摘要
本文根据水文现象的特性建议了两个网络模型———实时输出反馈网络（ＲＯＢＩＮ）和奢侈输出反馈网络（ＡＤＯＮＩＳ），并与水文模拟网络（ＨＹＭＮ）和传统水箱（ＴＡＮＫ）模型进行了比较．结果表明建议的两个网络模型是可行的．
In this paper, two types of neural networks for rainfall runoff modeling are suggested: real outputs back feeding to input layer network(ROBIN) and additional output neuron back feeding input layer system (ADONIS). They are compared with HYMN(a hydrologically based neural network) and TANK(a lumped conceptual water buckets model). The results show that the proposed network models can reasonably simulate daily runoff series.
出处
《水利学报》
EI
CSCD
北大核心
1998年第10期69-73,共5页
Journal of Hydraulic Engineering
基金
国家自然科学基金
关键词
神经网络
降雨
径流模拟
实时反馈
neural network,\ rainfall runoff modeling,\ real back feeding.
