Dk26766771

298 views

Published on

0 Comments
0 Likes
Statistics
Notes
  • Be the first to comment

  • Be the first to like this

No Downloads
Views
Total views
298
On SlideShare
0
From Embeds
0
Number of Embeds
3
Actions
Shares
0
Downloads
5
Comments
0
Likes
0
Embeds 0
No embeds

No notes for slide

Dk26766771

  1. 1. V. Usha Reddy, S. Deepika Chowdary / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 6, November- December 2012, pp.766-771 Detection And Mitigation Of Power Quality Disturbances Using DWT Technique And DVR V. Usha Reddy*, S. Deepika Chowdary** *(Department of EEE, S.V. University, Tirupati – 517 502) ** (Department of EEE, S.V. University, Tirupati – 517 502)ABSTRACT Power quality (PQ) is an issue that is different people. There is no agreed definition forbecoming increasingly important to electricity power quality, it may be defined as the problemsconsumers at all levels of usage. Improvement of manifested in voltage, frequency and the effect ofPQ has a positive impact on sustain profitability harmonics, poor power factor that results in mis-of the distribution utility on the one hand and operation/failure of customer equipment. Certaincustomer satisfaction on the other. To improve type of power quality degradation results in lossesthe power quality problems, the detection of PQ and thus losses in transmission and distributiondisturbances should be carried out first. The PQ system have come under greater scrutiny in recentproblems vary in a wide range of time & years. When wave shapes are irregular, voltage isfrequency, which make automatic detection of poorly regulated, harmonics and flickers are present,PQ problems often difficult to analyse the cause or there are momentary events that distort theof occurrence.There are numerous types of PQ usually sinusoidal wave, and the power utilization isdisturbances according to IEEE standards that degraded. This paper outlines the issue relating toare concerned very often. As it is evident that if power quality and their impact on Energythere is a problem then there will be a technique Conservation. Due to the advantages in technologyto solve that problem. As the same to detect the and the increasing growing of industrial/commercialPQ problems and their disturbances waveform facilities in regions, the power quality has been aautomatically, there are some techniques that are major concern among industries. Thus in order toproposed such as Fast Fourier Transform, Short maintain the consistent standard of power quality inTime Fourier Transform and Wavelet Transform the near feature with the increasing number ofTechnique. PQ disturbances also vary in a wide industries and commercial facilities; it would berange of time and frequency and Discrete wavelet extremely urgent and tackle the power qualityTransformation(DWT) with Fuzzy classifier have challengers. This is then lead to time wasted whileunique ability to examine the signal in time and trying to solve the problems. However certainfrequency ranges. In this paper the mitigation of appropriate measures can be taken to protect thedetected power quality problems can be done by affected equipment or to raise the quality of powerthe device Dynamic Voltage Restorer (DVR). supply. Power supply quality issues and resulting problems are the consequences of the increasing useKeywords—Power quality, Detection of of solid state switching devices, nonlinear and powerdisturbance, Discrete wavelettransform, Fuzzy electronically switched loads, unbalanced powerclassifier, DVR. system, lightning controls, computer and data processing equipment, as well as industrial plantI. INTRODUCTION rectifiers and inverters.A power quality problem In an ideal ac power system, energy is usually involves a variation in the electric servicesupplied at a single constant frequency and specified voltage or current, such as voltage dips andvoltage levels of constant magnitudes. However, this fluctuations, momentary interruptions, harmonicssituation is difficult to achieve in practice. The and oscillatory transient causing failure, or mis-undesirable deviation from a perfect sinusoidal operation of the power service equipment[1,2].waveform is generally expressed in the terms of With the common use of all kinds ofpower quality. The power quality is an umbrella electronic sensitive equipment, electric powerconcept for many individual types of power system quality, including voltage sag, voltage swell, voltagedisturbances such as harmonic distortion, transients, harmonics, and oscillatory transients has attractedvoltage variations, voltage flicker, etc. Of all power great attention. How to extract the features ofline disturbances harmonics are probably the most disturbances from a large number of power signalsdegenerative condition to power quality because of and how to recognize them automatically arebeing a steady state condition .Electric powerquality important for further understanding and improvingmay be defined as a measure of how well electric power quality. Hence, to improve the power quality,power service can be utilized by customers. The it is required to know the sources of power systemterm power quality means different things to disturbances and find ways to mitigate them. 766 | P a g e
