work will produce a set of standards that will improve the security and offer robust
protection to any external security attacks.
4.6.2.3 IP Networks
IP networks are packet switched networks that work in Datagram mode. The End-to-End
connection does not depend on the state of any one of the intermediate systems of the network.
If one of the intermediate systems fail, the information will be routed using another path so that
the final users will not be aware of this change. This scheme dramatically increases the overall
availability of the network, however at the expense of a non-deterministic delay.
Due to the fact that IP networks are based on statistical multiplexing and network access
techniques, they present a non-deterministic delay.
The quality of service, in terms of delay and packet loss, depends not only on the network
working principles but also on the offered traffic load. In the actual network implementations, the
End-to-End delay is not guaranteed. Nevertheless, if the network has been properly engineered
and includes a priority scheme, the delay experienced by the information with higher priority can
be limited to an acceptable value. In this way, the protection information flow is isolated from the
91 / 172
92 / 172
5 TELEPROTECTION INTERFACES
Four types of interfaces are commonly used for protection relaying:
- Contact based interfaces
- Analogue interfaces
- Electrical serial data interfaces
- Optical fibre interfaces
In the near future, Ethernet interfaces will likely be introduced. not only for intra-station
communication using LANs, but also for inter-station communication across a WAN.
The type of interface depends strongly on whether the teleprotection device is a separate
equipment or whether it is an integrated function of the protection relay. Little has been done in
the standardization area in respect of surge protection of copper-based Ethernet/LAN circuits
that operate in the electromagnetic hostile environment of power stations and substations, as
this technology has mainly been deployed in the office environment.
Interface co-ordination has to be ensured regarding:
- Type of interface (applicable standard)
- Data rate (digital) or bandwidth (analog)
- Signals to be used
- Signal flow direction
- Electrical insulation requirements
- EMC requirements
- Cable screening and signal ground connections
- Connector design and pin/signal allocation
And in particular for digital circuits:
- Clock provisioning for synchronous operation
- Low level data formats (asynchronous data format, synchronous operation)
- Data flow control
Unless all interface parameters are properly co-ordinated between devices, proper operation of
the protection scheme cannot be expected. The checklist in chapter 7.2.2 may serve as a
guideline.
93 / 172
94 / 172
ITU-T
V.11
Operating mode
RS 232c/d/e
RS 422a
Data rate
(typical)
up to 38.4kbit/s
asynchronous
V.24/V.28
EIA
up to 38.4kbit/s
RS 485
up to 38.4kbit/s
------
RS232c/d/e
up to 38.4kbit/s
V.11/X.24
G.703
co-directional
Table 5.3-1:
RS 422a
synchronous
------
up to 64kbit/s
or
n*64kbit/s
(n=1...32)
RS 485
up to 64kbit/s
or
n*64kbit/s
(n=1...32)
-----
n*64kbit/s;
n=132
Electrical
isolation
Not part of the
standard;
Electrical isolation
requires special
design
Not part of the
standard;
Electrical isolation
requires special
design
Not part of the
standard;
Electrical isolation
requires special
design
Not part of the
standard;
Electrical isolation
requires special
design
Not part of the
standard;
Electrical isolation
requires special
design
Not part of the
standard;
Electrical isolation
requires special
design
yes
The IEEE is presently (2000) developing a standard for nx64kbit/s relay-to-multiplexer communication.
The German VDE is presently (2000) developing a DIN/VDE standard for protection relay communication protocols.
95 / 172
to find suitable methods of providing noise immune methods of interconnecting services and
providing power supplies to the interface converter units within the substation environment.
Many electro-optical converters and DC power filters are available and offer advanced solutions
to the well-known problems experienced by protection and telecommunication engineers. The
majority of the units are proprietary and are only compliant when used in conjunction with the
approved proprietary protection solutions.
It is essential that the convergence of Substation control and Protection philosophies address
the need to have common international standards for protocols and interfaces to meet the
needs of standard substation network build.
96 / 172
topology and requires the use of a Hub equipped with a port for every device connected to the
LAN. The length of the cable from the attached device to the Hub can range up to 100 m.
Due to non-compliance with EMC requirements and test severity levels that have been
developed for protection relay I/O circuits, like the Fast Transient Test for example, the 10BaseT
interface should only be used with care for protection relaying, or when EMC requirements have
been agreed between user and manufacturer.
10BaseFL. - 10 Mbit/s Fibre Optic Media Ethernet
The 10BaseFL interface is a point to point connection that provides complete electrical isolation.
As in the previous case, a Hub (see 4.5.3.3) is required to interconnect every device to the LAN
segment. While Ethernet interfaces used in metallic media segments has protection circuit
designed for medium level electrical hazards, fibre optical interface provides total immunity from
electrical discharges including the effect of lightning strikes and ground current that can be
found in substations and specially in connection between different buildings of the same
substation. Another advantage of this type of interface is the distance it can span ranging up to
200 m from the host to the Hub.
The interconnection with the optical fibre requires a specific Media Attachment Unit (MAU) that
can be build into the Ethernet card or externally provided. In the latter case, the external unit is
connected to the Ethernet interface using the Attachment Unit Interface (AUI) which is based on
a 15 pin connector and can range up to 15 m.
100BaseTX - 100 Mbit/s Twisted-Pair Ethernet
The 100BaseTX interface operates over two pairs of wires, one pair is used for receive data
signals and the other pair is used for transmit data signals. The most popular wiring used today
is the Unshielded Twisted-Pair (UTP) cable though shielded twisted-pair cable can also be
used. The two wires of each pair must be twisted together for the entire length of the segment.
The two pairs are connected by means of an eight-pin RJ-45 type connector. This interface is a
point to point interface intended for star topology and requires the use of a Hub (see 4.5.3.3)
equipped with a port for every device connected to the LAN. The length of the cable from the
attached device to the Hub can range up to 100 m when a specific UTP Category 5 cable is
used.
As for previous similar metallic interfaces, this type of interface should only be use for wiring
inside a building.
Due to non-compliance with EMC tests and test severity levels that have been developed for
protection relay I/O circuits, the 100BaseTX interface should only be used with care for
protection relaying, or when more moderate EMC requirements have been agreed between
user and manufacturer.
100BaseFX - 100 Mbit/s Fibre Optic Media Ethernet
The 100BaseFX interface is a point to point connection that provides complete electrical
isolation and immunity to EMI. As in the previous case, a Hub (see 4.5.3.3) is required to
interconnect every device to the LAN segment. This interface uses two fibres, one for reception
and other for transmission. The fibres are connected to the device using SC, ST or FDDI type
fibre-optic connector.
97 / 172
IEEE802.3
Ethernet
Data rate
(Mbit/s)
Max. segment
length (m)
Medium
Topology
10Base2
10BaseT
10BaseFL
100BaseTX
100BaseFX
10
10
10
10
100
100
500
185
100
100
50-Ohm coax
(thick)
50-Ohm coax
(thin)
Unshielded
twisted-pair
wire
Optical fibre
pair
Bus
Bus
Star / Hub
Star /Hub
Table 5.3-2:
100
(UTP category
5 cable)
Unshielded
twisted-pair
wire
Star / Hub
100
Optical fibre
pair
Star / Hub
Warning:
LANs with Bus/Star/Hub topologies exhibit the risk of "single point-of -failures", unless
appropriate precautions like redundant LANs or some other form of protection against single
point-of-failures are implemented.
98 / 172
99 / 172
Table 6.1-1 and Table 6.1-2 merely permit the comparison of various systems and highlight
some critical issues that need to be addressed when designing a protection system using
telecommunications.
The requirements in Table 6.1-1 and Table 6.1-2 focus on digital systems. In numerical relays,
the teleprotection equipment - which caters for the data integrity - may be integrated as a
function into the protection relay. This is mainly the case for analog comparison protection
relays. Interface (a) is then not directly accessible for the user since it is relay internal. Thus,
interface (b) in Figures A1-1 to A1-5 in ANNEX A1 is of particular interest in the context of this
document. Requirements for the more traditional interface (a) are also given in the table, as
these are relevant for dedicated teleprotection equipment, or when the teleprotection function is
integrated into the telecommunication terminal equipment. Where possible, references to
existing standards are made.
6.1.1.1 Terminology and General Requirements
The basic requirements are summarized in Table 6.1-1 and Table 6.1-2. Terms and specific
requirements related to these tables are explained below in some detail. Definitions are found in
Chapter 6.1.1.2.
Propagation time
Signal propagation time across a telecommunication network is one of the most critical
parameters, as it should be kept to a minimum to ensure that a circuit can be tripped as quickly
as possible. Protection may therefore specify a maximum acceptable value measured at
reference point (b), see the figures in ANNEX A1. Validation of the actual value may be
performed for each link as part of the commissioning tests. The delay should also be measured
with the circuit manually switched to its alternate route(s).
Propagation time symmetry (differential delay)
Propagation time symmetry - i.e. equal propagation times for transmit- and receive direction - is
normally not required for state comparison schemes and command-based protection systems.
However, propagation time symmetry between transmit- and receive paths is likely to be a
critical issue for differential protection systems which measure the round trip delay and assume
that the one-way delay is half of this value. If transmit and receive paths have different delays,
this assumption is not valid. Non-equal propagation times may be caused by transmit and
receive signals being switched to different paths through the network, or by data buffering and
queuing in network nodes or traffic (over)load.
Particular attention has to be paid to ring topologies where the signal may be sent in the
opposite direction around the ring in case of a communication network fault. This may lead to
unacceptable differential delays which must be prevented, for example by ensuring that
independent switching of transmit and receive paths is disabled.
Maximum allowable values for differential delays depend on the particular protection /
teleprotection equipment design and should not be exceeded by the telecommunication system
(measured at interfaces (b), see ANNEX A1).
Propagation time variation
While propagation time variation - i.e. static or transient changes in propagation time - is
normally not an issue for state comparison schemes or command-based protection systems, it
is crucial for current differential protection where synchronous samples of the power frequency
waveform need to be compared at each end of the protected line.
Static changes in propagation time may arise due to signals being re-routed to different paths
through the network. Dynamic propagation time variation (jitter) may be the result of data
100 / 172
101 / 172
Channel squelching
Although communication protocols for protection signal transmission (= teleprotection protocols)
usually provide a high resilience against transmission errors, it is advisable that
telecommunication receive circuits are muted or clamped to a predefined state (normally "all
ones") when the error rate becomes excessive for an extended period. High error rates of up to
50% may for example be found in conjunction with a link failure, or data may be inverted due to
an equipment failure. Data muting or clamping may be accomplished at the receive side of
interfaces (b) or (a) in Figures A1-1 to A1-4 in ANNEX A1.
Terminal equipment addressing
Protection systems should incorporate some form of terminal equipment addressing in order to
prevent maloperation when communication links are looped back or when a signal should be
routed to the wrong terminal. Terminal addressing is of particular importance for differential
current protection, as a misdirected signal would simulate a differential current, which would
immediately trip the line. Address validation time should therefore be shorter than the protection
relays signal processing/decision time.
The address validation time is less critical for state comparison or command-based protection
systems, as temporary signal misdirection would normally only exchange guard signals between
teleprotection terminal units, thus not producing unwanted tripping.
The addressing facility is normally implemented in the teleprotection function (Figures A1-1 to
A1-4 in ANNEX A1).
Network management and configuration
Apart from precautions against random errors in the received data or against problems related
to automatic re-routing, appropriate security measures to protect from inadvertent human-made
errors need to be installed at the network management level. Password protection or channel
locking may be used for that purpose. For example, in the case of Main 1 and Main 2 circuits
being conveyed via the same network, it has to be ensured that there is no possibility of the
circuits following the same route and causing a single contingency to simultaneously fail all the
teleprotection associated with a particular line.
Synchronization
When analogue voice-frequency (VF) bearer services or asynchronous digital circuits are used,
synchronization of network with the teleprotection signaling devices is not fundamental to the
effective operation of the service.
When synchronous digital services are required the need for secure and stable synchronization
is imperative.
In most configurations the teleprotection signaling device is provided as a Data Terminating
Equipment (DTE). For the majority of schemes the bearer network is controlled from a master
clock source, and all DTEs slave the timing from the master. On few other applications the
DTEs may be required to take on the role of master clock.
As the number of digital teleprotection bearer services increase, the need to ensure that the
network providing the connectivity for these services is part of a well defined and secure
synchronized network, becomes an essential part of the power utilities strategy.
