Numerical Simulation of Dam-Break Flows Using Radial Basis Functions: Application to Urban Flood Inundation

Dam-break flows pose significant threats to urban areas due to their potential for causing rapid and extensive flooding. Traditional numerical methods for simulating these events struggle with complex urban landscapes. This paper presents an alternative approach using Radial Basis Functions to simulate dam-break flows and their impact on urban flood inundation. The proposed method adapts a new strategy based on Particle Swarm Optimization for variable shape parameter selection on meshfree formulation to enhance the numerical stability and convergence of the simulation. The method’s accuracy and efficiency are demonstrated through numerical experiments, including well-known partial and circular dam-break problems and an idealized city with a single building, highlighting its potential as a valuable tool for urban flood risk management.

Numerical Simulation and Risk Analysis of Coastal Inundation in Land Reclamation Areas:A Case Study of the Pearl River Estuary

Coastal inundation results in many human casualties and significant economic losses.In this study,an inundation model with an unstructured mesh was developed using the ADvanced CIRCulation(ADCIRC)and Simulating WAves Nearshore(SWAN)models to simulate 43 inundation events from 1998 to 2018.A combined wind-pressure field derived from the assimilation windpressure model was used to drive the coupled ADCIRC-SWAN model.On the basis of the model results,a sensitivity study of the influence of land reclamation on coastal inundation was conducted.The results of the study showed that coastline changes caused by reclamation significantly influence the distribution of coastal inundation,particularly in areas where narrow waterways,bays,and peninsulas are newly formed.Combining the extreme inundation events calculated using the Gumbel and Weibull distributions,the 10-year,50-year,100-year,and 200-year return coastal inundations were obtained for the Pearl River Estuary in China.The risk of coastal inundation was analyzed according to the probability of the inundation depth exceeding 1 m.A hazard grade zoning map was drawn to guide disaster relief and mitigation in the reclamation area.

Prediction of Debris-flow Danger Area by Combining Hydrological and Inundation Simulation Methods

Debris flows have caused serious human casualties and economic losses in the regions strongly affected by the Ms8.0 Wenchuan earthquake of 2oo8. Debris flow mitigation and risk assessment is a key issue for reconstruction. The existing methods of inundation simulation are based on historical disasters and have no power of prediction. The rain- flood method can not yield detailed flow hydrograph and does not meet the need of inundation simulation. In this paper, the process of water flow was studied by using the Arc-SCS model combined with hydraulic method, and then the debris flow runoff process was calculated using the empirical formula combining the result from Arc-SCS. The peak discharge and runoff duration served as input of inundation simulation. Then, the dangerous area is predicted using kinematic wave method and Manning equation. Taking the debris flow in Huashiban gully in Beichuan County, Sichuan Province, China on 24 Sep. 2oo8 as example, the peak discharge of water flow and debris flow were calculated as 35.52 m3·s-1 and 215.66 m3·s-, with error of 4.15% compared to the measured values. The simulated area of debris-flow deposition was 161,500 m2, vs. the measured area of 144,097 m2, in error of 81.75%. The simulated maximum depth was 12.3 m, consistent with the real maximum depth between lO and 15 m according to the field survey. The minor error is mainly due to the flow impact on buildings and variations in cross-section configuration. The present methodology can be applied to predict debrisflow magnitude and evaluate its risk in other watersheds inthe earthquake area.

上海城区动态洪水风险图应用系统及典型暴雨内涝分析

【目的】为了提高沿海超大城市对洪涝灾害的先知先觉能力,快速预测分析洪涝潮组合影响下的淹没风险时空分布,【方法】以水文-水动力暴雨洪水分析模型为核心模块,通过与外部的气象精细化降雨预报数据库、自动雨量站实时监测数据库及实时水情、闸泵运行数据库相关联,研发了上海城区动态洪水风险图应用系统。采用由雨量站点到气象格网和水力网格的降雨二级空间融合技术,实现了模型与气象预报降雨数据的有效耦合。利用数据挖掘提取和GIS空间分析技术构建了包含10项洪水风险要素要点的城市内涝预报专报自动生成方法,实现了对内涝风险的快速一站式概览。利用系统可以对城市暴雨、河道洪水和风暴潮等单一发生或遭遇组合引起的淹没分布进行快速实时预报计算,模拟和预测城市洪涝潮灾害的有关淹没特征数据和淹没动态过程。【结果】利用系统分析了2023年6月23—24日暴雨的内涝风险分布,将模型模拟的积水区域与积水监测站、灾情直报和热线灾报的积水点进行对比,结果显示在150处对比积水点中,有129处误差不超过20 cm,占86%。模型模拟的积水空间分布与实际情况基本接近。【结论】结果表明:系统满足汛期常态化、业务化运行需求,能够为城市洪涝风险的实时动态分析和防汛指挥决策提供重要工具。

