Banca de QUALIFICAÇÃO: AYRTON PINHEIRO DE ARAUJO SILVA

Uma banca de QUALIFICAÇÃO de MESTRADO foi cadastrada pelo programa.
STUDENT : AYRTON PINHEIRO DE ARAUJO SILVA
DATE: 17/12/2025
TIME: 14:00
LOCAL: Sala de reuniões do DCA
TITLE:

Fault Detection in Industrial Machines using Empirical Mode Decomposition, Cyclostationary Characteristics, and 1D-CNN

KEY WORDS:

Fault detection, EMD, 1D-CNN, Cyclostationary analysis, Machine learning.

PAGES: 44
BIG AREA: Engenharias
AREA: Engenharia Elétrica
SUMMARY:

The need to detect faults in industrial machines has been growing significantly due to the increasing complexity of production processes and the greater demand for reliability and availability of these systems. This has led to advancements in research based on machine learning and deep learning, accompanied by the development of new signal processing techniques. However, one of the main challenges in this area remains the treatment of signals, especially acoustic signals, which are highly noisy – a common situation in industrial environments due to the simultaneous presence of various equipment and noise sources. In this context, this work proposes an approach that integrates Empirical Mode Decomposition (EMD) and cyclostationary feature extraction as preprocessing steps, combined with automatic detection using One-Dimensional Convolutional Neural Networks (1D-CNN). This integration aims to improve fault detection performance, offering greater sensitivity to characteristic fault modulations and increasing robustness under conditions of intense noise. 

COMMITTEE MEMBERS:
Presidente - 1543191 - LUIZ FELIPE DE QUEIROZ SILVEIRA
Interno - 2579664 - ALLAN DE MEDEIROS MARTINS
Interno - 1141792 - RODRIGO PRADO DE MEDEIROS