Nonlinear Chaotic States Prediction Based on LS-SVM

摘要

Chaos phenomena can be found in many fields. Chaos prediction has played an important role in the study of chaos system. However, it is difficult to predict chaos. Previous studies have no perfect accuracy in forecasting, furthermore previous approaches have no well learning. A method is proposed to predict the states of chaos based on the algorithm of LS-SVM (least square support vectors machine) in this study. Our approach is based on reconstruct phase space coming from the Takens embedding theorem. In this approach, the data are divided into two parts; the first part is used to train the model, another part is used as the test set. The learning model can be obtained by moving the window, whose width is n, along the axis time. The n relates to the capacity of the input points, which has the best district. Theory analysis has also been done. The results show that the method based on LS-SVM, which has better performance, can be used effectively in chaos prediction by numerical experiments.