基于HAND的流域洪水淹没快速分析技术

快速分析计算流域洪水淹没,对于防洪具有重要作用,而基于水动力学模型的经典淹没算法在时效性方面存在不足,制约防洪“四预”功能实现和数字孪生流域建设。针对此,本文建立了基于HAND(Height Above Nearest Drainage)的形态淹没模型,用于快速分析流域洪水淹没情况,在乐安河流域上游进行了应用验证。结果表明,模型可在5秒内完成洪水淹没计算,对四场历史洪水和三场设计洪水淹没范围的总体模拟精度平均为0.95,达到水动力学模型的模拟精度;对淹没水深的模拟存在较大误差,平均误差为0.79 m。基于HAND的形态淹没模型可快速绘制全流域洪水淹没范围图,大幅提升影响预报和防洪预演的时效性水平。

基于DEM的水库下游洪水淹没演进过程研究

水库作为重要的水利工程在防洪减灾中发挥着重要作用,同时关系到下游广大人民的生命财产安全和经济可持续发展。为此,提出一种基于DEM的水库下游洪水淹没演进过程分析方法,并将其应用于周公宅水库下游区域洪水淹没演进过程研究中。结果表明,该方法能提高洪水淹没演进过程的分析效率,可实现水库下游洪水淹没演进过程模拟,且模拟结果符合山区河道的洪水淹没演进规律,可为水库洪水防御预案制定、下游洪水风险预警发布、避险转移工作开展提供数据参考和技术支撑。

汜水河洪涝风险模拟及灾害防控评价

暴雨所引发的洪涝灾害致灾机理复杂、破坏力度较大、预报预警困难,往往造成毁灭性的灾害。为了提高洪水预测精度,最大程度地降低洪涝灾害带来的影响,以荥阳汜水河流域为研究区域,基于MIKE FLOOD模块,构建了MIKE11与MIKE21耦合的洪涝仿真模型,采用河道实测流量、下游防洪堤高程、历史洪痕3个参数对模型进行了率定验证,模拟值与观测值吻合度较高。模拟设计了3个重现期情景,分析研究区域的洪水淹没分布特征,并划分风险等级。结果表明:研究区域洪水淹没深度和淹没面积随暴雨强度增强而增加,“7·20”情景下最大淹没水深高达9 m,汜水镇是所有情景下受淹情况最严重的地区。城镇与城镇周围环境存在淹没风险差异,3个暴雨重现期情景下城镇外围农业用地淹没风险皆最大,其最大淹没面积占淹没总面积的79.8%;淹没范围内各种土地利用类型占比由大到小依次为:农业用地、建筑用地、道路用地、其他类型。绘制“7·20”受灾的汜水镇洪水灾害转移路线图,为避难决策的制定提供科学依据,并提出汜水河不同洪水重现期情况下需要提高堤防和疏浚的措施,制定不同情景下防洪堤提升改造方案。

基于元胞自动机的降雨-径流模型对城市淹没过程的模拟

针对现有的基于元胞自动机(CA)的径流模型无法随时应用于复杂的城市环境,提出了一种基于CA的新型降雨-径流模型(RRCA)来模拟洪水淹没。该模型考虑了12种城市情景而非单一土地利用类型的精细径流生成过程,通过水资源供需的新过渡规则(WS-WD)模拟汇流过程。将RRCA与基于CA的洪水模型(E2DCA)进行比较,结果表明,RRCA和E2DCA与基准模型具有良好的一致性,R^(2)大于0.9,且RRCA在模拟效率和准确性上表现更好。因此,在城市内涝管理中,可通过精准模拟有效识别排水困难区域,从而及时采取相应措施。

北京系列高分辨率卫星在永定河流域洪涝灾害分析中的应用

2023年海河发生“23·7”流域性特大洪水,永定河发生1924年以来最大洪水。为了迅速掌握洪涝情况、及时做好洪水应对工作,在永定河流域洪涝灾害分析中应用北京系列高分辨率卫星遥感技术,通过灾前、灾后影像反复对比,采用目视解译法和经验法,人机交互识别确定阈值,快速获取水体淹没情况,识别出此次洪水的淹没范围。通过现场洪痕划定及复核工作,得出此次洪水的最大淹没范围,运用ArcGIS空间分析功能,提取出永定河河道最大淹没面积。淹没分析成果可为永定河后续综合治理提供支撑,推动流域灾后恢复重建工作。

