Ovcr-CurrcntRcIayModcIImpIcmcntationIor

RcaITimcSimuIation Hardvarc-In-thc-Loop,HIL)VaIidation
Muhammad Shoaib AImas, Rujiroj LccIaruji, and Luigi \anIrctti
Royal Institute of Technology (KTH)
Teknikringen 33, SE-lOO 44, Stockholm, Sweden
Email: [email protected]@kth.se [email protected]
Abstract-Digital microprocessor based relays are currently
being utilized for safe, reliable and efcient operation of power
systems. The overcurrent protection relay is the most exten
sively used component to safeguard power systems from the
detrimental efects of faults. Wrong settings in overcurrent relay
parameters can lead to false tripping or even bypassing fault
conditions which can lead to a catastrophe. Therefore it is
important to validate the settings of power protection equipment
and to confrm its performance when subject to difrent fault
conditions. This paper presents the modeling of an overcurrent
relay in SimPowerSystems (MATLAB/Simulink). The overcurrent
relay has the features of instantaneous, time defnite and inverse
defnite minimum time (IDMT) characteristics. A power system
is modeled in SimPowerSystems and this overcurrent relay
model is incorporated in the test case. The overall model
is then simulated in real-time using Opal-RT's eMEGAsim
real-time simulator to analyze the relay's performance when
subjected to faults and with different characteristic settings
in the relay model. Finally Hardware-in-the-Loop validation of
the model is done by using the overcurrent protection feature
in Schweitzer Engineering Laboratories Relay SEL-487E. The
event reports generated by the SEL relays during Hardware-in
the-Loop testing are compared with the results obtained from
the standalone testing and software model to validate the model.
A
CRONYMS
CT Current Transformer
HIT. Hardware-in-the-Loop
RT Real-Time
RTS Real-Time Simulation/Simulators
SIL Software-in-the-Loop
SPS SimPowerSystems
VT Voltage Transformer
I.
I
NTRODUCTION
A power system is vulnerable to faults, either due to
natural disasters or by misoperation of the system due to
operators' negligence. This can result in permanent damage
to power system components leading to considerable costs
for their replacement and in longer disconnections of power
supply to customers, which is highly undesirable. This sets
a requirement for a power system to sustain faults, while
protection systems should minimize the damage in important
components and the effect of faults as much as possible. This
is achieved by using power system protection techniques and
methodologies [1]. Microprocessor-based relays have their
own algorithms for monitoring the power system through
current and voltage inputs fom CTs and VTs respectively.
Protection relays are thoroughly tested using relay test sets
to confrm their reliability and safe operation before commis
sioning them in a substation.
With the advent of RTS and compact reconfgurable I/O
controllers , large power systems can be simulated and their
behavior can be analyzed in both steady state and faulted
conditions. Based on this analysis, protection system and
power system controllers can be developed for more effcient,
reliable and safe operation of power systems. RTS are cur
rently being used for HIT. validation, developing algorithms
for adaptive protection, designing system integrity protection
schemes (SIPS), remedial action schemes (RAS) and other
applications [2].
In this paper, a detailed model for overcurrent relays is
provided. To assess the performance of this model a test
system is designed in SPS and is simulated in RT using Opal
RT's eMEGAsim real-time simulator. Once the SIL model
is certifed, it is further validated by HIL simulation using
the overcurrent protection function of Schweitzer Engineering
Laboratories (SEL) relay SEL-487E. The results obtained
by the SIT. and the HIL are compared. SPS does not have
a dedicated library for protection functions. Therefore it is
necessary to validate the model with HIL so that the validated
software model can be used for accurate representation of
power system protection components. Similar approach for
protection relay model validation has been demonstrated in
[3] [4] using diferent modeling platform and RTS.
The remainder of the paper is arranged as follows. Section II
ofers a literature review on overcurrent relays. Section III
presents the details of the overcurrent relay modelled in
SPS. Section IV focuses on the design of test system in
SPS and incorporation of proposed relay model in the test
case. SIL simulation and results are discussed in Section V,
while standalone testing and HIL validation is explained in
Section VI. Finally in Section VII, conclusions are drawn and
future work is outlined.
II.
O
VERCURRENT
R
ELAYS
As the name states, an overcurrent relay provides protection
against over currents. This relay uses current inputs fom a
CT and compares the measured values with preset values.
Figure 1 shows the logical representation of an overcurrent
relay. If the input current value exceeds the preset value,
the relay detects an overcurrent and issues a trip signal
to the breaker which opens its contact to disconnect the
protected equipment. When the relay detects a fault, the
condition is called fault pickup. The relay can send a trip
978-1-4673-2421-2/12/$31.00 2012 111 4789
lHuI LuIIEHI
|L|uVuE
I 5gH
LOnIIOI
Fig. !. Logical representation of Over-Current Relay
signal instantaneously after picking up the fault (in the case
of instantaneous overcurrent relays) or it can wait for a
specifc time before issuing a trip signal (in the case of time
overcurrent relays). This time delay is also known as the
operation time of the relay, and is computed by the relay
on the basis of the protection algorithm incorporated in the
microprocessor [5].
. Classifcation of Over-Current Relays
Overcurrent relays are classifed on the basis of their
operation time, in the following three categories:
1) Instantaneous Overcurrent Relay: These relays instan
taneously send a trip command to the breaker as soon as the
fault is detected (input current greater than the preset value).
They do not have any intentional time delay. They are usually
implemented close to the source where the fault current level
is very high and a small delay in operation of relay can cause
heavy damage to the equipment. So an instantaneous relay is
used there to detect and respond to a fault in few cycles.
2) Defnite Time Overcurrent Relay: This type of over
current relay is used for backup protection (e.g. back up
protection for transmission line where primary protection is
distance relay). If the distance relay does not detect a line
fault and does not trip the breaker, then after a specifc time
delay, the overcurrent relay will send a trip command to the
breaker. In this case, the overcurrent relay is time delayed by
a specifc time which is just greater than the normal operating
time of the distance relay plus the breaker operation time.
3) Inverse Defnite Minimum Time (IDMT) Overcurrent
Relay: This relay has an inverse time characteristic. This
means that the relay operating time is inversely proportional
to the fault current. If the fault current is higher, the operating
time will be lesser [6]. It can be graded for a very large range
of operating times and fault currents [7]. The characteristics
of an IDMT overcurrent relay depend on the type of standard
selected for the relay operation. These standards can be
ANSI, IEEE, lAC or user defned. The relay calculates the
operation time by using the characteristic curves and their
corresponding parameters [8]. Any of the above mentioned
standards can be used to implement a characteristic curve
for an overcurrent relay. The overcurrent relay will then
calculate the operation time corresponding to that particular
characteristic curve.
In accordance with IEC 60255 or BS142, the characteristics
of IDMT relays are represented with the following equation:
where:
C
J

