Fault Detection and Classification in Microgrid Using Wavelet Transform and Artificial Neural Network

Abstract

In the proposed work, Wavelet Transform analysis and wavelet entropy methods have been used to classify various types of fault in a nine bus microgrid system. Both methods are compared and analyzed. The simulation result shows that the proposed method successfully identifies the fault type and phase involved in the fault. The proposed algorithm is validated for different locations and fault types on nine bus microgrid system. In addition to the above, wavelet analysis and wavelet coefficients are also used with the Artificial Neural Network (ANN) for detecting and classifying the faults. The different fault cases have different fault resistances and inception angles. The fault detection process is done by the summation of sixth level detail coefficients of current obtained using Discrete Wavelet Transform (DWT) based Multiresolution Analysis (MRA) technique for all the three phases while, for the classification of fault type, wavelet entropy calculations for each phase currents are acquired.