搜索

x
中国物理学会期刊

混沌时间序列全局预测新方法——连分式法

CSTR: 32037.14.aps.54.5062

A new method of global prediction for chaotic time series based on continued fractions

CSTR: 32037.14.aps.54.5062
PDF
导出引用
  • 拓展了多项式逼近理论,利用连分式法建立了混沌时间序列非线性全局预测模型,此模型替代混沌序列的动力学方程,实现对其动力学特性分析,达到预测目的. 理论分析和仿真实验表明,连分式法能够有效预测一些混沌序列,该方法预测精度高,并且能得到显式的预测表达式.

     

    A non-linear predictable global model for chaotic time series is built based on continued fractions after developing the method of polynomial approximants in theory. We can analyze the characteristics and realize the prediction by substituting the model for the equation of chaotic time series. The results of theoretical analysis and computer simulation have proved this method to be practically feasible for predicting a chaotic series accurately and give its explicit expression.

     

    目录

    /

    返回文章
    返回
    Baidu
    map