摘要
将径向基函数神经网络方法应用于河道洪水演算中,并利用最小二乘法来确定模型参数.结合河道洪水演算的具体方式,分别构建基于马斯京根方法和具有预见期的洪水演算方法的径向基函数神经网络模型.将该模型应用于两条天然河道的洪水演算中,计算结果表明,该模型运算快速,精度较高,具有较大的应用价值.
The radial basis functionneural network method is applied to channel flood routing. In combination with the specific modes of channel flood routing, radial basis functionartificial neural network models, based on the Muskingum method and flood routing method with a forecast period, are developed, and the least square method is used to determine the parameters of the models. The models are applied to flood routing for two natural river channels, and the results show that the models have the characteristics of fast calculation, high precision, and high value of application.
出处
《河海大学学报（自然科学版）》
CAS
CSCD
北大核心
2003年第6期621-625,共5页
Journal of Hohai University(Natural Sciences)
基金
河海大学科技创新基金资助项目(2002409743)
关键词
河道
洪水演算
径向基函数
人工神经网络
flood routing
Muskingum method
method with a forecast period
radial basis function
artificial neural network
