基于驱动-制约回归模型的区域需水预测研究  

Research on regional water demand forecast based on driven-inhibition regression model

作　　者：陈兴科[1,2] 

机构地区：[1]贵州大学资源与环境工程学院,贵州贵阳550003 [2]贵州省水利投资有限责任公司,贵州贵阳550002

出　　处：《人民长江》2013年第13期10-13,38,共5页Yangtze River

摘　　要：区域需水预测方法较多,但大部分预测方法均存在各种缺陷,已有研究成果也都不能适应最严格水资源管理制度的新要求。提出了基于计量经济学至上而下建模思维的驱动-制约(Drive-inhibition)回归模型,进行大尺度区域需水预测。该模型考虑了需水驱动侧、制约侧解释变量,包含了农田灌溉水有效利用系数、万元工业增加值用水量下降百分比两项用水效率控制指标,更加符合最严格水资源管理制度的新要求。以该方法预测贵州省各水平年的需水量,与国办发[2013]2号文件要求的用水总量控制指标相比,误差均在1%以内,预测结果精确度高,可以为大尺度区域需水预测提供参考。There are several methods for regional water demand forecast,but most of them have various defects and the existed research achievements can not meet the new requirements of the most stringent water management system.We introduce the driven-inhibition regression model based on the top-down thinking in econometrics,and use it to conduct water demand forecast in large-scaled area.The model takes into account explanatory variables of the water demand driver-inhibition factors and contains two control indexes of water-use efficiency,namely effective utilization coefficient of agricultural irrigation water and water-consumption decline percentage of industrial added value per 10 thousand Yuan,so it meets the new demand of the most stringent water management system.The error between the predicted water demand in Guizhou Province by this forecast method and the standard of the total amount control in the No.2 Document 2013 of the National Development Office is within 1%.The forecast result is more accurate and can be provided as a reference for regional water demand forecast.

关 键 词：驱动-制约回归模型 需水预测 最严格水资源管理制度 计量经济学 

分 类 号：TV213.4[水利工程—水文学及水资源]