High impedance fault diagnosis in electrical power distribution systems using signal processing techniques

Author(s)

Gabriela Nunes Lopes

• Document type: Master's Dissertation
• Press: São Carlos.
• Institution: Universidade de São Paulo (USP). Escola de Engenharia de São Carlos (EESC/SBD)
• Defense date: 2020-07-16

Examining board members

José Carlos de Melo Vieira Júnior; José Rubens Macedo Junior; Benemar Alencar de Souza

• Advisor: José Carlos de Melo Vieira Júnior

Abstract

The detection of High Impedance Fault (HIF) in Distribution Systems (DSs) is an arduous task that attracts the interest of companies and researchers worldwide. The random behavior, the possibility of ambiguity concerning other transients, and the small magnitude of the HIF current waveform, prevent the quick and effective identification by traditional protection methods. Furthermore, it is emphasized that the possible damage to the medium voltage network equipment, as well as risks to people and animals, are associated with the occurrence of HIFs and give rise to the search for new techniques and protection devices. However, there is no fully efficient solution for detecting this type of fault. Therefore, the goal of this dissertation is to analyze and make the diagnosis of HIFs through detection techniques based on Fourier Transform (FT), Wavelet Transform (WT), and Stockwell Transform (ST). In this regard, this work makes a critical and quantitative analysis of metrics usually used in the bibliography based on the aforementioned transforms, in addition to proposing two new algorithms based on ST. The results proved the need to define the limits of the techniques\' applicability and to carry out tests about other frequent events in DSs, including switching of non-linear loads, usually not addressed in the literature. Lastly, it is remarked that both proposed algorithms exhibited promising results, presenting a significant contribution to the studies of HIFs in DSs. (AU)

• FAPESP's process: 18/14242-0 - Development of a hybrid method for high impedance fault detection in power distribution systems
• Grantee: Gabriela Nunes Lopes
• Support Opportunities: Scholarships in Brazil - Master