Please use this identifier to cite or link to this item: http://repositorio.ufla.br/jspui/handle/1/56067
Title: Técnicas de inteligência computacional para estimação de sinais de oscilação de eixo em hidrogeradores
Other Titles: Computational intelligence techniques for estimation of shaft oscillation signs in hydrogenerators
Authors: Pereira, Daniel Augusto
Ferreira, Danton Diego
Lamin Filho, Paulo Cezar Monteiro
Pereira, Daniel Augusto
Keywords: Hidrogeradores
Vibrações
Inteligência computacional
Regressão linear
Regressão não linear
Redes neurais
Hydro generator
Vibrations
Computational intelligence
Linear regression
Nonlinear regression
Neural networks
Issue Date: 28-Feb-2023
Publisher: Universidade Federal de Lavras
Citation: OLIVEIRA, B. M. R. de. Técnicas de inteligência computacional para estimação de sinais de oscilação de eixo em hidrogeradores. 2022. 111 p. Dissertação (Mestrado em Engenharia de Sistemas de Automação)–Universidade Federal de Lavras, Lavras, 2023.
Abstract: This work presents computational intelligence techniques for hydro generator axis oscillation regression and comparison with methods through statistical analysis. With the continuous increase in demand for electricity in Brazil and the search for energy producers to increase availability, the reduction of "unplanned outages" resulting from failures was proposed. The failure of mechanical components in relation to the bearing of the hydro generator groups, which can be identified by the increase in "axis oscillation" levels in relation to the bearing, is of interest. The oscillation signals can be acquired (obtained and recorded) by the instrumentation installed in the generator unit and stored by the digital system of the plant. Based on standard values, it is possible to verify if the hydro generator unit will continue to operate reliably without the risk of failure and potential accidents. In this work, actual data from a generator unit located in Brazil were applied to compare computational intelligence techniques through simulations performed in the MATLAB environment and compared with each other.
URI: http://repositorio.ufla.br/jspui/handle/1/56067
Appears in Collections:Engenharia de Sistemas e automação (Dissertações)



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.