Z TMS
,` 1
s
(1)
artI art2 art3

Fig. 2. Functional Block Diagram of an Overcurrent Relay
J. Relay operation time.
C: Constant for relay characteristic.
Is: Current Set point.
I: Current Input to the relay.
0: Constant Representing Inverse Time Type (0 ` 0).
TMS: Time Multiplier setting controls the relay tripping time.
By using appropriate TMS settings, the grading of a protection
network system can be achieved [9]. The range of TMS is
normally 0.1 to 1.0.Different types of curves can be obtained
by varying 0 and C [10]. Table I below shows values for 0
and C corresponding to each curve:
TABLE I
OIIILRLN11YPLS CI !NvLRSLCHARAC1LRIS1ICSCURvLS
Relay Characteristic Tpe C C
Standard Inverse 0.02 0.14
Very Inverse 1 13.5
Extremely Inverse 2 80
Long Inverse 1 120
III. OVERCURRENT
R
ELAY
M
ODELING
F
OR
R
EAL-
T
IME
S
IMULATION
The detailed model of an overcurrent relay is implemented
in the SimPowerSystems (MATLAB/Simulink) Toolbox. The
reason for using the SPS Toolbox is because it operates in the
Simulink environment and is a dedicated tool for modeling
and simulating power systems [II]. Another motivation in
using SPS is because it is compatible with the OPAL-RT
eMEGAsim real-time simulator [12] which is used to simulate
the power system model in RT and for HIL validation of the
model.
The block diagram presented in Figure 2 shows the three
major functional components of an overcurrent relay. These
components and their functionalities are discussed next.
. Part 1
In order to present an accurate model of the microprocessor
based overcurrent relay, the current inputs which are analog
signals are converted to digital signals by Analog to Digital
Conversion. In Simulink this can be achieved by using a Zero
Order Hold Block. This ZOH block is available in the Discrete
Library of Simulink.
D. Part 2
A digital flter is used to extract the fundamental signal. A
down sampler is used in order to reduce the sampling rate.
Down sampling helps avoiding anti-aliasing [13] efects. The
digital flter used here is a low pass digital FIR flter [14].
4790

