A new approach to fault location in two-terminal transmission lines using artificial neural networks

This paper presents a new approach to fault location in two-terminal overhe ad transmission lines, using artificial neural networks (ANN's). The method presented enables the distance to be determined at which the fault occurs in a two-terminal transmission line using the fundamental components of 50/ 60 Hz of the fault and pre-fault voltage and current magnitudes, measured i n each phase of the reference end. The main characteristic of the method is the consideration during the training period of both correctly measured ma gnitudes and magnitudes affected by errors. Using properly validated fault simulation software has checked the accuracy of the method with both correc t initial data and initial data containing errors. (C) 2000 Elsevier Scienc e S.A. All rights reserved.