Species-based Quantum-behaved particle swarm for multi-peaks function optimization

This paper proposes an improved Quantum-behaved Particle Swarm Optimization using the notion of species for solving multi-peaks functions optimization problems. In the proposed Species-based QPSO (SQPSO), the swarm population is divided into paralleled species subpopulations based on their similarity. Each species is grouped around a dominating particle called the species seed. Over successive iterations, species are able to simultaneously optimize towards multiple optima by using QPSO, so each peaks are ensure to be searched equally, regardless of if they are global or local optima. Our experiments demonstrate that SQPSO can search peaks of function as much as possible with high accuracy and efficiency.