pOwQU
Fig. 5. Test Case Model developed in SimPowerSystems
C Part 3
Once the RMS value of the current is obtained, this
current is fed into the relay protection algorithm block. This
block compares the current value with the pickup value. If
the input current exceeds the pickup value, the relay will
compute the operation time for this scenario taking into
consideration the characteristic curve (Standard Inverse, Very
Inverse, Extremely Inverse or Long Inverse) and sends a trip
signal once the operation time is elapsed.
Figure 3 in p. 7 shows the overall Simulink model of the
overcurrent relay.
The protection algorithm implemented in the model is
shown in Fig. 4, p. 7. The overcurrent model is masked to
a single subsystem (block) and a graphical user interface is
provided where the user can select the type of overcurrent
relay (instantaneous, time defnite or IDMT) and can further
select the different characteristic curves for IDMT overcurrent
relay.
IV.
T
EST
C
ASE IN
S
IMPOWER
S
YSTEMS
Figure. 5 shows the test case model designed in SPS to
couple the overcurrent relay model and investigate its charac
teristics when subjected to faults. The major components of
the test case are:
Three phase voltage source, 50 Hz, 1 kV phase voltage.
Transmission Line (7-section), 100 km
Three phase fault block to introduce single phase to
ground, phase to phase and three phase to ground faults.
Circuit breaker to disconnect load with trip signals fom
the overcurrent relay model.
Three phase series RLC load of 1 MW.
Simulation time step " 50 j sec.
In order to guarantee a safe, reliable and fast operation
of overcurrent relays, settings such as pickup value, time
multiplier settings, characteristic curves, and others, have
to be chosen carefully. Otherwise, the relay can either not
detect the faulted conditions (and will not trip) or send a
false tripping command. Figure 6 shows some of the general
steps which are followed in selecting the pickup value of the
overcurrent relay. Following these steps, fault analysis of the
test system was conducted. The results are shown in Fig. 7.
Table II shows the settings of the overcurrent relay model
which are based on the results fom fault analysis (Fig. 7)
CaclulaIeu|| CalculaIeShorI Consider1urn3aIioIor
'oadCurrenI CircuiICurrenI CurrenI1ransIormer
SelecI|nsIanIaneous/ SelecIickupvalueoyusinIherelaIion
DeIiniIeor|DM1

