ABSTRACT

The largest physical length of transmission network in Pakistan makes it the most critical part of the power system. The fast recognition of fault and events in transmission line has a significant role in order to prevent equipment damage and collapse of power system. The signal-processing and computational-intelligence based techniques have been proposed in literature for automatic classification of faults and events in transmission network. In this thesis, wavelet transform based probabilistic neural network has been proposed for the identification and classification of faults in overhead transmission network. The symmetrical and unsymmetrical short circuit faults are individually created at various fault resistances and fault locations. The wavelet transform is used to extract t he features in order to distinguish the type of faults. The probabilistic neural network is used to automatically classify the type of faults. A real-time transmission network of Pakistan is proposed for simulation of faults. This algorithm has been created i n MATLAB/Simulink software. The simulation results show that the proposed algorithm is efficient and reliable for automatic classification of faults in electrical power system.