• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.18.2023.tde-01122023-122445
Document
Author
Full name
Murilo Marques Pinto
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 2023
Supervisor
Committee
Flauzino, Rogério Andrade (President)
Rabelo, Ricardo de Andrade Lira
Santo, Silvio Giuseppe Di
Title in Portuguese
Análise da aplicação de técnicas de inteligência artificial no diagnóstico de máquinas elétricas
Keywords in Portuguese
diagnóstico de hidrogeradores
diagnóstico de máquinas elétricas
inteligência artificial
sistemas especialistas probabilísticos
Abstract in Portuguese
Este trabalho apresenta o desenvolvimento de um sistema especialista probabilístico para monitoramento de condição de hidrogeradores, visando apoiar as equipes de manutenção e operação desses ativos. O sistema foi construído após extenso levantamento na literatura, validado por especialistas, que correlacionou os sintomas do equipamento aos modos de falha correspondentes. Inicialmente, foram identificadas as principais técnicas utilizadas no monitoramento de hidrogeradores, bem como os defeitos relatados na literatura, analisando suas causas e modos de falha. Em seguida, o sistema inteligente de diagnóstico foi desen- volvido e testado em casos de defeitos previamente encontrados na literatura, obtendo uma alta taxa de acerto de até 91%. A aplicação desse sistema nas rotinas de manutenção de usinas hidrelétricas traz benefícios econômicos e socioambientais significativos, como a redução dos custos de falhas e manutenções, dos riscos de acidentes de trabalho e dos impactos ambientais. Para trabalhos futuros, estão previstos estudos para levantar novas probabilidades de sintomas e modos de falha não contemplados nesta versão do trabalho, bem como a validação do modelo com mais casos de falha. Além disso, está planejado o teste do sistema com dados reais de operação e manutenção de uma usina hidrelétrica, dentro do projeto Pesquisa e Desenvolvimento ANEEL PD-00622-0119/2019.
Title in English
Analysis of the application of artificial intelligence techniques in the diagnosis of electrical machines
Keywords in English
artificial intelligence
electrical machine diagnostics
hydrogenerator diagnostics
probabilistic expert systems
Abstract in English
This work presents the development of a probabilistic expert system for monitoring the condition of Hydrogenerators, aiming to support maintenance and operation teams of these assets. The system was built after an extensive literature review, validated by experts, which correlated the equipment symptoms with corresponding failure modes. Initially, the main techniques used in monitoring Hydrogenerators were identified, as well as the defects reported in the literature, analyzing their causes and failure modes. Subsequently, the intelligent diagnostic system was developed and tested on cases of defects previously found in the literature, achieving a high accuracy rate of up to 91%. The application of this system in the maintenance routines of hydroelectric power plants brings significant economic and socio-environmental benefits, such as reducing costs from failures and maintenance, minimizing workplace accident risks, and mitigating environmental impacts. For future work, studies are planned to assess new probabilities for symptoms and failure modes not addressed in this version of the work, as well as the validation of the model with more failure cases. Additionally, the testing of the system with real operational and maintenance data from a hydroelectric power plant is scheduled within the Research and Development project ANEEL PD-00622-0119/2019.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2023-12-04
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.