基于ES残差优化的GM(1,1)模型预测城市用水量  

Prediction of Urban Water Consumption by GM(1,1) Model Based on ES Residual Optimization

作　　者：蒋白懿[1] 杲红双 孙志民 班福忱[1] JIANG Bai-yi;GAO Hong-shuang;SUN Zhi-min;BAN Fu-chen(School of Municipal and Environmental Engineering,Shenyang Jianzhu University,Shenyang 110168,China;Guangzhou Municipal Engineering Design&Research Institute Co.,Ltd.,Guangzhou 510060,China)

机构地区：[1]沈阳建筑大学市政与环境学院,辽宁沈阳110168 [2]广州市市政工程设计研究总院有限公司,广东广州510060

出　　处：《水电能源科学》2022年第12期85-88,共4页Water Resources and Power

基　　金：辽宁省教育厅项目(LJKZ0576);沈阳市科技计划项目(21-108-9-33)。

摘　　要：鉴于科学有效的城市用水量预测是进行水资源规划与管理的重要依据,以灰色预测模型为基础,提出应用二次指数平滑法(ES)对灰色预测模型进行残差优化的方法。以珠海市2010~2020年用水量为原始数据,使用基于ES残差优化的GM(1,1)模型对珠海市用水量进行预测分析,并与GM(1,1)模型、二次指数平滑法及其组合预测模型进行对比。结果表明,该模型预测精度相比其他方法提升明显,相对误差均小于5%,是一种有效的预测方式。Scientific and effective urban water consumption prediction is an important basis for water resources planning and management.Based on the grey prediction model,this paper proposed to optimize the residual of the grey prediction model by using the quadratic exponential sliding(ES).Taking the water consumption of Zhuhai from 2010 to 2020 as the original data,the water consumption was predicted by using GM(1,1)model based on ES residual optimization.Compared with the GM(1,1)model,second exponential smoothing method and combination model,the results show that the accuracy of this model is greatly improved,the relative error is less than 5%,and it is an effective prediction way.

关 键 词：灰色模型 指数平滑法 残差优化 用水量预测 

分 类 号：TV211.1[水利工程—水文学及水资源] F224.0[经济管理—国民经济]