l.2 xMox|mum|ooJCurrent:P|cku

characIerisIics vo|ue:.U xM|n|mum|ou|tCurrent

Fig. 6. Steps Involved in Selecting Pickup Values for Overcurrent Relays
TABLE II
1LS1 SYS1LM ANALYSIS AN RLLAYSL11INS
Full Load Current 330 A
Minimum Fault Current (Single
1760 A
Measurements Phase to Ground)
Maximum Fault Current (Three
3088 A
Phase to Ground)
Current Transformer Ratio 400:1
Relay Settings Pickup Value 1 A
Time Multiplier Setting 0.5
and the procedure mentioned in Fig 6. TMS was set to 0.5
to see a noticeable difference in operating time of the relay
when using different IDMT characteristic curves as shown in
Fig 8. From Table II it is obvious that the introduction of
three phase to ground fault will result in the CT's secondary
current equal to 7.72 A (i.e. 7.72 times the pickup value).
V.
S
OFTWARE-IN-THE-
L
oop
V
ALIDATION
The parameters of the overcurrent relay model were up
dated and the model was simulated ofine. For real-time SIL
validation steps presented in Figure 9 were followed. The trip
signal generated fom the Simulink model was sent to one of
the digital output of the simulator. This digital output was
looped back through one of the digital inputs. This digital
input was confgured to control the status of the breaker.
So when a fault is introduced and the relays picks up the
overcurrent condition, it sends a trip signal (digital signal 0'
for normal and for trip) to the digital output of simulator
which is coupled with the digital input of the simulator. This
digital input opens the breaker in the SPS model in real-time.
In the test case model, the three phase to ground fault
was applied at t=2 sec. The results for standard inverse,
very inverse and extremely inverse characteristic curves are
shown in Fig. 10 to 12. Only a single phase is shown in
the results as its a case of symmetrical fault and the other
two phases will be phase shifted by 120 degrees. Table III
shows the comparison of results fom proposed model with
the theoretical calculations.
VI.
H
ARDWARE-
I
N-THE-
L
oop
V
ALIDATION
In order to validate the software model and compare its per
formance with an actual relay, HIL simulation is carried out.
For this purpose a transformer diferential protection relay
fom Schweitzer Engineering Laboratories, SEL-487E [15],
is used. To amplify the low level current outputs from real
time simulator, amplifers fom Megger [] 6] are used. The
maximum current which can be sourced from Analog Outputs
4791
If\l)
ee
e e.e! . . e.e
.
_. _ o.os e.ez e.es o.os e.+
Fig. 7. Fault Analysis of the Test System (only Phase A is shown)
(Opoatiog Ti..o V. MuWple of Pickup Valuo fo TMS = O.B)

st n un o

ro
~vu:,+=es
~>.rn u.u

Ratio (Input Current to Relay f Pickup Value)
Fig. 8. Different IDMT characteristic curves for TMS 0.5. The vertical
line at ratio 7.72 is the region of interest in the test case
X
ti:igoa| ttom:imu+ole|a,
Voae|is:ootiguteatoooeottne
0igita|Oututottne :imu|atot
ti:igoa|iste:ei.eaatooeottne0igita|ioutot
tne :imu|atotwni:nis :ootiguteaiotne:imuol
Voae|tosnowttiiogiotne:ooso|e
0iitalOutput
and 0iital
|nputare
looped oack
Fig. 9. Steps for Real-Time SIL Validation of the Relay Model
........

....

_

.... ....
Fig. 10. Results for Relay Model with Standard Inverse Curve
of the simulator is 20 mA. In order to make sure that the
current level never exceeds this limit, the three phase current
signals are attenuated by a factor of 1000. As the maximum
fault current for this test case model is 3088 A (see Table II)
and the CT ratio is set to 400: 1. So the maximum fault
i:

4
x+ts+

......

0 1 3 4 6
......

..

...... ......
Fig. 11. Results for Relay Model with Very Inverse Curve
........ .......

4


....

0.6
0.4
1
0.
:
L 0

....

.
....
Fig. 12. Results for Relay Model with Extremely Inverse Curve
TABLE III
CCMPARISCN CI 1HLRLSUL1SIRCMPRCPCSL RLLAYMCLLSIL
vALIA1ICN vSILALCALCULA1ICNS
Summar of Resuls From Proposed Relay Model
(ault Applied at t=2sec)

Standard Very Extremely

Time (sec) Inverse Inverse Inverse
Operating Time 1.6780 1.0045 0.6827
Ideal Operating Time 1.6777 1.0040 0.6826
Trip Signal Time 3.6890 3.0160 2.6940
Ideal Trip Signal Time 3.6780 3.0040 2.6826
The ideal results are calculated using Eg. (1). Trip Signal Time Time at
which fault is applied Operating Time.
current which can be seen by the relay is 7.72A. With the
attenuation of this signal by a factor of 1000, the current fom
the Analog Outputs of the Simulator will be 7.72 H which
is well within the range of the current output capacity of the
simulator. In order to compensate for this attenuation, the
analog outputs of the simulator are fed to SMRT-l amplifers
(Megger) which amplify the signal by the factor of 1000.
This amplifed signal is then fed to the CT inputs of the
SEL-487E relay. In the software test model in Section IV, few
modifcations were made i.e. including vendor specifc blocks
fom RT-Lab [17] library to access the Analog Outputs of the
Simulator, as shown in Fig. 13.
In addition, one digital inputs of the simulator are coupled
with the digital outputs of SEL-487E, which is confgured to
change its contacts in case of a trip. The digital inputs of the
4792
Fig. 13. Test case model with modifcations for HIL simulation
dEGAsim .Cores,
OP 5600 I}O Extension Chasis
OP5949 Active Monitoring Panel
rjr
-

