Shunt Faults Detection on Transmission Line by Wavelet

ISSN 2349-7815
International Journal of Recent Research in Electrical and Electronics Engineering (IJRREEE)
Vol. 3, Issue 1, pp: (24-32), Month: January - March 2016, Available at: www.paperpublications.org
Page | 24
Paper Publications
Shunt Faults Detection on Transmission Line
by Wavelet
1
Divya Bisht, 2
Associate Prof. Arti Bhandakkar
1,2
Electrictrical and Electronics Department Shri Ram Institute of Technology Jabalpur, India
Abstract: Transmission line fault detection is a very important task because major portion of power system fault
occurring in transmission system. This paper represents a fast and reliable method of transmission line shunt fault
detection. MATLAB Simulink use for modeled an IEEE 9-bus test power system for case study of various faults. In
proposed work Daubechies wavelet is applied for decomposition of fault transients. The application of wavelet
analysis helps in accurate classification of the various fault patterns. Wavelet entropy measure based on wavelet
analysis is able to observe the unsteady signals and complexity of the system at time-frequency plane. The result
shows that the proposed method is capable to detect all the shunt faults.
Keywords: Transmission line faults, MATLAB Simulink, Daubechies Wavelet.
1. INTRODUCTION
Fault detection on transmission lines is very important to safeguard the power system. The most dangerous fault in power
system is the short circuit or shunt fault, it results either in an excess current or in the reduction of the impedance between
conductors or conductors and earth. In a transmitting system following types of fault possible: i) single line-to-ground
(LG) fault which is occurring 85%, ii) line-to-line (L-L) fault which is occurring 8%, and iii) double line- to-ground (L-L-
G) fault which is occurring 5%,and iv)triple line (L-L-L)fault having occurrence 2% of all shunt fault possible in
transmission line. Faults on power transmission lines need to be detected and located rapidly, classified correctly and
cleared as fast as possible for protection purpose which having these requirements : i) fault detection and ii) fault clearing,
which involve detection of a fault including determination of the fault location and subsequently fault classification.
When the type of fault is correctly identified, the possible remedial action can quickly be sorted out to solve the
problem.[1] An ability to detect faults in transmission lines as fast as possible is crucial, since they may compromise the
propagation of energy to customers and the functioning of the transmission network. Therefore, efficient fault detection
approaches are focused primarily on the analysis of short time intervals, or transient signals. In this context, the use of
wavelet transforms has emerged as a powerful tool for feature extraction, mainly due to its ability to focus on short time
intervals for the analysis of high-frequency components. Wavelets transforms can use varying time windows in order to
extract the coefficients of the mother wavelet. Short time windows are applied for high–frequency components, and long
time windows for low-frequency components [8].
In this paper an approach to presented Fault detection with the help of discrete wavelet analysis (DWT) in transmission
lines connecting a multi bus system. The use of the Wavelet Transform (WT) in power system analysis constitutes a

