Double Circuit Transmission Line Protection using Line Trap & Artificial Neural Network MATLAB Approach

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 07 | July -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 3015
DOUBLE CIRCUIT TRANSMISSION LINE PROTECTION USING LINE TRAP
& ARTIFICIAL NEURAL NETWORK MATLAB APPROACH
Abhilash G. Phiske1, Alok Ranjan2
1PG Scholar, Department of Electrical Engineering, Wainganga College of Engineering and Management , Nagpur
2Assistant Professor, Department of Electrical Engineering, Wainganga College of Engineering and Management ,
Nagpur
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The protection of double circuit transmission
lines could be a difficult task. This paper presents aprotection
technique supported the high-frequency transientsgenerated
by the fault to hide nearly the entire length of double circuit
line.
For this purpose, befittingly designed line traps area unit put
in at terminals of the protected line, and therefore
the Artificial Neural Network with appropriate range of
Neurons is employed to classify the faults supported the
frequency spectrum of the RMS current and RMS voltage
signals. In depth simulation studies indicate that
the projected approach is well capable of
discriminating differing types of faults and provides a
really quick, secure, and reliable protection technique. The
simulation model done in MATLAB simulink for system
analysis. In this model 300 km, 25 KV, 50Hz transmission line
power system model design with three zone bus bar system.
One end bus bar data measurement utilized for three phase
RMS voltage and current measurement. Also Neural Network
training done for designedpowersystemmodel usingMATLAB
Simulink.
Key Words: Line trap, ANN, RMS, fault classification, zone
protection.
1. INTRODUCTION
An overhead conductor is exposed to the surroundings
and therefore the chance of experiencing faults on the
conductor is usually more thanalternativemainparts.Oncea
fault happens on a conductor, it's vital to find it and notice its
zoneso as to createnecessary repairsand to revive power as
shortly as doable. Distance relaying has been wide used for
the protection of transmission lines. A distance relay must
perform the twin taskofprimaryandback-upprotection.The
first protection ought to be quick and with none intentional
time delay. Back-up protection ought to operate if and on
condition that corresponding primary relay fails. Distance
relays area unit given multiple zones of protection to satisfy
the demanding property and sensitivity needs. Zone one
provides the quickest protection with no intentional time
delay; the in operation time are often of the order of 1 cycle.
It is set to cover major portion of the line length owing
to the problem in identifying between faults that square
measure near remote bus. Zone two protectionsaredelayed
by co-ordination measure. Zone two is ready to
shield primary line
and additionally provides secondary protection to half
portion of the adjacent line with 0.25–0.4 s delay. Setting of
zone three is ready to hide complete primary and adjacent
line and up to quarter line additionally with further delay.
However, numerous conditions like remotein-feedcurrents,
fault-path resistance and shunt capacitance degrades the
performance of distance relays [1]. The present differential
protection theme hasbeen withsuccess applied toshield the
complete line. However, the relay settings square measure
tough to come to a decision as a result of line-charging
currents and unobvious current variation throughout high
resistance faults. Any composite voltage and current
measurements were accustomed improve relay sensitivity
[2]. For quick fault clearance toenhance systemstability, the
relaying schemes supported traveling wave square
measure planned [3, 4]. However, the techniques square
measure tough to discover close-in and zero voltage
faults. Numerous quite protection schemesfortransmission
lines are planned within the past for fault detection and
classification (phase selection) and distancelocation[5–12].
However, these techniques estimate the direction of fault
and its zone.
2. MATLAB SIMULATION MODEL
This project implementation will bedoneusingMATLAB
Simulink software. The major blocks will be design in
MATLAB Simulink as follows:
 Simulation of transmission line using distributed
parameter transmission line.
 Simulation of wavelet signal analysis block using
wavelet toolbox.
 Simulation of complete double circuit based power
system components using sim power system toolbox.
 Simulation of various transmission line faults and
unbalance load condition for analysis of design
protection scheme.
 Design of neural network for fault classification using
neural network toolbox.

 Training of neural network for various simulated fault
condition using neural network toolbox.
2.1. Main power system model
Figure 1 shows the complete MATLABsimulationmodel
of 25 KV, 60 Hz transmission line having three zone
separated by bus bar.
Fig -1: MATLAB simulation model of zone 1 of main power
system
The main power system model divided into three
zone having 120, 80 and 60 km of length. Figure 2 shows the
zone 2 of main power system model in which two generator
of 25KV, 100MVA fed transmission line system. Also
transmission line provided with three phase RLC load.

