变权组合预测模型在兴平市地下水动态预测中的应用  

Application of Changeable Weight Combination Forecasting Model in Predicting Xingping City's Groundwater

作　　者：张博炜[1] 刘俊民[1] 张斌[1] 吕玉娟[1] 

机构地区：[1]西北农林科技大学水利与建筑工程学院,陕西杨凌712100

出　　处：《中国农村水利水电》2012年第12期42-44,48,共4页China Rural Water and Hydropower

基　　金：国家科技支撑计划项目(2006BAD11B05)

摘　　要：针对地下水埋深变化离散性程度较大的兴平市,利用兴化漏斗区2000-2011年的地下水位埋深数据,采用基于指数预测法、线性回归预测法及灰色预测法的变权组合预测方法,对其进行地下水位埋深的模拟和预测。通过对兴平市地下水动态的预测,对比变权组合预测方法和单纯运用某一种单项预测方法在精度上的差别,证明了变权组合预测模型具有可行性和较高的精度,并在此基础上对兴平市未来地下水埋深进行了预测。In order to analyze the high dispersion degree of groundwater depth in Xingping City, changeable weight combination of index forecasting, linear regression prediction and grey prediction methods were introduced to predict the groundwater depth based on the groundwater depth data of Xingping funnel-shaped zone from 2000 to 2011. The predictive accuracy of single prediction method is compared with that of the changeable weight combination method. The results show that the combined method has a better predictive accuracy to predict the future groundwater depth in Xingping. This is a good reference for the application of the changeable weight combination forecasting model.

关 键 词：组合预测 变权 地下水位 地下水 

分 类 号：P641.74[天文地球—地质矿产勘探]