ISSN 2349-7815
Page | 25
Paper Publications
powerful tool for the diagnosis of fault events. The main advantage is that, as opposed to Fourier analysis, the WT has
good resolution in both time and frequency [4].
2. WAVELET TRANSFORM
A wavelet is a small wave, which deals with building a model for non-stationary signals, using a set of components. It has
become a well-known useful tool since its introduction, especially in signal and image processing [9]. Wavelet analysis is
a new development in the area of applied mathematics and was first introduced in seismology to provide a time dimension
to seismic analysis which the Fourier analysis lacked. Fourier analysis is ideal for studying stationary data but not well
suited for studying data with transient events which cannot be statistically predicted from past history of the data.
Wavelets are ideally suited for analysis of such non-stationary data. Wavelet transformation can be broadly categorized
as: (i) continuous wavelet transform and (ii) discrete wavelet transform. In the present study, discrete wavelet transform
(DWT) has been used as an effective tool for post processing and extraction of valuable features from the fault patterns
for discrimination and classification of various fault patterns [3],[6],[7]. Among the DWT family, the Daubechies wavelet
is considered to be one of the most effective DWT tool, and this has been used to extract valuable features from the
recorded waveforms. Daubechies wavelets are widely used in solving broad range of problems. The wavelets are
generated from a single basic wavelet (t),, namely, mother wavelet, by scaling and translation:
( )
√
( )
Where s is the scale factor, τ is the translation factor and the factor s-1/2 is the energy normalization across the different
scales. To obtain the DWT the parameters s and need to discretized. Discretizing and will yield
orthogonal basis functions for certain choices of :
( )
( ) ⁄
( )
C is a real coefficient, j and k are sets of integer indices and ( ) are the wavelet expansion functions. Function
( )can be expressed as a linear decomposition as follows:
( ) ∑ ∑ ( ) ∑ ∑ ∑ ( )
Where Jo is an integer. Equation [3] is a linear combination of wavelet coefficients, ( ,Dj,k ), a set of functions
called scalling function ( ) and Coefficients , and Dj,k are the Discrete Wavelet Transform (DWT) of ( )
[12].
The signals can be accurately reproduced with the wavelet analysis using relatively small number of components. The
analyzing wavelets are called the “mother wavelets” and it’s dilated and translated versions are called the “daughter
wavelets”. It has a digitally implementable counterpart called the discrete Wavelet transform (DWT). The generated
waveforms are analyzed with Wavelet multi resolution analysis (MRA) to extract sub-band information from the
simulated transients [3]. Daubechies four (db-4) wavelet is used in this work for the analysis as it closely matches the
signal to be processed which is of utmost importance in wavelet applications. Wavelet co-efficient of the signal are
obtained by the decomposition of a discrete fault current and voltage signals using Mallat’s algorithm [3], [11].

ISSN 2349-7815
Page | 26
Paper Publications
FIGURE 1-SIGNAL DECOMPOSITION BY WAVELET
The DWT acts as two filters defined by a low frequency filter h(n) and a high frequency filter g(n). Both filters have same
cut frequency fN/2. The function f(n) is divided into two parts, d1 (the high frequency part) and it is called Detail
component, and a1(the low frequency part), and is called Approximation. Decimation by 2 is done for removing
redundant information.. The algorithm is iterated for a1, obtaining a second level detail component d2 and a second level
Approximation a2, the process will be repeated. d1 have the frequency( fN – fN/2), d2 have( fN/2 – fN/4), dk have( fN/2n
– fN/2k+1 ) and an contains the frequencies lower than fN/2k+1 [4], [5], [13], [16]. Wavelets are functions that satisfy the
requirements of both time and frequency localization. The necessary and 16 sufficient condition for wavelets is that it
must be oscillatory, must decay quickly to zero and must have an average value of zero. In addition, for the discrete DWT
considered here, the Wavelets are orthogonal to each other [10].
3. WAVELET ENTROPY
To identify the transient signal, wavelet entropy measures is a feature, which have some unique capabilities[13]. The non
normalized Shannon entropy is as follows .The wavelet energy spectrum at scale j and instant k is:
| |
The nonnormalized Shannon entropy at scale j, in a moving data window goes through the detail coefficients shifting 128
samples at a time, is
∑
4. POWER SYSTEM MODEL
A nine bus power system [14] has been modeled in MATLAB Simulink. A schematic single line diagram of the nine bus
system under study has been presented in figure 2. The modeled system comprises of three generator, three transformers
and active and reactive loads connected to different buses. Detailed specification of the various components of the system
are provided in Appendix-I.

