Combined wavelet transfoms and neural network (WNN) based fault detection and classification in transmission lines

Transmission line protection is an important issue in power system engineering because 85-87% of power system faults are occurring in transmission lines. This project work presents a technique to detect and classify the different operating conditions in transmission lines to contribute quick and reliable operation of protection schemes. Discrimination among different transient conditions of transmission lines is achieved by combining wavelet transforms with neural network. MATLAB (7.3) / Simulink is used to simulate different operating signals in high voltage transmission lines namely single phase to ground fault, line to line fault, double line to ground, three phase short circuit, capacitor switching and breaker operation. The discrete wavelet transform (DWT) is applied for analysis of transients, because of its ability to extract information from the transient signal, simultaneously both in time and frequency domain. The data sets which are obtained from the DWT is used for training and testing the ANN architecture. In this proposed scheme three layer Back Propagations (BP) Neural network is used as classifier. It is concluded that this scheme discriminates the transients corresponding to single phase to ground, line to line fault, double line to ground, three phase short circuit, breaker operation and capacitor switching in high voltage transmission Line (HTL) in a very clear manner.