摘要
运用Mallat算法和Daubechies小波,对水文序列进行离散小波变换。通过离散小波变换,将水文序列分解成不同时间尺度的确定性序列和随机序列,为运用各种确定性模型和随机模型建立水文中长期耦合预报模型打下了基础。以长江寸滩站日平均流量和北碚站7月最大洪峰流量序列为例,进行了小波变换。通过对分解后的序列进行重构表明,结果是满意的。
The hydrological series, which is a complex dynamic process, contains certainty and uncertainty components. This paper introduced the ways for hydrological series decomposition based on discrete wavelet transform of Daubechies wavelet and Mallat algorithm. The hydrological series was decomposed into ascertain series and random series at multiple-time scales. It laid a foundation for establishment of combined model for medium-and-long term hydrological forecasting. This paper took the mean daily flow series at Cuntan Station on the Yangtze River and the maximum flow series in July at Beibei Station on the Jialing River as examples for wavelet decomposition and rebuilding. The results of rebuilding the series decomposed were satisfying.
出处
《中国农村水利水电》
北大核心
2007年第2期106-108,共3页
China Rural Water and Hydropower
基金
973国家重点基础研究规划项目(2003CB415205)。