  2. 2. V. Usha Reddy, S. Deepika Chowdary / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 6, November- December 2012, pp.766-771Signal processing techniques have been widely used lead to poor classification accuracy in the ANN andfor analysing power signals for the purpose of the FL methods, is the non-uniform time alignmentautomatic power quality (PQ) disturbance between the test signal and the prestored templates.recognition. Among the different signal processing The time alignment issue can be most efficientlytechniques used in the extracting of features of handled by applying the DWT algorithm, which hasdisturbances from a large number of power signals, been extensively used in the speech recognition. Thethe most widely used techniques are fast Fourier DWT is a template matching algorithm derived fromtransform (FFT) and the windowed Fourier dynamic programming. Conceptually, templatetransform which comprises of the short time Fourier matching is based on the comparison of the testtransform (STFT) and the wavelet transform (WT). signal against all of the stored templates in theThe FFT is ideal for calculating magnitudes of the dictionary. A measure of similarity is calculated, andsteady state sinusoidal signals but it does not have then used to achieve a recognition decision.the capability of copying with sharp changes and Among the power quality problems (sags,discontinuities in the signals. Although the modified swells, interruptions) voltage sags are theversion of the Fourier transform referred to as the mostsevere disturbances. In order to overcome theseSTFT can resolve some of the drawbacks of the problems the concept of custom power devices isFFT, it still has some technical problems. In the introduced recently. One of those devices is theSTFT technique, its resolution is greatly dependent Dynamic Voltage Restorer (DVR), which is the moston the width of the window function in which if the efficient and effective modern custom power devicewindow is of finite length, the technique covers only used in power distribution networks. DVR is aa portion of the signal, thus causing poor frequency recently proposed series connected solid state deviceresolution. On the other hand, if the length of the that injects voltage into the system in order towindow in the STFT is infinite so as to obtain the regulate the load side voltage. It is normally installedperfect frequency resolution, then all the time in a distribution system between the supply and theinformation will be lost. Due to this reason, critical load feeder at the point of common couplingresearchers have switched to wavelet transform from (PCC). Other than voltage sags and swellsthe STFT[3,4]. compensation, DVR can also added other features Some of the well-known wavelet like line voltageharmonics compensation, reductiontransforms are the continuous wavelet transform of transients in voltage and fault current(CWT) and the discrete wavelet transform (DWT). limitations[6,7].Wavelet analysis is based on the decomposition ofthe signal according to time-scale, rather thanfrequency, using basis functions with the adaptablescaling properties which are known as multi-resolution analysis. A wavelet transform expands a Fig: 1 Test systemsignal not in terms of a trigonometric polynomial butby wavelets, generated using transition (shift intime) and dilation (compression in time) of a fixedwavelet function. The wavelet function is localizedin time and frequency yielding wavelet coefficients II. Power Quality Problems In Distributionat different scales. Several types of wavelets have systemsbeen considered for detection, and localization of A wide diversity of solution to powerpower quality problems as both time and frequency quality problems is available to the both theinformation are available by multi-resolution distribution network operator and the end user. Moreanalysis. Most of the proposed systems uses either sophisticated monitoring equipment is readilythe Fourier transform (FT) or the wavelet transform affordable to end-users, who empower withfor feature extraction and the artificial neural information related to the level of power qualitynetwork (ANN) or fuzzy logic (FL) for event measurable quantities or occurrences.Most of theclassification. ANN have several drawbacks, more important international standards definesincluding their inherent need of a large numbers of power quality as the physical characteristic of thetraining cycles. The key benefit of FL is that its electrical supply provided under normal operatingknowledge representation is explicit in utilizing conditions that do not disturb the customer’ssimple “IF-THEN” relation. The wavelet transform processes. Therefore, a power quality problem existshas the capability to extract information from the if any voltage, current or frequency deviation resultssignal in both time and frequency domains in a failure or in a bad operation of customer’ssimultaneously. Recently, the wavelet transform has equipment. Power quality problems are common inbeen applied in the detection and classification of the most of commercial, industrial and utility networks.power quality events[5]. Natural phenomena, such as lightning are the most A major concern in the classification of frequent case of power quality problems. Switchingprocess of power quality disturbances, which may phenomena resulting in oscillatory transients in the 767 | P a g e
  3. 3. V. Usha Reddy, S. Deepika Chowdary / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 6, November- December 2012, pp.766-771electricalsupply, for example when capacitors are pass filters to analyze the low frequencies. Theswitched, also contribute substantially to power resolution of the signal, which is a measure of thequality disturbances. Also, the connection of high amount of detail information in the signal, ispower non-linear loads contributes to the generation changed by the filtering operations, and the scale isof current and voltage harmonic components., changed by up sampling and down samplingleading to costly interruptions of production. Power operations. For example, sub sampling by twodisturbances are caused by the generation, refers to dropping every other sample of the signal.distribution and use of power, and lightning. A Sub sampling by a factor n reduces the number ofpower disturbance can be defined as unwanted samples in the signal n times. The procedure startsexcess energy that is presented to the load. with passing this signal through a half band digital low pass filter with impulse response h[n].The various power quality disturbances are The convolution operation in discrete time is2.1 Voltage Sag: defined as follows: Voltage sag is a short term reduction in, RMS voltage in the range of 0.1 to 0.9 p.u. for the x[n]  h[n]   x[k ].h[n  k ] (3.1) k duration greater than half the mains cycle and lessthan 1 minute. It is specified in terms of duration andretained voltage, usually expressed as the percentage Half band low pass filtering removes half ofof nominal RMS voltage remaining at the lowest the frequencies, which can be interpreted as losingpoint during the dip. Voltage sag (dip) means that half of the information.The signal is thenthe required energy is not being delivered to the load subsampled by 2 since half of the number of samplesand this can have serious consequences depending are redundant. This doubles the scaleon the type of load involved. This procedure can mathematically be expressed as:2.2 Voltage Swell:  A voltage swell is increase in the RMS y[n]   h[k ].x[2n  k ] k  (3.2)voltage in the range of 1.1 to 1.8 p.u. for durationgreater than half the mains cycle and less than 1 DWT employs two sets of functions, calledminute. It is caused by load switching and capacitor scaling functions and wavelet functions, which areswitching. associated with low pass and high pass filters, respectively. The original signal x[n] is first passed2.3 Momentary Interruption: through a halfband highpass filter g[n] and a lowpass A type of short duration in which the filter h[n].complete loss of voltage on one or more phase This constitutes one level of decomposition and canconductors for a time period between 0.5 cycles and mathematically be expressed as follows:3 seconds. yhigh[k ]   x[n].g[2k  n]  (3.3) n ylow[k ]   x[n].h[2k  n]2.4 Impulsive Transients: A sudden non power frequency change in  (3.4) nthe steady state condition of voltage or current i.e.unidirectional polarity (primarily either positive ornegative). Where yhigh[k] and ylow[k] are the outputs of the highpass and lowpass filters, respectively, after subsampling by 2.The highpass and lowpass filtersIII.POWER QUALITY ANALYSIS USING are not independent of each other, and they areDISCRETE WAVELET TRANFORM related by:(DWT The DWT is considerably easier to g[ L  1  n]  (1) n .h[n]  (3.5)implement when compared to the CWT. The basicconcepts of the DWT will be introduced in this However, if the filters are not idealsection along with its properties and the algorithms halfband, then perfect reconstruction cannot beused to compute it. As in the previous chapters, achieved. Although it is not possible to realize idealexamples are provided to aid in the interpretation of filters, under certain conditions it is possible to findthe DWT. filters that provide perfect reconstruction. The most famous ones are the ones developed by Ingrid3.1 Sub band coding Daubechies, and they are known as Daubechies’ In the discrete case, filters of different cut- wavelets.off frequencies are used to analyze the signal atdifferent scales. The signal is passed through aseries of high pass filters to analyze the highfrequencies, and it is passed through a series of low 768 | P a g e
