灰色GM(1,N)自记忆神经网络组合预测模型研究  

Study on combined predictable model of Gray GM(1,N) self-memory and neural network

作　　者：段海妮[1] 沈冰[1] 莫淑红[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,陕西西安710048

出　　处：《水资源与水工程学报》2010年第5期93-95,共3页Journal of Water Resources and Water Engineering

基　　金：国家自然科学基金项目(50779052)

摘　　要：水文要素的预测具有不确定性,为了提高水文要素预测精度,将水文要素的多个相关因素在建模时加以考虑。本文通过建立考虑多个相关因素的灰色GM(1,N)自记忆模型,并与BP神经网络建立组合预测模型。利用新疆墨玉县年蒸发量实测资料,建立灰色GM(1,5)自记忆神经网络组合模型。研究表明:该组合模型拟合和预测效果较满意。Hydrological elements forecasting model has been improved continuously.The prediction of hydrological elements has uncertainty,in order to improve the prediction accuracy,we consider more related factors when building model.Gray GM（1,N） self-memory model that consider more related factors was established,and then it was combined with BP neural network model.Based on the annual potential evaporation in Moyu County,Xinjiang,the paper established gray GM（1,5） self-memory and neural network.The result showed that draught and forecast effect of the combined model was more satisfactory.

关 键 词：灰色GM(1 N) 自记忆 BP神经网络 年蒸发量 

分 类 号：P588.345[天文地球—岩石学]