抽水蓄能电站库区淹没与库岸稳定分析系统研发及应用

为进一步加快构建新型电力系统,促进抽水蓄能电站数字化转型。笔者通过运用地理信息系统(geographic information system,GIS)技术和数字高程模型(digital elevation model,DEM)建立抽水蓄能电站库区淹没与库岸稳定分析系统,并从系统的技术结构、三维可视化、库区淹没分析、库区浸没分析和库区塌岸分析等方面分析系统模块建设及其功能实现。结果表明,该系统可实现库区淹没范围、库区浸没范围和库区塌岸预测范围边界线的三维模拟分析及实时动态演示,相关的土地利用数据快速整合评估等。该系统可为合理确定抽水蓄能电站水库大坝位置,制定征地拆迁、土地赔偿、移民安置等方案提供科学的决策依据。

基于排水模型的城市排水应用研究

目前,城市排水系统存在诸多问题,包括排水不畅、内涝频发等,严重影响了城市的运行和居民的生活。本文研究了在城市排水系统中,排水模型的重要性,以及要运行模型需要准备的基础数据,并结合在真实项目中实践,剖析了排水模型在城市排水上的模拟分析功能及应用,包括对排水管网中淤积问题、雨污混接问题的分析,对内涝淹没区域和程度的推演,对水泵启停调度的模拟和优化等,并取得了较好的应用效果。

基于遥感智能解译技术的围堰溃决洪水淹没分析--以旭龙水电站为例

水电站围堰溃决洪水突发性强且非常规,应急调查时效性要求高。为分析水电站围堰溃决洪水风险,以旭龙水电站为例,提出了一套基于遥感智能解译技术的溃堰洪水淹没分析方法用于实物指标应急调查,并选取与居民财产安全密切相关的建筑物为典型对象,通过构建U-Net卷积神经网络模型,对旭龙水电站下游区域进行建筑物提取。结果表明:该方法可有效识别出建筑物分布情况,F-score指标精度在93%以上,在算法效率上也明显优于人工解译。

瀑布水库调度方式优化对下游防洪影响研究

为充分挖掘水库防洪潜力,提高水库防洪能力,本文以瀑布水库为例,通过合理降低汛限水位、设置防洪库容、充分利用水库闲置库容、优化调度运行等方式,结合补偿调节原理,系统分析了水库和径流特征,以及水库下泄洪水对下游镇区的防洪影响。该研究可供同类水库减少下游防洪压力,提升下游防洪安全提供参考。

Optimal Evacuation Scheme Based on Dam-Break Flood Numerical Simulation

The optimal evacuation scheme is studied based on the dam-break flood numerical simulation. A three- dimensional dam-break mathematical model combined with the volume of fluid （VOF） method is adopted. According to the hydraulic information obtained from numerical simulation and selecting principles of evacuation emergency scheme, evacuation route analysis model is proposed, which consists of the road right model and random degree model. The road right model is used to calculate the consumption time in roads, and the random degree model is used to judge whether the roads are blocked. Then the shortest evacuation route is obtained based on Dijstra algorithm. Gongming Reservoir located in Shenzhen is taken as a case to study. The results show that industrial area I is flooded at 2 500 s, and after 5 500 s, most of industrial area II is submerged. The Hushan, Loucun Forest and Chaishan are not flooded around industrial area I and II. Based on the above analysis, the optimal evacuation scheme is determined.

Experimental Study on the Dam-Break Hydrographs at the Gate Location

When studying the dam-break flow phenomenon,the basic hydrodynamic features of the dam-break flow at the gate location should be verified primarily.In this study,laboratory experiments were performed in a rectangular and horizontal flume with the same initial water head setting on the dry and wet downstream bed conditions.Water surface elevation was extracted through image analysis and validated by comparing with the data measured using a wave gauge.Temporal variation of the water surface elevation at the gate location,quantified in terms of high-speed video recorded images,can be divided into three stages,the sharp decreasing stage,the relatively steady stage,and the gradually decreasing stage.Applicability of several classic analytical solutions of the dam-break problem at the gate location was validated using present experimental data.Ritter's solution is effective for the dry bed condition while Stoker's solution could be applied to the wet bed case,and both are only applicable during the steady stage.Lin' solution reproduces the gate-site hydrographs well during both the relatively steady and the gradually decreasing stages,especially for the condition under which the down-upstream water depth ratio is smaller than 0.138.