It is important that the telecommunications engineer and the protection engineer have a clear
understanding of their equipment and its needs. Too often the service fails due to a lack of clear
and knowledgeable decisions made at the interface.
The difficulties likely to be encountered by synchronization problems are well documented in
[30]. This document should be considered as essential reading as part of this report.
102 / 172
103 / 172
Propagation time may be measured and quoted if interfaces (b) in Figures A1-1 to A1-4 are
accessible.
Transmission time
The transmission time of a teleprotection system is the time elapsed between the instant of
change of state at the teleprotection transmit input and the instant of the corresponding change
of state at the teleprotection receive output (Interfaces (a) in Figures A1-1 to A1-4), excluding
propagation time. ([27], [28]).
The nominal transmission time T0 is the transmission time measured under disturbancefree transmission conditions.
The maximum actual transmission time Tac is the maximum transmission time
encountered under disturbed transmission conditions for a defined dependability and
signal-to-noise ratio (SNR) or bit error rate (BER).
Overall operating time of a teleprotection system / Teleprotection operating time
The overall operating time TA of a teleprotection system is the time elapsed between the
instant of change of state at the teleprotection transmit input and the instant of the
corresponding change of state at the teleprotection receive output (Interfaces (a) in
Figures A1-1 to A1-5), including propagation time and additional delay due to
interference and/or noise [27], [28].
Transmission time (T0, Tac) and teleprotection operating time (TA) may be measured and quoted
if interfaces (a) in Figures A1-1 to A1-5 are accessible.
Protection operating time
The protection operating time TB is the time interval between the instant a specific set of values
of the input energizing quantities is applied under specific conditions at the input of the
protection system, including sensors or current and/or voltage transformers, and the instant
when the protection relay output circuits are operated [28].
Bandwidth
Analogue systems: The width of the frequency range used by the communication channel,
expressed in Hertz (Hz).
Digital systems: Sometimes used as a synonym for data rate.
Data rate
A measure for the information per unit time transferred across a certain reference point /
interface of a system, including any overhead. Normally expressed in bits per second.
Dependability
Dependability relates to the ability to issue and receive a valid signal (command or message) in
due time in the presence of interference and/or noise when a corresponding signal has been
transmitted [27], [28].1
For practical reasons the probability of a missing signal is normally measured. The reference
1
Dependability as defined for teleprotection does not necessarily have the same meaning to the operation of the scheme. For
example with blocking schemes, a missing blocking command is likely to occur in the presence of an external fault condition and
may cause unwanted tripping (=> lack of security of the protection scheme).
104 / 172
point for the measurement is interface (a) in Figures A1-1 to A1-5 (ANNEX A1).
A poor dependability gives rise to a failure to trip or a delayed trip in an intertripping (direct
tripping) or permissive tripping scheme. A poor dependability in a analog comparison protection
scheme may give rise to a failure to trip or a delayed trip, or may produce unwanted tripping.
Security
Security relates to the ability to prevent interference and noise from generating an unwanted
signal (command or message) at the receiving end when no corresponding signal has been
transmitted, [27], [28].1
For practical reasons the probability of an unwanted signal is normally measured. The reference
point for the measurement is interface (a) in Figures A1-1 to A1-5 (ANNEX A1).
With permissive tripping schemes, the risk of an unwanted tripping action due to inadequate
security of the teleprotection function is generally low, while in intertripping (direct tripping)
schemes each unwanted command will lead to an unwanted tripping action. A poor security in
an analog comparison protection scheme will typically produce unwanted tripping.
Data integrity
Data integrity relates to the probability that received data - that have passed certain error
detecting and/or error correcting procedures - are correct. Data integrity can be expressed as
residual error probability and may be categorized in classes, where each class is characterized
by its upper bound of residual error rates (IEC 60870-5-1). Data integrity is improved by using
protocols that detect, reject or discard data that have been corrupted due to bit errors
introduced by the telecommunication channel.
Security as defined for teleprotection does not necessarily have the same meaning to the operation of the scheme. For example
with blocking schemes, an unwanted command may lead, depending upon its duration, either to a delayed trip or to a failure to trip
(=> lack of dependability of the protection scheme).
105 / 172
Critical;
(< 10ms)
Critical;
(< 10ms)
Critical;
(< 10ms)
Less critical;
(< 30 ms)
Permissive
Underreach
Distance
Protection
Permissive
Overreach
Distance
Protection
Accelerated
Underreach
Distance
Protection
Direct
Intertripping
Command
Schemes
Table 6.1-1:
Critical;
(< 10ms)
Blocking
Overreach
Distance
Protection
Propagation
Time
Not required
Not required
Not required
Not required
Not required
Propagation
Time
Symmetry
Analog:
< 3kHz
Digital:
< 10kbit/s
Low 4
Analog:
< 3kHz
Digital:
< 10kbit/s
Low 4
Analog:
< 3kHz
Digital:
< 10kbit/s
Low 4
Analog:
< 3kHz
Digital:
< 10kbit/s
Low 4
Analog:
< 3kHz
Digital:
< 10kbit/s
Low 4
Bandwidth
or
Data Rate 1
Less critical;
(< 50ms)
Critical;
(< 25.. 30ms)
Critical;
(< 25.. 30ms)
Critical;
(< 25.. 30ms)
Critical;
Depending on
co-ordination
times
(< 25ms)
Overall
Operating Time
TA 1
High;
IEC 60834-1
High;
IEC 60834-1
High;
IEC 60834-1
Medium to high;
IEC 60834-1
High;
IEC 60834-1
Dependability
1; 2
1, 3
High;
IEC 60834-1
Medium;
IEC 60834-1
Medium to
high;
IEC 60834-1
Medium;
IEC 60834-1
Low to
medium;
IEC 60834-1
Security
Not critical to
the operation of
the scheme
Not critical to
the operation of
the scheme
Not critical to
the operation of
the scheme
Not critical to
the operation of
the scheme
Not critical to
the operation of
the scheme
Propagation
Time Variation
High;
IEC 60870-2-1
Class I3
Medium;
IEC 60870-2-1
Class I2
Medium;
IEC 60870-2-1
Class I2
Medium;
IEC 60870-2-1
Class I2
Medium;
IEC 60870-2-1
Class I2
Data Integrity
Misconnection critical when
present during fault
occurrence. Risk of nonselective tripping.
Address validation time less
critical, since system
normally in guard state
Misconnection critical when
present during fault
occurrence. Risk of delayed
tripping.
Address validation time less
critical, since system
normally in guard state
Misconnection critical when
present during fault
occurrence. Risk of delayed
tripping.
Address validation time less
critical, since system
normally in guard state
Misconnection critical when
present during fault
occurrence. Risk of missed
tripping.
Address validation time less
critical, since system
normally in guard state
Misconnection critical when
present during fault
occurrence. Risk of not
tripping and of tripping the
wrong line.
Address validation time less
critical, since system
normally in guard state
Impact of misconnection;
Need for terminal
addressing
General
Recovery
Time (nonredundant
systems) and
Availability
Transmission time, bandwidth, dependability and security are interrelated. One can for example improve one quantity at the expense of another. For example, for a fixed bandwidth, security can be improved at the
expense of transmission time and/or dependability. The particular choice depends on the actual system requirements and on practical limitations.
2
The interdependence between dependability and communication quality parameters (SNR or BER) is explained in IEC 60834-1. The figures suggest that for a BER of <10-6 the teleprotection shall not suffer a noticeable
dependability deterioriation. For a BER of 10-6 to 10-3 the teleprotection shall still be able to perform its function, although a certain loss in dependability is to be expected.
3
The interdependence between security and communication quality parameters (SNR or BER) is explained in IEC 60834-1. The figures suggest that the teleprotection shall guarantee a stated security against unwanted
operation irrespective of the actual BER or SNR (worst case scenario).
4
Although the requirement on bandwidth or data rate is low for command based protection schemes, a channel with 64 kbit/s capacity (or a multiple thereof) may be requested in digital systems for standardization reasons.
Recovery time: After a random communication failure less critical for command-based
non-unit protection schemes, as the system is normally in the guard state.
Availability: The overall system design must ensure that proper operation is not adversely
affected by the fault occurrence (electromagnetic interference)
Critical;
(< 10ms)
Critical;
(< 10ms)
Critical;
(< 10ms)4
Less critical;
(< 30ms)
Phase
Comparison
with time
stamp
Longitudinal
Current
Differential
w/o time
stamp
Longitudinal
Current
Differential
with time
stamp
Charge
Comparison
Table 6.1-2:
Critical;
(< 10ms)
Phase
Comparison
w/o time
stamp
Propagation
Time
Less critical
Less critical
Critical for
systems
based on
loop roundtrip delay
measurement
Less critical
Low to
medium;
Critical for
systems
based on
loop roundtrip delay
measurement
7.2kbit/s to
64kbit/s
n*64kbit/s
for multiterminal
lines
n*64kbit/s
for multiterminal
lines
64kbit/s
typical.
Analog:
< 3kHz
Digital:
< 10kbit/s
Typically
64kbit/s
Analog:
< 3kHz
Digital:
< 10kbit/s
Low to
medium;
Bandwidth
or
Data Rate 1
Propagation
Time
Symmetry
High;
IEC 60834-2
High;
IEC 60834-2
1 to 1.5 cycles
High;
IEC 60834-2
High;
IEC 60834-2
Dependability
Overall
Operating Time
TA 1
Medium to
high
Medium to
high;
IEC 60834-2
Medium to
high;
IEC 60834-2
Medium;
IEC 60834-2
Medium;
IEC 60834-2
Security
1; 3
Less critical;
(< 4ms)
Less critical;
(< 10ms)
Critical;
(< 0.1ms)
Less critical;
(< 10ms)
Critical;
(< 0.3ms)
Propagation
Time Variation
High
High
IEC 60870-2-1
Class I3
High
IEC 60870-2-1
Class I3
High
IEC 60870-2-1
Class I3
High
IEC 60870-2-1
Class I3
Data Integrity
Critical;
Risk of unwanted tripping;
Critical;
Risk of unwanted tripping;
Critical;
Risk of unwanted tripping;
Critical;
Risk of unwanted tripping;
Critical;
Risk of unwanted tripping;
Impact of misconnection;
Need for terminal
addressing
General
Recovery
Time (nonredundant
systems) and
Availability
Transmission time, bandwidth, dependability and security are interrelated. One can for example improve one quantity at the expense of another. For example, for a fixed bandwidth, security can be improved at the
expense of transmission time and/or dependability. The particular choice depends on the actual system requirements and on practical limitations.
The relationship between dependability and communication quality parameters (SNR or BER) is explained in IEC 60834-2. It is suggested that for a BER of less than 10-6 the teleprotection shall not suffer a noticeable
dependability deterioration. For a BER of 10-6 to 10-3 the teleprotection may still be able to perform its function, although a loss in dependability is to be expected.
The relationship between security and communication quality parameters (SNR or BER) is explained in IEC 60834-2. The figures suggest that the teleprotection shall guarantee a stated security against unwanted
operation irrespective of the actual BER or SNR (worst case scenario).
For pilot wire replacement a propagation time of less than 1ms is required.
These two types of protection system place differing demands on the communication network.
6.1.2.1 Time synchronization through GPS
Overall operating time
For a high-end protection, the overall operating time TA for a teleprotection system should be
less than 1/2 of a power frequency period. Any longer delay will adversely affect the total
operating time and fast fault clearing time of the protection.
Data integrity
The protocol used for teleprotection must detect errors before any unwanted functions occur.
Security against unwanted operation must normally be given priority over dependability.
Route switching
A protection using GPS as time base is not affected by the number and frequency of route
switching operations. However, the requirement of a maximum acceptable transmission time
must always be fulfilled to guarantee the stated performance for the protection.
Propagation time symmetry
A protection using GPS as time base is not affected if propagation times in the transmit and
receive direction are not the same, provided that the maximum propagation time is bounded
within useful limits.
6.1.2.2 Time synchronization through communication network
Overall operating time
For a high-end protection, the overall operating time TA for a teleprotection system should be
less than 1/2 of a power frequency period. Any longer delay will adversely affect the total
operating time and fast fault clearing time of the protection.
Data integrity
The protocol used for teleprotection must detect errors before any unwanted functions occur.
Security against unwanted operation must normally be given priority over dependability.
