Navegando por Assunto "Fast fourier transform"

Agora exibindo 1 - 1 de 1

Acesso aberto (Open Access)
Detecção de faltas de alta impedância em redes aéreas de distribuição: testes experimentais em laboratório e análises no domínio do tempo e da frequência
(Universidade Federal do Pará, 2023-07-04) SOUSA, Ádrea Lima de; VIEIRA, João Paulo Abreu; http://lattes.cnpq.br/8188999223769913
The present work deals with the analysis of high impedance faults (HIFs) in overhead distribution networks. A comprehensive and in-depth study is conducted on this type of fault, as well as on some existing methods to detect HIFs. Next, the laboratory developed for experimental testing of HIFs is explained, which replicates the three-phase, three wire medium voltage distribution network with solidly grounded system at the primary substation typically used in Brazil. From the laboratory, 127 tests were performed in 18 different scenarios, including shunt faults and series shunt faults on the source side. In order to perform a comparative analysis with the existing literature on high impedance fault detection based on harmonics, the harmonic spectrum was evaluated, based on even harmonics, odd harmonics (especially the third order), and the angle of the third harmonic, obtained through the Fast Fourier Transform (FFT). The main contribution of this work lies in evaluating the peculiarities of the harmonic spectrum of voltage and current signals obtained from experimental laboratory tests for different soil topologies and tree branch configurations, and their correlation with the signal in the time domain. This provides an important foundation for the development of HIF detection algorithms. The results obtained in the analysis revealed that shunt and series shunt high impedance faults exhibit distinct behaviors in the harmonic spectrum and time domain, and a large portion of the tests performed does not fall within the thresholds established in the literature. Furthermore, it should be noted that the evaluation of the harmonic spectrum of voltage was carried out, which is usually not considered in HIF detection algorithms.