Modeling and Simulation of Inverse Time Overcurrent Relay

Download as pdf or txt
Download as pdf or txt
You are on page 1of 5

2016 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), 22 October 2016, Shah Alam, Malaysia

Modeling and Simulation of Inverse Time


Overcurrent Relay using Matlab/Simulink
Nur Hazwani Hussin, Muhd Hafizi Idris*, Melaty Amirruddin, Mohd Saufi Ahmad, Mohd Alif Ismail, Farrah
Salwani Abdullah, Nurhakimah Mohd Mukhtar
Centre of Excellence for Renewable Energy (CERE)
School of Electrical System Engineering
Universiti Malaysia Perlis
Arau, Perlis, Malaysia
*[email protected]

Abstract—Overcurrent relay is an important relay used to measured in multiples of current setting or pickup current.
protect transmission and distribution feeders, transformer, bus When used in coordination with other overcurrent relays such
coupler etc. It can be used as main or backup protection relays. as to protect radial feeders, it can be coordinated with upstream
The modeling of overcurrent relay and other protection relays is or downstream relays easily by the use of Time Multiplier
important in the research on the effects of network parameters
and configurations on the operation of these relays. This research
Setting, TMS [3].
deal with modeling and simulation of inverse time overcurrent There are many works had been done to model and
relay using commonly used Matlab/Simulink software. The simulate protection relays using Matlab/Simulink software. As
developed model was tested with single line to ground fault and in [5], Matlab/Simulink was used to model Inverse Definite
phase to phase fault with various fault locations and fault Minimum Time (IDMT) overcurrent relay which based on
resistances to evaluate phase and earth detectors in sensing the digital signal processor (DSP). After the developed Simulink
fault current. Simulations results show the feasibility of the model was tested using simulation, the developed model then
software to be used to develop the relay. was embedded in TMS320F2812 DSP board. Simulink also
had been used to model overcurrent relay which used to
Index Terms—Overcurrent relay, inverse time, Matlab,
Simulation. protect the Photovoltaic (PV) system for irrigation [6]. The
developed model was investigated at normal conditions,
starting of the motor as well as at faulty conditions.
I. INTRODUCTION
Graphical user interface (GUI) of distance relay was
Protection relay is used to reduce harmful damage to both developed and linked with the model of distance relay created
electrical equipment and personnel when an electrical fault using Simulink [7]. The GUI was used to ease the
occurs. Protection relay is designed based on the basis of understanding among electrical engineering students on the
selectivity, reliability, speed and sensitivity [1]. One of operation of distance relay. Finally, Matlab/Simulink was used
protection relays used to protect the circuits in power system is to model the conventional distance relay as in [8]. The relay
overcurrent relay. When there is a fault in power system, model then was simulated, analyzed and evaluated in real time
overcurrent relay will detects and isolate the faulty section using Opal-RT’s Real-Time LABoratory (RT-LAB). The
from the rest of the system by measuring the value of current. authors also successfully evaluated the effects of in feed
It can be used as main or backup protection to protect the current for protection at zones 1 and 2.
transmission or distribution feeders, transformers, bus coupler
etc [2]. When it is used to protect the feeders and transformers, II. SCOPES OF THE PROJECT
it normally will be coordinated in sequence so that the relay The purpose of this research is to model, simulate and
nearest to the fault location will operate first. There are two evaluate inverse time ovecurrent relay using Matlab/Simulink
types of overcurrent relay operating speed which are software. If the results prove the feasible use of the software to
instantaneous time and delayed time [3]. The weakness of model and simulate the relay, the model can be extended to be
instantaneous type is it can’t discriminate the operation when used in further research in the future. Matlab/Simulink is
fault current at two or more locations is the same [4]. chosen because of the widespread use of this software as
For delayed time type, it can be divided into two categories research platform and its’ capability to model power system
which are definite time and inverse time. For definite time components [9]. To ease the modeling of the relay, the model
overcurrent relay, the operating time is independent of current has been simplified and bounded by limitation where only
where relay closest to the fault has shortest operating time. The standard inverse (SI) curve was modeled and simulated in this
problem with this type is the longest operating time is at the research.
electrical source where the fault current is the highest. While
for inverse time overcurrent relay, the higher the fault current III. THEORIES OF INVERSE TIME OVERCURRENT RELAY
level, the faster the operating time will be. Fault current is