Route switching
A protection using the communication system for time synchronization can not allow too many
or too frequent route switching operations. The maximum number or frequency of routeswitches depends both on manufacturer's algorithm for time synchronization and on how much
108 / 172
the transmitting and receiving times will change due to a route switch.
If the frequency of route switching is high but the change in propagation times is very little, then
the protection will not be affected at all.
A fixed, dedicated connection would of course eliminate this problem.
Propagation time symmetry
A protection using the communication system for time synchronization can tolerate some
difference between transmit and receive propagation times. Traditional current differential relays
normally assume equal transmit and receive propagation times, as the round-trip time is
normally measured and divided by 2. Any difference in the actual times will automatically end up
in a differential current and will directly affect both dependability and security. An example on
how security is affected is given below.
If the minimum allowed differential current limit is set to 20% of rated current, and this limit is
used up to 100% of rated current before the we start to stabilize for current through the line, we
can allow up to 1.2 ms difference between transmit and receive times, because
receive time = 4.2 ms, transmit time = 3 ms
differential time = 1.2 ms, and average time = 3.6 ms
This gives a time difference for the comparison of the current vectors of 0.6ms = 10.8 degrees
in a 50 Hz system, resulting in a differential current of sin(10.8) = 0.19*Irated.
To obtain a protection with high sensitivity the difference in receive and transmit time must be 5
to 10 times less than given in the example above. This results in a maximum time difference
between the transmit and receive direction of around 0.1 - 0.25 ms only.
109 / 172
110 / 172
and if the critical common component c gets into failure, both protection systems are required,
possibly leading to a more extensive outage. Therefore, such cross dependability and security
checks among network protection and control systems for complex power systems will become
significantly important. A procedure to avoid the inappropriate overlapping of telecommunication
channels is also needed for power system stabilizing protection systems as shown in Figure
6.1-1. A failure on the overlapped link may result in a halt or failure of the protection system
operation.
Figure 6.1-1:
111 / 172
112 / 172
by the declining health of communication circuits or equipment. Bad messages can also be
good messages sent to the wrong location by faulty communication switching equipment, or
intentionally misrouted, such as in a loop-back condition performed during communication circuit
testing. No matter what the cause, the digital protective relay or associated digital
communication system interface must detect and reject or correct bad messages (including
misrouted good messages) to prevent misoperation. It is common practice to alarm or
annunciate a communication problem to alert the appropriate communication system
maintenance personnel about existing or impending communication system problems. Some
relays log communication errors to help diagnose the nature and cause of communication
system problems. At the very least, the relay should provide a time-delayed output to alarm for
the continuous receipt of bad messages or complete loss of messages. In addition, modern
digital relays can perform other logic functions to modify their performance, or the performance
of the communication system when a problem occurs.
While it is important that the teleprotection function and telecommunication system send and
receive messages between protective relays as reliably and securely as possible, it is also
important that these functions not resend old messages and data that error checking
determined were not correctly received at the remote terminal. The teleprotection and
telecommunication systems should continue to send only the newest messages and data, and
discard old messages and data. Resending old messages may delay proper protection scheme
operation, or cause protection scheme misoperation and incorrect breaker tripping.
6.2.2.1 Analog comparison protection control and monitoring
Analog comparison protection schemes for digital communication circuits generally require
protocols with error check procedures to avoid unwanted operations due to errors in the
messages. Error checking may include parity check, check sum, and cyclic redundancy check
(CRC) sequence. In addition, other checks can be included to increase the security:
- Checking the length of the received messages
- Checking the time tag sequence of two consecutive messages
- Detecting changes in the delay time above a tolerance setting
When a message error is detected, the message is rejected and the protection is reset. Below a
predetermined message error rate, the protection remains stable but the operating time is
delayed due to the loss of messages. If the percentage of messages rejected exceeds a value,
e.g. 25%, during a defined period of time, the operating time becomes intolerable and the
protection is blocked until the message error rate decreases to acceptable values.
On blocking, a channel failure alarm is raised by the relay and sent to indicate a failure in the
communication channel. If the protection uses only one channel, the protection function failure
alarm is also raised. Some protection systems have a redundant channel through a different
communication path so the channel with a better performance is selected. In case of failure in
both channels, both channels and protection alarms are raised.
6.2.2.2 State comparison protection control and monitoring
State comparison protection schemes designed for digital communication circuits generally
include protocols with error checking to detect bad messages that can cause protection scheme
relay misoperation. Because each bit in a state comparison protection schemes message may
have a different function, and therefore a different priority, some schemes apply additional
security measures for each individual bit. For instance, a bit used to communicate a permissive
transfer trip does not need the same security as a bit used to communicate a direct,
unsupervised transfer trip. The protective relay scheme may therefore require that two or three
113 / 172
consecutive messages be received with the direct transfer trip bit asserted before taking action
to perform the direct transfer trip output. This is analogous to using a time delayed output in an
analog scheme to improve the security of an output.
Likewise, if a bad message is detected, the protective relay should reject and ignore the bad
message, and may assume the status of each bit in the expected message based on
predefined criteria. For instance, the relay may be programmed to assume the bit should be in
one of three states if the message is bad:
- a logical zero (0)
- a logical one (1)
- or it should retain the status the bit had in the last good message.
For additional security, the relay may require that several good messages be received after one
or more bad messages to ensure that the communication system has returned to a satisfactory
performance level before using the bits in the new messages.
6.2.2.3 Erroneous signal detection
Protective relays may detect erroneous permissive, deblocking, or blocking signals based on
signals arriving too soon after power system fault detection. Proper permissive, deblocking, and
blocking signals always incur some delay due to natural latencies in the relays and
telecommunication system. Premature signals can be generated by power system fault noise
induced on the communication system. While the teleprotection device may have filtering to
reject noise and pass only valid signals, filtering can be compromised by unique noise patterns.
6.2.2.4 Loop-back and misconnect detection
The telecommunication service provider commonly checks telecommunication circuits by
performing a loop-back test. During the loop back test, the signal sent by a relay, intended for
the relay at the remote terminal, is returned to the originating relay. State comparison protection
schemes can operate incorrectly when a power system fault occurs while the
telecommunication circuit is in a loop-back configuration. The teleprotection device and/or relay
should detect the communication circuit loop-back configuration and disable communication
assisted tripping until proper communication circuit connectivity is restored. The relay and/or
teleprotection device should also alarm during this condition to alert the appropriate operating,
maintenance, or telecommunication service personnel.
6.2.2.5 Actions on alarm conditions
Digital protective relays or communication schemes typically do not alarm on a single bad
message, but should alarm on a string of consecutive bad messages that could indicate a
permanent communication system failure. However, because individual bad messages, or small
groups of bad messages may indicate an impending communication system failure, the digital
protective relay or communication system should log the occurrence of bad messages, and
track the performance of the communication system over time by comparing the number of bad
messages to the number of good messages over a period of time. If the percentage of bad
messages exceeds a predefined threshold appropriate for that communication system, the relay
or communication system should alarm or annunciate the problem to alert communication
system maintenance personnel.
In addition to alarming, digital protective relays may also take additional action when
communication system problems persist. For instance, the protection scheme could:
- force the communication system to switch communication paths or activate a redundant
communication channel.
114 / 172
The second method employing time tagging is preferred, as the comparison of the respective
samples then does neither depend on equal signal propagation times for the 'go' and 'return'
direction (propagation time symmetry), nor on stringent limitations for signal propagation time
variations. See also chapters 6.1.1.1 and 6.1.1.2 for explanations and definitions.
As propagation time symmetry may be jeopardized when communication channels are re-routed
upon a failure in the communication network, or because signal propagation times may vary
with time due to signal buffering and queuing in the network, time tagging is a prerequisite for
protection using general communication networks, unless the network has been explicitly
engineered to comply with the timing requirements that are imposed by the protection relay.
Refer for example to Chapters 4.2, 4.6 and 6.1.
In conclusion, whatever method for time synchronization is chosen, the protection relay - when
used in conjunction with modern telecom networks - will have to cope with:
- Static propagation times
115 / 172
electrical hostile
(radio frequency
approach in the
service, to the
Many problems are caused by bad installation practices and by using inappropriate
telecommunication services for the required teleprotection signalling function.
This chapter attempts to capture a number of critical issues, and highlight the areas that may be
the source of service failure if adequate standards of installation are not implemented.
Fundamentally the protection requires a telecommunications bearer service that will provide
perfect performance during electrical network faults. This may only be two or three times per
year and requiring the allocated circuit traffic capacity for no more than a few seconds. The
protection engineer also expects the telecommunications bearer service to be monitored and
managed by the circuit provider. On the whole this can not be achieved.
As the monitoring of the bearer services provided must, by default, be a task performed by the
teleprotection function/device, manufactures and protection engineers require to develop alarm
strategies that are more in keeping with the bearer services provided and the needs of the
scheme.
To achieve this requirement teleprotection signaling devices and bearer circuits require to be
procured that will guarantee quality of service before, during and after any incident. The circuit
shall be designed to provide a secure and resilient service that will continue to function, within
specified limits, before, during and after an electrical fault or network disturbance. The total
116 / 172
connectivity, including local connections and local power requirements, must not compromise
the ability of the teleprotection function to perform its task.
When designing and provisioning the required network bearer service the following should be
considered:
- Is the bearer service immune to electrical disturbance?
- Can the bearer service provide the required operating characteristics?
- Can the bearer service provide the managed resilience and diverse routing in
compliance with service needs?
- Does the teleprotection function raise an alarm if the bearer service does not comply
with pre-defined characteristics?
- Assuming that we can provide immune external bearer services, how do we ensure that
we do not compromise the ability of the protection service to function by poor local
installation practices?
It is imperative that the methods used to interconnect and power the teleprotection signalling
device and the bearer service addresses the needs to comply with standards for EMC, RFI, rise
of earth potential and other electrical disturbances. If this is done the security and integrity of the
service will not be compromised
The existing arrangements developed over a number of years were introduced at the time
teleprotection signaling schemes were reliant on physical copper connections and / or voice
frequency bearer services.
The PTO (Public Telephone Operator) Bearer Service Providers, have over the last 20 years
significantly revised their networks and the services available on them. Copper with DC
connectivity is no longer available and the analogue network infrastructure, used to deliver voice
frequency services, has now been replaced by new digital network infrastructures. Most voice
frequency services are now derived from digital technologies.
Methods used to deliver the bearer service to the teleprotection device within the substation
have evolved rather than been developed to meet the changing needs of the service. Protection
services have become more sophisticated and in turn demands more from the bearer service
provided. The migration of service provision from pure analogue services to analogue derived
digital services has already caused problems with Differential Delay for analog comparison
Protection schemes. Guaranteed and consistent circuit parameters can no longer be assumed
on networks delivered by public service providers. Network and local considerations must be
addressed and understood.
In the typical electrical hostile environment found in substations it is necessary to ensure that
the equipments are compliant and the methods used to house, power and interconnect these
services will not compromise the effective operation of bearer or teleprotection device.
If consideration is given to the above likely risks, existing configurations can be made secure
and should provide trouble free service. Future developments and operational strategies should
however identify the best methods that are now available to minimize the likelihood of
teleprotection failure.
Proposals are now being developed that use the increased availability of optical fibre service
provisions and interface devices.
If external services can be provided over alternatively routed fibre and interfaced with
117 / 172
appropriate service equipments housed in EMC / RFI cubicles, the likelihood of bearer service
failure, during an electrical disturbance, will be greatly reduced.
If power requirements are designed to include appropriate filters and surge arrestors and if
cables are run in appropriate fashion, the interference experience by noise on DC power
supplies to equipment will be insignificant.
The connection from the delivered bearer service to the teleprotection device must consider the
local environmental risks and adopt the most appropriate interface. This connection from EMC /
RFI cubicle to teleprotection device may be best achieved by using Optical Fibre cable and the
appropriate converter units at each end.
When copper cable is being used best practice for screening and route choice, to reduce the
impact of induced voltage, may provide acceptable results.
A comprehensive treatment of EMC including practical design-, cabling- and installation guides
is found in [37] and in the IEC standards below.
