A Reliable Fault Classification Method in the AC Distribution System Based on the Bagged Tree Ensemble Classifier

Many typical faults in the such as LG, LLG, LLLG, and LLL, can distribution system, standard protection relays because the fault impedance is too excessive. Several approaches for overcoming these issues have been described in the literature. Nevertheless, these approaches were incapable of classifying fault types or determining their severity. In this paper, a fault classification prevention approach for AC distribution systems is proposed using Empirical wavelet transform (EWT) and Ensemble classifier techniques. The proposed technique has been tested against various defects such as LG, LLG, LLLG, and LLL situations. The precision is tested on an IEEE 13 bus system.