The digital output of relay is
connected to digital input of the
simulator. As soon as the relay
trips, it is detected by the Digital
Input of the simulator which opens
the breaker in the model being
simulated in real-time
trips due to over-current

MATLAB}Simulink
Design models for real
time simulation
The Analog outputs of the
Simulator are fed into the CT
Inputs of the SEL-487E
Fig. 14. Model-to-Data Work-Flow for HIL simulation. Solid Lines
indicate hardwired connections, dashed lines indicate digital data streams
over ethernet
simulator are confgured to change the breaker status in the
SPS model in real-time. Figure l3 shows the test case model
with additional blocks which are highlighted. The overall
workfow from building the model to simulating it in real
time and executing HIL validation is shown in Fig. 14.
In order to enable the overcurrent feature of the SEL-487E
SEL AcSELerator Quickset [18] is used. The characteristi
curve chosen for the validation is extremely inverse curve. The
same settings were incorporated in SEL-487E (Figure 15) as
in the software model (see Table II). A fault is introduced at
t=2sec. The ACSELERATOR Analytic Assistant [] 9] is used
Fig. 16. Results from Standalone test using Freja-300. Showing the tripping
tIme of 0.698 sec wIth TMS=0.5, pIckup value = 1 and characteristic curve
of extremely inverse. Scale is 1.93 Adiv
@ o:apntoc:\uses\msa|mos\oasHo\ 1u.cv
.U;
v,s
s,o
<,

o,o
1
.,s
s,o
!.
O IQ O OOD, = tOO
Fig. 17. Results from SEL-487E event report generated by HIL simulation.
The operating time calculated here is 34.5 cycles which corresponds to 0.69
sec
to display event report data as oscillography. In addition Freja
300 (Relay Test System) [20] is used to inject the secondary
currents to SEL-487E to confrm the tripping times obtained
fom HIL simulation. For standalone testing, the sofware
interface for Freja 300 i.e. Freja Win [21] is used to give
secondary injections of 0.98 A for the frst 2 seconds and then
a step increase in current to 7.72 A. The results are presented
in Fig. 16 showing that before the fault is applied, the current
input is equal to 0.98 A which is lesser than the pickup value
of 1 A. As the fault is applied at t=2 sec, the current rises
up to 7.72 A. At this point the relay picks up the fault and
its IDMT characteristic is activated which fnally generates
a trip signal after 0.698 sec. As the fault is applied at t=2
sec., and operating time of relay is 0.698 sec., so the tripping
time is equal to 2.698 sec. At t=2.698 sec, the relay D/O
changes its contacts and thus it opens the breaker in the SPS
model and the load is protected against overcurrent. Figure. 17
shows the oscillography fom the event report of SEL-487E
resulted fom HIL simulation. As soon as the relay trips,
SEL-487E generates an event report which can be viewed by
using ACSELERATOR Analytic Assistant. Figure. 18 shows
the operating time for SIL and HIL validations for diferent
characteristic curves.
4793
L
0
W
0
E
1.
1.
1.
8