IEC 61000-5-1:
IEC 61000-5-2:
Table 6.4-1:
118 / 172
119 / 172
Protection
Scheme
All State
Comparison
Schemes
(see 3.1.2)
Direct
intertripping
(see3.2.1.2)
Digital Current
Differential
(see 3.1.1)
Analog current
differential
(see 3.1.1)
Pilot wire
relays
(50/60Hz)
(see 3.1.1)
Phase
Comparison
(see 3.1.1.2)
Charge
Comparison
(see 3.1.1.3)
Licensed Radio
(Microwave)
(see 4.3.3.1)
Unlice
nsed
Radio
(see
4.3.3.2
)
Optical
fibre
(see
4.3.4)
Pilot wires
(see 4.3.1)
Voice
Frequency
Circuits
(see 4.1.2)
Analog
(SSB)
Digital
(QAM, MCM)
Analog
Digital
Digital
Digital
(1), (2)
OK
OK
(6)
OK
OK
(8)
OK
(1)
OK
OK
(6)
OK
OK
(8)
OK
(1)
Not
recommended
(3)
Not applicable
(3), up to
19.2kbit/s
(3)
(3), (8)
OK
OK
OK
(8)
Not
recommended
(3)
(3)
(3), (8)
(3)
OK
Not used
Not used
Not used
(4)
(4)
Not
used
(4)
(3), (5)
(3), (5)
(3), (5)
(3)
(3)
(3), (8)
(3)
(3)
OK
OK
OK
OK
(8)
OK
Table 7.1-1:
Not
recommended
(8)
Not
recommended
(8)
Notes:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
120 / 172
Protection
Scheme
All State
Comparison
Schemes
(see 3.1.2)
Direct
intertripping
(see3.2.1.2)
Digital Current
Differential
(see 3.1.1)
Analog Current
Differential
(see 3.1.1)
Pilot wire relays
(50/60Hz)
(see 3.1.1)
Phase
Comparison
(see 3.1.1.2)
Charge
Comparison
(see 3.1.1.3)
Frequency Division
Multiplexing
(see 4.4.1.1)
Wavelength Division
Multiplexing
(see 4.4.1.1)
Fixed TDM
(see 4.4.1.2)
Statistical TDM
(see 4.4.1.2)
Code Division
Multiplexing
(see 4.4.1.3)
OK
OK
OK
(2)
(3)
OK
OK
OK
(2)
(3)
OK
OK
(1)
(1), (2)
(3)
OK
OK
(1)
(1), (2)
(3)
Not used
OK
(4)
Not
recommended
Not
recommended
OK
OK
(1)
(1), (2)
(3)
OK
OK
(1)
(1), (2)
(3)
Table 7.1-2:
Notes:
(1)
(2)
(3)
(4)
OK if communication channel delay is kept within the relays delay compensation adjustment range.
Dynamically allocated bandwidth is not recommended unless one has full control over delay
No practical installations using Code Division Multiplexing for protection have been reported yet.
Possible, if delay can be kept below 1ms
121 / 172
Transport Networks
(see 4.5.1)
Protection
Scheme
All State
Comparison
Schemes
(see 3.1.2)
Direct Intertripping
(see3.2.1.2)
Digital Current
Differential
(see 3.1.1)
Analog current
differential
(see 3.1.1)
Pilot wire relays
(50/60Hz)
(see 3.1.1)
Service Networks
(see 4.5.2)
PDH
(see
4.4.2.1)
SDH /
SONET
(see
4.4.2.2)
Cell switched
(i.e. ATM)
(see 4.5.2.3)
Datagram
(IP)
(see 4.5.2.4)
Circuit switched
(POTS, ISDN)
(see 4.5.2.1)
Packet switched
(X.25, Frame
Relay)
(see 4.5.2.2)
OK
OK
(1)
Not recommended
(2), (4)
Not recommended
(3)
Not recommended
(4)
OK
OK
(1)
Not recommended
(2)
Not recommended
(3)
Not recommended
(4)
(7)
(5), (7)
(1)
Not recommended
(2), (6)
Not recommended
(3)
Not possible
(4)
(7)
(5), (7)
(1), (7)
Not possible
(8)
Not recommended
(3)
Not possible
(8)
OK
(9)
OK
(9)
(1), (9)
Not possible
(2)
Not recommended
(3)
Not possible
(4), (8)
Not recommended
(3)
Not possible
Analog: (4), (8)
Digital: (4)
Not recommended
(3)
Not recommended
(4)
Phase Comparison
(see 3.1.1.2)
(7)
(5), (7)
(1), (7)
Charge Comparison
(see 3.1.1.3)
(7)
(5), (7)
(1), (7)
Table 7.1-3:
Analog:
Not possible
Digital:
Not recommended
(2), (6)
Not recommended
(2), (6)
Notes:
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
No experience yet, but no obvious reason why it will not work with proper network design and ATM QoS guarantees, see
ANNEX A3.
Might be applicable in future WAN with proper network design with QoS guarantees. Today (1999), QoS standards are
still evolving.
Not recommended unless the circuit is permanently established. Call setup is too slow or may be rejected. Connection
may not be under the control of the service user or may be released inadvertently.
Delay is non-deterministic and may be excessive under heavy traffic load
OK, if requirement for propagation time symmetry is met, or protection terminals compensate for delay asymmetry
Protection devices may have to cope with significant propagation time and propagation time variation
OK if propagation time is within the relay's compensation range and delay variation is limited, see chapter 6, Table 6.1-2
Transparent transmission of analogue signals is not supported
If propagation time of less than 1ms end-to-end is guaranteed by the network
122 / 172
Table 7.1-4 broadly summarizes communication technologies and highlights some critical
issues that have to be addressed even when a properly designed communication
link/system/network is assumed.
Type of
communication
Critical issue
cost
Multiplexed PDH
circuits (point-to-point)
propagation time
IP based WANs
propagation time
propagation time
Ethernet/IP based
LANs
propagation time
Table 7.1-4:
Advantage for
protection
- ensures minimum
propagation time
- independent and
deterministic
- deterministic
behavior
- Network resilience
- deterministic
behavior with QoS
guarantees
- Network resilience
Configuration summary
123 / 172
Disadvantage / Risk /
Warning
Remedy
- single point-of-failure
- introduce redundancy
- single point-of-failure
- introduce redundancy
- use terminal
addressing
- time stamping in
protection relay
- proper telecom
network design and
system engineering
- use terminal
addressing
- channel crossover
- propagation time
variation due to rerouting
- channel crossover
- propagation time
variation
- non deterministic
behavior
- missing QoS
guarantees
- propagation time
variation
- EMC if copper
based LAN
- single point of failure
- time stamping in
protection relay
- proper network
design and system
engineering
- time stamping in
protection relay
- proper network
design and system
engineering
7.2.2 Interfaces
(see Table 7.2-1)
q
q
q
q
Type of interface
Electrical / optical parameters
- Connectivity
- Electrical (voltages)
- Optical (power budget)
Physical / mechanical
- Connectors
- Pins / pinout
Power requirements
- Converters
Cabling
- Shielding / grounding
- Type of cable / wire
- Cable length
- Who supplies
Environment
- EMC
- Distance
124 / 172
7.2.3 Contractual
q
q
q
q
Performance monitoring
Reporting
- Communication system
- Protection
Guarantees
Network policy
- Priorities
- Who decides
- Disaster recovery plan
Technical support
- Hot line
125 / 172
Pilot wire
input
Voltage
Current
Signal dynamic range
Impedance
Insulation / Isolation
Voltage
Current
Signal dynamic range
Impedance
Insulation / Isolation
Impedance
Receive level
Isolation/Insulation
Impedance
Transmit level
Isolation voltage
Insulation resistance
Max. voltage
ON/OFF threshold voltage /
current
Load current
Wetting voltage
Current rating (make, carry,
break)
Bandwidth / Frequency
spectrum
Bandwidth / Frequency
spectrum
Frequency restrictions?
Level restrictions?
Pick-up time
contact position
(N/O, N/C)
Crosstalk
(Near-end, far-end)
max. leakage
current
Type of interface (a) or (b) refers to the definition in the Figures A1-1 to A1-4 in ANNEX A1.
n-wire
(n = 2,3,4)?
output
4-wire
2-wire
4-wire
2-wire
Transistor input
Optocoupler input
Relay coil
n-wire
(n = 2,3,4)
input
output
input
output
Mechanical contact
Type of interface1
Contact interface
Electrical isolation
Shielding
Electrical isolation
EMC standards
Electrical isolation
EMC standards
Table 7.2-1:
input
output
(electrical)
Digital circuit
Fibre bandwidth
(MHz * km)?
Multimode fibre
Optical wavelength
Optical wavelength
Do pattern restrictions
apply? (e.g. o/1 density)
Data rate?
Do data pattern
restrictions apply? (e.g.
0/1 density)
Optical power
launched into fibre?
Synchronous
operation
Synchronous
operation
Asynchronous
operation
Clock master /
slave?
Clock provisioning:
Data format?
(e.g. Start, Stop,
Parity, Number of
bits)
Multimode fibre
Fibre bandwidth
(MHz * km)?
Contra-directional?
Co-directional? (preferred)
G.703
V.11/X.24
RS 485
RS 422
V24./V.28
RS232
Type of interface1
Type of interface (a) or (b) refers to the definitions in the Figures A1-1 to A1-4 in ANNEX A1.
Serial interface
Table 7.2-1 Part 2: Digital Serial Interfaces and Optical Fibre Interfaces
Electrical isolation
EMC standards
Connector type
(male / female .)
Connector pin-out
New transmission technologies, particularly in fibre, now provide abundant capacity for
any utility application. Commercial transmission solutions already provide capacity in
excess of 100 Gbit/s. Experimental work indicates that viable solutions with 10 times this
capacity will be available soon. The theoretical limit is still another order of magnitude
above todays experimental level.
Integration of services is a main development in the telecommunications industry. This
requires unified transport, switching and service-access systems. For this integration to
take place mechanisms to guarantee a certain level of Quality is required. Quality of
Service (QoS) -provisioning is now becoming an integral part of the network allowing any
service to be connected/maintained with a guaranteed QoS-profile.
These new telecommunications and data -communications technologies will mean that utilities
may be able to buy standard, off-the-shelf telecommunications products or even outsource their
communications needs. Within the Communications/Protection area of the utilities, the focus is
expected to shift from communication network design to a more facility managed approach.
128 / 172
in the fibre and exploring optical technology to switch individual bandwidth-slots in each node of
the network.
It has been shown that an ATM network may easily maintain and exceed the IEC Integrity level
3. Therefore security and, with a proper network topology, availability requirements may be met
with ATM networks.
Both absolute time delay and time delay variations are QoS-parameters that can be specified at
call-set-up in an ATM-network - provided the ATM-nodes are designed for this. In practice this
means that the Teleprotection messages are switched into high priority queues where no or little
other traffic may be waiting, and that the interface access is sufficiently fast ( > 2 Mbit/s).
However, the achieved time-domain granularity may not be sufficient in some networks to give
the desired QoS-profile, unless special considerations are made.
Several techniques may be employed to get around the time delay problems. They will need
careful validation before they can be considered to be mature.
ATM solutions are being implemented where the ATM technology accounts for a fair
portion of the transport mechanism, only the lowest layer (multiplex section and
downwards) is left to SDH components
ATM has through its Adaptation Layers provisions for direct access to services
129 / 172
A set of other and more detailed scenarios could be given. This, however, is outside the scope
of this brochure.
Essential for all these technologies is a Network Management system. Most systems on the
market today are Element Management systems strongly coupled to a particular vendors
equipment. There are some generic management platforms in existence today, but these
generally require substantial work to be developed and maintained in a specific network
environment consisting of multi-vendor equipment. Partly because of the complexity of the
network management system, unifying and simplifying the network technologies and limiting the
number of different equipment in a network is clearly an important task in smaller networks.
Three Network Management technologies are under development at present:
-
TMN is the only approach that gives a complete management view at present. The other two
platforms may also be suited for a network view but this requires some extra (often) networkspecific development. It will still take time before any of the platforms have reached a stage
where they can provide a multivendor/multitechnology network view without proprietary
development.
130 / 172
Figure 8.1-1:
Figure 8.1-2 shows a concept of an integrated substation and the relationship between
applications from the Utility Communications Architecture (UCA) proposed by EPRI [23]. UCA
provides integrated utilities computer communication protocols which enable a wide variety of
utility information and control systems to share data seamlessly including real-time
communications among corporate headquarters, power plants, control centres, substations,
distribution automation equipment, and customer sites. Since UCA covers not only SCADA
systems, RTUs and substation automation devices, but also microprocessor-based intelligent
electronics devices (IED) including numerical relays, protective relay communications are also
accommodated by such networks.
The siginificant work in the UCA proposal is to identify how equipment should interoperate, and
how IEDs (Intelligent Electronic Devices) publish and subscribe to each other's information.