Fig -2: MATLAB simulation model of zone 2 of main power
system.
Zone 2 of power system model consist of
transmission line model of 80km length and at each end of
line trap installed for high transient based protection ofline.
For impedance measurement of line trap, impedance
measurement block connected across line trap. That block
analyzed the impedance of line for different power system
frequency and career frequency shown in figure 3.
Table -1: MATLAB Simulink parameter specification of
power system
S.N. MATLAB simulation block Parameter specification of MATLAB block
1 Three-phase source 1,2,3,4 Phase to phase RMS voltage= 25 KV;
Frequency = 50 Hz;
Internal winding connection = start with neutral
grounding;
Three phase short circuit level at base voltage =
100 MVA;
Base voltage = 25KV;
X/R ratio = 2.
2 Three phase RLC load
1,2,3,4
Nominal phase to phase voltage = 11 KV;
Nominal frequency = 50 Hz;
Active power = 10 KW;
Inductive reactive power = 10 KVAr;
Capacitive reactive power = 100 VAr;
Star connected load open grounded
3 Double circuit line
1,2,3,4,5,6
Number of phases = 6;
Frequency used for RLC specification = 50 Hz;
Positive sequence resistance per unit length =
0.068 Ohm/Km;
Zero sequence resistance per unit length = 0.284
Ohms/km;
Positive sequence inductance per unit length =
1.31 mH/Km;
Zero sequence inductance per unit length = 4.02
mH/Km;
Positive sequence capacitance per unit length =
8.85nF/Km;
Zero sequence capacitance per unit length = 6.21
nF/Km;
Line length depend on zone.
Zone 1 (Line 1,2,) = 120Km; Zone 2 (line 3,4) = 80
Km; Zone 3 (Line 5,6) = 60Km.
4 Three phase fault Fault types = LG, LL, LLG, LLL (AG, BG, CG, ABG,
BCG, ACG, AB, BC, AC, ABC);
Fault resistance = 0.001 Ohm;
Transition time: Fault start time 0.4 sec and fault
end time 0.6 sec;
Snubber resistance = 1 MOhm.
5 Line trap subsystem R1=0.5 Ohm; L1= 2 mH; R2= 50 Ohm; C2= 5nF;C1=
50

Fig -3: MATLAB simulation model of zone 3 of main
power system
Fig -4: Bus bar 2 measurement subsystem
Figure 4 showsthebusbarparametermeasurement
subsystem, in which three phase RMS voltage of six phase of
line and three phase RMS current of line. That subsystem
measured the six phase voltage and current at normal
condition and abnormal fault condition for generation of
training data set for neural network training.
2.2. Neural Network subsystem
In this proposed approach Neural network use as
classifier for fault type classification and fault zone
identification. The data set for training neural network are
six phase rms voltage and current utilizedfortraining neural
network. After training neural network, neural network
block coupled with measurement subsystem for
classification. Figure 5 shows the neural network system
coupled with bus bar measurement subsystem. In future if
any fault condition occurs then that time measurement
subsystem send measure value to ANN input then ANN
classify the fault type and fault zone based on training data
set.
Figure 6 shows the simulation block of ANN1 for
transmission line fault type classification. The inputs for
ANN1 are six phase transmission line RMS voltage and RMS
current. Based on input ANN1 generatestheoutput based on
training data set.
Figure 7 shows the internal configuration of layer 1
of ANN1 in which total 10 number of neurons with weights
and bias value shown. The weight value updated based on
input training data set for generation of target value which
also decided by designer. All 10 neuronsconsistofactivation
function or transfer function which coupled using
multiplexer for generation of single decision based on
training data set.
Fig -5: Measurement subsystem coupled with ANN for
fault type classification and fault zone identification
Fig -6: ANN model for Fault type classification
Fig -7: Architecture of layer 1 of ANN for fault type
classification shows weights with value

3. MATLAB SIMULATION RESULTS
3.1. Result from line trap
Fig -8: Impedance of line trap at high frequency transients
The main function of the line trap is to present high
impedance at the carrier frequency band while presenting
negligible impedance at the power system frequency. The
high impedance is required to reduce the carrier signal
attenuation due to the division among the several
transmission lines terminated at the same bus.
Figure 8 shows that line trap is present high impedance at
the carrier frequency of 100 kHz while presenting negligible
impedance at the power supply frequency 50 Hz.
3.2. Power system voltage and current
Fig-9: Six phase RMS voltage and RMS current waveform
when LG (AG) fault takes place on line at 100km in zone 1
from reference bus bar B22
Figure 9 shows the six phase transmission line
voltages and current waveform when LG fault takesplaceon
line at 100km in zone 1 from reference bus bar 2. First and
third waveform shows the three phase voltage Vabc and
Va’b’c’ while second and third waveform shows the three
phase current of double circuit line Iabc and Ia’b’c’
respectively.
when LLG (B’C’G) fault takes place on line at 80 km in zone
1 from reference bus bar B22.
when LL (AB) fault takes place on line at 60 km in zone 2
from reference bus bar B22.
3.3. Result from Neural Network
Fig-12: ROC after successful training of ANN1 for
transmission line fault classification.
The receiver operating characteristic (ROC) is a
metric used to check the quality of classifiers. For each class
of a classifier, roc applies threshold values across the