4
Comparison ofSll and Hil Simulation for Different Characterlctlc Curves

= VEIyhVEI$E
~10h@1hvEr$E
+ OE8 L8 80h
W blL
0 H1
1.2
o
C

O
1
0.8

.I I.I05 I.I1

I.I15 I.I2 I.I25 I.I3 I.I35

Ratio (Input Current Relay /Pickup Value)
I.I4 I.I45
Fig. 18. Comparison of Operating Time by SIL and HIL validations
TABLE IV
I.I5
CCMPARISCN CI 1HLRLSUL1SIRCMPRCPCSLRLLAYMCLLAN HIL
Comparison of Results from Proposed Model and HIL
(Fault Applied at t * 2sec)
Curve Tme (sec) Ideal Cacultions SIL HIL
Standard Inverse
Operating 1.6777 1.6780 1.6850
Trip 3.6780 3.6890 3.7135
Very Inverse
Operating 1.0040 1.0045 1.0150
Trip 3.0040 3.0160 3.0650
Extremely Inverse
Operating 0.6826 0.6827 0.6980
Trip 2.6826 2.6940 2.7800
VII.
C
ONCLUSION AND
F
URTHER
W
ORK
By utilizing Opal-RT's eMEGAsim real-time simulator,
HIL testing of the proposed model is done. The results
for both proposed model and HIL are shown in Table IV.
Both the proposed software model and HIL gives almost the
same result. The difference between Proposed Model and
theoretical calculations is about 11 ms. This difference is due
to the changing of states of the digital I/O of the simulator and
the arc extinction process of the three phase circuit breaker
used in SPS model [22] which waits for the frst zero crossing
of the fault current to open the breaker once its external
controller changes fom 0' to 1'. In case of HIL validation,
there is some extra delay due to the response of the amplifers
and status changing of digital output contact of SEL-487E.
Opal-RT's simulator has the provision to stream out GOOSE
(61850-8-1) and Sampled Values (61850-9-2). The proposed
model can be used to validate the Station Bus [23] and Process
Bus [24] implementation with HIL, this will be reported in a
future publication.
A
CKNOWLEDGMENT
M.S. Almas is supported by the STRONg
2
r/d project, funded by Nordic Energy
Research.
R. Leelaruji is supported by EKC
2
- The Swedish Centre of Excellence in Electric
Power Engineering.
L. Vanfretti is supported in part by the STRONg
2
r/d project, funded by Nordic
Energy Research, by the STndUP )r Energy collaboration initiative and the KTH
School of Electrical Engineering.
The generosity of Schweitzer Engineering Laboratories, Pullman, WA, USA; for
their donation of protection relays, and MeggerfProgramma, Taby, Sweden; for their
donation of different hardware and technical support is deeply acknowledged.
R
EFERENCES
[1] 1. L. Blackburn and T. J. Domin, Prtective Relaying - Principles and
Applications, Third Edition. CRC Press, 2006 .
[2] Terzija, N and Valverde, G. and Deyu Cai and Regulski, F and Madani,
Nand Fitch, J. and Skok, S. and Begovic, M.M. and Phadke, A., "Wide
Area Monitoring, Protection, and Control of Future Electric Power
Networks," Prceedings of the IEEE, vol. 99, no. 1, pp. 80 -93, jan.
2011.
[3] Palla, S. ; Srivastava, A.K. ; Schulz, N.N. , "Hardware in the Loop Test
for Relay Model Validation." Arlington, Virginia, USA: Electric Ship
Technologies Symposium, 2007. ESTS '07. IEEE, May 2007.
[4] Saran, A. ; Kankanala, P. ; Srivastava, A.K. ; Schulz, N.N. , "Designing
and Testing Protective Overcurrent Relay using Real Time Digital
Simulation." Edinburgh, Scotland: Grand Challenges in Modeling and
Simulation, Summer Simulation Conference, June 2008 .
[5] General Electric, "Instruction Manual for Overcurrent Protection (MI
FIl)," available on-line: http://www.gedigitalenergy.com/multilin/.
