| Lyapunov指数的算法改进与加权预测 |
| |
| 引用本文： | 梁勇,孟桥,陆佶人.Lyapunov指数的算法改进与加权预测[J].声学技术,2006,25(5):463-467. |
| |
| 作者姓名： | 梁勇 孟桥 陆佶人 |
| |
| 作者单位： | 1. 南京财经大学工业工程系,南京,210003;东南大学无线电工程系,南京,210096
2. 东南大学无线电工程系,南京,210096 |
| |
| 摘 要： | 对在计算最大Lyapunov指数方面获得广泛应用的Rosenstein方法作出了改进。Rosenstein方法在计算时,只选取一个最近邻点,容易受到噪声的影响。Kantz提出用固定邻域内多点平均的方法来减少噪声的影响,但固定邻域在实际计算时存在困难。针对这个问题,提出了实用的可变邻域的取法。并且提出了将最大Lyapunov指数用于对预测模型进行加权,仿真表明预测效果得到了明显改善。
|
| 关 键 词： | 混沌 李雅普诺夫指数计算 算法改进 加权预测 |
| 文章编号： | 1000-3630（2006）-05-0463-06 |
| 收稿时间： | 2005-05-31 |
| 修稿时间： | 2005-05-312005-08-30 |
Improved calculation of Lyapunov exponent and weighted prediction |
| |
| LIANG Yong,MENG Qiao and LU Ji-ren.Improved calculation of Lyapunov exponent and weighted prediction[J].Technical Acoustics,2006,25(5):463-467. |
| |
| Authors: | LIANG Yong MENG Qiao and LU Ji-ren |
| |
| Affiliation: | Department of Industrial Engineering, Nanjing University of Finance & Economics, Nanjing 210003, China;Department of Radio Engineering, Southeast University, Nanjing 210096, China;Department of Radio Engineering, Southeast University, Nanjing 210096, China;Department of Radio Engineering, Southeast University, Nanjing 210096, China |
| |
| Abstract: | Rosenstein algorithm,widely used in calculating the maximal Lyapunov exponent,is improved.The Rosenstein algorithm selects only one neighboring point,therefore is susceptible to noise.Kantz selects several points of a fixed neighborhood to reduce the influence,resulting in some problems in practical use.A weighted method to select a variable neighborhood is given in this paper.The maximal Lyapunov exponent is used to be weights in the prediction model.Simulation indicates that the weighted prediction is more precise than the normal prediction model. |
| |
| Keywords: | chaos Lyapunov exponent weighted prediction |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
| 点击此处可从《声学技术》浏览原始摘要信息 |
|
点击此处可从《声学技术》下载全文 |
|