Substation devices and functions are represented as standardized object models in GOMSFE
(Generic Object Models for Substation and Feeder Equipment) in the UCA proposal. Modeling
131 / 172
provides a way to standardize information exchange between other models and devices.
As the IEEE 802.3 Ethernet technology is proposed for protection communications, its dynamic
performance has been evaluated by simulating the LAN traffic that would be generated by a
particularly severe incident in a typical substation. Under the reasonable assumption that
messages generated by a power system fault are spread over 1 millisecond, it was found that
[24], [29]:
- A 10 Mbit/s shared hub Ethernet network has limited performance, as it can deliver less
than 20 messages in 4ms.
- Three types of LANs (10 Mbit/s switched hub, 100 Mbit/s shared hub, and 100 Mbit/s
switched hub) all can deliver 100 messages within 4 ms if messages are spread over 1
ms.
This intra-substation communication project is co-ordinated with IEC activities to establish a
process bus standard where a serial unidirectional point-to-point link using Ethernet is employed
as an interim step (see also Chapter 8.3.1).
Substation
Controller
EMS
SCADA
Engineering
Workstation
Digital
Communications
Relays
Meters
Oscillography
Equipment
Diagnostics
Sensors
Figure 8.1-2:
These intra-substation networks can be used for transmitting protection signals among
protective relays inside a substation and for supervising and controlling protective relays. As
supervisory and control information can be also transmitted to other substations or control
centres using wide-area networks (WAN) connected to the intra-substation networks (LAN), the
issue is whether one can use the substation LAN for communicating between protection relays
in different substations or not.
8.1.7.2 Inter-substation communication
Between the LAN and the WAN a Gateway is normally needed for protocol conversion as
shown in Figure 8.1-3, because the protocols and network technologies (Chapter 4.5) utilized
with the LAN and WAN are typically not the same. Although the delay depends on the message
length and transmission speed of the WAN, one can broadly say that the main delays are
introduced by the substation LANs (random delay for accessing the shared medium), the
receiving Gateway (WAN-LAN protocol conversion) and by the propagation delay that is
introduced by the WAN.
In general, networking protocols and the data traffic they support can be characterized as being
either connection-oriented or connectionless.
With a connectionless protocol, no interaction between the terminals takes place prior to the
exchange of data, and no fixed path through the network is set up. Data packets are routed
132 / 172
through the network individually on a hop-per-hop basis. Connection-less thus also means that
data packets that belong to the same message may follow different routes with different signal
propagation delays through the network, and that the message transfer delay is determined by
the slowest packet. As data packets may not arrive in the same sequence as they have been
sent, they have to be re-assembled in the correct order at the receiving end. Because of its nondeterministic latency characteristic, connection-less type of communication has so far not been
utilized for the transmission of protection signals across a WAN.
With a connection-oriented protocol between two network terminals, the source and destination
terminals interact with one another prior to the transport of data in order to set up and secure a
fixed connection between them. All data packets or cells will follow the same route once the
path has been established. A connection may be requested by the terminals through a
standardized signalling protocol or dial-up procedure (e.g. PSTN or ISDN), or a permanent
connection may be established by the network operator on a contractual basis (leased lines), or
the network may provide a fixed and permanent connection by design ('hard-wired' circuits).
Since WANs that are based on PDH and SDH technology (see Chapters 4.4.2, 4.6.2.1) provide
a set of permanent communication channels with fixed bandwidths and minimum signal transfer
delays, they are today the first choice in the selection of network technologies for the fast and
reliable transmission of protection signals.
More on the subject of WANs and telecommunication networks is found in Section 4 of this
document.
In conclusion, LANs and Gateways can introduce substantial and unpredictable time delays
which may not comply with the total available time budget for protection signal transmission.
Considering the stringent requirements for bounded signal propagation times (upper limits,
propagation time variation and -symmetry) and Quality of Service guarantees, it seems more
advisable to link protection relays directly via PDH or SDH channels (and perhaps ATM with the
most stringent AAL Class 1), thus bypassing the LAN and the Gateway by using a deterministic
serial data link to access the WAN.
A very careful bandwidth and latency study would need to be performed in any case.
133 / 172
Figure 8.1-3:
134 / 172
Today:
Future:
Challenge:
LAN
P/C
P/C
P/C
Figure 8.2-1:
It is assumed that in substations, equipped which such a LAN, all the relevant information for
protection and control is available on the LAN and can be used by all protection and control
devices. This means that every relay can protect not only one bay, but it can perform protection
calculations with information from all other bays within the entire substation. Relays can even
135 / 172
send trip commands to the other circuit breakers of the various bays via the LAN connection. In
this situation, distributed processing of protection functions and distributed redundancy will be
possible in a substation. Therefore, separate redundant hardware modules for every single bay
can be avoided. The amount of hardware may even be reduced to two or three dedicated
devices that protect the entire substation and connected lines and transformers. In such a
system, substantial cost-savings in protection schemes will be obtained.
Developments in high-voltage networks
As mentioned before there is a future trend for building substations with LAN communications.
When this is realised in several substations, these can be interconnected by a WAN with high
transmission speed and high performance to suit the protection schemes. The following Figure
8.2-2 represents a WAN.
WAN
LAN - A
LAN - B
LAN - C
Figure 8.2-2:
LAN - D
In Figure 8.2-2 a high-voltage (sub-)network consists of various substation LAN's which are
interconnected by a WAN. In such a system, all information is available on every
protection/control device within this WAN network. Depending on the possibilities of the relays,
distributed processing of protection functions for analog comparison-, state comparison,
intertripping and wide-area protection can be carried out.
This communication structure creates new opportunities for future protection schemes on
network level. In the telecommunication industry, path protection and re-routing is an integrated
function. May be in the HV environment future protection can also co-operate with network
management like in the telecommunications discipline, because a lot of real time information is
available that can be used for fault clearing and automatic re-routing. This means that in every
substation bay information of all other bays in the network is available, so it may be used for
more advanced protection functionality and even for re-routing, because the information about
network topology and traffic load is also available.
Possibilities of (wide) area protection
In case of a failure in a HV network which is equipped with a area protection the following
options will become feasible:
136 / 172
Application examples
For the various options mentioned in chapter 5 a few examples of area protection functionality
can be described:
-
Petersen earthed HV-networks mostly include overhead lines and to some extent
cables. In case of an earth fault in the cable part, a persisting earthing fault condition is
not acceptable. In that situation a distributed area protection can determine the fault
location fast and accurately and trip the faulty cable.
Cross-country fault clearing can be better carried out by a second protection level, also
by an area protection. The first protection level trips the fault selective or if not, only one
of the two faults places. Because of the complexity of a cross-country failure, the fault
clearing can be more selective with the second protection level and if necessary, a
remaining earth fault can be tripped. Apart from that, the determination of the fault, that
has to be cleared (the first fault) can be based on the network topology.
A future reactive power protection could be build by processing voltage on HV busses
and reactive power flows in HV- (sub-) networks. A smart load scheduling application in
combination with reactive power protection can be used to minimise the number of
customers that will be switched off.
Conclusion
Application of LAN in substations in combination with a HV network WAN, with SDH, ATM and
real-time Internet Protocol communication challenges the design of future protection systems.
The existence of an information WAN on HV network level is the main advantage. This gives the
possibility of creating distributed processing and inherent redundancy. This means that the
number of protection devices, including wiring to the HV components, can be reduced without
loss of availability, dependability and security of the system, if the stringent real-time
requirements can be met by the proposed concept.
137 / 172
each line end independently from each other even when proprietary algorithms are used. Thus
only a minimum of co-ordination - if any - between relays is needed.
The situation is however less satisfactory for analog comparison protection schemes like current
differential protection, or for state comparison protection schemes when the teleprotection
function is integrated in the protection relay. Main reasons for incompatibility are proprietary
protection algorithms and/or proprietary communication protocols due to missing standards or
due to the choice of different options from existing standards.
The standards to be developed for protection relay interoperability would have to encompass at
least Layers 1 to 4 in the OSI reference model (Figure 4.5-3) when networking functionality shall
be included in the (tele-)protection device. True interoperability between protection relays would
require to develop standards for all layers of the OSI reference model, including the Application
Layer. Whether interoperability of protection relays from different vendors would justify the
development of a full set of complex standards is questionable. Nevertheless, it is becoming
clear that for future designs consideration should also be given to emerging standards in related
areas of utility communication, and that existing standards should be applied whenever possible
and economically justified.
Some ongoing - and for future teleprotection possibly relevant - standardization activities are:
-
As for today, protection systems using telecommunication are still some distance away from
"plug-and-play" solutions, particularly for analog comparison schemes. This is one of the areas
where forthcoming standardization may show promise.
138 / 172
9 CONCLUSIONS
The Telecommunications evolution and the IT revolution has resulted in the need to reconsider
tried and tested methods of delivering teleprotection schemes and their associated bearer
services. The roll out of new communication network technologies presents possible
opportunities for cost efficient and advanced solutions for teleprotection systems. Wider
bandwidths offer scope for the transmission of higher speed data for improved protection
functionality and potentially better reliability, combined with rich analytical information.
New network technologies can provide improvements in security, availability and resilience, as
modern networks are configurable from a centralized network management system and support
automatic re-routing in case of single link failures. Service availability may therefore be better
than for legacy network technologies or simple point-to-point configurations, and proper network
design can ensure that the risk of single point of failures is reduced.
Network technologies that permit delivery of messages to multiple locations, or collection of data
from multiple locations will be essential for novel schemes such as wide area protection or
system wide protection, stabilizing protection and centralized protection. The continued
development of transport modes that can offer this type of service must be recognized and
suitable design concepts considered.
Optical fibre offers the best solution in an electrical hostile environment and is generally
accepted as the most appropriate communication medium for developing service network
platforms for the Electricity Supply Industry.
New communication network technologies have however also brought up new problems and
new challenges to protection, mostly related to the non-deterministic signal transfer time
behavior. Propagation time, its variation and symmetry are all critical characteristics that require
careful consideration with respect to teleprotection requirements. The inherent problems
associated with latency and signal transfer delay variation of bandwidth on demand or "best
effort" techniques may be overcome with the availability of higher bandwidths in the core and
access network, together with controlled Quality of Service mechanisms. Quality of Service
guarantees and sufficiently low network latency may however only be available at significant
additional cost.
Measures to overcome problems related to signal transfer time variations introduced by the
communication network can also be taken in the protection relay, e.g. by time stamping of
measured quantities (samples). It is however obvious that some legacy protection and
teleprotection equipment may not work with, or be able to take advantage of new Telecom / IT
network technologies, because they have been designed for deterministic (with respect to
bandwidth and latency) communication channels.
The economic effectiveness of high capacity service connections with performance guarantees
will always be subject to business justification. Protection is a niche application that has specific
characteristic requirements associated with the bearer services provided by the core business
Telecommunications infrastructure or by the Public Telephone Operator (PTO). Bearer services
for Teleprotection often represent less than 0.001% of network services provided by third party
Telecom Service Providers. It is becoming clear that the Electricity Supply Industry cannot drive
the development of products and standards in the Telecommunications Industry.
The unique requirements for stringent signal propagation delays, delay variation and delay
139 / 172
symmetry are not common or even similar to any other customers needs. Future Telecom / IT
platforms and services offered by third party network service providers may not be able to
provide circuits that are compliant with existing and legacy teleprotection requirements.
Protection engineers must understand the changes that will be introduced by third party
providers, and develop solutions that are compatible with the new platforms being offered. If
these changes can not be accommodated, the Electricity Supply Industry will require to develop
its own Network for operational needs that can guarantee the long term availability of compliant
and resilient bearer services for the teleprotection and other mission-critical services needs.
Many utilities are already developing their own networks that are dedicated to their operational
needs using optical fibre infrastructure solutions. With an optical fibre infrastructure in place, the
availability of high capacity bandwidth at no additional cost is now implicit. Larger bandwidths
(more information within the acceptable delay) may open the door for new and improved
protection schemes such as multi terminal line protection, selectivity improvement for double
circuit lines and wide-area protection, to name a few.
The cost effectiveness and operational advantages of ensuring continued support of legacy
teleprotection services and the ability to accommodate new schemes on a compliant and
managed network are as yet unproven. Organizational changes that demand outsourcing and
facility management of assets may jeopardize the integrity and security of the
telecommunications platform that provides the essential elements of the electrical power
systems protection schemes.
