泉水月流量预报模型研究  

Study of the Forecasting Models for Monthly Discharge

作　　者：李力[1] 沈冰[1] 张晓伟[1] 

机构地区：[1]西安理工大学西北水资源与环境生态教育部重点实验室

出　　处：《西安理工大学学报》2008年第1期43-46,共4页Journal of Xi'an University of Technology

基　　金：国家自然科学基金资助项目(50779052,50579063)

摘　　要：泉水系统属非线性动力系统,泉水月流量序列具有季节波动的特点。借助季节变动模型建模思想,采用灰色趋势预测模型与神经网络模型,分别对其长期变动趋势与季节变动的非线性进行估计,经实例验证,表明提出的灰色季节神经网络耦合模型,其拟合、预测精度均优于独立应用灰色季节模型或神经网络模型的精度。The spring water system is a nonlinear dynamic system and its monthly discharge is characterized by seasonal fluctuations. With the help of the model establishing thought of seasonal fluctuations, the gray tendency predicting model and the neural network model are used to carry out the estimation of the long-term varying tendency and seasonal variation non-linearity individually. It is proved via the experiments that the fitness and prediction accuracy of the proposed gray seasonal neural netwok wupling model are superior to the independent application of gray seasonal model or neural network modec.

关 键 词：泉水 月流量 季节波动 灰色模型 神经网络 耦合 

分 类 号：TV121[水利工程—水文学及水资源]