Wavelet Analysis of Voltage Disturbances for Power Quality Applications

Abstract

This paper addresses the subject of stationary and non-stationary voltage disturbance problems in electric power system. The modified binary-tree wavelet decomposition technique with an appropriate wavelet filter is introduced as a powerful tool for detecting, classifying, and quantifying the voltage disturbances. Power quality (PQ) indices in terms of total rms, rms of individual frequency bands, duration of disturbance, and their dependent quantities such as voltage magnitude of disturbance and harmonic distortion can be measured directly from the wavelet transform coefficients. The proposed technique is simple, provides accurate value of PQ indices, and enables to discriminate among the type of similar voltage disturbance events. The proposed technique is validated by its application to various disturbance data and the measurement results, such as rms and harmonic distortion, are further compared with those obtained from the time domain and frequency domain methods. Keywords:Power Quality, Binary-Tree Wavelet Transforms, Voltage Disturbances, Harmonics. (6)(7), and some include the classification purposes (2)(4) (10), but none is intended to include the quantification purposes. Moreover, CWT techniques are criticized due to their computational redundancy and they are often difficult to be implemented in realtime system. DWT techniques are not suitable for harmonic analysis, because the resulted frequency bands do not have the same width and the results do not give easy insight in the time behavior of the harmonics(9). In addition, the choice of which wavelet to use in PQ monitoring system also plays an important role(11), because the wavelet affects the accuracy of PQ indices, such as rms, harmonic distortion, and voltage magnitude of disturbances. A technique to detect, classify, and quantify voltage disturbances is proposed. The technique is based on binary-tree wavelet transform which has been modified to fit the monitoring requirements, and an appropriate wavelet filter has been chosen to yield good quantification. As compared with the recent PQ monitoring system as illustrated in reference [2] and in reference [10] which used respectively the DWT and the CWT, the proposed technique (1) enables harmonic analysis, (2) provides more easy measurement and accurate value of PQ indices, and (3) can be used mainly for both stationary and non-stationary disturbances, besides the ability to discriminate among the type of similar disturbance events.This paper starts with a brief introduction to binarytree wavelet theory and it is followed by the rms measurement in wavelet domain. The calculation of magnitude of disturbance and harmonic distortion indices is also presented. The proposed technique is evaluated to several voltage disturbance data with stationary (harmonics and flat-top) and non-stationary (sag, swell, and

Classification of Power Quality Disturbances Using Wavelets and Support Vector Machine

“…Therefore, the selection of the most adequate wavelet mother function to be used in the analysis is one of the key factors in successful application of wavelets, not only in power quality applications. As general rule, for detection of short and fast (transient) disturbances, shorter filters are proposed as better, while for slow transient disturbances long filters are presented as particularly good [9], [14]. This means that selection of the best filter for detection and classification of PQ disturbances is not an easy task and in general depends from the application.…”

Efficient Feature Extraction and Classification of Power Quality Disturbances

“…Therefore, the selection of the wavelet mother function is one of the key factors for designing a successful wavelet application. As general rule, for detection of fast transient disturbances, shorter filters are proposed as better, while for slow transient disturbances long filters are presented as particularly good [21,22]. Thus, selection of best filter length for detection and classification is not an easy task.…”