• 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.3.2022.tde-27052022-074858
Document
Author
Full name
Fidel Ernesto Díaz Andino
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2022
Supervisor
Committee
Garcia, Claudio (President)
Barreto, Guilherme de Alencar
Braz, Helon David de Macêdo
Title in Portuguese
Estudo da influência de outliers univariados na identificação de sistemas SISO.
Keywords in Portuguese
Identificação de sistemas
Robustez
Valores atípicos
Abstract in Portuguese
Dados contaminados com outliers podem ser um problema na qualidade dos modelos estimados usando técnicas de identificação de sistemas. Estudam-se, neste trabalho, as definições dos diferentes tipos de outliers que comumente afetam os dados coletados de processos. Se faz um estudo sobre a influência destes valores na identificação de sistemas SISO. São apresentados dois caminhos comuns para abordar o problema. O primeiro, é o típico método de dois passos, detecção de outliers e logo após a sua remoção. Neste caso, são resumidas algumas técnicas que podem ser usadas. O segundo caminho é baseado no uso de estimadores robustos na modelagem. Neste âmbito é feita uma comparação do uso deste tipo de estimadores, quando os dados estão contaminados e comparando com o caso de não existir contaminação.
Title in English
Study of the influence of univariated outliers on SISO system identification.
Keywords in English
Outliers
Outliers detection
Outliers removal
Robust estimators
System identification
Abstract in English
Data contaminated with outliers can be a problem in the quality of models estimated using system identification techniques. In this work, the definitions of the different types of outliers that commonly affect the data collected from processes are studied. A study is carried out on the influence of these values on the identification of SISO systems. Two common ways to approach the problem are presented. The first is the typical two-step method, outlier detection and then removal. In this case, some techniques that can be used are summarized. The second way is based on the use of robust estimators in modeling. In this context, a comparison is made of the use of this type of estimator, when the data is contaminated, and comparing it if there is no contamination.
 
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
2022-05-27
 
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.