摘要
区域用水总量控制是落实最严格水资源管理制度的重要内容,提高用水总量预测水平是其关键支撑环节。分别利用ARIMA与GM(1,1)构建区域用水总量预测模型,并基于广州市时序用水特点,对其2002-2016年的用水总量进行模拟预测,在验证模型有效性基础上选用ARIMA(1,1,1)模型预测其2017-2020年用水总量。研究结果表明,ARIMA(1,1,1)对其用水总量模拟预测的相对误差呈现逐步收敛趋势,而GM(1,1)模型则具有先收敛后发散的特点,而结合平均与绝对相对误差验证ARIMA(1,1,1)适用于用水总量预测精度要求;通过预测发现广州市到2020年用水总量将面临超过其用水总量控制红线的危机,而从其2011-2015年用水结构中发现工业和农业用水仍是制约用水总量控制的关键点。因此,在后期水资源规划与调控中,广州市需要进一步加强工业用水效率和农业灌溉技术的提高。
Abstract
The total water use control is an important part of implementing the most stringent water resources management system. Increasing
the level of the total water use forecasting is the key support link. Based on the characteristics of seasonal that shown in water consumption
data in Guangzhou City from 2002 to 2016,ARIMA and GM ( 1,1) model were chosen to forecast the total water consumption. The results
showed that the relative error from ARIMA ( 1,1,1) model was gradually convergent; while the GM ( 1,1) model had the characteristics of
divergence after convergence. Therefore,combined with the average and absolute relative error verification,it was concluded that ARIMA
( 1,1,1) was suitable for the prediction accuracy of total water consumption. Through the forecast it is found that the total water consumption
in Guangzhou in 2020 will face the risk of beyond the red line of total amount water control. The fact that industrial and agricultural water use
was still the key constraints to total amount of water control was founded from the structure of water use data of Guangzhou in 2011-2015.
Therefore,Guangzhou needs to further strengthen increasing the industrial water use efficiency and the technique level of agricultural
irrigation in the future planning and control of water resources.
关键词
ARIMA 模型 /
GM( 1 /
1) 模型 /
用水总量 /
预测
{{custom_keyword}} /
Key words
ARIMA model /
GM( 1 /
1) model /
total water consumption /
forecast
{{custom_keyword}} /
基金
国家自然科学基金重点项目“面向智慧城市的水资源多元数据融合及建模方法研究”( U1501253) ; 广东省科技厅应用型研发基金专
项项目“水资源大数据综合应用平台研发及产业化”( 2016B010127005)
{{custom_fund}}
田 涛, 薛惠锋, 张 峰.
基于ARIMA与GM(1,1)的区域用水总量预测模型及应用——以广州市为例[J].节水灌溉, 2018(2): 61-65
TIAN Tao, XUE Hui-feng, ZHANG Feng.
Forecasting Models of Regional Water Consumption and Their Application
Based on ARIMA and GM ( 1,1) : A Case Study of Guangzhou City[J].Water Saving Irrigation, 2018(2): 61-65
{{custom_sec.title}}
{{custom_sec.title}}
{{custom_sec.content}}
参考文献
[1] 张玲玲,王宗志,李晓惠,等.总量控制约束下区域用水结构调控策略及动态模拟[J].长江流域资源与环境,2015,24( 1) : 90-96.
[2] Yadav B,Eliza K. A hybrid wavelet-support vector machine model forprediction of Lake water level fluctuations using hydro-meteorologicaldata[J]. Measurement,2017,( 103) : 294-301.
[3] Eslamian S A,Li S S,Haghighat F. A new multiple regression modelfor predictions of urban water use[J]. Sustainable Cities and Society,2016,( 27) : 419-429.
[4] Qaderi F,Babanejad E. Prediction of the groundwater remediationcosts for drinking use based on quality of water resource,using artificial neural network[J]. Journal of Cleaner Production,2017,( 161) : 840-849.
[5] Sempewo J I,Kyokaali L. Prediction of the future condition of a waterdistribution network using a markov based approach: a case study ofkampala water[J]. Procedia Engineering,2016,( 154) : 374-383.
[6] 袁朝阳,吴成国,张礼兵,等.集对分析聚类预测法在区域用水量中的应用[J].华北水利水电大学学报( 自然科学版) ,2015,36( 4) : 32-35.
[7] 孙 艳,刀海娅.自适应变异粒子群算法与支持向量机在农业用水预测中的应用[J].水资源与水工程学报,2015,( 3) : 231-236.
[8] 刘秀丽,邹庆荣.我国用水总量预测研究[J].系统科学与数学,2016,36( 10) : 1 643-1 651.
[9] 佟长福,史海滨,李和平,等.鄂尔多斯市工业用水变化趋势和需水量预测研究[J].干旱区资源与环境,2011,25( 1) : 148-150.
[10] 张志果,邵益生,徐宗学.基于恩格尔系数与霍夫曼系数的城市需水量预测[J].水利学报,2010,41( 11) : 1 304-1 309.
[11] 占 敏,薛惠锋,王海宁,等.贝叶斯神经网络在城市短期用水预测中的应用[J].南水北调与水利科技,2017,15( 3) : 73-79.
[12] Wu J,Xu S,Zhou R,et al. Scenario analysis of mine water inrushhazard using Bayesian networks[J]. Safety science,2016,( 89) :231-239.
[13] Shin H,Joo C,Koo J. Optimal rehabilitation model for water pipeline systems with genetic algorithm[J]. Procedia Engineering,2016,( 154) : 384-390.
[14] 张 静,余鹏翼,刘 娜.基于广州市水资源供需平衡分析的实证研究[J].生态经济( 学术版) ,2010,( 1) : 322-325.
[15] Calheiros R N,Masoumi E,Ranjan R,et al. Workload predictionusing ARIMA model and its impact on cloud applications’QoS[J].IEEE Transactions on Cloud Computing,2015,3( 4) : 449-458.
[16] 张 潇,夏自强,黄 峰,等.基于 SSA-ARIMA 模型的青弋江干流径流预测[J].中国农村水利水电,2015,( 3) : 6-9.
[17] 王 耕,王嘉丽,苏柏灵.基于 ARIMA 模型的辽河流域生态足迹动态模拟与预测[J].生态环境学报,2013,( 4) : 632-638.
[18] 邓聚龙.灰色系统理论教程[M].武汉: 华中理工大学出版社,1990: 14-18.
[19] Zhao H,Guo S. An optimized grey model for annual power load forecasting[J]. Energy,2016,( 107) : 272-286.
[20] 董承秀,陈继光.波动水资源数据灰幂模型预测中的维数选择[J].节水灌溉,2015,( 2) : 61-62.
[21] Astrm K J. Maximum likelihood and prediction error methods[J].Automatica,1980,16( 5) : 551-574.