Application of Hilbert--Huang transform and support vector machine for detection and classification of voltage sag sources

Abstract

Power quality disturbances, including voltage sag, swell, harmonics, flicker, and notch, are one of the main concerns for industries and electrical equipment. Among these disturbances, voltage sag, due to its irrecoverable economic effects on industries, is particularly important. In this paper, the detection and classification of voltage sag sources containing motor starting, short circuit, transformer energizing, and the reacceleration of motors after fault clearance using the Hilbert--Huang transform (HHT) and support vector machine (SVM) are studied. A voltage sag waveform includes several oscillating modes; for separating these oscillating modes, which are called intrinsic mode functions (IMFs), empirical mode decomposition is used. Next, by applying the HHT to these IMFs, some required features of each IMF are extracted. Finally, these features are given to the SVM for classification. The results of this classification method as compared with other methods show the high efficiency of the proposed method.