摘要
为解决传统方法难以可靠预报雨量站分布不均匀流域的次降雨径流量这一水文预报难题 ,探讨了人工神经网络模型用于该类水文预报问题的可能性。实例研究表明 ,以次暴雨量及其前期影响雨量为输入、次暴雨径流总量 (净雨量 )为输出的 BP网络模型 ,预报的相对误差比蓄满产流模型预报的相对误差平均低 9.2 % ,这说明 ,人工神经网络模型可作为雨量站分布不均匀等雨量观测存在系统偏差或不足流域的降雨径流预报模型。
This paper presents the possibility of using artificial neural network model to forecast runoff for rain gage unevenly distributed watersheds.Case study shows that the BP network model is significantly efficient for forecasting for rain gage unevenly distributed watersheds by using the total rainfall and the previous affected rainfall as model input,and net rainfall for runoff as model output.The forecasting relative error of the BP network model has an average value 9.2% lower than that of runoff yield at natural storage model.It is shown that artificial neural network model might be used to forecast or predict stream flow when the rainfall observation exhibits systematic errors.
出处
《西北农林科技大学学报（自然科学版）》
CSCD
北大核心
2002年第5期81-84,共4页
Journal of Northwest A&F University(Natural Science Edition)
基金
西北农林科技大学青年专项基金 (0 80 8)