The combined understanding of Electricity Supply Industry manufactures and the industries'
Protection / Telecommunication Engineers is needed to ensure that risks are minimized and the
most appropriate solutions are adopted.
140 / 172
(b)
Teleprotection
equipment
or
Teleprotection
function
(a)
(b)
Teleprotection
equipment
or
Teleprotection
function
Telecommunication system
Protection
equipment
or
Protection
function
Telecommunication system
Teleprotection system
Figure A1-1:
(a)
Protection
equipment
or
Protection
function
(b)
Teleprotection
equipment
or
Teleprotection
function
Figure A1-2:
Teleprotection
equipment
or
Teleprotection
function
(b)
Teleprotection
equipment
or
Teleprotection
function
Multiplexer
Teleprotection
equipment
or
Teleprotection
function
Multiplexer
(b)
Teleprotection
equipment
or
Teleprotection
function
Figure A1-4:
Protection
equipment
or
Protection
function
(a)
Protection
equipment
or
Protection
function
(a)
(b)
Figure A1-3:
Protection
equipment
or
Protection
function
(a)
Protection
equipment
or
Protection
function
(a)
(b)
(a)
(b)
Telecommunication network
or
WAN
Teleprotection
equipment
or
Teleprotection
function
141 / 172
Protection
equipment
or
Protection
function
Teleprotection equipment;
Teleprotection function:
Converts the information from the protection equipment (relay) into a signal suitable for
reliable transmission over a telecommunication link /system/network. Includes all functions
that are necessary to obtain the desired dependability, security, and data integrity.
Performs intermediate function between protection and telecommunication. May also
multiplex signals from several protection equipments/functions, status indications etc. onto
a single telecommunication channel.
Typical layouts:
May be integrated into the protection equipment, particularly for analog comparison
protection schemes. Interface (b) is accessible. Interface (a) is equipment internal and
normally not accessible.
Telecommunication system;
Telecommunication link:
Dedicated medium;
Dedicated link:
A medium / link / circuit / channel that carries only the teleprotection service.
Multiplexer:
A device which combines several signals or services onto a common medium for
transmission
Telecommunication network:
A set of communication and switching devices that work in a collaborating way to provide a
telecommunication service between access points distributed over a wide area
142 / 172
Fault inception
Fault clearing
TC
42 - 210 ms
Fault clearing time for a protection system
TB
12 - 130 ms
Protection operating time
TA
2 - 70 ms
Teleprotection operating time
(Overall operating time for a teleprotection system)
(a)
(a)
Propagation
time
(b)
Teleprotection
transmit
function
Fault
recognition
time
Time for
initiating
transmit
action
Telecommunication system
(-circuit / -link /
-network)
1)
10 - 30ms
Protection
equipment
Circuit breaker
Propagation
time 2)
0 - 5ms
Protection
equipment
(b)
Selection
and decision
time, incl.
O/P circuits
11040ms
3)
1 - 5ms
Additional
delay due to
disturbance
0 - 20ms
Relay
decision
time
01030ms
4)
Operating
time
including
arcing time
30 - 80ms
To
2 - 45ms
Nominal transmission time
(propagation time not included)
Tac
2 - 65 ms
Maximum actual transmission time
under disturbed channel conditions for a defined
dependability and signal to noise ratio or BER.
(propagation time not included)
Tpac
12 - 125 ms
Maximum actual protection operating time time
under disturbed channel conditions for a defined
dependability and signal to noise ratio or BER.
(propagation time not included)
Notes:
1)
2)
3)
4)
Including the physical medium (cable, fibre) and the telecommunication intermediate- and terminal equipment
0 ms applies when interface (b) is connected back to back. The propagation delay depends on the route length and on the number and type
of intermediate and terminal equipment involved.
5 ms may be difficult to achieve for networks. However, network latency will normally be offset by short times in the teleprotection receive
function (typ. 2 10 ms) for digital systems operating at 64 kbit/s or higher, such that the requirement for the overall operating time can be
met. See note 3)
Typical values for analog (narrowband) teleprotection systems are in the range 10 to 30 ms
Typical values for digital (wideband) teleprotection systems operating at 64kbit/s or higher are in the range 2 to 10 ms
0 ms can apply to intertripping (direct tripping equipment)
2 to 10 ms are typical for state comparison protection relays (time for phase selection and O/P circuits operation)
10 to 30 ms are typical for unit protection relays (time for signal comparison and O/P circuits operation)
Figure A1-5:
143 / 172
Few ms;
No guarantee
Few ms;
No guarantee
No guarantee
Not applicable
Not applicable
No data available
Not applicable
Critical;
No guarantee
No data available
(Similar to
SDH)
No data available
~0.6ms
Not applicable
Not applicable
Not applicable
Not applicable
Not applicable
Not applicable
(Pt-Pt links)
Not applicable
(Pt-Pt links)
Not applicable
(Pt-Pt links)
< 1ms
< 0.1ms
< 1ms
< 1ms
>> 1ms
No data available
No data available
>> 1ms
No data available
< 1ms
if the same route
< 1ms
if the same route
< 1ms
Propagation time
symmetry
(Differential delay)
< 1ms
if the same route
No data available
<< 1ms
<< 1ms
<< 1ms
<< 1ms
Propagation time
variation
< 10-6
< 10-6
exists
(Sync. failure)
exists
(Sync. failure)
exists
(Sync. failure)
~15min
~ 1ms for single
link failure;
depending on
system vendor
Not applicable
Not applicable
Not applicable
same as SDH if
SDH is transport
layer
Not applicable
exists
(Routing error)
exists
(addressing error)
exists
(addressing error)
> 64kbit/s
< 10
> 64kbit/s
-5
> 64kbit/s
on demand
> 64kbit/s
> 64kbit/s
< 10-5
< 10-5
Cell misinsertion
ratio (CMR) < 1/day < 10-6
(ITU-T I.356)
< 10-6
> 64kbit/s
> 64kbit/s
> 64kbit/s
< 10
< 10-3
No data available
No data available
No data available
No data available
> 64kbit/s
> 64kbit/s
< 10-6
> 64kbit/s
> 64kbit/s
few kHz;
<< 64 kbit/s
4 8kHz;
< 64kbit/s1
Bandwidth or
Capacity
< 10-3
< 10-3
Bit Error
Rate
(typical)
Not
applicable
-3
No data available
low
(human error)
low
(human error)
low
(human error)
low
(human error)
Channel crossover
probability
< 10-3
No data available
Not applicable
Not applicable
Not applicable
Not applicable
Routing recovery
time / Network
restoration time
For voice and data only. For protection signal transmission, the capacity is sufficient for binary protection commands (state comparison schemes) or for non-segregated phase comparison applications
(unit schemes).
IP networks
ATM networks
SDH networks
PDH networks
LEO Satellites
MEO Satellites
GEO Satellites
Pilot Wires
Propagation time
(from interfaces (b) to (b) in ANNEX A1)
Summary of transmission characteristics of various media and telecommunication systems. All values are approximate.
146 / 172
Throughput can be based both on average data rate and/or peak data rate. The ratio of
peak rate and average rate of data streams is known as burstiness.
Reliability relates to a certain probability of data loss which can be tolerated
Delay can be specified based on an absolute or probabilistic bound.
Delay variation or Jitter is the (short-term) variation in delay a message experiences.
Parameters are static if they are valid for the entire duration of the connection. If they can be
changed while a transmission is in progress, they are said to be dynamic.
147 / 172
ITU-T I.371:
ATM SERVICE
CATEGORY
ATM TRANSFER
CAPABILITY
Traffic characteristics,
Purpose
-
(no equivalent)
Real-time / time-delay
sensitive data with QoS
guarantees
Low cell delay variation
Circuit emulation
Bursty real-time data
Efficient use of network
resources for delay
sensitive data
Statistical multiplexing
Application examples
Videoconferencing
Telephony
Audio/Video Distribution
Teleprotection
Voice communication
with bandwidth
compression and silence
suppression
Interactive compressed
video
LAN interconnection
SCADA / EMS
Future teleprotection?
Response-time critical
transaction processing
applications (e.g.,
banking transactions)
Multimedia E-mail
No teleprotection
LAN interconnection
LAN emulation
TCP/IP
E-Mail
File transfer
No teleprotection
E-Mail
File transfer
Messaging
TCP/IP
No teleprotection
Traffic Parameters
A source traffic parameter describes an inherent characteristic of a source. The following
parameters are considered for the purpose of defining the Service Categories :
- Peak Cell Rate (PCR)
- Sustainable Cell Rate (SCR)
- Maximum Burst Size (MBS) and Burst Tolerance (BT)
- Minimum Cell Rate (MCR)
- Cell Delay Variation Tolerance (CDVT)
- QoS Parameters
The traffic contract defines how the network should react when parameters that characterize the
traffic are exceeded. A traffic contract may be established either when the network user first
subscribes to a network service, or dynamically via user interface signaling or network
management negotiation.
QoS Parameters
The QoS parameters selected to correspond to a network performance objective may be
negotiated between the end-systems and the network, e.g., via signalling procedures, or can be
taken as default. One or more values of the QoS parameters may be offered on a per
148 / 172
connection basis:
- Peak-to-peak Cell Delay Variation (CDV)
- Maximum Cell Transfer Delay (maxCTD)
- Cell Loss Ratio (CLR)
A number of additional QoS parameters have been identified, but their negotiation is not
foreseen but are assigned at call set-up, e.g.:
- Cell Error Ratio (CER)
- Severely Errored Cell Block Ratio (SECBR)
- Cell Misinsertion Rate (CMR).
A3.2.1
149 / 172
150 / 172
ATM Service
Category
Traffic
Parameters
Delay and
Delay Variance
(maxCTD, CDV)
Bandwidth
Use of
Feedback
Control
CBR
PCR, CDVT
YES
YES
YES
NO
rt-VBR
YES
YES
YES
NO
nrt-VBR
YES
NO
YES
NO
ABR
YES
NO
YES
YES
UBR
PCR, CDVT
NO
NO
NO
NO
A3.2.2
ATM may use various technologies as its transport vehicle (physical layer). Frequently,
SDH/SONET transport systems will be used. Errors introduced by the transport system as well
as other impairments will have a negative impact on the QoS. The Table below shows some
sources of degradation for the ATM QoS parameters.
QoS Parameters
General
network
design
ATM specific
Link / network /
medium
parameters
CER
Cell Error
Ratio
Propagation
Time
Propagation
Time
Variation
Bit Errors /
Media Errors
Switch
Architecture
Buffer
Capacity
Traffic Load /
Resource
Allocation
Number of
Nodes / Hops
Network /
Hardware
Failures
CLR
Cell Loss
Ratio
CMR
Cell
Misinsertion
Ratio
maxCTD
Maximum
Cell Transfer
Delay
CDV
Cell Delay
Variation
X
(except CBR)
X
(except CBR)
X
(except CBR)
X
(except CBR)
The following Table shows the expected impact of QoS parameters onto general
communication performance parameters at interface (b) in the Figures in ANNEX A1, and their
possible adverse impact onto the (tele)protection function. It is noted that the table is
speculative as little experience yet exists with protection signal transmission over ATM.
151 / 172
Communication performance
parameters
BER
Jitter
Bit Error Rate
A3.2.3
CER
Cell Error Ratio
CLR
Cell Loss Ratio
CMR
Cell
Misinsertion
Ratio
maxCTD
Maximum Cell
Transfer Delay
CDV
Cell Delay
Variation
(Tele-)Protection parameters
Dependability
Security
Transmission
Time
X
X
X
X
(X)
(X)
(X)
Applications Summary
The Table below is an attempt to sum up the indications outlined in this section as related to
typical power utility applications. It is not intended to create a restrictive correspondence
between the identified application areas and ATM-layer services, and should therefore not be
taken restrictively.
Application areas for ATM service categories:
APPLICATION AREA
CBR
rt-VBR
nrt-VBR
ABR
UBR
fair
fair
optimum
fair
not suitable
good
good
not suitable
not suitable
not suitable
optimum
good
not suitable
not suitable
not suitable
fair
fair
good
optimum
good
good
fairoptimum ?
not suitable
not suitable
not suitable
good
fairoptimum ?
not suitable
not suitable
not suitable
The ratings (optimum - good - fair - not suitable) refer to the efficiency/cost advantages the use
of ATM may have for the stated application areas. The ratings given for protection signal
transmission are speculative and apply subject to the condition that native ATM interfaces are
available and that the requirements for the end-to-end signal propagation time can be fulfilled
(see Figure A1-5).
