摘要
针对函数优化问题,提出了一种基于混沌的聚类粒子群优化算法。该算法利用混沌序列产生粒子的位置和速度,并与粒子群优化算法产生的粒子位置进行比较,选择好的粒子位置。同时通过谱系聚类方法进行聚类,并且给出新的速度更新公式。最后将算法应用到5个典型的函数优化问题中,并与其它改进的粒子群算法进行比较分析。数值结果表明,该算法提高了全局搜索能力、收敛速度和解的精度。
A clustering particle swarm optimization algorithm based on chaos （CPSOC） is proposed. In this algorithm, the chaotic sequence is used for producing the position and the speed for every particle. Through the particle＇ s position compare with the position that particle swarm optimization algorithm has produced, chooses the preferable position for every particle. At the same time the hiera- rchical clustering method is introduced and then a new formula is gived for renewal speed of the particle in the CPSOC algorithm. Finally the CPSOC algorithm is applied to five typical fimction optimization problem and is comparatively analyzed with other variant particle swarm optimization algorithm. The numerical results show that the CPSOC algorithm improves the capacity of global searching optimal solution, convergence speed and computational precision of solution.
出处
《计算机工程与设计》
CSCD
北大核心
2011年第2期685-688,735,共5页
Computer Engineering and Design
基金
国家自然科学基金项目(10871033)
辽宁省教育厅科学技术研究基金项目(2008004)
关键词
粒子群优化算法
混沌优化
聚类
谱系聚类
非线性规划
particle swarm optimization algorithm
chaotic optimization
clustering
hierarchical clustering method
nonlinear programming
