基于广义回归神经网络模型的径流预测研究

Research of Runoff Prediction Based on Generalized Regression Neural Network Model

为提高径流预报精度,构建了基于广义回归神经网络模型的径流预测模型并将其应用于浙江丽水小溪流域中,对白岩测站50年月平均径流资料进行模拟研究,并与BP神经网络模拟结果进行对比。结果表明,广义回归神经网络预测精度较BP神经网络模型高,更接近径流监测实测值,平均相对误差为11.06%,且预报结果比较稳定,为径流时间序列模拟提供了一种更优的建模方法。

In order to improve the accuracy of runoff prediction,this paper studied a runoff prediction model based generalized regression neural network model（GRNN）.And the model was applied to Xiaoxi river basin in Lishui,Zhejiang and simulate 50-years＇ monthly runoff data in Baiyan Station.Compared with the results of the BP neural network,it shows that the GRNN displays higher accuracy and the prediction results are closer to the observation data with an average relative error of 11.06%.Additionally,the results appear good stability.Therefore,GRNN can be considered as a better model in runoff simulation.