Application of particle swarm optimization based neural network to fault classification

A new scheme fault classification for line protection in high voltage transmission systems is proposed. The scheme uses particle swarm optimization (PSO) algorithm to train a feed-forward neural network (FNN). The goal is the enhancement of the convergence rate, learning process and fill up the gap of local minimum point. The proposed algorithm is tested on the 345 kV, 3 phases, 100 km line for equivalent transmission system with lumped-parameter presentation, considering variations in fault location and fault resistance. Simulation studies carried out using MATLAB/Simulink show that the proposed scheme is very promising and gives high accuracy.