Transmission Line Boundary Protection UsingWavelet Transform and Neural Network
Thread Rating:
  • 0 Vote(s) - 0 Average
  • 1
  • 2
  • 3
  • 4
  • 5
seminar class
Active In SP
**

Posts: 5,361
Joined: Feb 2011
#1
23-02-2011, 04:38 PM


Transmission Line Boundary Protection UsingWavelet Transform and Neural Network
Abstract

Two of the most expected objectives of transmissionline protection are: a) differentiating precisely the internal faultsfrom external, and b) indicating exactly the fault type usingone end data only. This paper proposes an improved solutionbased on wavelet transform and self-organized neural network.The measured voltage and current signals are preprocessed firstand then decomposed using wavelet multi-resolution analysis toobtain the high frequency details and low frequency approximations.The patterns formed based on high frequency signalcomponents are arranged as inputs of neural network #1, whosetask is to indicate whether the fault is internal or external.The patterns formed using low frequency approximations arearranged as inputs of neural network #2, whose task is to indicatethe exact fault type. The new method uses both low and highfrequency information of the fault signal to achieve an advancedline protection scheme. The proposed approach is verified usingfrequency-dependent transmission line model and the test resultsprove its enhanced performance. A discussion of the applicationissues for the proposed approach is provided at the end wherethe generality of the proposed approach and guidance for futurestudy are pointed out.
I. INTRODUCTION
Aperfect transmission line protection scheme is expectedto: a) differentiate the internal faults from external preciselyso that only the faulted line will be removed; b) providethe exact fault type selection so that advanced single-poletripping and reclosing schemes can be implemented. Thereliability of those two functions is highly desirable so thatthe impact of faults on system stability is reduced.The traditional line protection schemes based on fundamentalfrequency components of the fault generated transientvoltage and current signals, can be classified into two categories:a) non-unit protection and b) unit protection. The nonunitprotection schemes use one end transmission line datawhile the unit protection schemes usually use data from twoends. The non-unit protection such as distance relay, can notprotect the entire length of the primary line because it cannot differentiate the internal faults from external occurringaround the multi-zone boundaries. Backup protection may beintroduced as a trade-off approach for protecting the entire length of the transmission line. For unit protection such aspilot protection, it usually requires a communication link totransmit the blocking or transfer tripping signals. Therefore,the reliability of the protection scheme highly relies on thereliability of the communication link. The cost of the communicationlink also needs to be taken into account.Recently, new techniques using high frequency componentsof the fault generated transient signals were studied andsome useful solutions were obtained [1]–[4]. An approachcalled “boundary protection” for solving the disadvantagesof conventional non-unit protection schemes was proposed[2], [5]. This approach introduces a possibility of preciselydifferentiating the internal faults from external using data fromone end only. In this case, the relay at one end can protect theentire line length with no intentional time delay.Regarding the fault type selection or classification, thetraditional method is based on the fundamental frequencyphasors. The feature formed by a nonlinear ratio betweenvoltage and current phasors is compared to the threshold tofind out the faulted phase [6]. This kind of method is affectedby the different conditions such as remote-end infeed, faultresistance, mutual coupling of parallel lines, etc. An alternatesolution is to use artificial intelligence schemes such as neuralnetwork based algorithm [7], [8].This paper introduces a new approach based on wavelettransform and a self-organized neural network to realize accurateboundary protection and fault type classification using oneend transmission line data. The new method retains both lowfrequency and high frequency component of the fault signalto achieve high reliability and selectivity of the protectionscheme. It inherits many advantages from different techniquesit utilizes.The paper first presents, in Section II, the background ofboundary protection. Brief introduction of wavelet transformand self-organized neural network algorithm are then providedin Section III and Section IV respectively. Section V describesthe entire design procedure of the new protection scheme,followed by a performance study in Section VI. A discussionof the application issues for the proposed approach is providedin Section VII. Conclusion of the paper is summarized at theend.
II. BACKGROUND OF BOUNDARY PROTECTION
The principle of boundary protection was studied in [2].The previous work is explored in this section to providebackground of the new approach introduced in this paper. Thesystem shown in Fig. 1 is a typical multi-line system. We assume the relay is installed at the bus 2 to protect the line2 − 3 shown in the figure. A fault on the lines will generatewideband transient voltage and current signals. The signalswill travel in both directions with reflections and refractionsat the discontinuity points, which are usually the buses andfaults. The bus of the power system is always connected tomany power system apparatus and they usually represent thecapacitance at high frequency. This effect is shown in Fig. 1.For an external fault F2 close to the bus 3, the high frequencyportion of the fault current signal I2 will be shunted to earth(in I0) significantly due to the bus capacitance. The higherthe frequency, the more significant portion of the currentsignal will be shunted. From the viewpoint of the relay, themagnitude of high frequency portion of the fault current signalI1 is reduced. In contrast, for the internal fault F1 close tothe bus 3, the fault current of the entire frequency band canbe seen by the relay. That means, if other fault conditions(fault type, fault resistance, fault angle) are identical, we candifferentiate the internal fault F1 from the external fault F2by comparing the high frequency portions of their signals.Similarly, the same method can be used to differentiate thefaults at F3 and F4. Using the voltage signals, we can stilldifferentiate faults at F1 and F2 but can not differentiate faultsat F3 and F4 because the voltage measurements of the relayare obtained from bus 2.The feature differences of the faults on different line sectionsseen by the relay at bus 2 in Fig. 1 can be summarizedas follows:• For faults on the primary line, the energy of high frequencyportion of the voltage and current signals will beseen as “big” values.• For faults on the backward line, the energy of highfrequency portion of the voltage signals will be seen as“big” values while the energy of high frequency portionof the current signals will be seen as “small” values.