ISSN 2349-7815
Page | 27
Paper Publications
Figure-2 Nine bus power system model
5. SIMULATION OF SHUNT FAULTS
As mentioned earlier different types of faults were simulated using the fault breaker module in the MATLAB Simulink. In
the present study, four different types of short-circuit faults were simulated, which include L-G fault, L-L fault, L-L-G
fault and symmetrical three phase fault involving all the three phases of transmission lines (L-L-L). The letters A, B, C
and G stands for. These faults were simulated at different buses of the system and continuous monitoring and recording of
current is carried out at the Bus-6.Inspection and comparison of these results with the healthy waveforms reveals
considerable difference between healthy and faulty waveforms. In Figure-3,5,7,9and 11 given the waveform of current
and in Figure-4,6,8,10 and 12 given DWT and entropy at different conditions as healthy and different fault condition.
Fault timing is taking T1= 0.2sec (in fault time) and T2= 0.3sec (fault out time). It is clear from the waveform at the
healthy condition waveforms are regular in shape and DWT and WE shows zero value means no fault detection. Table-I
shows statistical data of bus 6 at healthy and different fault conditions. From highlighted area it is clear which type of
fault occurring. At healthy condition there is no fault detection shows by zero values. At fault condition only faulted phase
gives the reading, no faulty phases gives zero. These differences are helpful in proper identification of the faults. Table-2
and 3 shows statistical data of bus5 and 8. Faulted phase data highlighted in the table.

ISSN 2349-7815
Page | 28
Paper Publications
Figure-3 Waveform of current for phase A, B and C at healthy condition Figure-5 Waveform of current at L-G fault
Figure-4 DWT and WE for phase A, B and C at healthy condition Figure-6 DWT and WE AT L-G fault

ISSN 2349-7815
Page | 29
Paper Publications
Figure-7 Waveform of current at L-L-G fault Figure-10 DWT and WE at L-L-G fault
Figure-8 DWT and WE at L-L fault Figure-11 Waveform of current at L-L-L fault

ISSN 2349-7815
Page | 30
Paper Publications
Figure-9 Waveform of current at L-L-G fault Figure-12 DWT and WE at L-L-L fault
TABLE-I STATISTICAL FAULT CURRENT ANALYSIS AT BUS 6
TABLE-II STATISTICAL FAULT CURRENT ANALYSIS AT BUS 5
faults Phase A PhaseB Phase C
T1 T2 T1 T2 T1 T2
DWT WE DWT WE DWT WE DWT WE DWT WE DWT WE
No fault 0 0 0 0 0 0 0 0 0 0 0 0
LG 24 -3793 8.6 -340 0 0 0 0 0 0 0 0
LL 18.5 -2000 10 -500 18.5 -2000 10 -500 0 0 0 0
LLG 26 -4560 12 -800 11 -600 8 -287 0 0 0 0
LLL 29 -6000 15.7 -1400 7.5 -276 4.4 -60 21 -2850 11.2 -600
faults Phase A Phase B Phase C
T1 T2 T1 T2 T1 T2

ISSN 2349-7815
Page | 31
Paper Publications
TABLE-III STATISTICAL FAULT CURRENT ANALYSIS AT BUS 8
6. CONCLUSION
In this paper a wavelet analysis based approach has been proposed to detect and classify various types of faults in a multi
bus power system. A case study has been conducted on IEEE 9-bus test system, where different types of faults LG, L-L,
L-L-G and L-L-L fault were artificially simulated in different buses of the network. All these faults could be correctly
detected and classified with the help of discrete wavelet analysis and entropy using Daubechie-4 up to level 2
decomposition. The case study the 9-bus system reveals that the proposed technique can yield 100% fault detection and
classification accuracy.
REFERENCES
[1] Subhra Jana, Abhinandan De, “Transmission line fault detection and classification using Wavelet Analysis”, 2013
Annual IEEE India Conference (INDICON).
[2] Manohar singh, Dr. B.K Panigrahi,. Dr. R. P. Maheshwari, “Transmission Line Fault Detection and Classification”,
proceedings of ICETECT 2011.
[3] Dr. A. K. Sinha1, kranthi kiran chowdoju2,’ Power System Fault Detection Classification Based On PCA And
PNN’’, proceedings of ICETECT 2011.
[4] F. Martín, Member, IEEE; J.A. Aguado, Member, IEEE; M. Medina; J. Muñoz, “Classification of faults in double
circuit lines using Wavelet transforms”2008 IEEE.
[5] G. Sudha , T. Basavaraju, “A Comparison Between Different Approaches For Fault Classification in Transmission
Lines” IET-UK International Conference on Information and Communication Technology in Electrical Sciences
(ICTES 2007), Dr. M.G.R. University, Chennai, Tamil Nadu, India. Dec. 20-22, 2007. pp.398-403.
No fault 0 0 0 0 0 0 0 0 0 0 0 0
LG 19 -2200 7.5 200 0 0 0 0 0 0 0 0
LL 15 -1280 8.4 -330 15 -1280 8.4 -330 0 0 0 0
LLG 20 -2750 10 -520 9.3 -400 6.8 -190 0 0 0 0
LLL 23.3 -3584 13.2 -940 6.7 -180 4 -45.6 16.6 -1600 9.2 -400
Faults Phase A PhaseB Phase C
T1 T2 T1 T2 T1 T2
No fault 0 0 0 0 0 0 0 0 0 0 0 0
LG 10 -500 5 90 0 0 0 0 0 0 0 0
LL 8 -294 5.4 -100 8 -294 5.4 -100 0 0 0 0
LLG 11 -600 6.5 -175 5.3 -100 4 -50 0 0 0 0
LLL 12.4 -800 8 -292 4 -45 2.7 -15 8.4 -314 5.4 -100

