Artificial intelligent based fault location technique for EHV series-compensated lines

This paper proposes the use of fuzzy neural networks (FNN) to solve the fault location problem for series-compensated lines. The technique is based on a hybrid intelligent model that integrates artificial neural networks (ANN) and a fuzzy logic system (FLS). The frequency components of the instantaneous three phase voltages and currents derived at the fault locator-end of the line are used to train an ANN to classify the fault type, and a separate FNN is used to accurately locate a fault on EHV series-compensated lines.