摘要
为了预报黄河内蒙古段的流凌和封开河日期,以多元线性回归理论与人工神经网络理论为基础,设计开发了黄河内蒙古段冰情预报系统预报内蒙古段各个站点的流凌和封开河日期。选用黄河内蒙古段水文、气象和冰情信息建立数据库,应用ADO OLEDB技术调用数据库,建立符合相关性要求的人工神经网络模型和多元线性回归模型,从而实现预报识别检验。使用2003—2004年度至2007—2008年度5年的资料进行预报检验,并与实际开河日期对比分析,预报结果合格率为80%和67.7%,分别达到甲等和丙等预报精度,表明系统可以应用于黄河内蒙古段的冰情预报。
In order to forecast the date of ice-run, freeze-up and break-up, the forecasting system of ice conditions in Inner Mongolia section of the Yellow River has been designed and developed at every station based on the theory of multiple linear regression and artificial neural network theory. After selecting hydrological, meteorological and ice information the system has established a database, which can be called by applying ADO OLEDB technology. Additionally, the models of artificial neural network and multiple linear regression are used to realize the prediction, identification, and inspection. The overall results based on the data of 5 years from 2003-2004 to 2007-2008 employed in the prediction test, the qualified rate of the forecast results is respectively 80% and 67.7%. These, respectively reaches the first and third class forecasting scheme, indicating that the system can be applied to forecast the date of ice-run, freeze-up and break-up in Inner Mongolia section of the Yellow River.
出处
《水利水电科技进展》
CSCD
北大核心
2014年第4期62-65,81,共5页
Advances in Science and Technology of Water Resources
基金
国家自然科学基金(51369017
51369021)
内蒙古水利厅科技计划(201007)
关键词
冰情预报系统
人工神经网络
多元线性回归
黄河内蒙古段
forecasting system of ice condition
artificial neural network
multiple linear regression
Inner Mongolia reach of Yellow River
