摘要
根据全信息理论,充分挖掘降雨变量语法信息和语义信息,在降雨信息传递指数基础上对淮河流域蚌埠站以上区域99个雨量站进行模糊聚类,利用降雨量贴近度指标对分类予以显著性检验,最后依据平均信息传递指数对其调整后得到最佳分类。各子区域内的降雨信息既在降雨信息的统计特征上体现出显著的相似性,又在降雨量上具有较高的贴近度,从而达到子区域内降雨信息具有较大同质性、子区域间具有较大异质性的研究目标。
The rainfall information transmission index was used for fuzzy clustering of 99 gauging stations upstream of the Bengbu Station in the Huaihe River Basin,based on the comprehensive information theory through deep exploitation of the syntax information and semantic information of rainfall variables. The rainfall proximity index was used for significance testing,and optimal clustering was obtained according to the average information transmission.The rainfall information of each sub-region showed remarkable similarity in statistical characteristics and high proximity in rainfall values,indicating significant homogeneity in each sub-region and heterogeneity between subregions.
出处
《河海大学学报（自然科学版）》
CAS
CSCD
北大核心
2013年第6期477-481,共5页
Journal of Hohai University(Natural Sciences)
基金
武汉大学水资源与水电工程科学国家重点实验室开发研究基金(2010B067)
关键词
降雨信息
区域化
信息熵
模糊聚类
全信息
rainfall information
regionalization
information entropy
fuzzy clustering
comprehensive information
