Proposal for a distance relaying scheme based on wavelet assisted neural network

In the present scenario of market driven business, power supply has become more like a commodity. Reliable and quality power need to be ensured to meet customer requirements. In such a situation, it is extremely important that transmission line faults be identified accurately, reliably and in quick time. Fault information from relays need to be used effectively for locating the fault. Advanced signal processing tools such as discrete wavelet transform (DWT) can be used very effectively for parameterization and characterization of the fault signals. On the other hand, properly configured neural network (NN) can be utilized for classification of the faults based on the DWT signal. The present contribution uses electromagnetic transient program (EMTP) for simulation of a model transmission system and MATLAB for performing DWT and NN. Various types of faults have been simulated at different locations along the transmission line and an attempt has been made to correctly identify and locate the fault.