基于残差修正GM(1,1)灰色动态模型群的城市用水量预测研究  

Prediction of Urban Water Consumption Based on Residual Modified GM(1,1)Grey Dynamic Model Group

作　　者：李振 张健 LI Zhen;ZHANG Jian(Jining Hydrological Center,Jining 272000,China)

机构地区：[1]济宁市水文中心,山东济宁272000

出　　处：《河南科技》2024年第22期90-94,共5页Henan Science and Technology

基　　金：济宁市重点研发计划软科学项目(2023JNZC010)。

摘　　要：[目的]探究残差修正GM(1,1)灰色动态模型群在城市用水量预测中的应用。[方法]在传统单一GM(1,1)灰色模型的基础上,采用对数曲线对预测结果进行残差拟合修正,分别构建传统单一GM(1,1)灰色模型和残差修正GM(1,1)灰色动态模型群,利用济宁市2013-2022年实际城市用水量数据进行实例验证和预测分析。[结果]相对于传统单一GM(1,1)灰色模型,残差修正GM(1,1)灰色动态模型群拟合精度和预测精度均有明显提升,传统单一GM(1,1)灰色模型拟合相对误差为0.22%~4.22%,预测相对误差为0.91%~2.74%,预测平均相对误差为1.62%,预测均方误差为0.143,残差修正GM(1,1)0灰色动态模型群拟合相对误差在0.02%~3.09%,预测相对误差为0.75%~2.42%,预测平均相对误差为1.32%,预测均方误差为0.102。[结论]济宁市未来用水量呈下降趋势,这与济宁市严格落实最严格水资源管理制度、刚性约束用水相关指标、用水效率得到提升的情况吻合。残差修正GM(1,1)灰色动态模型群能够满足城市用水量预测。[Purposes]This paper aims to explore the application of residual modified GM(1,1)grey dy-namic model group in urban water consumption prediction.[Methods]On the basis of the traditional single GM(1,1)gray model,the logarithmic curve was used to fit and correct the residual results,and the traditional single GM(1,1)gray model and the residual modified GM(1,1)gray dynamic model group were constructed,the actual urban water consumption data of Jining City from 2013 to 2022 were used for case verification and prediction analysis.[Findings]Compared with the traditional single GM(1,1)gray model,the fitting accuracy and prediction accuracy of the traditional single GM(1,1)gray model group were significantly improved,and the fitting relative error of the traditional single GM(1,1)gray model was 0.22%~4.22%,the prediction relative error was 0.91%~2.74%,the prediction average relative error was 1.62%,the prediction mean square error was 0.143,and the residual modified GM(1,1)0 gray dynamic model group fitting relative error was 0.02%~3.09%,the relative error of prediction is 0.75%~2.42%,the average relative error of prediction is 1.32%,and the mean square error of prediction is 0.102.[Con-clusions]Through the prediction results of Jining′s water consumption from 2023 to 2030,it can be seen that Jining′s future water consumption is on a downward trend,which is consistent with Jining′s strict implementation of the strictest water resources management system,rigid constraints on water-related in-dicators,and the improvement of water use efficiency.And the residual modified GM(1,1)gray dynamic model group can meet the prediction of urban water consumption.

关 键 词：残差修正 灰色动态模型群 用水量预测 济宁市 

分 类 号：TV213.9[水利工程—水文学及水资源]