A FUZZY-EXPERT SYSTEM FOR CLASSIFICATION OF SHORT DURATION VOLTAGE DISTURBANCES

One of the important aspects in power quality assessment is automated detectionand classification of power quality disturbances which requires the use of artificial intelligenttechniques. This paper presents the application of fuzzy-expert system for classification ofshort duration voltage disturbances which include voltage sag, swell and interruption. Toobtain unique features of the voltage disturbances, fast Fourier transform analysis and rootmean square averaging technique are utilized so as to determine the disturbance parameterssuch as duration, maximum and minimum rms voltage magnitudes. Based on theseparameters, a fuzzy-expert system has been developed to set the fuzzy rules incorporatingfive inputs and three outputs. The system is designed for detecting and classifying the threetypes of short duration voltage disturbances, so as to determine whether the disturbance isinstantaneous, momentary and non sag, swell and interruption. To verify the accuracy of theproposed system, it has been tested with recorded voltage disturbances obtained frommonitoring. Tests results showed that the developed fuzzy-expert system gives a correctclassification rate of 98.4 %.