One of the fundamental properties of an antenna is its ability to radiate or receive more energy in one given direction over all the others. This property is called the directivity. The optimization of the directivity usually requires a great deal of attention when an antenna is being designed. There are a number of iterative analytical and experimental procedures available for the optimization of the directivity, however they can be manually intensive, and very time consuming when computer simulation is employed. / Optimization of antenna parameters has been hindered by the need of techniques that prove to be reliable, robust and can search and find a global maximum. Past efforts have focused on techniques confined to small search volumes due to the time overheads of searching. With the increasing utility of the personal computer, techniques have emerged which can search large volumes efficiently and economically. Genetic Algorithms are one such technique. / Genetic Algorithms are an optimization technique based on the mechanics of natural selection, which combines the biological concepts of survival of the fittest among string structures. They operate on a population of candidate solutions and are able to change a number of parameters simultaneously while testing the solutions for goodness-of-fit. They also possess memory so that a good solution can be saved and tested from generation to generation. In this way they are able to quickly find and maintain the best solution to the problem. / A Genetic Algorithm is used to optimize the directivity of a linear array of dipole radiators. Mutual and self-coupling is taken into consideration through the use of the Method of Moments. The inter-element spacing as well as the radiator length are allowed to vary. This gives the optimization many degrees of freedom. The arrays so optimized are verified using a standard industrial antenna software simulation program. The optimized array achieves a directivity of approximately 1.5 dB better than published data for a uniform array of the same size. There is an overall reduction in the length of the array of one wavelength and the currents on the radiating elements are realisable. The final product is a basic computer aided design package capable of optimizing the directivity of a linear antenna array with the only inputs needed being the frequency of operation and the number of dipole elements. / Thesis (MEng(ElectronicsEngineering))--University of South Australia, 2005.

Thesis (PhD) -- University of South Australia, 1998

Thesis (PhD in Computer and Information Science)--University of South Australia, 1995

Thesis (PhD in Computer and Information Science)--University of South Australia, 1995

Thesis (PhD in Computer and Information Science)--University of South Australia, 1995

Assembly Sequence Planning (ASP) is part of Assembly Planning. The assembly sequence is the most important part of an assembly plan. Assembly has an important share in both lead time and cost of a product. Therefore, its optimisation is necessary to ensure the competitivity of manufactured goods. The aim of this thesis is the optimisation of assembly sequences for mechanical products, for real/realistic problems and constraints. This thesis represents an integrated approach in assembly sequence planning and optimisation. It tackles real problems by building the generality in the models. The ASP problem is a large scale, highly constrained, combinatorial problem, with an extraordinarily diverse character. Assembly can address sequential or non-sequential, linear or non-linear, monotone or non-monotone, coherent or non-coherent assembly plans or any combination of those, involving rigid, elastic, non elastic, solid, liquid or gaseous components or subassemblies. To be applicable in practice and useful, an assembly sequence planning and optimisation algorithm has to be general enough to accommodate any type of assembly plan and component. For this reason, modelling becomes critically important. A model has been developed for the assembly process, to determine what the assembly process is in mathematical terms. A second model has been developed to model/represent assembly plans as chromosomes that encode any type and combination of assembly plans. Another model has been developed for modelling/representing products for assembly. This model constitutes the database containing all information necessary for generating feasible assembly sequences, for any type of component and subassembly. A framework has been developed for the definition of a fitness function to assess the quality of an assembly sequence and plan from optimisation criteria. Solving the ASP problem (prior to its optimisation), implies generating a sequence to assemble an n-part product given its description and a number of supplementary constraints. A guided search algorithm has been developed to solve the ASP problem. To optimise the ASP, Genetic Algorithms (GA) were used in this research. The GA has a classic structure and modified genetic operators: it only generates and manipulates legal and feasible chromosomes. An initial population of feasible chromosomes is generated through guided search. This population, then, undergoes transformations over a number of generations, through crossover and selection. The crossover, based on the guided search algorithm, is also designed to produce only legal chromosomes. The selection is a classical operation, through a weighed roulette algorithm. It operates on an extended population of parent and children chromosomes. The output of the GA is a population of chromosomes with a high fitness value, corresponding to optimal/near optimal assembly sequences, from which the best one is selected. A number of examples are used in each chapter to illustrate each significant aspect considered. A final example illustrates the application of the whole algorithm to produce optimised assembly sequences for an industrial-size product. / thesis (PhD)--University of South Australia, 2003.

Clustering is the search for those partitions that reflect the structure of an object set. Traditional clustering algorithms search only a small sub-set of all possible clusterings (the solution space) and consequently, there is no guarantee that the solution found will be optimal. We report here on the application of Genetic Algorithms (GAs) -- stochastic search algorithms touted as effective search methods for large and complex spaces -- to the problem of clustering. GAs which have been made applicable to the problem of clustering (by adapting the representation, fitness function, and developing suitable evolutionary operators) are known as Genetic Clustering Algorithms (GCAs). There are two parts to our investigation of GCAs: first we look at clustering into a given number of clusters. The performance of GCAs on three generated data sets, analysed using 4320 differing combinations of adaptions, establishes their efficacy. Choice of adaptions and parameter settings is data set dependent, but comparison between results using generated and real data sets indicate that performance is consistent for similar data sets with the same number of objects, clusters, attributes, and a similar distribution of objects. Generally, group-number representations are better suited to the clustering problem, as are dynamic scaling, elite selection and high mutation rates. Independent generalised models fitted to the correctness and timing results for each of the generated data sets produced accurate predictions of the performance of GCAs on similar real data sets. While GCAs can be successfully adapted to clustering, and the method produces results as accurate and correct as traditional methods, our findings indicate that, given a criterion based on simple distance metrics, GCAs provide no advantages over traditional methods. Second, we investigate the potential of genetic algorithms for the more general clustering problem, where the number of clusters is unknown. We show that only simple modifications to the adapted GCAs are needed. We have developed a merging operator, which with elite selection, is employed to evolve an initial population with a large number of clusters toward better clusterings. With regards to accuracy and correctness, these GCAs are more successful than optimisation methods such as simulated annealing. However, such GCAs can become trapped in local minima in the same manner as traditional hierarchical methods. Such trapping is characterised by the situation where good (k-1)-clusterings do not result from our merge operator acting on good k-clusterings. A marked improvement in the algorithm is observed with the addition of a local heuristic.

Thesis (Ph.D.)--Tufts University, 1998. / Adviser: Steven H. Levine. Submitted to the Dept. of Engineering Design. Includes bibliographical references (leaves 152-167). Access restricted to members of the Tufts University community. Also available via the World Wide Web;

Thesis (M.S.)--Auburn University, 2007. / Abstract. Vita. Includes bibliographic references (ℓ. 52-58)

Thesis (Ph. D.)--University of Hong Kong, 2008. / Also available in print.