摘要
基于粒子分工与合作的思想,提出一种自适应粒子群优化(PSO)算法。该算法为不同的粒子分配不同的任务,对性能较好的粒子使用较大的惯性权,对性能较差的粒子采用较小的惯性权,加速系数根据惯性权自适应调整。将标准PSO算法中的全局最优位置与个体最优位置分别替换为相关个体最优位置的加权平均,更好地平衡了算法的全局与局部搜索能力,提高了算法的多样性与搜索效率。5个经典测试函数的仿真结果及与其他PSO算法的比较结果验证了该算法的有效性。
Based on the idea of specialization and cooperation, an adaptive Particle Swarm Optimization(PSO) algorithm is proposed. In the new algorithm, different particles are assigned specific tasks. Better particles are given larger inertial weights, while worse ones are given smaller inertial weights. And the particle’s acceleration coefficients are adaptively adjusted according to its inertial weight. Besides, the personal best position and global best position in standard PSO algorithm are respectively replaced by the weighted mean of some relevant personal best positions. These strategies improve the PSO algorithm at the aspects of diversity and the balance of exploration and exploitation. The efficiency of the new algorithm is verified by the simulation results of five classical test functions and the comparison with other PSO algorithms.
出处
《计算机工程》
CAS
CSCD
北大核心
2008年第7期181-183,共3页
Computer Engineering
关键词
粒子群优化
自适应参数
分工
平衡点
多样性
Particle Swarm Optimization(PSO)
adaptive parameters
specialization
equilibrium point
diversity