基于空域相关的RBF神经网络混沌时间序列预测研究

与现有预测方法比较,神经网络在混沌时间序列预测中具有优势。利用RBF神经网络对混沌Lorenz时间序列的预测进行仿真研究,仿真结果表明：在单步直接预测、单步间接预测、多步直接预测和多步间接预测中,多步间接预测是其中最有效的方式。

Prediction of Chaotic Time Series Based on Spatial Correlation Filtering and RBF Neural Network

The significance of chaotic time series is introduced and the predictability of chaotic time series is theoretically analyzed.Then,the deficiency of the existed prediction methods and the predictability limitation of chaotic time series are discussed,and the advantages of neural network in chaotic time series prediction are shown.Meanwhile,the chaotic Lorenz time series prediction is studied by simulation in RBF neural network.The simulation shows that the multi-step indirect predication is the most effective method in the four prediction methods,such as single-step direct predication,single-step indirect predication,multi-step direct predication and multi-step indirect predication.