摘要
在人工神经网络水文模型的研究中,往往加入前期径流以提高模型的预报精度.针对由此带来的问题,通过耦合总径流线性响应模型,建立一种基于人工神经网络的实时预报模型.通过引入总径流线性响应模型的模拟径流作为模型输入,模型的模拟模式能够提供较长的预见期,同时加入误差校正模型的实时预报模式也能够取得较高的模型精度.采用3个不同流域的流量资料对模型进行率定与校核.结果表明,模型能够取得较高的预报精度,显示了良好的适用性.
Current artificial neural network (ANN) models can obtain high accuracy in flood simulation and forecasting with short effective real-time; therefore they can't be used in operational flood forecasting. A modified ANN model is proposed and developed by using the output of the total runoff linear response (TLR) model as the model input. The data from three different catchments are selected to test and compare the models. The results show that the proposed model not only can obtain high accuracy in flood forecasting, but also has longer effective real-time. The modified ANN model can be used in operational flood forecasting.
出处
《武汉大学学报(工学版)》
CAS
CSCD
北大核心
2007年第1期33-36,41,共5页
Engineering Journal of Wuhan University
基金
国家自然科学基金(编号:50409008)
国家"973"前期专项(编号:2003CCA00200)资助
关键词
水文模型
洪水预报
总径流线性响应模型
人工神经网络模型
hydrological model
flood forecasting
total runoff linear response model
artificial neural network model