• For faults on the forward line, the energy of high frequencyportion of the voltage and current signals will beseen as “small” values.It should be emphasized that the above statements are basedon the assumption that all other fault parameters are thesame and the “big” and “small” value are indicating relativenumbers. The absolute values are dependent on fault type, faultresistance, fault angle, etc.In [2], the author uses a specially designed multi-channelfilter to extract the transient current signals for two signaloutputs If1, If2 with center frequency at 80kHz and 1kHzrespectively. Then the ratio of the energy spectrum for If1, If2is calculated and compared to a threshold to find out whetherthe fault is internal or external. The advantage of this methodis justified by the result from a performance study.Still some issues are remaining in this method: a) Thedirection of the external faults can not be distinguished sinceonly the current signal is used. The method also has nophase selection function available; b) The theoretical basis forselection of the center frequency of the extracted features andselection of the thresholds is not apparent; c) The reliabilityof the method is unknown since only high frequency signalis used. It may be affected by the disturbance from noise,switching, lightning, etc; d) There are no extensive studiesprovided for the performance evaluation under various faultconditions. As mentioned earlier, the boundary condition arehighly dependent on fault type, fault resistance, fault angle,etc.This paper provides a new boundary protection schemeaimed at solving those issues. First of all, the voltage andcurrent signal will both be used; this can provide more informationabout the direction of the fault point. The new schemeuses wavelet transform as the feature extraction tool thus thereis no need to design extra filters. Wavelet transform has astrong capability of extracting the signal component underdifferent frequency bands while retaining the time domaininformation. Secondly, the extracted features will be handledusing a self-organized neural network algorithm [7]. With itsstrong capability of generalization and training mechanism, itcan be used as an alternative solution when theoretical basisfor dealing with the fault generated high frequency signalcomponents is not well defined. The neural network basedalgorithm is also capable of implementing a superior faultclassification scheme. Finally, the new scheme will use boththe low and high frequency components of the fault signal toeliminate impact from non-fault disturbances. The reliabilityand robustness of the method will be verified by an extensivestudy for various kinds of faults.
.III. WAVELET TRANSFORM
Wavelet analysis is a relatively new signal processing tooland is applied recently by many researchers in power systemsdue to its strong capability of time and frequency domainanalysis [9], [10]. The two areas with most applications arepower quality analysis and power system protection [11]–[13].The definition of continuous wavelet transform (CWT) fora given signal x(t) with respect to a mother wavelet (t) is:CWT (a, b) =1√a Z 1−1x(t) _t − ba _dt (1)where a is the scale factor and b is the translation factor.For CWT, t, a, b are all continuous. Unlike Fourier transform,the wavelet transform requires selection of a motherwavelet for different applications. One of the most popularmother wavelets found for power system transient analysis inthe literature is Daubichies’s wavelet family. In this paper, thedb5 wavelet is selected as the mother wavelet for detecting theshort duration, fast decaying fault generated transient signals.The application of wavelet transform in engineering areasusually requires discrete wavelet transform (DWT), which assume the relay is installed at the bus 2 to protect the line2 − 3 shown in the figure. A fault on the lines will generatewideband transient voltage and current signals. The signalswill travel in both directions with reflections and refractionsat the discontinuity points, which are usually the buses andfaults. The bus of the power system is always connected tomany power system apparatus and they usually represent thecapacitance at high frequency. This effect is shown in Fig. 1.For an external fault F2 close to the bus 3, the high frequencyportion of the fault current signal I2 will be shunted to earth(in I0) significantly due to the bus capacitance. The higherthe frequency, the more significant portion of the currentsignal will be shunted. From the viewpoint of the relay, themagnitude of high frequency portion of the fault current signalI1 is reduced. In contrast, for the internal fault F1 close tothe bus 3, the fault current of the entire frequency band canbe seen by the relay. That means, if other fault conditions(fault type, fault resistance, fault angle) are identical, we candifferentiate the internal fault F1 from the external fault F2by comparing the high frequency portions of their signals.Similarly, the same method can be used to differentiate thefaults at F3 and F4. Using the voltage signals, we can stilldifferentiate faults at F1 and F2 but can not differentiate faultsat F3 and F4 because the voltage measurements of the relayare obtained from bus 2.The feature differences of the faults on different line sectionsseen by the relay at bus 2 in Fig. 1 can be summarizedas follows:• For faults on the primary line, the energy of high frequencyportion of the voltage and current signals will beseen as “big” values.• For faults on the backward line, the energy of highfrequency portion of the voltage signals will be seen as“big” values while the energy of high frequency portionof the current signals will be seen as “small” values.• For faults on the forward line, the energy of high frequencyportion of the voltage and current signals will beseen as “small” values.It should be emphasized that the above statements are basedon the assumption that all other fault parameters are thesame and the “big” and “small” value are indicating relativenumbers. The absolute values are dependent on fault type, faultresistance, fault angle, etc.In [2], the author uses a specially designed multi-channelfilter to extract the transient current signals for two signaloutputs If1, If2 with center frequency at 80kHz and 1kHzrespectively. Then the ratio of the energy spectrum for If1, If2is calculated and compared to a threshold to find out whetherthe fault is internal or external. The advantage of this methodis justified by the result from a performance study.Still some issues are remaining in this method:
a) Thedirection of the external faults can not be distinguished sinceonly the current signal is used. The method also has nophase selection function available;
b) The theoretical basis forselection of the center frequency of the extracted features andselection of the thresholds is not apparent;
c) The reliabilityof the method is unknown since only high frequency signalis used. It may be affected by the disturbance from noise,switching, lightning, etc;
d) There are no extensive studiesprovided for the performance evaluation under various faultconditions.
As mentioned earlier, the boundary condition arehighly dependent on fault type, fault resistance, fault angle,etc.This paper provides a new boundary protection schemeaimed at solving those issues. First of all, the voltage andcurrent signal will both be used; this can provide more informationabout the direction of the fault point. The new schemeuses wavelet transform as the feature extraction tool thus thereis no need to design extra filters. Wavelet transform has astrong capability of extracting the signal component underdifferent frequency bands while retaining the time domaininformation. Secondly, the extracted features will be handledusing a self-organized neural network algorithm [7]. With itsstrong capability of generalization and training mechanism, itcan be used as an alternative solution when theoretical basisfor dealing with the fault generated high frequency signalcomponents is not well defined. The neural network basedalgorithm is also capable of implementing a superior faultclassification scheme. Finally, the new scheme will use boththe low and high frequency components of the fault signal toeliminate impact from non-fault disturbances. The reliabilityand robustness of the method will be verified by an extensivestudy for various kinds of faults.


