Bioluminescent Swarm Optimization Algorithm

摘要: Evolutionary Computation (EC) is a research area of metaheuristics mainly applied to real-world optimization problems. EC inspired by biological mechanisms such as reproduction, mutation, recombination, natural selection and collective animal behavior. Two branches can be highlighted: Algorithms (EA) comprising Genetic (Goldberg, 1989), Programming (Koza, 1992), Differential Evolution (Storn & Price, 1997), Harmony Search (Geem et al., 2001), others; Swarm Intelligence (SI) Ant Colony Optimization (ACO) (Dorigo Stutzle, 2004) Particle (PSO) (Kennedy Eberhart, 2001; Poli 2007) others. The ACO metaheuristic1 the foraging behavior ants. On other hand, PSO metaheuristic2 motivated coordinated movement fish schools bird flocks. Both approaches have been successfully in vast range problems (Clerc, 2006; Dorigo 2004). In recent years, new SI algorithms were proposed. They common inspirations, bacterial (Passino, 2002), slime molds life cycle (Monismith Mayfield, 2008), various bees behaviors (Karaboga Akay, 2009), cockroaches infestation (Havens mosquitoes host-seeking (Feng bats echolocation (Yang, 2010), fireflies bioluminescense (Krishnanand Ghose, 2005; 2009; Yang, 2009). This work proposes swarm-based evolutionary approach based on bioluminescent fireflies, called Bioluminescent (BSO) algorithm. BSO uses two basic characteristics Glow-worm (GSO) algorithm proposed 2005): luciferin attractant, stochastic neighbor selection. However, goes further introducing features as: adaptive step sizing, global optimum attraction, leader movement, mass extinction. Besides, hybridized with local search techniques: unimodal sampling single-dimension perturbation. All these makes powerful for hard Experiments done analyze sensitivity control parameters. Later, extensive experiments performed using several benchmark functions high