152 / 172
are transmitted, the applications must set-up a path and reserve the resources. This working
principle resembles the one used in circuit switched networks therefore being adequate for nonelastic applications such as voice telephony, or Protection communication in the future. The
complete implementation of this architecture requires QoS aware applications or the addition of
Gateways between the legacy applications and the Integrated Services core since the network
has to be informed about the profile of the traffic offered and the QoS requirements of the user.
Differentiated Services is based on the definition of different classes of services. The packets
are marketed differently depending on the service class their application/flow belongs to.
Therefore, Differentiated Services is essentially an evolution of the static-priority scheme.
Whereas Int-serv architecture guarantee and End-to-End QoS, that is to say, application-toapplication, the Diff-serv architecture guarantees the QoS only in its own domain.
Differentiated Services is an evolving architecture and in fact, still an immature technology
under discussion by different working groups of the IETF.
The provision of QoS for the Teleprotection application is a must, since both the bandwidth and
the delay of the virtual channel established throughout the network to support the protection
relay communication has to be guaranteed in a deterministic way.
Existing protection relays do not include IP Ethernet interface; therefore, we have to focus on
the new and evolving technologies. In this field, the new Utility Communication Architecture
UCA is the most relevant example. The UCA architecture is based on the use of standard
protocol stacks. Two tracks with different profiles have been defined including both the ISO and
the TCP/IP approaches.
Data interchange with devices in real-time networking environments is accomplished by adding
the following specific components:
-
Generic Object Models for Substation and Feeder Equipment, (GOMSFE) which defines
a set of object models for use across a broad range of typical utility devices;
Generic Object Oriented Substation Event, (GOOSE) which allows a device to broadcast
status information on the local segment;
Common Application Service Models, (CASM) which defines a standardized set of
abstract services supporting the UCA object models, as well as the methods of mapping
the services to the Manufacturing Message Services (MMS).
None of these application elements have been designed to specify their QoS requirements to
the network. Therefore, the provision of Teleprotection service will require static QoS
requirements that will have to be introduced in the Network Management Centre.
Three classes of services could be identified in IP QoS networks: Guaranteed Service,
Controlled Load and Best-Effort.
The first one is used to support those applications that are non-tolerant to uncontrolled delays
and/or losses. The second is used to support elastic applications, that is to say, applications
that can control the traffic flow that they are offering to the network and therefore can tolerate
some changes in delay and throughput, whereas that the third one is used by those applications
that do not require QoS.
Teleprotection service is associated with the Guaranteed Service class. Nevertheless, special
153 / 172
care has to be taken to verify that the network is providing deterministic guarantees. To achieve
this, the network architecture requires a set of components to be provided.
The principle of QoS assurance is based on an interaction between the user and the network.
This interaction is expressed by means of a Traffic Contract or agreement that includes the
commitments of both parties.
The total delay has two parts: a fixed delay caused by the transmission delay of the path and a
queuing delay caused by the intermediate nodes. The fixed delay depends only on the chosen
path and is not determined by the QoS mechanism but by the setup mechanism. The queuing
delay is determined and controlled by the QoS mechanism provided by the network. The
queuing delay is primarily a function of two parameters the token bucket size and the data rate.
These two values are under the application control, and therefore, the application can estimate
the delay and if it is larger than required, the bucket size and the rate can be modified to
achieve a lower delay.
The end-to-end QoS is based on an assured bandwidth provided by the network that, when
used by a shaped flow, produces a delay bound service with no queuing loss. This service
scheme can only control the maximal queuing delay without providing any kind of control on the
jitter since the delay will change from the transmission delay of the path to the transmission
delay plus the queuing delay.
The traffic profile is specified by means of the Traffic Specification or TSpec whereas that the
QoS requirements are specified by means of the Service Request Specification or RSpec. Both
specifications form the two parts of the Traffic Contract.
TSpec defines the flows traffic pattern allowed in terms of the average packet rate, maximum
burst rate and size, and the packet size. It has to be considered that these parameters define
the maximum allowed not the actual traffic profile. Excess traffic could be directly discarded or
transmitted as Best-effort depending on the traffic policy of the network.
RSpec specifies the QoS a flow requires. That is to say, the bounded maximum End-to-End
delay expressed in this case by means of the bandwidth reservation required by the flow
defined by means of TSpec. The network should reserve enough resources to guarantee that
no losses due to network congestion will ever occur. A detailed specification of the QoS
parameters including formats and range of values can be found in the RFC 2212 Specification
for Guaranteed Quality of Service. The same document also describes internal network
architecture to achieve end-to-end QoS.
TSpec and RSpec are calculated by a QoS reservation algorithm embedded in the application.
The algorithm takes into account the actual QoS requirement and the propagation delay of the
path to carry out the calculation of the TSpec and RSpec parameters That once accepted by the
admission control will be delivered to the network elements.
Other QoS aspects such as service availability are not defined in the Traffic Contract. They
have to be achieved by means of the proper network design.
154 / 172
Parameter
Bucket depth b
Bucket rate r
Peak rate p
Minimum policed unit m
Datagrams shorter than
accounted as length m
Maximum datagram size M
Table 8.3-1:
Units
Bytes/sec
Bytes/sec
Bytes/sec
m
are Bytes
m<M
Bytes
M<=MTU
TSpec Parameters
Parameter
Transmission rate R
Units
Range
1 to 40 Tbytes/sec
R>r
Byte/sec
Slack term s
Difference between the desired delay sec
and the delay obtained using R
Table 8.3-2:
Range
1 to 40 Tbytes/sec
1 to 40 Tbytes/sec
1 to 40 Tbytes/sec
0 to 232-1sec
RSpec Parameters
Although IP networks can offer deterministic guarantees to those flows that require it, only the
maximum end-to-end delay can be controlled. The difference between the propagation delay
and the bounded delay that includes the queuing delay cannot be controlled, thereby obtaining
a considerable delay jitter that can impair the global quality of the virtual connection. This jitter
could be increased by the imprecision of the queue schedulers in the packet switches. That is to
say, the algorithm that controls the process of packet switching can only offer the calculated
performance for infinitesimal short packets. The longer the packets the bigger the imprecision in
the delay control. It has to be considered that the packet length of any other flow switched in the
same node will affect the delay jitter of our flow. Due to this, the use of IP networks to support
Teleprotection services should be limited to well-controlled domains in which all the above
mentioned factor fall into our control.
ATM technology, which is based on a short and fix packet length called cell, can be envisioned
as feasible solution to support the Teleprotection service since it can guarantee and effectively
control both the delay and the delay variation.
The combination of IP technology that offers cost-effective access interfaces and the ATM
technology that offers network wide reliable delay control might probably be one future
approach to support the Teleprotection service in the broadband environment.
Lit:. [31], [32], [33], [34], [35], [36].
155 / 172
following association:
-
Guaranteed Service
Controlled Load
Best Effort
CBR or rtVBR
nrtVBR or ABR (with a minimum cell rate)
UBR or ABR
The details of service mapping as well as the QoS and traffic parameters could be found in the
RFC 2381. Interoperation of Controlled-Load Service and Guaranteed Service with ATM, [36].
Traffic and QoS parameters are defined in the IP environment in terms of bytes/s whereas that
in ATM are defined in cells/s. The corresponding mathematical transformations that takes into
account overhead introduced by the different size of IP and ATM headers can be found in RFC
2381, [36]. Since these functions are included in the standards, they are always included in the
Gateways that interconnect IP and ATM networks.
TITLE
Terms and definitions related to quality of service and network performance
including dependability
General quality of service parameters for communication via public data
networks
Information technology - Quality of Service - Framework
Information technology - Quality of Service Guide to methods and
mechanisms
(presently at the stage of draft)
General aspects of quality of service and network performance in digital
networks, including ISDNs
ISO/IEC
EQUIVALENT
none
none
13236
13243
none
I.356
none
I.371
none
I.731
Q.2723.1
156 / 172
none
none
Figure 8.3-1:
A4.1.1
The internal timing synchronization scheme between two terminals is implemented in the
transmission or teleprotection equipment, and timing synchronization signals are transmitted in
the teleprotection channel to self-adjust the internal clocks of the terminals by sending a
message back and forth between the terminals. There are basically two types of message
transmissions including timing pulses between master (or reference) and slave (or
synchronizing) terminals: round-trip and mutual (or two-way) transmission methods. However,
many variations exist for implementation.
In the round-trip transmission method shown in Figure 8.3-2(a), a reference timing pulse at the
master terminal is transmitted to the slave terminal. The transmitted reference pulse is returned
to the master terminal. The returned reference pulse is delayed by round-trip transmission
through the outgoing and incoming transmission lines (1 and 2). Transmission delay is
measured at the master terminal, and the data is transmitted to the slave node. The slave node
executes delay compensation corresponding to a half of the round-trip delay, (1 + 2)/2 to
achieve timing synchronization.
157 / 172
In the mutual or two-way transmission method, both terminals mutually transmit reference
pulses. Each terminal measures the time difference between the transmission of its own pulse
and the reception of the opposite terminal's pulse; TM at the master terminal and TS at the
slave terminal. TM is transmitted to the slave terminal which executes delay compensation
corresponding to a half of the difference between the two delays, (TM - TS)/2. The delay
compensation is initiated upon reception of the master node's pulse by the slave terminal. A
synchronization error of a few microseconds between two PDH-based synchronization devices
was experimentally obtained in multi- (four or five) repeater microwave.
Figure 8.3-2:
Figure 8.3-3 shows another implementation. A terminal acts as the reference clock for the
system. A numbered message is sent from the synchronizing terminal at time tA1. It is received
at time tB1 at the reference terminal and returned back at time tB2. It is received at the
synchronizing terminal at time tA2. The times tB1 and tB2 are sent with the next message to the
synchronizing terminal. The difference t between the clocks in the synchronizing and the
reference terminal can now be calculated by the synchronizing terminal as
t =
t A1 + t A2' t B1' + t B 2
2
2
The clock in the synchronizing terminal can now be adjusted by a fraction of t until t becomes
zero. The synchronizing and reference terminal clocks are synchronized and the samples of
current can be compared at the same sampling instant. Since the clocks are crystal controlled,
they maintain synchronism for long times of transmission interruptions [16].
Figure 8.3-3:
Another implementation is shown in Figure 8.3-4. In this case the terminals also use the same
data polling technique as described above for the measurement of the channel delay time.
Every terminal calculates the time delay with regard to the other ones as
158 / 172
1 = 2 =
(t A2' t A1 ) (t B 2 t B1' )
2
With the measurement of the time delay, the local sampling timing can be adjusted as
t AB 2 = t A2 .
Figure 8.3-4:
Although this method can achieve synchronization between the terminals, an asynchronous
operation where the internal clocks of the terminals dont need to be synchronized can be also
performed. The sampling time of the received current vector values from the remote terminal
can be measured as tAB2. Since the current samples have not been taken at the same sampling
instant, a vector transformation in software is required to rotate the remote vector by an angle
corresponding to the time tAB2 - tA2, and then to compare with the local value sampled at tA2.
These procedures assume the same time delay in both directions, so in
systems where both directions can be switched via different routes (SDH and
the difference in time delay for data transfer in both directions will introduce
determination of the differential current [15], [5]. Requirements on differential
solutions are discussed in chapter 6.
