Study on Fault Location for High Voltage Overhead Transmission Lines Based on Neural Network System

A distributed & hierarchical NN (DHNN) system based on the integrated module architecture and hierarchy architecture is proposed in this paper. The DHNN system adequately uses the powerful function of artificial neural networks at aspects of pattern identification, nonlinear-approaching, associative memory et al. Its information processing mechanism coincides with the processing law of classification-sketchiness-accuracy in human biologic NN system. This system not only can deal with the advanced information required by fault location for HV overhead transmission lines but accurately locate the fault sites. Thus this method for fault location presented in this paper can thoroughly eliminate the disadvantages in other extant fault location methods, such as convergence to the false root or divergence in the procedure of iteration, which result in the great location error in practice. This paper pioneers a new direction to the study and application on fault location. Results from theoretical analyses and simulations by EMTP show that fault location precision of this method can completely satisfy practical requirements.