A traveling-wave fault location technique for three-terminal lines based on wavelet analysis and Recurrent Neural Network using GPS timing

One of the most important features of smart distribution networks is handling fault situations in an efficient way. This paper describes a fault location algorithm for three-terminal transmission lines based on wavelet transform and Artificial Neural Network (ANN). Because of small size data base, Recurrent Neural Network (RNN) was utilized and for the purpose of synchronized time tagging, the Global Positioning System (GPS) with the highly-accurate timing capabilities is used. All the possible fault types are generated using the ATP-EMTP and results are discussed. Extensive simulation studies indicate that proposed network estimate fault location in different conditions with average error percentage less than 0.15% though practical limitations.