A4.1.2
communication
ATM networks)
an error in the
time delay and
159 / 172
GPS
receivers
Slave
stations
Data
TSE-M
SDH
Time
signal
CSE-S
Slave
stations
S Time
signal
TSE-S
SDH
Time
signal
Figure 8.3-5:
160 / 172
LIST OF FIGURES
FIGURE 2.1-1:
FIGURE 2.1-2:
FIGURE 2.1-3:
FIGURE 2.2-1:
FIGURE 2.4-1:
FIGURE 2.4-2:
FIGURE 2.4-3:
FIGURE 2.5-1:
FIGURE 3.1-1:
FIGURE 3.1-2:
FIGURE 3.1-3:
FIGURE 3.1-4:
FIGURE 3.1-5:
FIGURE 3.1-6:
FIGURE 3.1-7:
FIGURE 3.1-8:
FIGURE 3.1-9:
FIGURE 3.1-10:
FIGURE 3.1-11:
FIGURE 3.1-12:
FIGURE 3.1-13:
FIGURE 3.1-14:
FIGURE 3.1-15:
FIGURE 3.2-1:
FIGURE 3.2-2:
FIGURE 3.2-3:
FIGURE 3.3-1:
FIGURE 3.3-2:
FIGURE 3.3-3:
FIGURE 3.4-1:
161 / 172
FIGURE 8.3-3:
FIGURE 8.3-4:
FIGURE 8.3-5:
162 / 172
LIST OF TABLES
TABLE 3.1-1: STATE COMPARISON PROTECTION SCHEMES .....................................................................................34
TABLE 4.3-1:
ADVANTAGES AND DISADVANTAGES OF PILOT WIRES .......................................................................60
TABLE 4.3-2: ADVANTAGES AND DISADVANTAGES OF POWER LINE CARRIER LINKS................................................62
TABLE 4.3-3: ADVANTAGES AND DISADVANTAGES OF RADIO LINKS .......................................................................65
TABLE 4.3-4: ADVANTAGES AND DISADVANTAGES OF OPTICAL FIBRE LINKS ..........................................................67
TABLE 4.3-5: ADVANTAGES AND DISADVANTAGES OF SATELLITE LINKS.................................................................69
TABLE 4.4-1: PDH - PLESIOCHRONOUS DIGITAL HIERARCHY LEVELS ....................................................................72
TABLE 4.4-2: SDH - SYNCHRONOUS DIGITAL HIERARCHY LEVELS ........................................................................73
TABLE 5.3-1: SERIAL DATA INTERFACES .................................................................................................................95
TABLE 5.3-2: COMMON PHYSICAL LAN INTERFACES ..............................................................................................98
TABLE 6.1-1: REQUIREMENTS FROM PROTECTION ON TELECOMMUNICATION AND TELEPROTECTION: STATE
COMPARISON SCHEMES. FOR TERMS AND DEFINITIONS REFER TO CHAPTERS 6.1.1.1 AND 6.1.1.2. ...................106
TABLE 6.1-2: REQUIREMENTS FROM PROTECTION ON TELECOMMUNICATION AND TELEPROTECTION: ANALOG
COMPARISON SCHEMES. FOR TERMS AND DEFINITIONS REFER TO CHAPTERS 6.1.1.1 AND 6.1.1.2. ...................107
TABLE 6.4-1: IEC PUBLICATIONS FOR EMC AND INSTALLATION ..........................................................................118
TABLE 7.1-1: PROTECTION SCHEMES VS. MEDIA...................................................................................................120
TABLE 7.1-2: PROTECTION SCHEMES VS. MULTIPLEXING TECHNIQUES ................................................................121
TABLE 7.1-3: PROTECTION SCHEMES VS. NETWORK TECHNOLOGIES ....................................................................122
TABLE 7.1-4: CONFIGURATION SUMMARY .............................................................................................................123
TABLE 7.2-1: CHECKLIST FOR INTERFACE CO-ORDINATION BETWEEN PROTECTION / TELEPROTECTION /
TELECOMMUNICATION DEVICES ........................................................................................................................127
TABLE 8.3-1: TSPEC PARAMETERS ........................................................................................................................155
TABLE 8.3-2: RSPEC PARAMETERS ........................................................................................................................155
163 / 172
BIBLIOGRAPHY
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
164 / 172
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
165 / 172
ABBREVIATIONS
AAL
ABR
ADSS
AM
ANSI
ATM
AUI
BER
BFP
B-ISDN
bit/s
CAC
CASM
CB
CBR
CDM
CDMA
CDT
CDV
CDVT
CER
CES
CLR
CMR
CPU
CRC
CSMA/CD
CT
CVT
dB
DCE
DCS
Diff-serv
DSL
DTE
DTT
DWDM
DXC
EDFA
EHV
EIA
EMC
EMI
EMS
EPRI
FDDI
FDM
FDMA
166 / 172
FM
FR
GEO
GI
GLONASS
GOMSFE
GOOSE
GPS
GSM
HV
Hz
IEC
IED
IEEE
IETF
Int-serv
IP
ISDN
ISO
ITU
kbit/s
LAN
LD
LED
LEO
MAN
MAU
Mbit/s
MBS
MCM
MCR
MDT
MEO
MM
MMS
MODEM
MUT
NTP
OPGW
OSI
PCM
PCR
PDH
PEP
PES
PLC
POTS
PSK
PTO
QAM
QoS
Frequency Modulation
Frame Relay
Geosynchronous Earth Orbit (Satellite)
Graded Index (optical fibres)
Global Navigation Satellite System
Generic Object Models for Substation and Feeder Equipment
Generic Object Oriented Substation Event
Global Positioning System
Global System Mobile
High Voltage
Hertz
International Electrotechnical Commission
Intelligent Electronic Device
The Institute of Electrical and Electronics Engineers
Internet Engineering Task Force
Integrated services
Internet Protocol
Integrated Services Digital Network
International Standards Organization
International Telecommunications Union
kilobits per second
Local Area Network
Laser Diode
Light Emitting Diode
Low Earth Orbit (Satellite)
Metropolitan Area Network
Media Attachment Unit
Megabits per second
Maximum Burst Size
Multi-Carrier Modulation
Minimum Cell Rate
Mean Down Time
Medium Earth Orbit (Satellite)
Multi-Mode (optical fibres)
Manufacturing Message Services
Modulator - Demodulator
Mean Up Time
Network Time Protocol
Optical Ground Wire
Open Systems Interconnection
Pulse Code Modulation
Peak Cell Rate
Plesiochronous Digital Hierarchy
Peak Envelope Power
Power Engineering Society (of IEEE)
Power Line Carrier (equipment)
Plain Old Telephone Service (System)
Phase Shift Keying
Public Telephone Operator
Quadrature Amplitude Modulation
Quality of Service
167 / 172
RFC
RFI
RSpec
RSVP
RTCP
RTP
RTU
SCADA
SCR
SDH
SM
SNMP
SNR
SOH
SONET
SS
SSB
STM(-N)
TCP
TDM
TDMA
TE
TMN
TSpec
UBR
UCA
UDP
UTP
VAC
VBR
VBR-nrt
VBR-rt
VC
VDC
VF
VP
WAN
WDM
168 / 172
INDEX
2
data integrity..........................................................101
definition ...........................................................105
datagram ..................................................................82
datagram networks...................................................81
deblocking .........................................................30, 41
deblocking overreach distance protection ...............41
dedicated protection ring
SDH/SONET.......................................................74
delay ......................................... See propagation time
delay compensationSee propagation time
compensation
dependability ...............................................33, 55, 86
definition ...........................................................104
differential delay...... See propagation time symmetry
differential protection ........................................47, 58
digital circuits ..........................................................57
digital communication systems ...............................56
digital hierarchies
PDH, SDH...........................................................72
directional distance relay.........................................33
directional overcurrent relay....................................33
diversity...................................................................57
space, frequency..................................................64
DWDM
Dense Wavelength Division Multiplex ...............70
fading.......................................................................63
fault clearing............................................................13
fault clearing system................................................17
FDM
frequency division multiplex...............................69
frequency division multiplex...............................60
fibre-optic cables .....................................................65
frequency modulation........................................26, 63
FSK
frequency shift keying.........................................30
full-wave phase comparison ....................................30
169 / 172
installation .............................................................116
interface ...................................................99, 112, 126
contact .................................................................93
EMC....................................................................94
fibre-optic............................................................95
LAN / Ethernet..............................................85, 93
protection / telecommunication... 26, 30, 56, 93, 99
serial....................................................................94
VF - voice frequency...........................................94
interference26, 35, 36, 38, 39, 55, 59, 60, 61, 62, 63,
64, 65, 67, 88, 90, 95, 104, 105, 116, 160
internal timing synchronization .............................157
interoperability
of protection relays............................................137
intertripping underreach distance protection ...........34
intra-substation networks.......................................132
IP networks..............................................................91
ITU-T ......................................................................73
J
LAN
local area network ...........................75, 82, 96, 132
topologies............................................................83
LD
laser diode ...........................................................66
LED
light emitting diode .............................................66
line traps ..................................................................61
load shedding...........................................................53
lock out signal .........................................................47
loop-back.................................................................57
M
OPGW
optical ground wire .............................................65
optical fibre interface...............................................95
optical fibres ............................................................65
optical transmitters ..................................................66
OSI reference model................................................78
over reaching ...........................................................34
overall operating time..............................................55
definition ...........................................................104
P
170 / 172
interface...............................................................94
pilot-wire relay ........................................................94
PLC
analog, digital......................................................60
channel impairments ...........................................61
coupling...............................................................61
modes, propagation modes..................................61
power line carrier ................................................60
reliability .............................................................60
plesiochronous.........................................................72
power line carrier...............................................40, 41
power system faults .................................................12
predictive out-of-step protection .............................53
propagation time29, 39, 53, 55, 58, 67, 68, 75, 86, 88,
89, 100, 104, 106, 107, 110, 115, 132, 144, 149,
152
compensation ........................ 24, 30, 115, 157, 158
definition ...........................................................103
difference ..........................................................133
symmetry74, 75, 86, 89, 100, 106, 107, 108, 109,
115
variation ...................... 68, 100, 101, 106, 107, 115
propagation time symmetry.....................................58
definition ...........................................................103
propagation time variation........... 75, 81, 90, 100, 155
definition ...........................................................103
protection
communication dependent ..................................19
communication-aided..........................................19
protection function
introduction .........................................................18
protection operating time
definition ...........................................................104
protection scheme....................................................19
analog comparison ........................................19, 21
state comparison............................................19, 33
protection switching
SDH/SONET.......................................................74
protection system.....................................................17
PSK
phase shift keying................................................63
Q
QAM
quadrature amplitude modulation..................61, 63
R
Radio
licensed, unlicensed ............................................63
reactor protection.....................................................48
repeater ....................................................................85
requirements
from wide-area protection.................................110
re-routing .................................................................57
time coordination ..............................................101
risks (for protection)
delay related21, 29, 33, 52, 55, 56, 57, 58, 59, 61,
64, 67, 68, 71, 74, 75, 80, 82, 86, 89, 90, 91, 92,
100, 101, 103, 104, 108, 115, 119, 128, 129,
133
maloperation..............................................106, 107
multiplexing ........................................................69
network related............................................58, 133
PLC related .........................................................62
rented circuits......................................................56
re-routing.....................................................58, 101
security related ..................................................105
signal crossover...................................................58
signal loopback ...........................58, 102, 113, 119
S
satellites ...................................................................67
saturation ...........................................................22, 24
SDH.................................................................88, 133
synchronous digital hierarchy .......................73, 78
security ........................................................33, 55, 86
definition ...........................................................105
self-healing
SDH/SONET.......................................................89
serial interface .........................................................94
service networks ......................................................78
shared protection ring
SDH/SONET.......................................................74
signal quality ...........................................................55
signal transfer delay.................................................55
slips..........................................................................88
SNR
signal-to-noise ratio.............................................55
SONET
synchronous optical network...............................73
splitting protection...................................................50
spread spectrum.......................................................71
squelching..............................................................102
stabilizing angle.......................................................29
stabilizing protection ...............................................53
standardization.......................................................138
starters .....................................................................29
state comparison protection .....................................33
statistical multiplexing.............................................71
ATM....................................................................89
synchronization......................................................102
synchronous transport module.................................73
system-wide protection............................................49
T
TDM
fixed, synchronous ..............................................71
statistical, asynchronous......................................71
time division multiplex .................................61, 70
TDMA
time division multiplex access ............................64
telecommunication system
definition ...........................................................103
introduction, purpose...........................................17
171 / 172
usage of ...............................................................19
teleprotection equipment / function
definition ...........................................................103
teleprotection function
introduction, purpose ..........................................17
teleprotection system
definition ...........................................................103
time coordination.....................................................51
time delay variation .................................................30
time stamping ........................................................101
times
operating times..................................................143
timing synchronization for simultaneous sampling
..........................................................................157
transformer protection .............................................47
transmission time.....................................86, 106, 107
definition ...........................................................104
maximum actual................................................104
nominal .............................................................104
transport networks .............................................77, 88
WDM
Wavelength Division Multiplex ..............65, 66, 70
wide-area current differential protection .................52
wide-area protection ..........................................51, 57
requirements on telecommunication .................110
wide-area timing synchronization............................52
Z
zone ..... 15, 22, 35, 36, 38, 39, 42, 43, 46, 49, 51, 115
172 / 172