基于小波消噪的水文序列复杂性和趋势分析

Analysis of Hydrological Sequence Complexity and Trend Based on Wavelet De-noising

考虑到噪声淹没了水文序列的真实变化特性,运用小波消噪方法对水文序列进行消噪处理,然后根据符号动力学理论对水文系统的复杂性进行分析和研究。分别计算出呼伦湖流域小波消噪前后的径流序列的复杂度空间变化和动态变化结果。水文序列经过史坦无偏风险估计的软阈值方法消噪处理后,得到的消噪水文序列与原序列具有同等的光滑性和相似性。据此特点对呼伦湖流域克鲁伦河、乌尔逊河和海拉尔河进行水文序列趋势成分分析。

Considering the condition that the true change features in hydrological time series are drowned by noise,the wavelet de-noising method is used to deal with hydrological time series,and then the complexity of the hydrological system is analyzed and researched by symbolic dynamics.Both the complexity of spatial variation and time-varying in Hulun basin runoff series before and after wavelet de-noising are calculated.The hydrological sequence of de-noising,which was processed by the Stein unbiased risk estimation of soft-threshold method,had same smoothness and similarity with original hydrological time series.On the basis of the characteristics,the hydrological time series trend of Herlen,Wuerxun and Hailar River in Hulun basin is analyzed and researched.