[6] IEEE Tutorial Course, "Microprocessor Relays and Protection Sys
tems," Course Text 88EH0269-1PWR, IEEE Service Center, Piscat
away, NJ, 1988.
[7] J.P. Whiting, and D. Lidgate, "Computer Prediction of lDMT Relay
Settings and Performance for Interconnected Power Systems," vol. 130.
IEEE Proceedings for Generation, Transmission and Distribution, May
1983.
[8] Walter A. Elmore, "Protective Relaying Theory and Applications," ABB
Power T&D Company Inc., Tech. Rep., 1994.
[9] Schneider Electric, "Protective Relay Settings- Understanding the
IEC based IDMT Settings of Phase Overcurrent Protection," ABB
Power T&D Company Inc., Tech. Rep., 1994, available on-line:
http://selink.schneider-electric.com.sglselink.
[10] Benmouyal, G. ; Meisinger, M ; Burnworth, J. et aI., "IEEE Standard
Inverse-Time Characteristic Equations for Overcurrent Relays," IEEE
Trnsactions on Power Deliver, vol. 14, no. 3, pp. 868 - 872, July,
1999.
[11] The MathWorks, "SimPowerSystems User's Guide," available on-line:
http://www.mathworks.com/help.
[12] Opal-RT, "eMEGAsim PowerGrid Real-Time Digital Hardware in the
Loop Simulator," available on-line: http://www.opal-rt.coml.
[13] Edmund. O. Schweitzer III, and Daqing Hou, "Filtering For Protective
Relays," vol. 130. Atlanta, Georgia: 47th Annual Georgia Tech
Protective Relay Conference, April 1993.
[14] Zocholl, S.E., "Integrated Metering and Protective Relay Systems,"
IEEE Trnsactions on Industr Applications, vol. 25, no. 5, pp. 889
- 893, Sep-Oct, 1989.
[15] SEL, "Protection Relays by Schweitzer Engineering Laboratories,"
available on-line: http://www.selinc.com/protection/.
[16] "Current and Voltage Amplifers by Megger," available on-line:
http://www.megger.com/cae/story/lndex.php?ID=527 .
[17] Opal-RT, "RT-LAB Real Time Simulation Sofware," available on-line:
http://www.opal-rt.com/productrt -lab-professional.
[18] SEL, "SEL AcSELerator Quickset - Tool for modifying SEL Relays
Settings," available on-line: http://www.selinc.comiSEL-50301.
[19] , "SEL AcSELerator Analytic Assistant SEL-5601 Event Analysis
Sofware," available on-line: http://www.selinc.com/SEL-5601!.
[20] Megger, "Freja 300 - Relay Test System," available on-line:
http://tinyurl.comFreja300.
[21] , "Freja Win - Graphical Interface for Freja-300 Relay Test
System," available on-line: http://tinyurl.com/rejaWin.
[22] The MathWorks, "Three Phase Circuit Breaker Block
Avaliable in SimPowerSystems Library," available on-line:
http://tinyurl.com/3ph-brkr-sps.
[23] IEC Standard, "Communication networks and systems in substations
- Part 8-1: Specifc Communication Service Mapping (SCSM) - Map
pings to MMS (ISO 9506-1 and ISO 9506-2) and to ISO/lEC 8802-3."
[24] , "Communication networks and systems in substations - Part 9-2:
Specifc Communication Service Mapping (SCSM) - Sampled values
over ISO/lEC 8802-3."
4794
LullCl lul IlCm llC
5CCCO3ly5lOC CIllC
L.ll5 l5 3 33lCg
5lg3l
1
l
N3
LCvCll5 3 lul 5lg3l
Wlll CClluCu5 53mlC
CW-53mllg lC
llmC lC 3 Culul 5lg3l
3vClO 3ll-3ll35lg
Wlll Ol5ClClC 53mlC
CIICCl5
I
llmC
.CW 355 Olgll3l lH
IlllCl 1
LCm3l3lCl lC ClCCK
WlCllCl llClul
CullCl lCvCl l5 glC3lCl
ll3 llC lCKu v3luC
Cl Cl _
>-IuCllClulg 3 3lgCllllm. ll
mCllCl5 llC lul 3O ClCCK5 lIllC
lul l5 Cl Cl. [l.C. lul CullCl
glC3lCl ll3lCKu). lC >-IuCllC
glvC5 CulllCllmC 3l WClll lll5 lul
CullCl gCC5 3UCvC llC lCKu v3luC
'
Lulul CI lll5 UlCCK l5
llC llmC WllCl l35
Cl35CO5lCC llClul
CullCl l35 CxCCCOCO
llC lCKu v3luC