download full report
pserc.wisc.edu/documents/publications/papers/2006_general_publications/boundary_protection_zk.pdf
Reply
seminar tips
Super Moderator
******

Posts: 8,857
Joined: Oct 2012
#2
29-11-2012, 12:00 PM

to get information about the topic "TRANSMISSION LINE FAULT DETECTION AND CLASSIFICATION USING WAVELET TRANSFORM AND NEURAL NETWORKS" full report ppt and related topic refer the link bellow

topicideashow-to-transmission-line-fault-detection-phase-selection-using-ann

topicideashow-to-transmission-line-boundary-protection-usingwavelet-transform-and-neural-network
Reply

Important Note..!

If you are not satisfied with above reply ,..Please

ASK HERE

So that we will collect data for you and will made reply to the request....OR try below "QUICK REPLY" box to add a reply to this page

Quick Reply
Message
Type your reply to this message here.


Image Verification
Please enter the text contained within the image into the text box below it. This process is used to prevent automated spam bots.
Image Verification
(case insensitive)

Possibly Related Threads...
Thread Author Replies Views Last Post
  wireless electricity transmission ppt jaseelati 0 320 09-02-2015, 02:38 PM
Last Post: jaseelati
  live line maintenance of transmission lines ppt jaseelati 0 281 23-01-2015, 02:24 PM
Last Post: jaseelati
  elastic optical network wiki jaseelati 0 262 22-01-2015, 02:07 PM
Last Post: jaseelati
  network intrusion detection system project report doc jaseelati 0 366 13-01-2015, 01:15 PM
Last Post: jaseelati
  switchgear and protection ppt jaseelati 0 206 09-01-2015, 02:44 PM
Last Post: jaseelati
  transmission lines and waveguides by bakshi free download jaseelati 0 225 27-12-2014, 02:14 PM
Last Post: jaseelati
  network intrusion detection system project report jaseelati 0 245 23-12-2014, 03:08 PM
Last Post: jaseelati
  transmission lines and waveguides by dhananjayan jaseelati 0 232 18-12-2014, 03:41 PM
Last Post: jaseelati
  transmission lines and waveguides by bakshi pdf free download jaseelati 0 235 11-12-2014, 03:23 PM
Last Post: jaseelati
  the client/server network strategy can handle very large networks efficiently. jaseelati 0 266 11-12-2014, 02:17 PM
Last Post: jaseelati