基于改进RBF网络模型的中长期水文预报  

Mid and long term hydrological forecast based on improved RBF network model

作　　者：左亚会[1] 徐敏月 韩阳[1] 

机构地区：[1]江苏省灌溉总渠管理处,江苏淮安223200

出　　处：《江苏水利》2017年第2期9-11,16,共4页Jiangsu Water Resources

摘　　要：针对历史径流资料中各预报因子及预报年份关联度,在分析预报年份非汛期中各月份径流量预报因子基础上,进行较大关联度代表年份筛选。同时,基于Matlab软件构建融合灰色关联度的RBF网络预报模型,探索预报目标年汛期径流量,采用改进RBF网络模型实施汛期径流总量预报具有较强适用性与准确度。Concerning about the correlation between forecast factor and forecast year of historical runoff data, on the basis of analyzing runoff forecast factors of each month during non-flood season, larger correlation representative years have been filtrated. In addition, RBF network model integrating grey relational degree has been established based on Matlab software. Runoff forecasting method of target year during flood season has been explored. The improved RBF network model has strong applicability and accuracy in runoff volume forecasting during flood season.

关 键 词：水文预报 灰色关联度 RBF网络模型 

分 类 号：TV124[水利工程—水文学及水资源]