Sie sind auf Seite 1von 7

Testing Teleprotection Schemes by Transient Network Simulation

Boris Bastigkeit
OMICRON electronics GmbH
boris.bastigkeit@omicron.at

Introduction
Absolute selectivity and short tripping times of the line
protection in meshed networks can be achieved by
using line differential protection - or distance protection
with teleprotection schemes. As a result of the exchange of information between the protection devices at
the line ends, fault tripping is also quicker in the range
of the last 10-20% of the line, where instantaneous
tripping is not performed in conventional distance protection. Testing of such distributed protection systems
with communication features is performed according to
the state of the art using GPS synchronized end-to-end
testing. The example of "testing distance-teleprotection
systems" is used in this article to demonstrate how the
simulation of faults on the protected line by means of a
network model can substantially simplify testing. Hence,
not only the correct parameter setting and function of
the protection device are ensured but also the plausibility of the calculated parameter settings is verified.

Recently network model software has become available


that enables easy network simulation also in substations. When one wants to open up to the idea of using a
network model in a substation, it is important to avoid
being discouraged by the concept of "transient network
simulation". An easy-to-use tool may even simplify the
testing task by using a network model. The following
application report describes the practical implications of
this.

Until now, using a transient network model for testing


has rather not been a topic in the field of commissioning
and routine testing. Typical users of network models are
universities, specialized research institutes, engineering
departments of protection equipment manufacturers,
protection laboratories of major power utility companies,
etc. Network models are reputed to be expert tools that
are complex and difficult to handle. This is often due to
the fact that the available network models are very
powerful and allow modelling of very complex networks.

Line data
Line type: Overhead line
Grounding: Compensated network
Nominal voltage: 110kV
X (primary): 7.175 ohms
R (primary): 3.44 ohms

However, if we consider that a test of protection devices


by transient network simulation a most realistic test,
there is a growing wish for a simple network model
which is also suitable for commissioning and routine
testing. The main advantage of such a method is that it
offers a very high degree of reliability that the protection
device fulfils its designated main function and that the
parameter settings have also been calculated correctly.
If one knows the protection concept and how the protection equipment should respond to certain faults,
testing is possible independently of the protection device type used.

Voltage transformer: : 110000/100


Current transformer: : 600/1
CT starpoint towards busbar

Description of the distance-teleprotection system


A field test was carried out to verify the usability of a
transient network model for the purpose of routine or
commissioning testing of a distance teleprotection
scheme. The protected line and the protection concept
are described below.

CT/VT data
Side A:

Side B:
Voltage transformer: :110000/100
Current transformer: :600/1
CT starpoint towards busbar

OMICRON electronics GmbH 2010

Protection Concept/ Test Concept


Distance protection devices with communication channels for teleprotection are used.
Teleprotection method: A release signal is sent to the opposite end when a fault is detected in zone 1 or in the extended
zone. The extended zones are released for instantaneous tripping when the opposite end sends the release signal. If
there is no pick-up, a received release signal is returned to the other end (echo scheme).

Fig. 1: Grading of distance protection zone 1 and extended


zone / fault locations

The following test cases (1 5) are typically carried out. They describe the function of the teleprotection scheme in detail
and are the basis for designing a test plan.
(1) Fault at the middle of the line: Both ends feed into
the fault. Teleprotection communication channel active.
The protections in A and B detect the fault as a zone 1
fault and both perform instantaneous tripping.
(2) Double-end feeding, teleprotection communication
channel active.
Fault near busbar B. The protection in A detects the
fault as a zone 2 fault and issues a "Send PSIG". Upon
the PSIG receipt, the extended zone is released in A
and instantaneous tripping is performed.
(2a) Single-end feeding, teleprotection communication
channel active.
Fault near busbar B. The protection in A detects the
fault as a zone 2 fault and issues a "Send PSIG". Since
protection does not pick up in B because of single-end
feeding, B returns an echo signal upon the PSIG receipt. Upon the PSIG receipt, the extended zone is
released in A and instantaneous tripping is performed.
(2b) Double-end feeding, teleprotection communication
channel failed. Fault near busbar B. The protection in A
detects the fault as a zone 2 fault. The protection in B
detects the fault as a zone 1 fault and performs instantaneous tripping. Due to a PSIG failure, both ends have
to process the zone timers.
(3) Fault beyond busbar B. The protection in A detects
the fault as a zone 2 fault and issues a "Send PSIG".
The protection in B detects the fault in reverse direction,
trips with the time for the reverse zone, and must not
issue a "Send PSIG". Since the protection in A does not
receive a PSIG, tripping is performed in zone 2.