  4. 4. V. Usha Reddy, S. Deepika Chowdary / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 6, November- December 2012, pp.766-771 IV.SIMULATION RESULTS AND OUTPUTS CA1 CD1 CA1 CD1 Fig 4.3 Extraction of coefficients for Impulse Transientsusing DWT Fig 4.1Extraction of coefficients for sag using DWT CA1 CD1CA1 CD1 Fig 4.4Extraction of coefficients for Momentary Interruptions Fig 4.2 Extraction of coefficients for swell using DWT 769 | P a g e
  5. 5. V. Usha Reddy, S. Deepika Chowdary / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 6, November- December 2012, pp.766-771Table 4.1: Matlab Results for Sag Percentage Load First peak Of Sag (%) (MVAR) coefficients 10 0.44 2.075 20 0.98 3.514 30 1.64 4.5 40 2.55 6.2871 50 3.84 8.679 60 5.68 11.815 70 8.95 16.672 80 15.5 24.045 90 35.4 126.23 Fig 4.6: Simulation wave form withusing DVRTable 4.2: Matlab Results for Swell Percentage Of Load First peak Swell (%) (MVAR) coefficients 10 0.175 59.92 20 0.235 58.6 30 0.31 59.8 40 0.41 60.76 50 0.558 61.58 Fig 4.7: Simulation Full wave form withusing DVR 60 0.82 62.28 70 1.9 63.18 V. CONCLUSION The discrete wavelet transform termed as DWT is used in this project as powerful analysis toolTable 4.3: Matlab Results for Impulse Transients for monitoring , detecting and classification of power quality disturbances . The DWT gave the Load First peak First peak perfect time – frequency plot by which it became (MVAR) coefficient time (sec) easier to analyse the power quality disturbances 20 97.1816 0.1102 which has occurred in the power system network.The proposed method is simple andTable 4.4: Matlab Results for Momentary effective methodology for the detection andInterruptions classification of power quality disturbances. By applying the DWT to the distorted voltage signals of the disturbances, the DWT coefficients are extracted, Fault First peak First Peak in which the cD1 (Detail coefficients at level1) is Impedance coefficient time(sec) used to analyse the disturbance. The first peak of the (ohms) cD1 is taken for all the four PQ disturbances i.e., the 0.001 6.805 0.2002 magnitude of the first peak is taken and given as the input to the fuzzy classifier. A MATLAB program has been developed for the detection and classification of PQ disturbances where it gives the information about the time duration of the disturbance, magnitude of the coefficients (cD1) and classifies what kind of PQ disturbance has occurred. Several power quality disturbances such as voltage sag, voltage swell, impulse transients and momentary interruptions have been analysed and recognized using the Discrete Wavelet Transform(DWT). This paper has presented the power quality problems such as voltage sags, swells andFig 4.5: Simulation wave formwithout using DVR interruptions and mitigation techniques of custom 770 | P a g e
  6. 6. V. Usha Reddy, S. Deepika Chowdary / International Journal of Engineering Research and Applications (IJERA) ISSN: 2248-9622 www.ijera.com Vol. 2, Issue 6, November- December 2012, pp.766-771power electronic device DVR. The design andapplications of DVR for voltage sag results arepresented.A new PWM-based control scheme hasbeen implemented to control the electronic valves inthe two-level VSC used in the DVR. As opposed tofundamentalfrequency switching schemes alreadyavailable in the MATLAB/ SIMULINK, this PWMcontrol scheme only requires voltage measurements.The simulations carried out showed that the DVRprovides relatively better voltage regulationcapabilities.REFERENCES [1] T.Lachman, A.P.Memon, T.R.Mohomad, Z.A.Memon, “Detection of Power Quality Disturbances Using Wavelet Transform Technique”, International Journal For Science & Arts, Vol .1, No .1, 2010. [2] Penna, C., “Detection and Classification of Power Quality Disturbances Using the Wavelet Transform, M. Sc. Dissertation”, June 2000, University Federal deUberlandia, Brazil. [3] Gaouda, A. M., Salama, M. M. A., Sultan M. R., Chikhani, A. Y., “Power Quality Detection and Classification Using Wavelet- multiresolution Signal Decomposition”, IEEE Transactions on Power Delivery, Vol. 14, No. 4, pp. 1469- 1476, October 1999. [4] Surya Santoso, Edward J. Powers, W. Mack Grady, Peter Hofmann, “Power Quality Assessment via Wavelet Transform Analysis”, IEEE transactions on Power Delivery,Vol. 11, No.2, April 1996. [5] Ibrahim W.R.Anis and Morcos M.M., 2002. Artificial intelligence and advanced mathematical tools for power quality applications: a survey, IEEE Trans. on Power Delivery,vol. 17, pp. 668-673. [6] S.V Ravi Kumar, S. Siva Nagaraju “ Simulation of DVR and D-STATCOM in power systems”, APRN Journal of Engineering And Applied Sciences , vol.2, No 3 jun. 2007. [7] Haque, M.H., “Compensation of distribution system voltage sag by DVR and D-STATCOM”, Power Tech Proceedings, IEEE Porto, vol.1, pp.10-13, Sept. 2006. 771 | P a g e

×