无为大堤防洪保护区河道建模方法研究
|
摘要
使用MIKE软件对地形数据进行网格剖分时,河道易被概化失去其原有的行洪能力。【目的】寻找合适的河道建模方法。【方法】通过总结三角形网格方案、混合型网格方案、一二维耦合方案共3种建模方案,并分别建立模型,对结果进行了对比分析。【结果】三角形网格方案建模时,河道易被概化失去过水能力;混合型网格方案建模时,河道过水能力接近实际,但河岸与保护区接边处易发散;一二维耦合方案建模时,河道过水能力较好,但易忽略堤防薄弱处的洪水风险。三角形网格方案宜在计算精度要求高、计算时间充足时,将河道网格加密后使用;混合型网格方案建模时,河岸处理过程复杂,但计算速度较加密后的三角形网格方案约提高2倍,是替代三角形网格的重要方案之一;一二维耦合方案建模时,计算速度较加密后的三角形网格方案约提高5倍,但计算精度取决于断面间距的选取,宜在河道堤防高程差别不大时使用。【结论】河道作为保护区内的主要行洪通道,建模方案的选择会改变其行洪能力的强弱,进而必然影响洪水风险分析结果,可根据分析区域的特点选择合适的建模方法。
【Objective】Meshing the terrain in use of the MIKE model to simulate water flow in river could give rise to errors in calculating the ability of the channel to conduct water, and this paper aims to compare the performance of different mesh methods.【Method】We compared the pros and cons of triangular mesh network, the hybrid grid network and the one-dimension two-dimension coupling scheme.【Result】The triangular mesh resulted in errors in calculating water flow in the channel. The hybrid network was adequate to describe water flow in the river, but led to a dispersive interface between the river bank and the protection area. The two-dimensional coupling scheme captured water flow in the river, but underestimated the risk of flood in the embankment area. Triangular mesh was computationally demanded but gave accurate results by refining the meshes in the river. The hybrid grid scheme was computationally complicated, but was two time faster than the triangle grid scheme. The two-dimensional coupling scheme was five times faster than the triangle grid scheme, but its accuracy depends on spacing of the cross-section and it was suitable only when the river embankment is not much higher than the river bed.【Conclusion】The channel acts as the main flood channel in flood-controlled protected area, and different meshes could give different results and they are case-dependent and therefore should be used with care.
【Objective】Meshing the terrain in use of the MIKE model to simulate water flow in river could give rise to errors in calculating the ability of the channel to conduct water, and this paper aims to compare the performance of different mesh methods.【Method】We compared the pros and cons of triangular mesh network, the hybrid grid network and the one-dimension two-dimension coupling scheme.【Result】The triangular mesh resulted in errors in calculating water flow in the channel. The hybrid network was adequate to describe water flow in the river, but led to a dispersive interface between the river bank and the protection area. The two-dimensional coupling scheme captured water flow in the river, but underestimated the risk of flood in the embankment area. Triangular mesh was computationally demanded but gave accurate results by refining the meshes in the river. The hybrid grid scheme was computationally complicated, but was two time faster than the triangle grid scheme. The two-dimensional coupling scheme was five times faster than the triangle grid scheme, but its accuracy depends on spacing of the cross-section and it was suitable only when the river embankment is not much higher than the river bed.【Conclusion】The channel acts as the main flood channel in flood-controlled protected area, and different meshes could give different results and they are case-dependent and therefore should be used with care.
引文
[1]程晓陶.关于洪水管理基本理念的探讨[J].中国水利水电科学研究院学报, 2004, 2(1):36-43.
[2]李娜,向立云.国外风险图制作比较及对我国洪水风险图制作的建议[J].水利发展研究, 2005(6):28-32.
[3] DOTTORI F, FIGUEIREDO R, MARTINA M L V, et al. INSYDE:A synthetic, probabilistic flood damage model based on explicit cost analysis[J].Natural Hazards&Earth System Sciences, 2016,16(12):1-23.
[4]向立云.关于我国洪水风险图编制工作的思考[J].中国水利, 2005(17):14-16.
[5]魏凯,梁忠民,王军.基于MIKE21的濛洼蓄滞洪区洪水演算模拟[J].南水北调与水利科技, 2013, 11(6):16-19.
[6]张锦堂,李京兵,方泓,等. 2016年安徽省长江流域暴雨洪水特性分析[J].人民长江, 2017(4):66-69.
[7]王蕾.巢湖流域历代防洪思想与措施研究[D].武汉:华中师范大学, 2013.
[8]韩平.巢湖流域洪涝灾害损失评估研究[D].芜湖:安徽师范大学, 2013.
[9] Danish Hydraulic Institute. MIKE 21:A Modeling System for Rivers and Channels Reference Manual[R]. Denmar:DHI Water and Environment, 2007.
[10]陈涛,李光耀,韩旭.散乱模型的四边形网格剖分方法[J].机械工程学报, 2009, 45(5):121-127.
[11]江顺亮,万志国,周国发.注塑流动与传热分析的四边形单元控制体积法[J].计算力学学报, 2008, 25(3):359-363.
[12]江顺亮,余都.三角形四边形混合网格上的数据后处理方法[J].南昌大学学报(工科版), 2009, 31(3):238-243.
[13]徐石磊,甘忠,孙新申,等.不规则边界曲面的四边形-三角形混合网格展开法[J].机械设计与制造工程, 2007, 36(11):79-82.
[14]姜晓明,李丹勋,王兴奎.基于黎曼近似解的溃堤洪水一维-二维耦合数学模型[J].水科学进展, 2012, 23(2):214-221.
[2]李娜,向立云.国外风险图制作比较及对我国洪水风险图制作的建议[J].水利发展研究, 2005(6):28-32.
[3] DOTTORI F, FIGUEIREDO R, MARTINA M L V, et al. INSYDE:A synthetic, probabilistic flood damage model based on explicit cost analysis[J].Natural Hazards&Earth System Sciences, 2016,16(12):1-23.
[4]向立云.关于我国洪水风险图编制工作的思考[J].中国水利, 2005(17):14-16.
[5]魏凯,梁忠民,王军.基于MIKE21的濛洼蓄滞洪区洪水演算模拟[J].南水北调与水利科技, 2013, 11(6):16-19.
[6]张锦堂,李京兵,方泓,等. 2016年安徽省长江流域暴雨洪水特性分析[J].人民长江, 2017(4):66-69.
[7]王蕾.巢湖流域历代防洪思想与措施研究[D].武汉:华中师范大学, 2013.
[8]韩平.巢湖流域洪涝灾害损失评估研究[D].芜湖:安徽师范大学, 2013.
[9] Danish Hydraulic Institute. MIKE 21:A Modeling System for Rivers and Channels Reference Manual[R]. Denmar:DHI Water and Environment, 2007.
[10]陈涛,李光耀,韩旭.散乱模型的四边形网格剖分方法[J].机械工程学报, 2009, 45(5):121-127.
[11]江顺亮,万志国,周国发.注塑流动与传热分析的四边形单元控制体积法[J].计算力学学报, 2008, 25(3):359-363.
[12]江顺亮,余都.三角形四边形混合网格上的数据后处理方法[J].南昌大学学报(工科版), 2009, 31(3):238-243.
[13]徐石磊,甘忠,孙新申,等.不规则边界曲面的四边形-三角形混合网格展开法[J].机械设计与制造工程, 2007, 36(11):79-82.
[14]姜晓明,李丹勋,王兴奎.基于黎曼近似解的溃堤洪水一维-二维耦合数学模型[J].水科学进展, 2012, 23(2):214-221.