观测数据分析中几种方法的探讨(三)混沌时间序列的预测  

DISCUSSION ON SEVERAL NEW METHODS FOR ANALYZING OBSERVATIONAL DATA OF DAMS Part Three Predication of Chaotic Time Series

作　　者：汪树玉[1] 刘国华[1] 包志仁[1] 杜王盖[1] 

机构地区：[1]浙江大学水工结构及水环境研究所,浙江省杭州市310027

出　　处：《水电自动化与大坝监测》2003年第4期45-47,63,共4页HYDROPOWER AUTOMATION AND DAM MONITORING

摘　　要：介绍了混沌时间序列分析中根据单个时间序列重建相空间 ,用关联维方法和李雅普诺夫指数来辨识观测数据中存在混沌的可能性 ,以及全局和局部预测方法。This paper introduces the reconstruction of phase space in one time series,the identification of possible existence of chaos by the correlation dimension method and Lyapunov exponent,and the global and local method of predication of chaotic time series. These concepts and approaches will significantly influence dam monitoring and nonlinear predication.

关 键 词：大坝观测 混沌现象 重建相空间 关联维法 李雅普诺夫指数 非线性预测 

分 类 号：TV698[水利工程—水利水电工程] O212[理学—概率论与数理统计]