摘要
系统地介绍了微粒群优化算法(PSO)和遗传算法(GA)的基本原理、发展和应用的状况,比较了两者的原理特点,列举了各种微粒群优化算法和遗传算法的改进算法。介绍和总结目前出现的两种算法思想结合的局部混合与全局混合两种方式,并用图表给出了说明。分析了两种混合方式的局限性,提出对具体问题找出计算速度和计算精度的平衡点来改进算法。最后做了总结和展望,指出微粒群算法的应用需进一步拓展,和其他算法结合是提高其性能的主要方向。
The basic theories,development and applications of particle swarm optimization and genetic algorithm are introduced. Some models of improved PSO algorithms are outlined. Characteristics of PSO and GA are compared. Two methods of hybrid of PSO and GA at present are summarized: global combination of the two algorithms or partial combination are illustrated with flowchart. Limitation of the two hybrid methods is analysed. It is pointed out that hybrid algorithms can be improved with a balance between speed and accuracy of computation. Finally,it is pointed out application of PSO needs to be extended, and hybrid methods with other algorithms is seen as a good way to improve PSO algorithm.
出处
《控制工程》
CSCD
2005年第S2期93-96,共4页
Control Engineering of China
关键词
微粒群优化算法
遗传算法
进化算法
混合
群智能
particle swarm optimization
genetic algorithm
evolutionary algorithm
hybrid swarm
intelligence
