中文摘要：

针对种群多样性对粒子群算法的性能影响,提出了一种基于差异进化思想的粒子群算法。该算法采用多生态子群社会结构,利用一种新的全信息粒子作为信息交互的渠道,通过进化过程中的种群衰落监控指导子群间的差异融合,有利于优秀个体的产生,增加粒子间的差异性,提高种群整体品质和算法的收敛性能。最后对八个测试函数进行实验仿真,并与六个改进粒子群算法进行多方面对比。实验结果表明,该算法有效地保持了种群的多样性,在保证收敛速度的同时大幅提高了算法的收敛精度,从理论和实验仿真两个方面证明了算法有很强的全局搜索能力。

英文摘要：

As the performance of the particle swarm optimization was affected by the population diversity,this paper developed the particle swarm optimization based on the differential evolution.The new algorithm used the multi-ecological subgroups structure and presented a new full-informed particle to link the subgroups,the population decline monitor guided the ecological subgroups to differential fusion dynamically.The new algorithm was beneficial to obtain better particle,it also could increase the otherness between the particles,the whole quality and the convergence performance.Finally,it applied the new algorithm to eight test functions and compared with six extended particle swarm optimization.The result shows that the new algorithm has a potent affect on population diversity,improves the convergence accuracy with fast speed,the theory and experiment both support that the new algorithm has strong global searching ability.