基于EMD方法的混沌时间序列预测

将经验模态分解(EMD)方法引入到非线性数据处理中，提出用EMD分解后的数据进行混沌预测的方法.通过Duffing方程和Lorenz系统的非线性响应预测实例表明，EMD分解后的信号和原始信号相比具有较小的最大Lyapunov指数，可提高预测时间和长时预测精度.关键词：EMD混沌预测

Prediction of chaotic time series based on EMD method

In order to improve the nonlinear response prediction precision in a long period, the empirical mode decomposition (EMD) method is introduced in the nonlinear prediction. Here, the EMD method is used to decompose the signal, the rosenstein method is used to calculate the largest Lyapunov exponent (LLE), and then the prediction results are obtained on the basis of the LLE. The simulation results of Duffing equation, Lorenz system and cracked rotor system show that the EMD's signals have smaller LLE than the original signal. In this way, the maximum prediction time of a nonlinear signal can be obtained.