A mechanism based on Artificial Bee Colony to generate diversity in Particle Swarm Optimization

Particle Swarm Optimization (PSO) presents fast convergence for problems with continuous variables, but in most cases it may not balance properly exploration and exploitation behaviours. On the other hand, Artificial Bee Colony (ABC) presents an interesting capability to generate diversity when employed bees stagnate in a certain region of the search space. In this paper we put forward a mechanism based on the ABC to generate diversity when all particles of the PSO converge to a single point of the search space. Then, the swarm entities can switch between two pre-defined behaviours by using fuzzy rules depending on the diversity of the whole swarm. As the basis of our proposal, we utilize the Adaptive PSO (APSO) approach because it presents the capability to properly weight the terms of the velocity equation depending mainly on the current diversity of the entire swarm. We name our proposal ABeePSO, which was evaluated and compared to other well known swarm based approaches in all benchmark functions recently proposed in CEC 2010 for large scale optimization. Our proposal outperformed previous approaches in most of the cases.