Synchrosqueezing Wavelet Transform based Identification of Transient Events in AC Microgrid

Abstract

This paper deals with the identification of transient events in AC microgrid using continuous wavelet transform-based synchrnosqueezing transform. The traditional time-frequency (TF) transforms always generate low-resolution Time-Frequency Responses (TFR). However, Synchrosqueezing transform is a reassignment method that provides a high-resolution time-frequency spectrum while being invertible. In this paper, the focus is put on the wavelet-based synchrosqueezed transform (SWT) that represents the joint application of time-frequency transformation and T-F enhancement. The transient events considered in this study, include single line-ground (L-G) fault, three-phase fault (dead short circuit), load switching, grid disconnection, and distributed generation (DG) switching. SWT is also compared with other time-frequency transforms, namely, continuous wavelet transform (CWT) and S-transform. SWT gives a more concentrated and sharp TFR which improves the readability of the spectrum. Secondly, the effectiveness of the synchrosqueezing wavelet transform (SWT) over other time-frequency transforms including continuous wavelet transform (CWT), S-transform is verified by computing Renyi entropy as a performance indicator.