期刊导航

论文摘要

具有随机停滞行为的粒子群优化算法

A Novel Particle Swarm Optimization with Stochastic Stagnation

作者:姜海明(电子科技大学 光电信息学院,四川 成都 610054);谢康(电子科技大学 光电信息学院,四川 成都 610054);任诚(电子科技大学 光电信息学院,四川 成都 610054)

Author:(School of OptoElectronic Info., Univ. of Electronic Sci. and Technol. of China, Chengdu 610054, China);(School of OptoElectronic Info., Univ. of Electronic Sci. and Technol. of China, Chengdu 610054, China);(School of OptoElectronic Info., Univ. of Electronic Sci. and Technol. of China, Chengdu 610054, China)

收稿日期:2005-09-25          年卷(期)页码:2006,38(4):117-121

期刊名称:工程科学与技术

Journal Name:Advanced Engineering Sciences

关键字:随机停滞;粒子群优化;群智能;接受概率

Key words:stochastic stagnation; particle swarm optimization; swarm intelligence; acceptation probability

基金项目:国家杰出青年基金资助项目(60588502);电子科技大学青年科技基金资助项目(JX04028)

中文摘要

为了提高算法的优化性能,通过借鉴模拟退火算法(SA)和遗传算法(GA)的思想,在基本粒子群优化(PSO)算法的基础之上,引入了一个称为接受概率的关键参数,改写了原算法中粒子飞翔的速度公式,使粒子以一定的概率随机在解空间的某一方向上产生停滞行为,提出了一种新颖的粒子群优化方法——随机停滞粒子群优化(SSPSO)。数值计算结果表明,合理地选取接受概率的大小,该算法能在保持原算法稳定性的同时,明显提高算法的优化效率。最后,通过与传统的搜索算法、SA和GA的类比,对SSPSO的性能进行了深入分析。

英文摘要

In order to improve the capability of the algorithm, based on the analysis of basic particle swarm optimization (PSO) and inspired by some ideas of simulated annealing (SA) and genetic algorithms (GA), a novel PSO with stochastic stagnation,SSPSO, was presented, in which a key parameter,acceptation probability, was introduced and the formula for flying velocity of the particle was changed, resulting in the particles stochastically stagnating in some directions. The simulations indicated that the presented algorithm can improve the search efficiency obviously while the stabilization was maintained if only the acceptation probability was selected properly. At last, a further analysis was made contrasting with traditional search algorithm SA and GA.

关闭

Copyright © 2020四川大学期刊社 版权所有.

地址:成都市一环路南一段24号

邮编:610065