Abstract
An adaptive quantum-behaved particle swarm optimizer(AQPSO) is proposed to ensure a good balance between exploration and exploitation searches of the algorithm. In the proposed algorithm, some indicators to identify the searching states of a particular particle, the whole swarm, and the iterative process are proposed and used to design an adaptively tuning mechanism for the Contraction-Expansion Coefficient. Moreover, a partially regenerated swarm scheme is designed to select promising particles and to regenerate unpromising ones when the swarm is trapped into a local optimal point. The proposed AQPSO is used to solve the TEAM workshop benchmark problem 22, and compared with existing QPSO. The numerical results have demonstrated that the proposed algorithm outperforms the existing QPSO, especially in view of the global search ability.