灰绿耦合雨洪系统多目标优化建模与应用  

Multi-objective optimization modeling and application of grey-green coupling stormwater system

作　　者：李江云[1] 李瑶[1] 胡子欣 LI Jiangyun;LI Yao;HU Zixin(School of Civil Engineering,Wuhan University,Wuhan 430072,China)

机构地区：[1]武汉大学土木建筑工程学院,湖北武汉430072

出　　处：《水资源保护》2022年第6期49-55,80,共8页Water Resources Protection

基　　金：广东省水利科技创新项目(2017-33)。

摘　　要：以中山市某开发区为例,以灰绿基础设施总成本、径流系数和节点总溢流量为目标函数,以子汇水区内涝风险等级作为罚函数构建灰绿耦合排水系统多目标优化模型,基于NSGA-Ⅱ算法,编写SWMM接口程序,调用SWMM进行雨洪模型计算及方案寻优,对24 h设计暴雨条件下的灰色调蓄设施和绿色LID设施的规模进行多目标优化设计及综合评估。结果表明:绿色LID设施径流系数和节点总溢流量削减量均随其成本的增加呈近似线性下降的趋势,并随重现期增加而加快;灰色调蓄设施容积与节点总溢流量曲线存在明显拐点,拐点后规模增大的边际效益趋0;提出的灰绿耦合基础设施优化模型及求解程序对初始种群设置技术要求低,且能稳定收敛,可为城市雨洪管理方案的制定提供支持和依据。Taking a development zone in Zhongshan City as an example,taking the total cost of gray-green infrastructure,runoff coefficient and total node overflow as the objective function,and taking the waterlogging risk level of sub catchment area as the penalty function,a multi-objective optimization model of gray-green coupling drainage system was constructed.Based on NSGA-II algorithm,the SWMM interface program was compiled,and the SWMM was constructed to calculate the rain flood model and optimize the scheme.The scale of gray storage facilities and green LID facilities under 24h design rainstorm conditions was optimized and comprehensively evaluated.The results show that the runoff coefficient and the reduction of total node overflow decrease linearly with the increase of the cost of green LID facilities,and accelerate with the increase of the return period.There is an obvious inflexion point in the curve between the volume of storage facilities and the total node overflow,and the marginal benefit of the scale increase after the inflexion point tends to zero.The proposed grey-green coupling infrastructure optimization model and solution program have low technical requirements for initial population setting and can converge stably,providing support and basis for the formulation of urban stormwater management scheme.

关 键 词：雨洪系统 排水系统 灰色调蓄设施 绿色LID设施 NSGA-Ⅱ算法 多目标优化 

分 类 号：TU992[建筑科学—市政工程]