基于FLOW3D的梅溪洪濑段桥梁雍水三维数值模拟

Numerical simulation of Meixi River bridge backwater based on 3D hydrodynamic model FLOW3D in Honglai Town

为探究桥梁雍水与桥洞堵塞对洪泛区淹没情况的贡献,以山区河道梅溪洪濑段上的8座桥梁为例,借助FLOW3D模型构建精细化三维河道与桥梁模型。模型经过验证,模拟值的绝对误差平均值小于0.05 m,Nash系数大于0.77,表明模拟结果良好。分别计算各桥不同重现期下的雍水值,模拟桥洞堵塞程度的影响。结果表明:从整体来看,上游桥梁雍水值大部分大于下游桥梁点位;七号桥的雍水影响最大,在20 a、50 a和100 a重现期,对桥前洪泛区最大淹没水深的贡献比分别达到15.1%、18.5%与22.7%;桥孔堵塞程度增加的比例与桥前水位增量基本呈线性关系;相对于50 a无堵塞的情况,七号桥堵塞20%对桥前洪泛区最大淹没水深的贡献比又增加了21%,桥前平均水位甚至大于100 a无堵塞的洪水水位。本研究可以为梅溪洪濑段实际桥梁安全防护与河道沿岸洪泛区防洪减灾提供参考。

The calculation of bridge backwater of piers played an important role in both the riverbanks and the floodplain.Previous studies showed that the bridge backwater of piers could not be ignored.However,in the simulation of floodplain inundation caused by river overbank,the influence of bridge was not considered in river flood simulation.Bridge hole clogging is a common phenomenon in narrow rivers of mountainous areas,it is,therefore,necessary to study the influence of bridge hole clogging.For this reason,the contribution of bridge backwater of piers and bridge clogging to flooding is explored,which provides the foundation for the actual bridge maintenance and protection and floodplain disaster reduction.A catastrophic flood took place in the Honglai Town of the downriver reaches of the Meixi River catchment during the typhoon"Meranti"in 2016.As a result,the inundation depth of the floodplain around the Meixi River was nearly 3 m.The typhoon"Meranti"caused a total of 6.881 billion direct losses according to statistics from the government.The bridge backwater of piers may play an important role in the flooding during the typhoon"Meranti".Therefore,eight bridges along the Meixi River in Honglai Town were selected.To relatively veritably reflect the different shapes and structures of bridges and riverbeds,a refined three-dimensional model was built by three-dimensional hydrodynamic software(FLOW3 D).According to the measurement result by unmanned aerial vehicle oblique photography,the surface data was obtained with 1 meter DEM resolution in the Honglai Town.After that,the river channel and adjacent floodplain three-dimension models were generated.In addition,associated with field investigation of each bridge,the refined three-dimensional model of each bridge was also depicted in equal proportion by FLOW3 D.The model calculated the river water level changes and the backwater of piers of each bridge in five return periods.The observed water level data of different locations along the Meixi River was used to validate the FLOW3 D model in 10-year,20-year,and 50-year return periods.The maximum relative error between the simulated value and the data value is 0.19 meters in a single point.The simulated value using the average water level of each cross-section might be one of the reasons for explaining the error.As a whole,most of the simulated water level values were larger than the observed data values.In addition,it was verified that the absolute errors of the abovementioned three return periods are less than 0.1 meters.The average absolute errors of the 10-year,20-year,and 50-year return period were 0.02 meter,0.05 meter and 0.04 meter,respectively.The Nash coefficients were 0.83,0.77,and 0.81 respectively,which indicated that the simulated values fit the observed data well.Based on the verified FLOW3 D model,the backwater of piers for each bridge in different return periods were calculated.The influence of bridge hole clogging degree was also simulated.The results show that most of the upstream bridge′s backwater values were greater than the downstream bridges.Among them,the backwater of bridge number seven had the greatest impact.The river water will overflow to the floodplain near the bridge when the return periods larger than 10 year since its obstruction effect can even reach to 47%in channel cross-section.When the return period was more than 20-year,its contribution to the floodplain′s maximum inundation depth was more than 15%in front of the bridge number seven.In almost all return periods,the average velocity under bridge number seven was greater than the calculated critical velocity,which may cause scouring of the riverbank.The increasing proportion of the bridge hole clogging degree was linear with the increase of water level in front of the bridge.Compared with the situation without clogging,the contribution ratio with 20%clogging to the floodplain maximum inundation depth in front of the bridge will increase 21%.The average water level in front of bridge number seven with 20%clogging in 50-year was larger than that in 100-year without clogging.The variations of backwater of piers in front of each bridge along the Meixi River were calculated and the inundation contribution of each bridge to its adjacent floodplain was revealed.The flow simulated velocity and calculated velocity under each bridge were compared for all five return periods and all bridges can be used to simply estimate flood scouring to the riverbank.The effect of bridge hole clogging on upstream water level rising was determined.It is implied that reducing the obstruction in channel cross-section will have a relatively significant effect on decreasing adjacent floodplain inundation extent.It is suggested that removing the retaining part of piers extending into the river channel of bridge number seven.If no corresponding measures can be taken,monitoring equipment should be installed at bridge number seven to prevent the bridge from clogging.The abovementioned findings can provide important references for flood control,disaster reduction,and bridge safety.