Avaliação do impacto dos tipos de equipamentos de medição no desempenho da detecção de defeitos em sistemas de distribuição utilizando redes neurais

Abstract

In this work we developed a methodology for the detection of single-phase fault impedance in the medium voltage distribution systems and the impact is assessed using different types of equipment in the performance of the fault detection. The choice of single-phase fault detection is based on the fact that these are the vast majority in distribution systems. The proposed methodology can identify single-phase faults with arc resistance ranging from 0 to 20 ohms, however, the methodology did not aim to detect high impedance faults. Furthermore, the proposed method is able to correctly detect the flaws considering the variation in system load. The detection success rate achieved by the proposed method was greater than 95% in all simulated cases. And in this work, to perform the detection of faults in distribution systems were used artificial neural networks, where entries are electrical parameters of voltage and current can be fasorias and / or non-phasor measurements. This variation was used to analyze the performance of the fault detection using different types of measuring equipment.