• 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
 
 
Habilitation Thesis
DOI
10.11606/T.59.2013.tde-03062013-083436
Document
Author
Full name
Renato Tinós
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
Ribeirão Preto, 2012
Committee
Morales, Eduardo Alex Hernandez (President)
Liang, Zhao
Rezende, Solange Oliveira
Ruiz, Evandro Eduardo Seron
Zuben, Fernando José von
Title in Portuguese
Computação Evolutiva em Ambientes Dinâmicos
Keywords in Portuguese
Algoritmos Genéticos
Computação Evolutiva
Problemas de Otimização Dinâmica
Abstract in Portuguese
Algoritmos Evolutivos (AEs) são meta-heurísticas populacionais inspiradas em princípios básicos da evolução natural e de outros paradigmas biológicos. Diversos aplicações de AEs ocorrem em ambientes dinâmicos, nos quais a função de avaliação, as variáveis de decisão e/ou as restrições do problema mudam durante o processo de otimização. Em tais problemas, AEs tradicionais geralmente não apresentam desempenho satisfatório. Esta tese trata do problema do uso de Computação Evolutiva em ambientes dinâmicos sob diversos ângulos. Na primeira parte do trabalho, uma visão geral sobre o problema tratado é apresentada. As duas partes seguintes desta tese aprofundam os temas estudados, trazendo com detalhes exemplos de técnicas práticas e teóricas, todas propostas pelo autor desta tese. Três AEs especialmente desenvolvidos para ambientes dinâmicos são apresentados: o Algoritmo Genético com Taxa de Mutação Dependente do Gene, o Algoritmo Genético com Imigrantes Aleatórios Auto-Organizado; e os Algoritmos Evolutivos com Mutação q-Gaussiana. Com relação aos aspectos teóricos do problema estudado, é apresentada, entre outras, a análise de Algoritmos Genéticos pelo enfoque dos sistemas dinâmicos.
Title in English
Evolutionary Computation in Dynamic Environments
Keywords in English
Dynamic Optimization Problems
Evolutionary Computation
Genetic Algorithms
Abstract in English
Evolutionary Algorithms (EAs) are population meta-heuristics inspired by the principles of natural evolution and other biological paradigms. Many real world problems occur in dynamic environments, where the evaluation function, decision variables and/or the constraints of the problem change during the optimization process. When the optimization problem changes during the evolutionary process, traditional EAs generally do not present good performance. This thesis addresses the problem of the use of Evolutionary Computation in dynamic environments from different aspects. In the first part of the work, an overview of the problem addressed is presented. The next two parts present in details examples of practical and theoretical techniques for the investigated topic, all proposed by the author of this thesis. Three EAs specially developed for dynamic environments are presented: the Genetic Algorithm with Gene Dependent Mutation Rate, the Genetic Algorithm with Self-Organized Random Immigrants, and the Evolutionary Algorithms with q-Gaussian Mutation . Concerning the theoretical aspects of the problem, it is presented, among others, the analysis of Genetic Algorithms for dynamic environments using the dynamical systems approach.
 
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.
tese.pdf (1.27 Mbytes)
Publishing Date
2013-06-03
 
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 © 2001-2020. All rights reserved.