Design, implementation and testing of an artificial neural network based fault direction discriminator for protecting transmission lines

This paper describes a fault direction discriminator that uses an artificial neural network (ANN) for protecting transmission lines. The discriminator uses various attributes to reach a decision and tends to emulate the conventional pattern classification problem. An equation of the boundary describing the classification is embedded in the multilayer feedforward neural network (MFNN) by training through the use of an appropriate learning algorithm and suitable training data. The discriminator uses instantaneous values of the line voltages and line currents to make decisions. Results showing the performance of the ANN-based discriminator are presented in the paper and indicate that it is fast, robust and accurate. It is suitable for realizing an ultrafast directional comparison protection of transmission lines