A Current Based Hybrid Algorithm using Discrete Wavelet Transform and Hilbert Transform for Detection and Classification of Power System Faults in the Presence of Solar Energy

Abstract

This presents a current based algorithm using complex features evaluated by the use of discrete wavelet transform (DWT) and Hilbert transform (HT) for identification and classification of transmission line faults with solar energy. The proposed algorithm presents the fault indexes based on the current signal which is used for investigations of faults. The current signal is processed using discrete wavelet transform at a sampling frequency (SF) of 3.907 kHz for obtaining proposed W-index using detailed coefficients corresponding to first decomposition level. Same current signal is also processed using HT at a SF of 3.907 kHz for obtaining proposed H-index from absolute magnitudes of HT output. Proposed fault index (FI) is achieved by multiplying the W-index and H-index. A threshold value of proposed current based FI is selected for discriminating faulty phase from healthy conditions. Faults investigated in the study include the faults line to ground (LG), double line (LL), double line with ground (LLG) and fault involving all the three phases and ground (LLLG). Performance of algorithm is validated in MATLAb/Simulink software.