中国物理B ›› 2003, Vol. 12 ›› Issue (11): 1213-1217.doi: 10.1088/1009-1963/12/11/305

• • 上一篇    下一篇

Improving the prediction of chaotic time series

陈天仑1, 李克平2, 高自友2   

  1. (1)Department of Physics, Nankai University, Tianjin 300071, China; (2)Institute of Systems Science, Northern Jiaotong University, Beijing 100044, China
  • 收稿日期:2003-03-04 修回日期:2003-06-16 出版日期:2003-11-16 发布日期:2005-03-16
  • 基金资助:
    Project supported by the Funds for Outstanding researchers from the National Natural Science Foundation of China (Grant No 70225005), and Research Award Program (2001) for Outstanding Young Teachers in Higher Education Institutions of Ministry of Educatio

Improving the prediction of chaotic time series

Li Ke-Ping (李克平)a, Gao Zi-You (高自友)a, Chen Tian-Lun (陈天仑)b   

  1. a Institute of Systems Science, Northern Jiaotong University, Beijing 100044, China; b Department of Physics, Nankai University, Tianjin 300071, China
  • Received:2003-03-04 Revised:2003-06-16 Online:2003-11-16 Published:2005-03-16
  • Supported by:
    Project supported by the Funds for Outstanding researchers from the National Natural Science Foundation of China (Grant No 70225005), and Research Award Program (2001) for Outstanding Young Teachers in Higher Education Institutions of Ministry of Educatio

PDF (PC)

483

赞

摘要： One of the features of deterministic chaos is sensitive to initial conditions. This feature limits the prediction horizons of many chaotic systems. In this paper, we propose a new prediction technique for chaotic time series. In our method, some neighbouring points of the predicted point, for which the corresponding local Lyapunov exponent is particularly large, would be discarded during estimating the local dynamics, and thus the error accumulated by the prediction algorithm is reduced. The model is tested for the convection amplitude of Lorenz systems. The simulation results indicate that the prediction technique can improve the prediction of chaotic time series.

Abstract: One of the features of deterministic chaos is sensitive to initial conditions. This feature limits the prediction horizons of many chaotic systems. In this paper, we propose a new prediction technique for chaotic time series. In our method, some neighbouring points of the predicted point, for which the corresponding local Lyapunov exponent is particularly large, would be discarded during estimating the local dynamics, and thus the error accumulated by the prediction algorithm is reduced. The model is tested for the convection amplitude of Lorenz systems. The simulation results indicate that the prediction technique can improve the prediction of chaotic time series.

Key words: chaotic time series, local Lyapunov exponent, neighbouring point, neural network

中图分类号:  (Time series analysis)

  • 05.45.Tp
05.45.Pq (Numerical simulations of chaotic systems)