Please use this identifier to cite or link to this item:
Type: Thesis
Title: Theory and applications of bio-inspired algorithms.
Author: Wagner, Markus
Issue Date: 2013
School/Discipline: School of Computer Science
Abstract: Evolutionary algorithms, which form a sub-class of bio-inspired algorithms, mimic some fundamental aspects of the neo-Darwinian evolutionary process. They simultaneously search with a population of candidate solutions and associate an objective score as a fitness value for each one. The algorithms then select among the population to favour those solutions that are more fit. The next generation (i.e. a new population) consists of replicates of the fitter solutions that have been genetically mutated and crossed over in a biological metaphor: the decision variables were perturbed such that they inherit characters of their parents, as well as change in random ways. For the past decades, the algorithms’ success has led to strongly practical-oriented interests. Although the theory of them is far behind the knowledge gained from experiments, there are theoretical investigations about some of their properties. This thesis spans theoretical investigations, theory-motivated algorithm engineering, and also the real-world application of evolutionary algorithms. First, we analyse different algorithms that work with solutions of variable length. We show theoretically and experimentally that certain design choices can have drastic impacts on the ability of an algorithm to find optimal solutions. Second, motivated by recent theoretical investigations, we design a framework for solving problems with conflicting objectives. We demonstrate that it can efficiently handle problems with many such objectives, which most existing algorithms have difficulties dealing with. Finally, we consider the problem of maximising the energy yield of wind farms. Our problem-specific algorithm achieves higher quality results than existing approaches, and it allows for an optimisation within minutes or hours instead of days or weeks.
Advisor: Neumann, Frank
Michalewicz, Zbigniew
Dissertation Note: Thesis (Ph.D.) -- University of Adelaide, School of Computer Science, 2013
Keywords: optimisation; renewable energy; computational complexity
Provenance: Copyright material removed from digital thesis. See print copy in University of Adelaide Library for full text.
Appears in Collections:Research Theses

Files in This Item:
File Description SizeFormat 
01front.pdf119.25 kBAdobe PDFView/Open
02whole.pdf7.33 MBAdobe PDFView/Open
  Restricted Access
Library staff access only341.46 kBAdobe PDFView/Open
  Restricted Access
Library staff access only7 MBAdobe PDFView/Open

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