Experimental testing of the artificial neural network based protection of power transformers

This work presents a novel technique to distinguish between magnetizing inrush and internal fault currents of a power transformer. The proposed differential algorithm is based on an artificial neural network (ANN) and unlike the existing relaying techniques, this method is independent of the harmonic contents of the differential current. A novel neural network is designed and trained using the back-propagation algorithm with experimental data. After training the network, simulation and on-line tests are carried out to evaluate the performance of the ANN based algorithm under different fault and energization conditions. Both simulation and experimental results are quite satisfactory.