Algoritmo VMO continuo con topologías estructuradas
Fecha
2014-07-08
Autores
Lugones Acosta, Román Eduardo
Título de la revista
ISSN de la revista
Título del volumen
Editor
Universidad Central “Marta Abreu” de Las Villas
Resumen
Las meta-heurísticas poblacionales constituyen, en la actualidad, una poderosa herramienta para la solución de problemas complejos de optimización. Investigaciones han revelado que no existe la mejor de las meta-heurísticas para solucionar cualquier problema complejo, por lo que la búsqueda de nuevas estrategias en la exploración del espacio de búsqueda para una mejor calidad de las soluciones se ha convertido en un objetivo deseado. De manera que la presente investigación estará encaminada a implementar y validar un modelo basado en una meta heurística poblacional, VMO, la cual será implementada siguiendo una topología estructurada, con el objetivo de evitar estados de convergencia prematura o estancamiento. Las topologías a usar son: en un primer algoritmo, topologías con k-subpoblaciones sin solapamiento; mientras que en la segunda etapa topologías con k-subpoblaciones con solapamiento. Ambas con el objetivo de presentar un modelo más robusto. Los resultados de esta investigación son la obtención de distintas variantes VMO_TOP1 y VMO_TOP2 con mejor comportamiento que el modelo basado en VMO clásico.
Population meta-heuristics are currently a powerful tool in the solution of complex optimization problems. Some investigations have shown that there is not a best of meta-heuristics for solving any complex problem. Therefore seeking for new strategies in the exploration of the search space in order to achieve a better quality in the solutions has become a desired goal. So this investigation will be focused in the implementation and validation of a model based in a population meta-heuristic, VMO, which will be implemented following a structured topology aiming to avoid premature convergence or stemming states. The used topologies are: in the first algorithm topologies with k-subpopulations without overlapping. While in a second stage we use topologies with k-subpopulations with overlapping. Both are used with the goal of presenting a more robust model. The results of this investigation allowed to obtain distinct variants VMO_TOP1 and VMO_TOP2 with a better performance than the classic VMO based model.
Population meta-heuristics are currently a powerful tool in the solution of complex optimization problems. Some investigations have shown that there is not a best of meta-heuristics for solving any complex problem. Therefore seeking for new strategies in the exploration of the search space in order to achieve a better quality in the solutions has become a desired goal. So this investigation will be focused in the implementation and validation of a model based in a population meta-heuristic, VMO, which will be implemented following a structured topology aiming to avoid premature convergence or stemming states. The used topologies are: in the first algorithm topologies with k-subpopulations without overlapping. While in a second stage we use topologies with k-subpopulations with overlapping. Both are used with the goal of presenting a more robust model. The results of this investigation allowed to obtain distinct variants VMO_TOP1 and VMO_TOP2 with a better performance than the classic VMO based model.
Descripción
Palabras clave
Algoritmos Poblacionales, Topología Estructurada, Modelo VMO