507 by ides editor

Proc. of Int. Conf. on Control, Communication and Power Engineering 2010

Detection and Classification of Power Quality Disturbances using Wavelets S.Durga Prasad1 MIEEE, CH.V.V.S Bhaskara Reddy2 DEE, A.U College of Engineering, Andhra University, Visakhapatnam-3, A.P Email: 1sdp4u@in.com, 2chbr_eee@yahoo.com Abstract - This paper presents a novel method to detect and classify the power quality disturbances using wavelets. The proposed method uses different wavelets for a particular class of disturbance. This method uses wavelet filter banks in an effective way and does multiple filtering to detect the disturbance. The classifier has been tested on various disturbances viz. voltage sag, swell, momentary interruption, capacitor switching and single line to ground fault. A quantitative comparison of results shows the set of wavelets used for detection of the disturbances. This method is tested for a large class of test conditions simulated in MATLAB/SIMULINK. Index Terms- Power Quality Disturbances, Transforms ,Voltage Swell, Sag and Transients.

Wavelet

I.INTRODUCTION The increased requirements on supervision, control and performance in modern power systems make power quality monitoring a common practice for utilities. Studies of power quality phenomena have emerged as an important subject in recent years due to renewed interest in improving the quality of the electric supply. As sensitive electronic equipment continues to proliferate the studies of power quality will be further emphasized. New tools are required to extract all relevant information from the recordings in an automatic way. Without determining the existing levels of power quality, electric utilities cannot adopt suitable strategies to provide a better service. Therefore an efficient approach of justifying these electric power quality disturbances is motivated. Several research studies regarding the power quality have been conducted. Their aims were often concentrated on the collection of raw data for a further analysis, so that the impacts of various disturbances can be investigated. Sources of such disturbances can be located or further mitigated. However, the amount of acquisition data was often massive in their test cases. Such an abundance of data may be time consuming for the inspection of possible culprits. A more efficient approach is thus required in the power quality assessment. The implementation of the discrete Fourier transform by various algorithms has been constructed as the basis of modern spectral analysis. Such transforms were successfully applied to stationary signals where the properties of signals did not evolve in time. However, for those non-stationary signals any abrupt change may spread over the whole frequency axis. In this situation, the Fourier transform is less efficient in tracking the signal dynamics [1]. A point -topoint comparison scheme has been proposed to

II.DISCRETE WAVELET TRANFORM The discrete wavelet transform (DWT) is one of the three forms of wavelet transform. It moves a time domain discritized signal into its corresponding wavelet domain.

discover the dissimilarities between consecutive cycles[2]. This approach was feasible in detecting certain kinds of disturbances but fail to detect those disturbances that appear periodically. With the introduction of new network topologies and improved training algorithms, neural network technologies have demonstrated their effectiveness in several power system applications. Once the networks have been well trained, the disturbances that correspond to the new scenario can be identified in a very short time [3]. This technique has also been applied in the power system applications. However, it can only be applied to detect a particular type of disturbances. When encountering different disturbances, the network structure has to be reorganized, plus the training process must be restarted. A method of detecting power quality disturbances based on neural networks and wavelets has been proposed [4]. In this method, the fundamental component is removed using wavelets and the remaining signal corresponding to disturbances is processed and given as input to ANN. However, this method fails to detect voltage sag/swell and also new ANN’s have to be developed for different rated load voltages and sampling frequencies. Recently with the emergence of wavelets it has paved a unified framework for signal processing and its applications. Fourier transforms rely on a uniform window for spreaded frequencies. Wavelet transforms can apply various lengths of windows according to the amount of signal frequencies. Characteristics of nonstationary disturbances were found to be more closely monitored by wavelets. The transient behavior, cavities and discontinuities of signals can be all investigated by wavelet transforms For example, if there is an instantaneous impulse disturbance, which happens at a certain time interval it may contribute to the Fourier transform, but its location on the time axis is lost. However, by wavelets both time and frequency information can be obtained. In other words, the wavelet transform are more local. Instead of transforming a pure ‘time domain’ in to a pure’ frequency domain ’, the wavelet transforms find a good compromise in time - frequency domain.

