Computer Engineering and Applications ›› 2009, Vol. 45 ›› Issue (9): 67-70.DOI: 10.3778/j.issn.1002-8331.2009.09.019

• 研究、探讨 • Previous Articles     Next Articles

System identification based on QPSO algorithm

SHEN Jia-ning,SUN Jun,XU Wen-bo   

  1. School of Information Technology,Southern Yangtze University,Wuxi,Jiangsu 214122,China
  • Received:2008-01-24 Revised:2008-04-18 Online:2009-03-21 Published:2009-03-21
  • Contact: SHEN Jia-ning

运用QPSO算法进行系统辨识的研究

沈佳宁,孙 俊,须文波   

  1. 江南大学 信息工程学院,江苏 无锡 214122
  • 通讯作者: 沈佳宁

Abstract: An opening and practical solution,that is Quantum-behaved Particle Swarm Optimization algorithm,is applied to system identification.Not only parameters of QPSO are few and randomicity of QPSO is strong,but also QPSO covers with all the space of solution and guarantee global convergence of algorithms.Experiment results of emulator show that QPSO algorithm provides a much better effect than GA algorithm and PSO algorithm on linear system identification and nonlinear system identification.

Key words: system identification, Quantum-behaved Particle Swarm Optimization(QPSO) algorithm, linear system, nonlinear system, Hammerstein model, Wiener model

摘要: 引入了一种广泛而实用的方法——基于量子行为的粒子群算法的理论应用于系统辨识领域,QPSO算法不仅参数个数少,随机性强,并且能覆盖所有解空间,保证算法的全局收敛性。仿真实验结果表明,QPSO算法具有比GA算法及PSO算法更强的线性系统辨识能力和非线性系统辨识能力。

关键词: 系统辨识, 量子粒子群优化算法, 线性系统, 非线性系统, Hammerstein模型, Wiener模型