978-1-5090-4186-2/16/$31.00 ©2016 IEEE 40


2016 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), 22 October 2016, Shah Alam, Malaysia

Figure 1 shows the characteristic curves of inverse time Where;


overcurrent relay. There are four curves in the figure where top = Operating Time (s)
each curve can be expressed by an equation to calculate the TMS = Time Multiplier Setting (% or pu)
operating time. Table I shows the equations of operating time If = Fault Current (A)
which represent the characteristic curves of inverse time Ip = Current Setting or Pickup Current (A)
overcurrent relay. For the same multiples of current setting,
long time inverse curve has the longest time to operate while IV. MODELING AND SIMULATION
extremely inverse curve has the fastest time to operate the Table II shows the overall parameters used for the power
relay [3]. The current setting (or the pickup current) for each system network and relay setting. The CT ratio been chosen is
curve is represented by Eq. (1). 1200/1 A depending on the value of calculated full load
current. The length of the line is 50 km. The capacitance of the
line is neglected. The PS parameters for phase and earth faults
are different where PS for phase fault normally set higher than
full load current while PS for earth fault normally set lower
than full load current to detect high impedance fault when
involving ground fault which normally undetected by phase
overcurrent relay. The TMS for both phase and earth faults are
set to the same value. In this model, only Standard Inverse (SI)
curve used to detect phase and earth faults. However, the
model can easily be extended to other types of curve.

TABLE II. NETWORK AND RELAY SETTING PARAMETERS

Parameters Values
Source Short Circuit Level (VA) 500,000,000
X/R Ratio 5
Source VLL (V) 132,000
Line CT Ratio 1200/1 A
Line Length (km) 50
Line Positive Sequence Resistance, R1 (Ω/km) 0.045531917
Line Zero Sequence Resistance, R0 (Ω/km) 0.151489359
Line Positive Sequence Inductance, L1 (H/km) 0.0006176566224
Line Zero Sequence Inductance, L0 (H/km) 0.001533982723
Line Full Load Current, IFL (A) 1233 A
3 Phase Line Capacity (VA) 282,000,000
Fig. 1. Characteristics of inverse time overcurrent relay [3] Plug Setting (PS) for Phase 1.2
Time Multiplier Setting (TMS) for Phase 0.35
Ip = PS x CTsec (1) Plug Setting (PS) for Earth 0.4
Where; Time Multiplier Setting (TMS) for Earth 0.35
PS = Plug Setting (% or pu) Line VLL (V) 132,000
Network Frequency (Hz) 50
CTsec = Rated Secondary Current of Current Transformer
Type of Curve Standard Inverse
(CT) in Ampere (A). Load Active Power (W) 175,000,000

TABLE I. OPERATING TIME EQUATIONS [3]


Figure 2 shows the overall model for the network and relay
which was created using Matlab/Simulink software. The
network chosen is a simple spur transmission line which is
connected to a Thevenin equivalent three phase power source
at one end while the other end is connected to a three phase
load. Figure 3 shows the blocks inside OC Relay Subsystem.
All phases and earth subsystems have their own curve to
calculate the operation time and send the trip signal. For EF
subsystem, the input to it is the phasor summation of red,
yellow and blue currents where when there is a ground fault,
there will be residual or imbalanced current flow to the earth.
This will not happen for phase to phase fault. Subsystem from
any phase or earth will send the trip signal faster than others
when the current to it is higher than others. This is why all the
trip signals is ‘OR’ gated. All phases and earth subsystems will
compute the operation time depending on the value of fault

41
2016 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), 22 October 2016, Shah Alam, Malaysia

current entering them. Figure 4 shows the block properties developed for the phase and earth subsystems.

Fig. 2. Overall diagram of Simulink model

Figure 5 shows the blocks inside each phase and earth


subsystem. All these blocks represent the equation for SI curve.
It can be seen that, the timer only will be started when the fault
current is higher than the pickup current. When the timer
elapse (more than) the calculated operation time, trip signal
will be sent.

