| 1,305 | 26 | 201 |
| 下载次数 | 被引频次 | 阅读次数 |
针对城市水资源短缺和城市生态环境需水研究的迫切性，选用6种预测模型开展城市生态需水预测研究。为探究适合北京市城市生态环境用水预测的模型，以中国城市统计年鉴2004—2019年的北京市相关统计数据为基础，在时间序列预测方法上比较了灰色模型、分数阶灰色模型，在系统分析预测方法上比较了多元线性回归、主成分回归、BP神经网络和灰色神经网络方法。总结并归纳了城市生态环境用水量的概念和计算方法，分析和选取6种不同模型的输入变量，分别开展模型的训练和测试模拟，2004—2016年为训练期，2017—2019为测试期。通过交叉检验得出各类模型预测结果的得分，结果显示分数阶灰色模型得分为0.081 4,优于其他5种模型的得分，说明该模型在六种方法中对北京市城市生态环境用水短期预测的适用性较好。利用分数阶灰色模型对北京市未来2 a的城市生态环境用水进行预测，2021和2022年生态环境用水量分别为19.79亿m3和22.16亿m3。北京市的城市生态环境用水在未来短期时间内将保持高速增长的趋势。
Abstract：Aiming at the urgency of the study on the urban water resources shortage and eco-environmental water demand, 6 mathematical models are selected for carrying out the prediction of the urban eco-water demand. In order to explore a model suitable for the prediction of the urban eco-environmental water use in Beijing and on the basis of the relevant statistical data of Beijing between 2004—2019 from China Urban Statistical Yearbook, the multi-gray model and fractional gray model are compared in the aspect of time series prediction method, while the multiple linear regression, principal component regression, BP neural network and gray neural network are compared in the aspect of the systematic analysis prediction method. The concept and calculation method of the amount of the urban eco-environmental water use are summarized, and then the input variables of the six different models are analyzed and selected to carry out the model trainings and testing simulations respectively with the training period of 2004—2016 and the testing period of 2017—2019. The prediction scores of all kinds of the models are obtained through the relevant cross validations and the results show that the score from the fractional gray model is 0.081 4 and better than those of the other five models, which indicates that within the six methods, this method has a better applicability for the short-term prediction of the urban eco-environmental water use in Beijing. The prediction of the urban eco-environmental water use in Beijing obtained from the fractional gray model shows that the eco-environmental water uses in 2021 and 2022 are to be 1.979 billion m3 and 2.116 billion m3 respectively; that is to say, the urban eco-environmental water use in Beijing is going to maintain a trend of rapid increase within a short-period of time in the days to come.
[1] 李萌，白丹，肖平.城市生态用水概念及西安市生态用水分析[C]//周孝德，沈冰.水与社会经济发展的相互影响及作用：全国第三届水问题研究学术研讨会论文集.西安：中国水利水电出版社，2005:413-416.LI Meng,BAI Dan,XIAO Ping.Concept of urban ecological water and analysis of Xi′an ecological water [C]//ZHOU Xiaode,SHEN Bing.Geophysics:The interaction and effect of water and social and economic development — Proceedings of the Third National Symposium on Water Issues.Xi′an:China Water and Power Press,2005:413-416.
[2] 董云程，周明，杜坤，等.城市需水量预测方法与模型综述[J].软件导刊，2019,18(12):1-5.DONG Yuncheng,ZHOU Ming,DU Kun,et al.Review of forecasting methods and models of urban water demand [J].Software guide,2019,18(12):1-5.
[3] 郑志磊，郑航，刘悦忆，等.基于优化模型的城市供水多水源配置研究[J].水利水电技术，2020,51(9):58-64.ZHENG Zhilei,ZHENG Hang,LIU Yueyi,et al.Optimization model-based study on allocation of multi-water source for urban water supply[J].Water Resources and Hydropower Engineering,2020,51(9):58-64.
[4] 李琴，马耀光.宝鸡市用水结构及用水量预测分析[J].干旱地区农业研究，2013,31(6):238-242.LI Qin,MA Yaoguang.Analysis on structure and forecast of water consumption in Baoji City.Agricultural Research in the Arid Areas,2013,31(6):238-242.
[5] 白鹏，龙秋波.3种用水量预测方法在京津冀地区的适用性比较[J].水资源保护，2021,37(2):102-107.BAI Peng,LONG Qiubo.Comparison of applicability of three forecasting methods of water consumption in Beijing-Tianjin-Hebei region [J].Water Resources Conservation,2021,37(2):102-107.
[6] 黄军.合肥庐阳经济开发区需水预测及合理性分析[J].水资源开发与管理，2020(8):50-53.HUANG Jun.Forecast and rational analysis of water demand in Luyang Economic Development Zone,Hefei [J].Water Resources Development and Management,2020(8):50-53.
[7] 薛林丽，袁星宇，刘宗显.城市需水预测方法比较[J].人民珠江，2018,39(1):63-66.XUE Linli,YUAN Xingyu,LIU Zongxian.Comparison of forecasting methods of urban water demand [J].Pearl River,2018,39(1):63-66.
[8] 辛珂，李文竹，刘心.基于GA-ELM算法的城市短期用水预测[J].电脑知识与技术，2020,16(13):217-220.XIN Ke,LI Wenzhu,LIU Xin.Prediction of short-term urban water consumption based on GA-ELM algorithm[J].Computer Knowledge and Technology,2020,16(13):217-220.
[9] GUO Guancheng,LIU Shuming,WU Yipeng,et al.Short-term water demand forecast based on deep learning method[J].Journal of Water Resources Planning and Management,2018,144(12):04018076.
[10] PEYMAN Y,SHABANI S,MOHAMMADI H,et al.Gene expression programing in long term water demand forecasts using wavelet decomposition[J].Procedia Engineering,2017,186(2) :544-550.
[11] SINA S,CANDELIERI A,ARCHETTI F,et al.Gene expression programming coupled with unsupervised learning:a two-stage learning process in multi-scale,short-term water demand forecasts[J].Water,2018,10:142-154.
[12] 张艺聪.基于BP神经网络的郑州农业灌溉用水预测[J].科学技术创新，2021(2):96-98.ZHANG Yicong.Prediction of irrigation water in Zhengzhou based on BP neural network [J].Science and Technology Innovation,2021(2):96-98.
[13] QIN T L,YAN D H,WANG G,et al.Water demand forecast in the Baiyangdian Basin with the extensive and low-carbon economic modes[J].Journal of Applied Mathematics,2014,2014:673485.doi:10.1155/2014/673485.
[14] LOPEZ F.R,PUIG V,RODRIGUEZ R H,et al.Multi-model prediction for demand forecast in water distribution networks[J].Energies,2018,11(3):660.doi:10.3390/en11030660.
[15] LIANG Z,ZOU R,CHEN X,et al.Simulate the forecast capacity of a complicated water quality model using the long short-term memory approach[J].Journal of Hydrology,2020,581:124432.
[16] 张达.北京市水资源绿色发展指数综合评价[J].合作经济与科技，2021(13):22-23.ZHANG Da.Comprehensive evaluation of green development index of water resources in Beijing [J].Cooperative Economy & Science & Technology,2021(13):22-23.
[17] 魏彦昌，苗鸿，欧阳志云，等.海河流域生态需水核算[J].生态学报，2004,24(10):2100-2107.WEI Yanchang,MIAO Hong,OUYANG Zhiyun,et al.Primary estimate of ecological water requirement in Haihe Basin[J].Acta Ecologica Sinica,2004,24(10):2100-2107.
[18] 张国政，申君歌.基于多周期时间序列的灰色预测模型及其应用[J].统计与决策，2021,37(9):14-19.ZHANG Guozheng,SHEN Junge.Gray prediction model based on multi-period time series and its application [J].Statistics & Decision,2021,37(9):14-19.
[19] 吴利丰.分数阶灰色预测模型及其应用研究[D].南京：南京航空航天大学，2015.WU Lifeng.Fractional order grey forecasting models and their application[D].Nanjing:Nanjing University of Aeronautics and Astronautics,2015.
[20] SAHOO A,NANDKEOLY R.Entropy generation in convective radiative flow of a Casson nanofluid in non-Darcy porous medium with Hall current and activation energy:The multiple regression model[J].Applied Mathematics and Computation,2021,402:125923.
[21] 董巧玲.不同误差影响模型下总体最小二乘法在多元线性回归中的应用研究[D].太原：太原理工大学，2016.DONG Qiaoling.Research on the application of total least squares method in multiple linear regression under different error influence model [D].Taiyuan:Taiyuan University of Technology,2016.
[22] 刘玉金.基于主成分分析与多元线性回归分析的灌溉水利用效率影响因素分析[D].呼和浩特：内蒙古农业大学，2014.LIU Yujin.Analysis of influencing factors of irrigation water use efficiency based on principal component analysis and multiple linear regression analysis [D].Huhhot:Inner Mongolia Agricultural University,2014.
[23] 张宏，马岩，李勇，等.基于遗传BP神经网络的核桃破裂功预测模型[J].农业工程学报，2014,30(18):78-84.ZHANG Hong,MA Yan,LI Yong,et al.Prediction model of walnut breaking work based on genetic BP neural network [J].Transactions of the Chinese Society of Agricultural Engineering,2014,30(18):78-84.
[24] 于志军，杨善林，章政，等.基于误差校正的灰色神经网络股票收益率预测[J].中国管理科学，2015,23(12):20-26.YU Zhijun,YANG Shanlin,ZHANG Zheng,et al.Stock returns prediction based on error-correction grey neural network[J].Chinese Journal of Management Science,2015,23(12):20-26.
[25] 李小燕.灰色神经网络预测模型的优化研究[D].武汉：武汉理工大学，2009.LI Xiaoyan.Research on optimization of grey neural network prediction model [D].Wuhan:Wuhan University of Technology,2009.
[26] 左其亭，胡德胜，窦明，等.基于人水和谐理念的最严格水资源管理制度研究框架及核心体系[J].资源科学，2014,36(5):906-912.ZUO Qiting,HU Desheng,DOU Ming,et al.Research framework and core System of the strictest water resources Management System based on the concept of human-water harmony [J].Resources Science,2014,36(5):906-912.
[27] 王慧敏.落实最严格水资源管理的适应性政策选择研究[J].河海大学学报(哲学社会科学版),2016,18(3):38-43.WANG Huiming.Research on adaptive policy selection for implementing the strictest water resources management [J].Journal of Haihe University (Philosophy and Social Sciences edition),2016,18(3):38-43.
基本信息:
DOI：10.13928/j.cnki.wrahe.2022.03.012
中图分类号:TV213.4;X171.1
引用信息:
[1]黄天意,周晋军,李雅君,等.六种预测模型在北京市城市生态环境用水短期预测中的比较[J],2022,53(03):119-133.DOI:10.13928/j.cnki.wrahe.2022.03.012.
基金信息:
北京市自然科学基金青年项目(8214046);; 清华大学水沙科学水利水电工程国家重点实验室及宁夏银川水联网数字治水联合研究院联合开放研究基金资助课题(sklhse-2021-Iow07);; 国家重点研发计划项目(2016YFC0401401,2017YFC1502701)