Banca de QUALIFICAÇÃO: DAVI MOREIRA SILVA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
DISCENTE : DAVI MOREIRA SILVA
DATA : 09/12/2025
HORA: 13:30
LOCAL: https://meet.google.com/btu-dnsz-sji
TÍTULO:

Analysis and diagnosis of SF6 high-voltage circuit breakers through a monitoring system using signal processing techniques and intelligent systems.

PALAVRAS-CHAVES:

Monitoring system; Feature extraction; SF6 high-voltage circuit breakers; Signal processing; Intelligent diagnostics;

PÁGINAS: 63
GRANDE ÁREA: Engenharias
ÁREA: Engenharia Elétrica
SUBÁREA: Sistemas Elétricos de Potência
RESUMO:

The circuit breaker is one of the main components of the Power System (PS), playing a fundamental role in both routine operational maneuvers of substations and in protecting the system in fault situations, by automatically isolating the faulty section of the circuit. Therefore, predictive maintenance of circuit breakers plays a key role in ensuring the availability, continuity, and reliability of substations, contributing to greater predictability of circuit breaker interruptions throughout their lifespan. Thus, continuous monitoring of various circuit breaker interruptions becomes essential for proactively diagnosing their condition. With the storage, analysis, and processing of this monitoring data, it is possible to anticipate failures, resulting in economic, operational, and strategic gains. In this context, this work presents preliminary results from the analysis of a dataset found in an SF6 Online Circuit Breaker Monitoring System (SIMOD) developed in a research and innovation project (R&D) ANEEL PD-00394-2119/2021 between the Federal University of Itajubá and the company Eletrobras. More specifically, data from a laboratory-developed SIMOD prototype implemented in a 345kV SF6 switchgear at the Eletrobras Circuit Breaker Workshop in São José da Barra, Minas Gerais, Brazil, were found. Based on various laboratory tests conducted under different operating conditions and simulated fault scenarios, correlations between the extracted variables are investigated to identify profiles that can aid in diagnosing the health of circuit breakers. This data will serve as the basis for training Artificial Intelligence (AI) models, which will later be applied to real-time diagnosis of circuit breakers in the prototypes under development in this project.

MEMBROS DA BANCA:
Presidente - 2132806 - AURELIO LUIZ MAGALHAES COELHO
Interno - 2362923 - GIOVANI BERNARDES VITOR
Interno - ***.230.456-** - LUIZ CARLOS BAMBIRRA TORRES - UFOP
Externo ao Programa - 2225499 - IVAN PAULO DE FARIA - UNIFEI