Fig. 3. OC relay subsystem


Fig. 5. Phases and earth subsystems

V. SIMULATION RESULTS
This section discusses the results from the simulation. From
the model, all faults were initiated at 0.5 s. Two types of fault
which are single line to ground (SLG) fault and line to line
(LL) fault were taken to be studied. Faults were simulated at
different locations, FL and with different fault resistances, Rf to
vary the value of fault current. The results which were focused
are the operation times, top for all phases and earth detectors
and whether the fault current, If for each phase and earth is
higher than the pickup current, Ip or not. From the relay setting
parameters, Ip for each phase and earth were calculated as
follows:
i. Earth, Ip = 0.4 x 1200 = 480 A
ii. Phase, Ip = 1.2 x 1200 = 1440 A
Fig. 4. Block properties for phase and earth subsystems

42
2016 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), 22 October 2016, Shah Alam, Malaysia

Table III shows the results of fault current, operation time multiple of earth’s fault current over pickup current (2061/480
and trip signal for faults at different locations and different = 4.294) was higher than for phase (2604/1440 = 1.808). Thus,
fault resistances. Case 1 to 3 are for SLG faults (RG) while the breaker was isolated and tripped at 2.204 s (≈ 0.5 s +
case 4 to case 6 are for LL faults (RY). For case 1, it can be 1.692 s = 2.192 s). Figure 6 and 7 show the rms red phase
seen that both red phase and earth detected the fault due to low current and rms earth return current for case 1 respectively
fault resistance (1 Ω) and both fault currents exceeded the while Fig. 8 shows the trip signal which was sent to trip the
pickup currents. However, the operation time was faster for breaker after the operation time 1.692 s had been elapsed.
earth detector compared with phase detector because the

TABLE III. SIMULATION RESULTS

Case top for


Fault FL Rf If for Faulted If for If > Ipic top for Trip
No. Faulted
Type (km) (Ω) Phase (A) Earth (A) (Yes/No) Earth (s) Signal (s)
Phases (s)
Phase: Yes
1 RG 10 1 2604 2061 4.335 1.692 2.204
Earth: Yes
Phase: Yes
2 RG 40 1 1907 1356 9.809 2.487 3.0041
Earth: Yes
Phase: No
3 RG 40 40 1374.6 533.15 53.889 22.831 23.4691
Earth: Yes
R = 2590 Phase: R = Yes, Y = Yes R = 4.132
4 RY 10 1 ≈0 0.1252 4.5068
Y = 1920 Earth: No Y = 8.504
R = 2130 Phase: R = Yes, Y = Yes R = 6.352
5 RY 40 1 ≈0 0.132 6.8849
Y = 1480 Earth: No Y = 88.235
R = 1890 Phase: R = Yes, Y = No R = 8.992
6 RY 40 20 ≈0 0.122 9.5475
Y = 1395 Earth: No Y = 78.705

RED PHASE CURRENT (RMS) TRIP SIGNAL


3500 1.5

If = 2604 A
3000

2500
1
CURRENT (A)

2.204s
2000
0 OR 1

FAULT ISOLATION AT 2.204s

1500

0.5
1000

FAULT INCEPTION AT 0.5s


500

0 0
0 0.5 1 1.5 2 2.5 0 0.5 1 1.5 2 2.5
TIME (s) TIME (s)

Fig. 6. Rms red phase current for case 1 Fig. 8. Trip signal for case 1

NEUTRAL CURRENT (RMS) For case 2, the fault currents for both red phase and earth
4000 detectors were smaller compared with case 1. This was because
OPERATION TIME = 1.692s the fault location for case 2 (in km) was greater than case 1
3000
although the fault resistance value was the same, thus the
2000 impedance for the fault current to flow was higher. However,
both phase and earth fault currents still greater than each
CURRENT (A)

1000
If = 2061 A
pickup current and the breaker was tripped at 3.0041 s (≈
2.487 + 0.5 s = 2.987 s) by the trip signal sent by the earth fault
0
FAULT ISOLATION AT 2.204s
-1000 detector. For case 3, the fault location was still the same with
-2000
FAULT INCEPTION AT 0.5s
case 2. However, the fault resistance was increased to 40 Ω. It
can be seen that, fault currents for both phase and earth
-3000
detectors dropped significantly. Only the earth detector sensed
0 0.5 1
TIME (s)
1.5 2 2.5
the fault because the fault current was still greater than pickup
Fig. 7. Rms earth return current for case 1 current. The breaker was tripped at a very long time which was
23.4691 s (≈ 22.831 s + 0.5 s = 23.331 s).