(4) Double-end feeding, teleprotection communication


channel active. Fault near busbar A. The protection in B
detects the fault as a zone 2 fault and issues a "Send
PSIG". The protection in A detects the fault as a zone 1
fault, initiates instantaneous tripping and issues a "Send
PSIG". Upon the PSIG receipt, the extended zone is
released in B and instantaneous tripping is performed.
(4a) Single-end feeding, teleprotection communication
channel active. Fault near busbar A. The protection in B
detects the fault as a zone 2 fault and issues a "Send
PSIG". Since protection does not pick up in A because
of single-end feeding, A returns an echo signal upon the
PSIG receipt. Upon the PSIG receipt, the extended
zone is released in B and instantaneous tripping is
performed.
(4b) Double-end feeding, teleprotection communication
channel failed. Fault near busbar A. The protection in B
detects the fault as a zone 2 fault. The protection in A
detects the fault as a zone 1 fault and performs instantaneous tripping. Due to a PSIG failure, both ends have
to process the zone timers.
(5) Fault beyond busbar A. The protection in B detects
the fault as a zone 2 fault and issues a "Send PSIG".
The protection in A detects the fault in reverse direction,
trips with the time for the reverse zone, and must not
issue a "Send PSIG". Since the protection in B does not
receive a PSIG, tripping is performed in zone 2.

OMICRON electronics GmbH 2010

Testing with a transient network model:


A new easy to use transient simulation test tool can be
used to establish a field compatible test plan. The field
test is carried out in the course of a cyclic routine test.
After conventional end-to-end testing is carried out, the
new method - using the transient network model - is
applied. The test plan is prepared with an appropriate
tool for automatic testing. The following describes how
a test plan is created using the new method.

this, for example, by defining the phase angles of the


two sources slightly out of phase. The amount of the
fault currents for 2-phase and 3-phase faults is controlled simply by the value of the positive sequence
impedances of the sources.

Selecting the network type


The first action in the transient network model is the
selection of the network type to be used for the test.
The appropriate network configuration that is selected is
"single line". This is one line with double-end feeding.
When the circuit breakers are opened this model also
simulates a line with single end feeding. Both cases are
required for the test.

Fig. 3: Transient network model - Configuring the sources

Specifying the line data

Fig. 2: Transient network model - Selecting the "single line"


configuration

Entering the line impedances correctly in the network


model is the most important step for testing distance
protection devices. The network model supports both
entering the line data as primary values or as secondary values. In this example secondary data are used,
therefore the primary line data have to be converted to
secondary data using the voltage (V) and current (I)
transformer ratios.

X sec = Xprim
Setting the sources
As the next step the sources of the network model are
configured. The described transient network model
offers the advantage that the pre-assignment of the
individual parameters of the sources has already been
done for typical cases and it is - especially for testing
distance protection - usually not necessary to make any
changes to the default settings. The only adjustment
recommended for the given example is to set the zero
sequence impedance of the sources to the maximum
impedance since the case discussed is a compensated
network which does not permit a high fault current in
case of a ground fault and causes a voltage offset.
However, since in the case described only 2-phase
faults are simulated the zero sequence impedance of
the sources is not relevant. If it is desirable that a load
current flows in the prefault condition, one can achieve

rI
rV

R sec = Rprim

Xsec ... Secondary Reactance


Xprim ... Primary Reactance
Rsec ... Secondary Resistance
Rprim ... Primary Resistance

Secondary line data (positive sequence):


X1 (sec): 3.914
R1 (sec): 1.876
Z1 (sec): 4.34
Phi Z1: 64

OMICRON electronics GmbH 2010

rI
rV

The data for the zero sequence impedance of the line


are not available. In compensated networks, this is only
relevant in case of a cross country fault. However, this
is not the focus of the test. Testing ground faults is
more important in networks with other starpoint handling. If the zero sequence impedance should not be
known in such a case, it can be measured. [5] For the
purposes of the trial, the following arbitrary assumptions
have been made:
Z0(sec): 16
Phi Z0: 85

Fig. 5: Transient network model - Configuring the fault


quantities

The fault inception angle is not relevant in the example


discussed. The fault resistance (arc resistance) is defined as metallic fault (0 ).

Fig. 4: Transient network model - Entering the data of the line


to be protected

To provide automatic assessment of the test steps, it is


possible to define which response is to be expected
from the protection (for example, instantaneous tripping,
send release). This can be done first for the relay at the
A end. Later, these settings have to be adjusted accordingly in the test plan copied for the B end for cases
where the protections at the both ends are expected to
react differently.

Test sequence
In accordance with the protection concept as described
at the beginning of this article, the various test steps (for
example, one transient network model module per test
step) are now generated in the test plan by copying the
transient network model test module defined so far. The
following items are defined for each test step:

Fault location / line affected


Fault type
Prefault duration
Fault duration
Postfault duration

Fig.6: Transient network model - Definition of the


measurement conditions for the automatic assessment of the
test steps

In each transient network model module inserted,


"GPS" is defined as the trigger for the start of the test.
This allows the generation of time-synchronized test
signals at both ends of the line. State of the art test
equipment allows to synchronize test signals via GPS
with an accuracy in the range of a few microseconds.

