摘要
地下水系统是一个复杂的随机系统。根据地下水位与其影响因素之间存在的非线性映射关系,建立了RBF网络地下水动态预测模型,并与BP网络动态预测模型相比较。实验表明,前者预测精度较高,具有一定的推广价值。
Groundwater system is a complex and random system.On the basis of the nonlinear relationship between groundwater level and its main influential factors,a model for the dynamic prediction of regional groundwater level has been set by means of a three-layer forward radial basis function artificial neural network(RBF-ANN).Afterward,this model is compared with the one by means of back propagation neural network.The result shows that the former has a higher accuracy and worth being popularized.
出处
《舰船电子工程》
2008年第8期149-151,共3页
Ship Electronic Engineering
关键词
RBF网络
地下水位
动态预测
RBF network
groundwater level
dynamic prediction
