A Combined Approach of Empirical Wavelet Transform and Artificial Neural Network based High Impedance Fault Classification in Micro-grid system

High impedance faults (HIFs) in the microgrid system are difficult to identify by the conventional over-current relays due to the random, asymmetric nature and less level of arc fault current. In this paper, a combined approach of Empirical wavelet transform(EWT) and Artificial neural network(ANN) is used for fault classification in the microgrid system. The extracted features of residual current by using EWT as it provides both the frequency and time information are used for training and testing of the ANN for the distinction of HIF from no-fault cases. The accuracy of the proposed method is validated on a 5-bus Micro-grid system which is simulated in PSCAD/EMTDC software.