Raising flood forecasting precision based on improving artificial neural network algorithm

被引：0

作者：

Yuan, XM ^{[1
]}

Li, HY ^{[1
]}

Liu, SK ^{[1
]}

Liu, HB ^{[1
]}

机构：

[1] China Inst Water Resources & Hydropower Res, IWHR, Beijing, Peoples R China

来源：

FORECASTING AND MITIGATION OF WATER-RELATED DISASTERS, THEME C, PROCEEDINGS: 21ST CENTURY: THE NEW ERA FOR HYDRAULIC RESEARCH AND ITS APPLICATIONS | 2001年

关键词：

flood forecasting; raising precision; neural network; improving algorithm;

D O I：

暂无

中图分类号：

TU [建筑科学];

学科分类号：

0813 ;

摘要：

Flood disaster is becoming more frequently day by day and the loss is always increasing. It is very important for us to adopt advanced flood forecasting technology to improve flood control decision accurately and reliably. The flood forecasting method based on artificial neural network is introduced in this paper. The improved BP neural network model is discussed here. The structural patterns and methods to-converge of neural network are mainly investigated. The basic method of neural network used in the forecasting of rainfall runoff, river flood and river system flood is studied. In the meantime, the algorithm of correcting errors of flood peak values is adopted in learning neural network models. The models are proved by the examples of forecasting rainfall runoff and Huang He water-sand process.

引用

页码：175 / 183

页数：9