Fault classification in a transmission line using discrete wavelet transform and artificial neural networks

Abstract

Occurrence of fault in transmission line can cause loss of power to consumer. When fault occur, it is important to have a system that can identify the types of fault thus can make fast corrective action. The Matlab/Simulink is used to simulate four different types of fault signals in transmission line which are Single Line to Ground (SLG), Double line to Ground (DLG), Line to Line (LL) and Three Phase Fault. The mother wavelet daubechies4 (db4) is used while using DWT for decomposition of these signals to obtain coefficient. The coefficient data sets which are obtained from the DWT is used as inputs for training and testing the ANN architecture. The overall accuracy of 95% is achieved using the DWT and BPNN technique after 62 iterations. The high success rate means the system have low mean squared error and is reliable to use for fault classification. The results indicate that the proposed scheme can correctly classify almost every possible fault types.