Proc. of Int. Conf. on Control, Communication and Power Engineering 2010

“rebuilding” the corresponding wavelet coefficients, along the different resolution levels. This procedure will provide the approximated (aj (n)) and the detailed (dj (n)) version of the original signal as well as the corresponding wavelet spectrum. III. THE APPROACH DEVELOPED By using the DWT and observing the particular features of the several decomposition levels of a signal, some important conclusions of it can be drawn. This information can be used to detect, to locate and to classify the disturbance. A digital program was developed and implemented in the wavelet toolbox of the MATLAB platform, through five steps, as follows: Step1: Evaluation of the wavelet coefficients of the signal in study. Step2: Evaluation of the square of the wavelet coefficients found at step 1. Step3: Calculation of the distorted signal energy, in each wavelet coefficient level. The “energy” mentioned above is based on the Parseval’s theorem: “the energy that a time domain function contains is equal to the sum of all energy concentrated in the different resolution levels of the corresponding wavelet transformed signal”. This can be mathematically expressed as:

Fig. 1 Sub –band codification scheme of a signal

This is done through a process called “sub-band codification”, which is done through digital filter techniques. In the signal processing theory, to filter a given signal f(n) means to make a convolution of this signal. This is illustrated in Fig. 1: the f(n) signal is passed through a low-pass digital filter (hd(n)) and a highpass digital filter (gd(n)). After that, half of the signal samples are eliminated. This is indicated by the symbol in Fig. 1. Basically, the DWT evaluation has two stages. The first consists on the wavelet coefficients determination. These coefficients represent the given signal in the wavelet domain. From these coefficients, the second stage is achieved with the calculation of both the approximated and the detailed version of the original signal, in different levels of resolutions, in the time domain. At the end of the first level of signal decomposition (as illustrated in Fig. 1), the resulting vectors yh (k) and yg (k) will be, respectively, the level 1 wavelet coefficients of approximation and of detail. In fact, for the first level, these wavelet coefficients are called cA1 (n) and cD1 (n), respectively, as stated bellow[5] cA1(n)= ∑ k f ( n).hd (− k + 2n)

(1.a)

cD1(n)= ∑ k f ( n).gd ( − k + 2n)

(1.b)

∑

n =1

(2)

Where:

f(n): Time domain signal in study N: Total number of samples of the signal N

∑ f (n) 2 : Total energy of the f(n) signal

n =1 N

∑ aj ( n) 2 :

Total energy concentrated in the level “j “of

n =1

the approximated version of the signal ∑ Nj=1

. ∑ dj (n) 2 : Total energy concentrated in the detailed n =1

version of the signal, from levels “1” to “j”. Step4: In this stage the steps 1, 2 and 3 are repeated for the corresponding “pure sinusoidal version” of the signal in study. Step5: The total distorted signal energy of the signal in study (found in step 3) is compared to the corresponding one of the pure signal version (evaluated in step 4). The result of this comparison is a deviation that can be evaluated by (3):

⎡ en _ dist( j) − en _ ref ( j) ⎤ dp( j)(%) = ⎢ ⎥ *100 en _ ref (k ) ⎣ ⎦

Next, in the same way, the calculation of the approximated (cA2 (n)) and the detailed (cD2 (n)) version associated to the level 2 is based on the level 1 wavelet coefficient of approximation (cA1 (n)). The process goes on, always adopting the “n -1” wavelet coefficient of approximation to calculate the “n” approximated and detailed wavelet coefficients. Once all the wavelet coefficients are known, the discrete wavelet transform in the time domain can be determined. This is achieved by

(3)

where: j: wavelet transform level dp(j) (%): Deviation between the energy distributions of the signal in study and its corresponding fundamental sinusoidal wave signal, at each wavelet transform level. 194

f ( n) 2 = ∑ aj (n) 2 + ∑ Nj=1 . ∑ dj ( n) 2 :

Proc. of Int. Conf. on Control, Communication and Power Engineering 2010

en_dist(j):energy distribution concentrated in each wavelet transform level of the signal in study. en_ref(j): energy distribution concentrated in each wavelet transform level of the correspondent fundamental component of the signal in study. en_ref(k):energy concentrated at the level k (which concentrates the highest energy) of the corresponding fundamental component of the signal in study.