interval [0,1] to outputs. For each threshold, two values are
calculated, the True Positive Ratio (the number of outputs
greater or equal to the threshold, divided by the number of
one targets), and the False Positive Ratio (the number of
outputs less than the threshold, divided by the number of
zero targets).
Fig-13: Confusion matrix after successful training of ANN1
for transmission line fault classification
Figure 13 shows that 93%data areperfectlyclassify
the fault and remaining fault case data not classify using
neural network 1. It means that for remaining 7% data set
neural network was in confusion state for classify the fault.
Fig-14: ROC after successful training of ANN2 for
transmission line fault zone identification
Figure 15 shows that 61.5 % data are perfectly
classify the fault zone and remaining fault case data not
classify using neural network 2. It means that for remaining
48.5 % data set neural network was in confusion state for
classify the fault zone.
Fig-15: Confusion matrix after successful training of ANN2
for transmission line fault zone identification.
4. CONCLUSION
The protection of double circuit line difficult due to self
mutual inductance and mutual inductance effect of
transmission line conductor during loading condition. This
mutual inductance affects theprotectionoftransmission line
using distance relay approach due to the over reach and
under reach operation of line. For eliminating this difficulty
ANN approach for classification of line and line zone
classification utilized. The proposed approach using ANN
classify the transmission line faultupto93%whilefaultzone
identification done upto 75% only. Also line trap installedat
main protection zone of double circuit line for providing the
protection from high transient surge.
REFERENCES
[1] Aggarwal, R. K., Xuan, Q. Y., Johns, A. T., Li, F., & Bennett,
A. (1999). A novel approach to fault diagnosis in multicircuit
transmission lines using fuzzy ARTmap neural
networks. IEEE transactions on neural networks, 10(5),
1214-1221.
[2] Dash, P. K., Pradhan, A. K., Panda, G., & Liew, A. C. (2000,
January). Digital protection of power transmission lines in
the presence of series connected FACTS devices. In Power
Engineering Society Winter Meeting, 2000. IEEE(Vol. 3, pp.
1967-1972). IEEE.
[3] Eissa, M. M., & Masoud, M. (2001).Anovel digital distance
relaying technique for transmission line protection. IEEE
Transactions on Power Delivery, 16(3), 380-384.
[4] Gracia, J., Mazon, A. J., & Zamora, I. (2005). Best ANN
structures for fault location in single-and double-circuit
transmission lines. IEEE transactions on power
delivery, 20(4), 2389-2395.
[5] Jain, A., Thoke, A. S., & Patel, R. N. (2008). Fault
classification of double circuit transmission line using

artificial neural network. International Journal of Electrical
Systems Science and Engineering, 1(4), 750-755[6] Jain, A.,
Thoke, A. S., Koley, E., & Patel, R. N. (2009, December). Fault
classification and fault distance location of double circuit
transmission lines for phase to phase faults using only one
terminal data. In Power Systems, 2009. ICPS'09.
International Conference on (pp. 1-6). IEEE.
[7] Jain, A., Thoke, A. S., & Patel, R. N. (2009, December).
Double circuit transmission line faultdistancelocationusing
artificial neural network. In Nature & Biologically Inspired
Computing, 2009. NaBIC 2009. World Congress on (pp. 13-
18). IEEE.
[8] Jain, A., Thoke, A. S., & Patel, R. N. (2009, December).
Double circuit transmission line faultdistancelocationusing
artificial neural network. In Nature & Biologically Inspired
Computing, 2009. NaBIC 2009. World Congress on (pp. 13-
18). IEEE.
[9] Apostolopoulos, C. A., & Korres, G. N. (2011). A novel
fault-location algorithmfordouble-circuittransmissionlines
without utilizing line parameters. IEEE Transactions on
Power Delivery, 26(3), 1467-1478.
[10] Yadav, A., & Swetapadma,A.(2014).Improvedfirstzone
reach setting of artificial neural network-based directional
relay for protection of double circuit transmission lines. IET
Generation, Transmission & Distribution, 8(3), 373-388.

Double Circuit Transmission Line Protection using Line Trap & Artificial Neural Network MATLAB Approach

More Related Content

What's hot (20)

Similar to Double Circuit Transmission Line Protection using Line Trap & Artificial Neural Network MATLAB Approach (20)

More from IRJET Journal (20)

Recently uploaded (20)

Double Circuit Transmission Line Protection using Line Trap & Artificial Neural Network MATLAB Approach