高级检索

    庄敏慧, 王振雷, 钱锋. 基于正弦载波的粒子群算法[J]. 华东理工大学学报(自然科学版), 2009, (3): 428-434.
    引用本文: 庄敏慧, 王振雷, 钱锋. 基于正弦载波的粒子群算法[J]. 华东理工大学学报(自然科学版), 2009, (3): 428-434.
    Particle Swarm Optimization Based on the Sine CarrierWave[J]. Journal of East China University of Science and Technology, 2009, (3): 428-434.
    Citation: Particle Swarm Optimization Based on the Sine CarrierWave[J]. Journal of East China University of Science and Technology, 2009, (3): 428-434.

    基于正弦载波的粒子群算法

    Particle Swarm Optimization Based on the Sine CarrierWave

    • 摘要: 针对传统粒子群算法(Traditional Particle Swarm Optimization, TPSO)存在的易陷入局部最优、收敛速度慢等缺点,提出了一种基于载波的粒子群算法(carrier-wave Particle Swarm Optimization,CWPSO)。根据正弦函数具有的自变量连续变化而值域不变的特点,该算法设计了以载波自变量变化确定粒子搜索位置的新方法,从而极大地提高了全局搜索能力。同时对于搜索到的可能极值点,通过载波扩展的方法进行局部寻优,以进行精确搜索。对一系列测试函数的寻优结果表明:CWPSO算法不仅都能找到最优值,且寻优时间仅为TPSO算法和惯性权值线性下降的改进PSO算法(Line-WPSO,LWPSO)的1/3~1/5;同时,CWPSO具有对寻优问题维数不敏感的优点,大大扩展了该算法的适用范围。

       

      Abstract: The traditional PSO algorithm(TPSO) is easily trapped in the local optimum and converges slowly. Coping with the above shortcomings, a novel PSO algorithm based on the carrierwave, CarrierWave PSO (CWPSO) is presented in this paper. By means of the sine wave's character that the value domain is steady while the independent variable is continuously changing, the CWPSO algorithm uses the changing independent variables of every carrierwaves to fix the positions of every particle so as to obviously improve the global searching ability. At the same time, for the possible extreme points, a precise search process based on the carrierwave extending is taken. By the simulation on a series of benchmark functions, it is shown that the CWPSO algorithm can not only find all the optimum values, but also spends only about 1/3 to 1/5 times as the TPSO and LWPSO algorithms. Furthermore, the CWPSO algorithm is not sensitive to the dimensions of the optimumsearching problems such that the applied field is greatly enlarged.

       

    /

    返回文章
    返回