Comparisons on ways of magnitude characterization of power quality disturbances

Abstract

Power quality disturbances are normally recorded by dedicated PQ monitors as sampled points in time. Magnitude is accepted as a significant detection and general classification index in field devices or later assessment analysis. Suitable methods of magnitude characterization are one of the bases of PQD measuring and monitoring. The magnitude of PQD can be determined in several different ways. The RMS voltage, peak voltage and fundamental voltage component are introduced to quantify the voltage severity levels in number of published standards. The paper presents the algorithms of the mentioned approaches implemented in Matlab for quasi real-time situation. The algorithms are FFT-based and/or wavelet transform based. They are validated with the example of the radial test system field voltage sag waveforms provided by the IEEE. The depth-duration characterization of voltage sags is illustrated with the mentioned different approaches. The most appropriate approach is suggested at the end of the paper according to the practical monitoring or measuring requirements. We develop a new PQ monitor with DSP chip infrastructure where all the methods are applied. The proper window length selection and technique to avoid oscillation is also discussed in the rest of the paper. The descriptive information obtained from the magnitude characterization can be used further to extract features for PQD detection and classification.