摘要
为了提高粒子群优化(PSO)算法的计算精度和计算效率,避免"早熟",给出了文化粒子群优化算法.该算法模型将PSO纳入文化算法框架,组成基于PSO的主群体空间和知识空间,两空间具有各自群体并独立并行演化.下层主群体空间定期贡献精英个体给上层知识空间,上层知识空间经演化后,定期贡献精英个体给下层主群体空间,于是形成"双演化双促进"机制,从而实现增加PSO的群体多样性.在以卫星舱和印刷电路板布局设计为背景的算例中进行了数值验证,结果表明对于该算例,该方法的计算精度和计算效率比遗传算法、PSO算法高.
A cultural-based particle swarm optimization （CBPSO） algorithm is proposed to improve the computational accuracy and efficiency of PSO and avoid premature. This algorithm model consists of a PSO-based main population space and a knowledge space, which respectively has its own population to evolve independently and parallel. The lower level main population space （PSO population） contributes elite individuals to the upper level space （knowledge population） periodically, and the upper level space continually evolves these elite individuals and then contributes elite individuals to the lower level space. The mechanism of dual evolution and dual promotion improves the population diversity, and avoids premature. Two examples originated from the layout design of satellite module and integrated circuit show that CBPSO exhibits better computational efficiency and accuracy than GA and PSO.
出处
《大连理工大学学报》
EI
CAS
CSCD
北大核心
2007年第4期539-544,共6页
Journal of Dalian University of Technology
基金
国家自然科学基金资助项目(503350405057503160674078)
国家"八六三"计划资助项目(2006AA04Z109)
关键词
演化计算
粒子群优化
文化算法
布局设计
evolutionary computation
particle swarm optimization
cultural algorithm
layout design