.CW3@ Ouif j
ll5 UlCCK CCm3lC5 llC
CCl3llC llmC CCmulCO Wlll
llC llmC Cl3tCO UyllCllmCl

Cl

O
l
-
Cl

.C

3@

lll

Cl
CWmlC xll3CllglmCCI 3ull
lmCl
Ouif
HCl3llC3l
>CCC
LCl3lCl

lC IlO l3lIC lvlOC2

Oeai.oo1.me-----

LCvClllg C N.>
>CCC2
ll5 UlCCK l353 5WllCl. lC Culul CI5WllCll5 2ClC
L3lCul3llg LCl3llC lmC
lC35C WlCllCCullCl lul l5 lC55Cl ll3lCKu
v3luC.lC CululCI llCUlCCK l5 llC 3Clu3l 5lmul3llC
llmC l C35C llC CullCl CxCCCO5 llClCKu v3luC
Fig. 3. Detailed Model of Overcurrent Relay Implemented in MATLAB/Simulink
YES
|nstantaneous
YES Timer>De!inite
Timeset byuser
N3
CIaracterisicCurve=
N

Very |nverse
YES
3|erationTime
>=Timer
1kll
Fig. 4. Protection Algorithm Implemented in the Overcurrent Relay Model
4795
>CCC
ll5 UlCCK lmlCmCl5llCm3llCm3llC3l
Cu3llC lC CCmulC

llC CCl3llC llmCCCllC5COlg lC llC

lyC CI Cl3l3ClCll5llC CulvC ClC5C Uy
llC u5Cl
CbLerl0rQuICkbe|Q L:\Lumtnl9 n0bElO0g5\lulgIvptfIUnl\bL\RCbtlPrl0r\Qulfkb0l{Rey.rU0
dmgSt0It0r lEreuUdVcrLurrdnIP1 [bLL~41J5bEllWQSrIVer)
>1
|0I
g)8|1
Q 8Uw
@ ca.eol|d0||U8
|o|l1|am|Jaa
Inverse Tme Overcurrent Elements
1nEr5E1lmEUVcrL0tr0 UmcN
51001 l,' Ti Oe,,e E01 oaeq
Q vHccnT5
5!P mo!cTm0tucNrt m
0emdioOe0e
Q H0dUdLlLe1
b!LO Wo0T>0ou0U..Cov0w
Q w~5
@ wnp1
Q W||gU
@ W||gW
Q W||A
1C
: tU!,CC
5!1%\ W|TmCrEw01 Tm0
10,5
I,
J!I
IR! l0wweootw:t mse.te
@ v+p+e::e:
@ 0v+e:
vnVo||gLlN:
d |k4
Q vnr0cb
Q Um|0Lmt
@ 0-weocet
I:Iy
@ L0q
Q wt@
i!l
@ ut02
'
_t,N
5IVI In<TmUro|kUTw\0O
IPLT0
[nVdrSdlmdVerLutrdnL Lkmdm 0
5!C museTi0utureN0Z8eQ
lIM |
!H Imm1m0mu1mdtt m
I +&&
tsrteate:u:tkset-c.,ce.emeatsnd5ctlI
Lcmm||0lw|ock|OIe|m
JI| > J U W A
|
L0I cm||e|huNmN1mJwO|rn|tmt
''
_aN,!1d
c4bLmLu1eNUmLeNt|oWr0||>|p
_bdllWQSLdIl0r~ JldrcnDlVdrLutrdnl~UM1(stt-
Q dm|Lmt
Me|@Lk _
.-
.

IL-487E Dm| . 2Ul W
|

bdllmgS|0tl0r :creUdlUVcrLurtdnlM1 [bLl4b1t 05bcl

-e i
is
~0t@

[j 0l01 N&ao0to(V; _
I
-.-0mxtqO9! 0SIwt
e ux Olp,.
@ OLMdUu
Q Mr|aedBm1_
V|8Xs> Dmrcrw:5.2.0.1 0a|c. 211-0-1202:0:0
RLbtlEtt0rQultkbEt@ L\0LumEt6n0 %m9\LutglV\l|Cll00D
18tt0Lmi
cL!1LhHLn8it|LwmJe
Off 5 1 U w K
P P P P
c!1LHHL0|vle|m
| V

Fig. 15. Settings of SEL-487E relay using AcSELerator Quickset. (a) Inverse time overcurrent element enabling with extremely inverse curve, Pickup value
of 1 and TMS of 0.5, (b) Trip logic for the IDMT confguration, (c) Confguring D/O of relay to change its contacts for IDMT trip (d) Enabling CT of relay,
(e) Enabling Event Recording, (f Enabling IDMT feature for Terminal U input of SEL-487
4796