摘要
识别合流制管网溢流特征是城市水环境治理工作的重要依据,也是城市给排水研究的难题。为此,基于支持向量机算法与SWMM模型进行耦合分析场降雨事件合流制溢流特征。以北京市通州区海绵城市建设试点区域内一片老城区为例,通过建立SWMM模型构造数据集训练得到SVM模型,对历史127场降雨事件进行溢流特征识别,在训练期、测试期精度分别达到93%、89%,表明基于支持向量机的算法精度可靠,结果符合实际,能作为一种新的手段识别合流制管网溢流特征,为海绵城市规划设计、防灾减灾与水环境治理工作提供了理论依据。
Combined sewer overflow(CSO)feature is an important basis for the urban water environment treatment,and it is also a difficult problem for urban water supply and drainage research.Support vector machine(SVM)algorithm and SWMM model are coupled to predict whether the overflow occurs or not.Taking a built-up area in the pilot area of sponge city construction in Tongzhou District of Beijing as an example,the SVM model is trained based on the SWMM model simulation results,and the overflow characteristics of 127 rainfall events are identified.The accuracy during the training period and test period can reach 93 percent and 89 percent,respectively.The results show that the approach based on SVM algorithm is reliable.It can be used as a novel means to identify the CSO characteristics and provide a theoretical basis for the planning and design of sponge city,disaster prevention and water environment management.
作者
张宇航
于磊
马盼盼
潘兴瑶
杨涛
宋磊
ZHANG Yu-hang;YU Lei;MA Pan-pan;PAN Xing-yao;YANG Tao;SONG Lei(Beijing Water Sciences and Technology Institute,Beijing 100048,China;College of Hydrology and Water Resources,Hohai University,Nanjing 210098,China;Beijing Hydrologic Center,Beijing 100089,China;Institute of Water Resources and Hydro-electric Engineering,Xi' an University of Technology,Xi' an 710048,China)
出处
《水电能源科学》
北大核心
2019年第6期87-90,共4页
Water Resources and Power
基金
国家水体污染控制与治理科技重大专项(2017ZX07103-002,2017ZX07103-007)
北京市优秀人才培养资助(青年骨干个人)
北京市科委重大项目(D161100005916001,D161100005916003,Z161100001116104)
关键词
合流制溢流
海绵城市
支持向量机
SWMM
判别方法
应用
combined sewer overflow
sponge city
support vector machine
SWMM
discrimination method
application