ISSN 2349-7815
Page | 32
Paper Publications
[6] Yi-Hua Liu, Guang-Zhou Zhao“Power System Fault Detection based on Stationary Wavelet Packet Transform and
Hilbert Transform”2008 Congress on Image and Signal Processing .
[7] Shyh-Jier Huang,Cheng-Tao Hsieh Ching-Lien Huang, “Application of Morlet Wavelets to Supervise Power System
Disturbances,” IEEE Transactions on Power Delivery, Vol. 14, No. 1, January 1999.
[8] Behnam Mahamedi and Jian Guo Zhu “A Novel Approach to Detect Symmetrical Faults Occurring During Power
Swings by Using Frequency Components of Instantaneous Three-Phase Active Power” IEEE Transactions on Power
Delivery, Vol. 27, No. 3, July 2012.
[9] V. S. Kale, S. R. Bhide ,P. P. Bedekar “ Faulted Phase Selection on Double Circuit Transmission Line using
Wavelet Transform and Neural Network” 2009 Third International Conference on Power Systems, Kharagpur,
INDIA December 27-29 ICPS-195.
[10] El Sayed Mohamed Tag Eldin “Fault Location for a Series Compensated Transmission Line Based on Wavelet
Transform and an Adaptive Neuro-fuzzy Inference System” 978-1-4244-6981-9/10 ©2010 IEEE.
[11] David C. Robertson, Member, Octavia I. Camps, Member “Wavelets and Electromagnetic Power System
Transients” IEEE Transactions on Power Delivery, Vol. 11, No. 2, April 1996.
[12] P. Venugopal Rao, Shaik Abdul Gafoor, and C. Venkatesh,“ Detection of Transmission Line Faults by Wavelet
Based Transient Extraction” ACEEE Int. J. on Electrical and Power Engineering, Vol. 02, No. 02, August 2011.
[13] Smita Shrivastava, Shailendra Jain, R.K. Nema “Wavelet Entropy: Application in Islanding Detection” WSEAS
Transactions on Power system, Issue 3, Volume 7, July 2012.
[14] Swaroop Kumar.Nallagalva,Mukesh Kumar Kirar, Dr.Ganga Agnihotri “Transient Stability Analysis of the IEEE
9-Bus Electric Power System” International Journal of Scientific Engineering and Technology (ISSN : 2277-1581)
www.ijset.com, Volume No.1, Issue No.3, pg : 161-166 01 July 2012.

Shunt Faults Detection on Transmission Line by Wavelet

More Related Content

What's hot (18)

Viewers also liked (12)

Similar to Shunt Faults Detection on Transmission Line by Wavelet (20)

Recently uploaded (20)

Shunt Faults Detection on Transmission Line by Wavelet