2006 | OriginalPaper | Buchkapitel
Hybrid Particle Swarm Optimization: An Examination of the Influence of Iterative Improvement Algorithms on Performance
verfasst von : Jens Gimmler, Thomas Stützle, Thomas E. Exner
Erschienen in: Ant Colony Optimization and Swarm Intelligence
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
In this article, we study hybrid Particle Swarm Optimization (PSO) algorithms for continuous optimization. The algorithms combine a PSO algorithm with either the Nelder-Mead-Simplex or Powell’s Direction-Set local search methods. Local search is applied each time the PSO part meets some convergence criterion. Our experimental results for test functions with up to 100 dimensions indicate that the usage of the iterative improvement algorithms can strongly improve PSO performance but also that the preferable choice of which local search algorithm to apply depends on the test function. The results also suggest that another main contribution of the local search is to make PSO algorithms more robust with respect to their parameter settings.