摘要
运用自回归方法、多元线性回归方法和人工神经网络方法分别对汛期和非汛期的日径流量进行了预测,汛期预报因子又分别用有降水因子和无降水因子进行了预测。预测结果表明:非汛期的预测精度较高,汛期预测效果较差。另外,在汛期,有降水因子的预测结果要比没有降水因子预测效果好。
Autogressive method, multiple linear regression and artificial neural networks method are used to forecast the daily run-off in flood season and in low water season. In flood season, the forecasting factors are daily rainfall and no rainfall respectively.The forecasting result shows that the forecasting precision in low water season is better than that in flood season. And in flood season, the forecasting result with rainfall and prevenient daily flow factors is better than that with only. prevenient daily flow.
出处
《水科学与工程技术》
2009年第6期11-14,共4页
Water Sciences and Engineering Technology
关键词
自回归模型
多元线性回归
人工神经网络模型
径流预测
autogressive model
multiple linear regression
artificial neural networks model
runoff forecasting