Test plan for end B


Once one has defined all the test steps and the corresponding measurement conditions, one can produce the
test plan for the B end by creating a copy of the test
plan for the A end. As already mentioned it is necessary
to adjust the measurement conditions in the test plan
for the end B for the test cases (3) and (5).

OMICRON electronics GmbH 2010

Outputs/hardware configuration
The test plan for end B needs to be configured that a
connected test set generates the simulated currents for
end B in the corresponding hardware configuration
dialog.

CMGPS

CMGPS

Test Set
End A

PC-A with
Test SW

travel time. Once a standardized test plan has been


created - for the next line, with an identical protection
concept, but different line parameters, the test plans are
adjusted in less than 30 minutes.

Test Set
End B

PC-B with
Test SW

Fig. 9: Test plan for end B

Fig. 7: End-to-end test using transient network model

Once one has defined which test set is to correspond


with which network node in the simulation software, one
can verify the assignment again on the "Outputs" tab of
the software. The "Outputs" tab allows also defining
how the current transformer starpoint is connected either towards the busbar or towards the line. One can
also activate current transformer models for the simulation of CT saturation effects. [6] In the example described, simulation of CT saturation is not applied.

Fig. 10: Transient network model Time signal view: Test


result for test case 5 / End B

Fig. 8: Transient network model - Outputs tab

Testing, practical experience


As described above, the preparation of a test with a
transient network model can be done very efficiently.
When using transient network model test plans, there is
not much difference regarding the test sequence as
compared to conventional end-to-end tests. Conducting
field tests with the transient network model method, it
was found that it takes about 1-2 hours to create the
test plan for both ends once the line and protection
concept data have been identified. The actual test in the
field took about 2 hours (including setting up and removing the test equipment and GPS), not including

Fig. 11: Transient network model Visualization of impedance


trajectories for test case 5 / End A and End B

OMICRON electronics GmbH 2010

Summary

Literature

End-to-end testing of distance protection systems with


teleprotection can be prepared and carried out very
easily using a transient network model that can be
applied in the substation. A field trial evidenced that the
correct function of the protection system can be verified
efficiently by entering only a few data and within a minimum of time. This shows that transient network simulation, which has previously rather been an issue for
research and development, may now also become an
issue for routine testing and commissioning.

[1] Oliver Gludowatz, Prfung eines Leistungsschutzes mit Signalvergleich; OMICRON User Meeting
2000, Mnchen
[2] Boris Bastigkeit, "Neue Mglichkeiten bei der Prfung von Leitungsdifferentialschutz-Einrichtungen";
OMICRON User Meeting 2008, Lbeck
[3] Dr. Peter Meinhardt, "Zeitsynchronisierte Prfung
von Leitungsdifferentialschutz"; OMICRON User
Meeting 2006 Weimar

The major advantages are:

End-to-end
testing
verifies
the
complete
teleprotection scheme
Very realistic testing by applying transient
simulation
High degree of test automation and standardization
The test verifies proper function of the protection
for the most important operational conditions and
fault conditions
The plausibility of the calculated protection settings
is verified
Testing is independent of the type of protection
device used (black box test)
It is not required to know any relay parameters for
this kind of test - only a few network parameters
(line-data, CT, VT) and
the protection concept
need to be known
The test software calculates secondary test
quantities automatically from primary data and from
CT/VT data
Time savings when preparing the test plans as
compared to the conventional end-to-end test
method
Familiar test environment the transient network
model behaves like easy to use, state of the art
protection test tools in contrast to tools like EMTP
etc.

[4] Dr. Fred Steinhauser, "Netzsimulation auf dem PC


Neue Mglichkeiten fr die Schutzprfung";
OMICRON User Meeting 2002, Lindau
[5] Alexander Dierks, Harry Troskie, Michael Krger,
"Messung von Freileitungsparametern zur Verbesserung der Selektivitt von Distanzschutzrelais",
OMICRON User Meeting 2005, Friedrichshafen
[6] Dr. Peter Meinhardt, "Test des Einflusses von
Stromwandlersttigung und -dimensionierung mit
der Netzsimulationssoftware NetSim"; OMICRON
User Meeting 2008, Lbeck

OMICRON electronics GmbH 2010

About the Author


Dipl.-Ing. Boris Bastigkeit (IEEE, OVE) was born in
1969 in Kempton Park / South Africa. He received his
diploma (master degree) in Electrical Engineering at the
Technical University of Graz (Austria) in 1998. He
joined OMICRON electronics in Austria 1997 where he
held various positions in development and sales before
he was appointed to his current position as head of
product management in the field of testing solutions for
protection and measurement systems. He is member of
the working group IEC TC 95 MT4.
boris.bastigkeit@omicron.at

OMICRON electronics GmbH 2010