月径流序列的混沌特征识别及Volterra自适应预测法的应用  

Identification of chaos of monthly runoff and prediction of runoff time series using Voterra adaptive method

作　　者：李红霞[1] 许士国[1] 范垂仁[2] 

机构地区：[1]大连理工大学土木水利学院,辽宁大连116024 [2]长春自然灾害预测研究服务中心,吉林长春130022

出　　处：《水利学报》2007年第6期760-766,共7页Journal of Hydraulic Engineering

基　　金：国家自然科学基金重点项目(50139020);国家重点基础研究计划"973"项目(2006CB403405)

摘　　要：在混沌动力系统相空间重构的基础上,利用关联维数法和最大Lyapunov指数法,对月径流时间序列进行混沌特性识别。然后结合自适应技术的实时递推特性和Volterra级数的非线性表征能力,利用Volterra自适应滤波法可对径流时间序列进行预测。通过江桥站和丰满水库实际月径流序列的预测结果表明,月径流序列中存在着一定的混沌特征。应用Volterra自适应法可以有效地对水文时间序列进行预测,与加权一阶局域预测法相比,能够实现更高精度的多步预测。Based on the reconstruction of the chaotic dynamic space, the correlation dimension method and Lyapunov exponent method are applied to identify the chaos of monthly runoff series. On this basis the Voherra adaptive method combining the updating capability of adaptive techniques with Volterra series is used to predict the monthly runoff data. The analysis on the observation from two typical hydrological stations reveals the presence of low - dimensional chaos nonlinearity of data obtained in the runoff dynamics. The comparison of the observation data with the predicted runoff shows that the Volterra adaptive method is feasible.

关 键 词：月径流序列 混沌 关联维数 最大LYAPUNOV指数 Volterra自适应预测 

分 类 号：P333[天文地球—水文科学]