摘要： 提出一种新的双子群粒子群优化(PSO)算法。充分利用搜索域内的有效信息，通过2组搜索方向相反的主、辅子群之间的相互协同，扩大搜索范围。在不增加粒子群规模的前提下，提高解高维最优化问题的精度，降低粒子群优化算法陷入局部最优点的风险。3种典型函数的仿真结果及与2种经典PSO算法的比较结果验证了该算法的有效性。

关键词: 粒子群优化, 子群, 收敛性

Abstract: A new two-subpopulation Particle Swarm Optimization(PSO) algorithm is proposed. The information of search space is taken full advantage in the algorithm. The search rang is extended through main subpopulation particle swarm and assistant subpopulation particle swarm, which search direction are inversed completely. Without increasing the size of particle swarm, the optimal precision of high dimension functions is improved and the risk of trapping into local optima is decreased effectively. The efficiency of the algorithm is verified by the simulation results of three benchmark functions and the comparison with two classical PSO algorithms.

Key words: Particle Swarm Optimization(PSO), subpopulation, convergence

中图分类号: 

• TP18