王俊岭,胡颖梦,张昕喆,吴宾

• 1:北京建筑大学环境与能源工程学院
• 2:国网河南省电力公司郑州供电公司
• 3:中国林业科学研究院

摘要(Abstract)：

为了分析管网漏损影响要素和供水管网不可避免漏失水量之间存在的复杂联系及对应关系,建立适用于我国不可避免漏失水量的确定方法。通过采集、筛选实际管网的漏损统计数据,对独立计量区域(DMA)展开夜间最小流量分析,并分析研究区域内管网的影响因素,将管龄、管材、用户数量及管网压力设定为相关影响参数。在此基础上,分别建立BP神经网络模型以及经过遗传算法优化后的BP神经网络模型(GA-BP神经网络),并对优化前后二者神经网络模型拟合结果对比分析。最终将优化后的模型应用于不同DMA进行漏损分析。结果显示,优化后的BP神经网络不仅训练效率更高,且误差平均降低了3.02%。将该模型应用于实际管网三个不同DMA时,可以发现DMA内均存在可控漏损水量,以经济可行为前提,分别可节约水量10 512 m~3/a、12 702 m~3/a、37 580.4 m~3/a。研究成果可为DMA漏损控制提供参考。

关键词(KeyWords)： BP神经网络;遗传算法;可行性评估;不可避免漏失;线性回归分析;人工智能算法;DMA;影响因素

Abstract:

Keywords:

基金项目(Foundation): 国家水体污染与治理科技重大专项课题(2017ZX07501-002-05)

