martes, 29 de septiembre de 2009

GLOBAL AND LOCAL SELECTION IN DIFFERENTIAL EVOLUTION FOR CONTRAINED NUMERICAL OPTIMIZATION

Global and Local Selection in Differential Evolution for Constrained Numerical Optimization
Efrén Mezura-Montes and Carlos A. Monterrosa López


emezura@lania.mx
cmonterrosa@gmail.com

Laboratorio Nacional de Informática Avanzada (LANIA A.C.) Rébsamen 80, Centro, Xalapa, Veracruz, 91000

Instituto Tecnológico de Tuxtla Gutiérrez
Depto. de Ingeniería en Sistemas Computacionales
Carretera Panamericana Km. 1080, Tuxtla Gutiérrez, Chiapas



ABSTRACT

The performance of two selection mechanisms used in the most popular variant of differential evolution, known as DE/rand/1/bin, are com- pared in the solution of constrained numerical op- timization problems. Four performance measures proposed in the specialized literature are used to analyze the capabilities of each selection mech- anism to reach the feasible region of the search space, to find the vicinity of the feasible global op- timum and the computational cost (measured by the number of evaluations) required. Two para- meters of the differential evolution algorithm are varied to determine the most convenient values. A set of problems with different features is cho- sen to test both selection mechanisms and some findings are extracted from the results obtained.

Keywords: Constrained Numerical Optimiza- tion, Differential Evolution, Selection Mecha- nisms.


THE PAPER WAS PUBLISHED AT "Journal of Computer Science & Technology", in October 2009
AND YOU CAN DOWNLOAD HERE

No hay comentarios: