Analysis the Kinematics of Particle Swarm Optimization Algorithm

被引:0
作者
Li, Hongliang [1 ,2 ]
Hou, Chaozhen [1 ]
Zhou, Shaosheng [2 ]
Shao, Changbin [2 ]
机构
[1] Beijing Inst Technol, Dept Automat Control, Beijing 100081, Peoples R China
[2] Qufu Normal Univ, Inst Automat, Qufu 273165, Peoples R China
来源
2008 7TH WORLD CONGRESS ON INTELLIGENT CONTROL AND AUTOMATION, VOLS 1-23 | 2008年
关键词
Particle swarm optimization (PSO); Kinematics; Random;
D O I
10.1109/WCICA.2008.4594303
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
This paper introduced a particle optimization kinematics equation based on the simplified model of particle swarm optimization algorithm (PSO), which is a vibration equation on the balance point; gave the particle optimization kinematics principle of basic PSO algorithm; analyzed the relationship between parameters chosen and vibration angular frequency and amplitude of particle movement. Initial condition and parameters chosen decide the particle movement optimization search scope and moving characters. As to kinematics, it is concluded that particle optimization point compressed on vibration peak while sparse near the balance point according to PSO. When initial parameters are not zero, particles movement will not stop at a local minimum point.
引用
收藏
页码:8722 / +
页数:3
相关论文
共 7 条
  • [1] The particle swarm - Explosion, stability, and convergence in a multidimensional complex space
    Clerc, M
    Kennedy, J
    [J]. IEEE TRANSACTIONS ON EVOLUTIONARY COMPUTATION, 2002, 6 (01) : 58 - 73
  • [2] Eberhart RC, 2001, IEEE C EVOL COMPUTAT, P81, DOI 10.1109/CEC.2001.934374
  • [3] Kennedy J, 1995, 1995 IEEE INTERNATIONAL CONFERENCE ON NEURAL NETWORKS PROCEEDINGS, VOLS 1-6, P1942, DOI 10.1109/icnn.1995.488968
  • [4] KENNEDY J, 1998, 7 INT C EV PROGR, P581
  • [5] Liang J. J., 2004, P INT C SYST MAN CYB
  • [6] Ozcan E., 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406), P1939, DOI 10.1109/CEC.1999.785510
  • [7] Ozcan Ender, 1998, Intelligent Engineering Systems through Artificial Neural Networks, V8, P253