43
2016 IEEE International Conference on Automatic Control and Intelligent Systems (I2CACIS), 22 October 2016, Shah Alam, Malaysia

In case 4, a fault was introduced which connected between VI. CONCLUSION


red and yellow phases of the line. From the results, fault The inverse time overcurrent relay was successfully
current at red phase was higher than yellow phase. No fault modeled and simulated using Matlab/Simulink software. This
current flow to the earth because it was not a ground fault. shows that the software is capable to be used to model any type
Breaker was tripped at 4.5068 s (≈ 4.132 s + 0.5 s = 4.632 s) of relays. However, because Matlab is equation solver based
by the red phase detector. Although there was operation time software, the simulation time is depending on the complexity
calculated by earth detector, the trip signal was not sent by of the model. Higher complexity of the model will take longer
earth detector because the fault current was too low (≈ 0) time for the simulation to finish. Appropriate simulation solver
compared with its pickup current. Figure 9 shows the red must be chosen to reduce the simulation time. The model
phase, yellow phase and earth return currents for case 4 while developed in this research can easily be extended to other types
Fig. 10 shows the red phase trip signal. of curve such as extremely inverse, very inverse, and long time
With the same fault resistance as case 4, fault location was inverse. Besides that, definite time feature also can be added to
changed to 40 km for case 5. It can be seen that both fault cater for saturated current transformer when the fault current is
currents at red and yellow phases decreased. However, both red too high.
and yellow phase detectors still can detected the fault currents
because both fault currents more than pickup current for phase ACKNOWLEDGMENT
detector. The breaker was tripped at 6.8849 s by red phase
detector (≈ 6.352 s + 0.5 s = 6.852 s). Finally for case 6, fault The authors would like to thanks the staffs of School of
location was maintained at 40 km but fault resistance was Electrical System Engineering, Universiti Malaysia Perlis and
increased to 20 Ω. It can be seen that fault currents for both red any person who contributed directly and indirectly to this
and yellow phases dropped quite significantly. Fault current for research.
yellow phase dropped below the pickup current for phase fault.
Thus, only red phase detector sensed the fault and sent the trip REFERENCES
signal. [1] H. J. Altuve, K. Zimmerman, and D. Tziouvaras, “Maximizing
Line Protection Reliability, Speed, and Sensitivity”, Annual
RED PHASE, YELLOW PHASE & EARTH CURRENTS (RMS) Georgia Tech Protective Relaying Conference, 20-22 April
3500
2016.
OPERATION TIME = 4.132 s
[2] P. R. Subcommittee, “Protective Relaying Philosophy and
3000
Design Guidelines PJM Relay Subcommittee,” 2013.
2500 [3] Fecime, “Chapter 9: Overcurrent Protection for Phase and
Earth Faults”, Network Protection & Automation Guide, pp.
CURRENT (A)

2000 123–151, 2008.


[4] Z. Wang, J. Li, W. Yang, and Z. Shi, “Impact of Distributed
1500 If yellow = 1920 A If red = 2590 A
Generation on the Power Supply Reliability”, IEEE PES
FAULT INCEPTION AT 0.5 s Innovative Smart Grid Technologies, 2012.
1000
FAULT ISOLATION AT 4.5068 s
[5] Y. L. Goh, A. K. Ramasamy, A. A. Z. Abidin, and F. H. Nagi,
500 “Modeling of overcurrent relay using digital signal processor”,
Earth current = 0 A
IEEE Symposium on Industrial Electronics & Applications
0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
(ISIEA), pp. 367–370, 2010.
TIME (s) [6] R. Kamel and M. Khalaf, “Analysis of Overcurrent Numerical-
Fig. 9. Red phase, yellow phase and earth return currents for case 4 Relays for Protection of a Stand-Alone PV System”, Saudi
Arabia Smart Grid Conference (SASG), pp. 1–6, 2014.
TRIP SIGNAL [7] M. H. Idris, S. Hardi, and M. Z. Hasan, “Teaching distance
1.5
relay using Matlab/Simulink graphical user interface”, Procedia
Engineering, 2013, vol. 53, pp. 264–270.
[8] M. Pazoki, Z. Moravej, M. Khederzadeh, and N. K. C. Nair,
“Distance protection of transmission line with infeed based on
1
real-time simulator”, Australasian Universities Power
4.5068 S Engineering Conference (AUPEC), pp. 1–5, 2014.
0 OR 1

[9] C. Mathworks, “Simscape TM Power Systems TM Reference


(Simscape TM Components) R 2016 a”, 2016.
0.5

0
0 0.5 1 1.5 2 2.5 3 3.5 4 4.5
TIME (s)

Fig. 10. Trip signal for case 4

44

You might also like