计算机工程与应用 ›› 2008, Vol. 44 ›› Issue (2): 80-82.

• 学术探讨 • 上一篇    下一篇

小波混沌神经网络模拟退火参数研究

徐耀群1,2,岳海燕2   

  1. 1.哈尔滨商业大学 系统工程研究所,哈尔滨 150028
    2.哈尔滨工程大学 数学系,哈尔滨 150001
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2008-01-11 发布日期:2008-01-11
  • 通讯作者: 徐耀群

Research on simulated annealing parameters in wavelet chaotic neural network

XU Yao-qun1,2,YUE Hai-yan2   

  1. 1.Institute of System Engineering,Harbin University of Commerce,Harbin 150028,China
    2.Harbin Engineering University,Harbin 150001,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2008-01-11 Published:2008-01-11
  • Contact: XU Yao-qun

摘要: 小波混沌神经网络已经成功地解决了函数优化和组合优化问题。研究了分段指数退火函数的Morlet小波混沌神经元模型,给出了分段小波混沌神经元的倒分岔图和Lyapunov指数图。在小波混沌神经网络的基础上,加入了分段指数退火函数,提出了一种新的改进的小波混沌神经网络,并把它应用到函数优化和组合优化问题中。仿真结果表明,改善了小波混沌神经网络的寻优能力,改进的小波混沌神经网络优于原来的小波混沌神经网络。

关键词: 小波混沌神经网络, Lyapunov指数, 模拟退火参数, TSP

Abstract: Wavelet chaotic neural networks have successfully solved function and combinatorial optimization problems.Morlet wavelet chaotic neural units with the annealing function of subparagraph index are studied.The reversed bifurcation and Lyapunov exponent figures are respectively given.On the basis of wavelet chaotic neural network,the annealing function of subparagraph index is introduced into network,a new reformative wavelet chaotic neural network is presented.Then it is applied to function and combinatorial optimization problems.The simulation results show that the search-optimization capacity of wavelet chaotic neural network has been improved and the reformative wavelet chaotic neural network is superior to the primary wavelet chaotic neural networks.

Key words: wavelet chaotic neural network, Lyapunov exponent, simulated annealing parameter, TSP