Improving Particle Swarm Optimization by hybridization of stochastic search heuristics and Self-Organized Criticality

The objective of this thesis is to investigate how to improve Particle Swarm Optimization by hybridization of stochastic search heuristics and by a Self-Organized Criticality extension. The thesis will describe two hybrid models extending Particle Swarm Optimization with two aspects from Evolutionary Algorithms, recombination via breeding and gene flow restriction via subpopulations. A further hybrid combining Particle Swarm Optimization with Genetic Algorithms and Hill-Climbing using an individual based Life-Cycle stage transition scheme will also be investigated. The thesis will also describe how to improve the search of Particle Swarm Optimization using Self-Organized Criticality to add diversity and control parameters. The results presented in this thesis show the potential of Particle Swarm Optimization hybrids and extensions. Hybridizing the Particle Swarm Optimization with a Genetic Algorithm inspired crossover operator increased the reliability in multimodal optimization but also enables the incorporation of domain knowledge, previously not possible in Particle Swarm Optimization. Generally Particle Swarm Optimization has a good convergence and with the Self-Organized Criticality extension described the convergence is even faster without compromising an as good final result. With faster convergence one can reach the precision requirement of a given search problem much faster reducing the number of objective function evaluations. Problem dependent performance is a major drawback in most stochastic search algorithms. Hybridizing Particle Swarm Optimization, Genetic Algorithms and Hill-Climbing using a simple self-adaptive transition scheme diminishes this drawback.