Fig.3 some voltage signals with PQ disturbances

Fig.2 Network used for creating various PQ disturbances

The “dp(j) (%) x wavelet levels” curve of every particular power quality disturbance has an unique pattern that can be used to identify the problem in the voltage waveform. IV. SIMULATION RESULTS Network as shown in Fig. 2 is used for simulation as well as for visualizing disturbances in MATLAB/SIMULINK. After obtaining the disturbances data, Multi-resolution signal decomposition (MSD) wavelet analysis is applied on the PQ signal for extracting the useful information relating the event. Thereafter, Parseval’s energy theorem is used to determine the energy of the PQ signal corresponding to each detailed level. Parseval theorem relates the energy of the distorted signal to the energy in each of the expansion components and their wavelet coefficients. All the procedures related in section 3 were repeated to all Fig. 3 curves. The final results are summarized in the “Distribution of energy deviation” curves illustrated in Fig. 4 For instance, Fig. 4(a) does not show any deviation from the Sinusoidal signal because it refers exactly to the Fig. 3(a) pure sinusoidal. However, the

Fig.4 Deviation in energy distributions related to fig.3voltages

other curves (Fig. 4(b),...,4(h)) indicate different deviation patterns from the sinusoidal waveform. These features are unique for each disturbance studied an they can be adopted as “patterns” or “signatures” for each disturbance. PQ events detection capabilities of different wavelets: The following wavelets have been considered for comparison of performance evaluation using the proposed method. They are DMeyer, Daubecius, Sym5, Coif, Bior and Haar wavelets as shown in Table 1. From the table it is clear that, Sym8 and Coif detects Voltage sag accurately. Db4 detects voltage swell accurately.Sym8 and Db10 detects voltage flickers accurately. Dmey detects harmonics accurately and Db3 detects Impulsive Transients accurately.

Proc. of Int. Conf. on Control, Communication and Power Engineering 2010 TABLE.1

REFERENCES

PERFORMANCE CAPABILITIES OF WAVELETS

S.No 1 2 3 4

Power Quality Event Voltage Sag Voltage Swell Flickers Harmonics

Wavelet Sym8,Coif Db4 Sym8,Db10 Dmey

Impulsive Transients

Db3

[1] Malay, S. and W.L. Hwang, 1992. “Singularity detection and processing with wavelets”. IEEE Trans.Information Theory, 38: 617-643. [2] Ghosh,A.andG.Ledwich,2002."Power Quality Enhancement Using Custom Power Devices”. Kluwer Academic Publishers, USA. [3] Wilkinson, W.A. and M.D. Cox, 1996. “Discrete wavelet analysis of power system transients.” IEEE Trans. Power Systems, 11: 2033-2044 [4] Santoso, S., E.J. Powers, W.M. Grady and A.C. Parsons, 2000. “Power quality disturbance waveform recognition using wavelet-based neural classifier”. I. Theoretical foundation. IEEE Trans. Power delivery, 5: 222-228. [5] Gaouda, A. M., Salama, M. M. A., Sultan M.R., Chikhani, A. Y., “Power quality detection and Classification using wavelet-multi resolution signal decomposition”, IEEE Transactions on Power Delivery, Vol. 14, No. 4, October 1999, p. 1469- 1476.

V. CONCLUSION Nowadays the detection and classification of transient phenomena in the power supply lines are of ever growing importance. Correct detection of undesired transient disturbances is essential for electrical utilities. It is also important for customers principally in verifying that the received power is above the level of quality defined by the service contract. This paper has shown an approach that, by using some wavelet transform features, it is able to detect and to locate in time, as well as to classify a transient disturbance. It was shown that the power quality disturbances have unique deviations in their curves of energy from their corresponding pure sinusoidal waveform curve of energy. This feature is adopted to provide a classification of the type of disturbance. Also different types of wavelets that are accurately detect Particular type of Power Quality disturbances presented in this paper.