作者(Author): 王俊岭,胡颖梦,张昕喆,吴宾

DOI: 10.13928/j.cnki.wrahe.2021.12.020

参考文献(References)：

• [1] FARLEY M,WYETH G,GHAZALI Z B,et al.The manager′s non-revenue water hand book,a guide to understanding water losses [M].London:International Water Association publishing,2008.
• [2] KINGDOM B,LIMBERGER R,MARIN P,et al.The challenge of reducing non-revenue water (NRW) in developing countries.How the private sector can help:A look at performance-based service contracting,water supply and sanitation sector board discussion paper series No.8 [M].Washington,DC:The World Bank,2006.
• [3] THORNTON J,STURM R,KUNKEL G,Water loss control,seconded [M].New York City:McGraw-Hill Companies,2008.
• [4] HELENA A,HIMER W,BAPTISTA J M.Performance indicators for water supply services.second edition [M].London:International Water Association publishing,2006,30-45.
• [5] CAVAZZINI G,PAVESE G,ARDIZZON G.Optimal assets management of a water distribution network for leakage minimization based on an innovative index[J].Sustainable Cities and Society,2020,54:101890.
• [6] 张现国，汪慧贞，王俊岭，等.供水管网漏损评价指标筛选与计算实例[J].给水排水，2011,47(1):158-161.ZHANG Xianguo,WANG Huizhen,WANG Junling,et al.Selection and calculation example of water supply pipeline network leakage evaluation index[J].Water & Wastewater Engineering,2011,47(1):158-161.
• [7] 代焕芳，刘书明.国内外供水管网漏损评价指标初探[J].给水排水，2016,52(S1):258-261.DAI Huanfang,LIU Shuming.Preliminary study on leakage evaluation index of water supply pipe network at home and abroad[J].Water & Wastewater Engineering,2016,52(S1):258-261.
• [8] 代焕芳，刘书明，吴雪.供水管网背景漏失指数研究[J].中国给水排水，2019,35(11):59-62.DAI Huanfang,LIU Shuming,WU Xue.Research on background leakage index of water supply network[J].China Water& Wastewater,2019,35(11):59-62.
• [9] LENZI C,BRAGALLI C,BOLOGNESI A,et al.Infrastructure leakage index assessment in large water systems[J].Procedia Engineering,2014,70:1017-1026.
• [10] 董驹萍，吴珊，刘阔，等.DMA不可避免漏失量计算值与实测值对比分析[J].给水排水，2014,50(S1):366-369.DONG Juping,WU Shan,LIU Kuo,et al.Comparison analysis of calculated and measured values of inevitable loss of DMA[J].Water & Wastewater Engineering,2014,50(S1):366-369.
• [11] 孙福强.国内外供水管网漏损管理技术与指标浅析[J].城镇供水，2013(6):64-66.SUN Fuqiang.Analysis of leakage management technology and indicators of domestic and foreign water supply pipe network[J].City and Town Water Supply,2013(6):64-66.
• [12] 徐强，焦静，赵顺萍，等.供水管网漏损评价指标对比与改进[J].中国给水排水，2016,32(20):14-18.XU Qiang,JIAO Jing,ZHAO Shunping,et al.Comparison and improvement of water supply pipeline network leakage assessment indicators[J].China Water & Wastewater,2016,32(20):14-18.
• [13] 李飞，陶涛.供水管网漏损评估与控制方法[J].中国给水排水，2012,28(18):35-39.LI Fei,TAO Tao.Leakage assessment and control method of water supply pipe network[J].China Water & Wastewater,2012,28(18):35-39.
• [14] 常田，刘书明，王敏，等.基于BP神经网络的城市供水管网健康状态评估[J].给水排水，2016,52(6):138-141.CHANG Tian,LIU Shuming,WANG Min,et al.Health condition assessment of urban water supply network based on BP neural network [J].Water & Wastewater Engineering,2016,52(6):138-141.
• [15] 王珞桦，李红卫，吕谋，等.基于BP神经网络深度学习的供水管网漏损智能定位方法[J].水电能源科学，2019,37(5):61-64.WANG Luohua,LI Hongwei,LYU Mou,et al.Intelligent locating method of water pipeline leakage based on deep learning of BP neural network [J].Water Resources and Power,2019,37(5):61-64.
• [16] 王韬.基于BP神经网络的供水管网爆管风险预测模型研究[D].重庆：重庆大学，2019.WANG Tao.Research on burst risk prediction model of water supply network based on BP neural network [D].Chongqing:Chongqing University,2019.
• [17] 万盛萍.GA优化BP神经网络[J].软件导刊，2007(3):24-25.WAN Shengping.GA optimizes BP neural network[J].Software Guide,2007(3):24-25.
• [18] 罗杰，苏兵，翟乐育.基于BP神经网络的空中无人通信平台作战效能评估[J].指挥控制与仿真：2021,43(4):21-25.LUO Jie,SU Bing,ZHAI Leyu.Evaluation of operational effectiveness of unmanned aerial communication platform based on BP neural network[J].Command Control and Simulation:2021,43(4):21-25.
• [19] 墨蒙，赵龙章，龚嫒雯，等.基于遗传算法优化的BP神经网络研究应用[J].现代电子技术，2018(9):41-44.MO Meng,ZHAO Longzhang,GONG Miwen,et al.Research and application of BP neural network based on genetic algorithm optimization[J].Modern Electronic Technology,2018(9):41-44.
• [20] 张立仿，张喜平.量子遗传算法优化BP神经网络的网络流量预测[J].计算机工程与科学，2016,38(1):114-119.ZHANG Lifang,ZHANG Xiping.Quantum genetic algorithm optimizes network traffic prediction of BP neural network[J].Computer Engineering and Science,2016,38(1):114-119.
• [21] HO C I,LIN M D,LO S L.Use of a GIS-based hybrid artificial neural network to prioritize the order of pipe replacement in a water distribution network[J].Environmental Monitoring & Assessment,2010,166(1-4):177-189.
• [22] 邵圆媛.嵌套BP/GMS神经网络模型在供水管网漏损预测中的研究[D].重庆：重庆大学，2018.SHAO Yuanyuan.Research on nested BP/GMS neural network model in leakage prediction of water supply network[D].Chongqing:Chongqing University,2018.
• [23] 付波.住宅区热力站负荷预测研究[D].北京：北京建筑大学，2019.FU Bo.Research on load forecasting of thermal stations in residential district[D].Beijing:Beijing University of Architecture and Architecture,2019.
• [24] 田旭光，宋彤，刘宇新.结合遗传算法优化BP神经网络的结构和参数[J].计算机应用与软件，2004(6):69-71.TIAN Xuguang,SONG Tong,LIU Yuxin.Combined with genetic algorithm to optimize the structure and parameters of BP neural network[J].Computer Applications and Software,2004(6):69-71.
• [25] 杨猛.基于遗传算法与人工神经网络的加热炉建模方法研究[D].合肥：中国科学技术大学，2017.YANG Meng.Research on modeling method of heating furnace based on genetic algorithm and artificial neural network[D].Hefei:University of Science and Technology of China,2017.