贵州省正安县“2020·6·12”山洪灾害成因分析与启示

2020年6月12日凌晨,贵州省正安县白花河流域突降大暴雨至特大暴雨,致碧峰集镇所在地发生特大山洪灾害。为探寻此次山洪灾害成因,在资料收集和现场勘测基础上,分析了灾情、雨情和水情,还原了山洪动态淹没过程,并利用基于二维浅水方程开发的洪水演进模型软件对洪水演进过程进行了模拟和验证,得出以下主要结论:(1)特殊的地形地貌和独特的地理位置为灾害发生提供了基础条件;(2)短历时超强降雨叠加较高的前期累积降雨量是诱发灾害关键因素;(3)河道现状防洪能力不足与洪峰流量过大的矛盾使得灾害发生成为必然;(4)非常住人口与流动人口较多,妇女和儿童自救能力差,部分群众避险意识不强,极大增加了灾害损失。在此基础上,提出了科学规划顶层设计、完善山洪灾害群测群防体系、细化防洪减灾动态管理等启示和建议。研究成果对防洪减灾、洪水风险分析和灾情评估等山洪灾害防治工作具有一定的借鉴和指导意义。

基于仿真平台的矿井突水淹没路径和逃生路径规划

井下环境错综复杂,与矿井水害不确定性及非线性等叠加,水害淹没情况千变万化,难以预测。针对如何获取矿井水灾发生时各个时间点巷道内的淹没情况,并根据淹没情况快速确定逃生路径的难点问题,采用“木桶原理”构建淹没路径树,确定巷道水流分岔方向,分析矿井突水淹没递进规律,模拟井下巷道内的水流状况,仿真直观展示实时水流变化、水位高度和逃生路径。研究表明:根据构建的淹没路径树,通过递归得到的水位等淹没参数,作为设计逃生路径的基础,微分积分递进动态设计逃生路径,解决了因井下复杂环境和不同时刻的突水状况不同而无法实时规划最优逃生路线的难题。

县级行政区洪水风险区划与防治区划案例研究

研究洪水风险区划与防治区划方法在不同地区的适用性具有重要科学意义与实用价值。以江苏省徐州市铜山区为对象,采用洪水风险区划与防治区划编制技术要求中的方法确定,洪水模拟计算方案,计算综合风险度FRI值,绘制洪水风险区划图,最后绘制洪水灾害防治区划图。研究结果表明,洪水风险区划符合铜山区洪水危险性空间分布实际情况,洪水灾害防治区划与铜山区水利治理规划重点任务方向一致。

基于PGIS和水文-淹没模型的社区洪水风险分析——以浙江省台州市大田河流域为例

以浙江省大田河流域为案例,基于参与式地理信息系统(PGIS)开展野外调查,构建居民建筑及其室内财产的脆弱性曲线;利用HEC-HMS水文和HAND淹没模型,模拟6种洪水淹没情景,开展承灾体建模、暴露与潜在损失分析,从而实现社区尺度的多情景洪水风险分析。结果表明,基于PGIS的问卷调查和家庭访谈可有效收集社区本地知识并辅助完成风险分析。从2年一遇到500年一遇洪水,受淹建筑足迹由6.72%增至61.44%,总的暴露资产由23.24亿元增至222.48亿元;结合脆弱性分析得出,洪水潜在损失由3.86亿元上升为54.19亿元。社区的洪水年期望损失为4.89亿元,有必要采取综合性防洪适应措施,以应对未来可能加剧的洪水灾害风险。该研究可为开展社区参与式的洪水风险分析提供借鉴。

基于元胞自动机的实时降雨下流域洪水淹没过程分析

目前实时降雨下流域洪水淹没过程分析复杂,难以满足实际工程应用要求。虽然目前推广性较强的无源淹没分析方法计算简便,但是其实际物理意义不强。为此,基于实时降雨预报和地理信息系统(GIS)技术,建立了基于元胞自动机的洪水淹没分析模型,借助GIS技术生成数字高程模型格网数据,并结合流域下垫面条件求解洪水淹没范围及淹没水深。最后,以江西省宜春市袁州流域为例,考虑植被截留量和土壤下渗量这两种降雨损耗,利用径流系数表示降雨产生径流的损耗过程,进而采用建立的元胞自动机洪水淹没分析模型计算得到4个不同时刻(10 min、30 min、1 h和2 h)的袁州流域洪水淹没数据。同时,利用无源淹没分析方法和基于MIKE SHE的坡面流模型计算结果验证了建立的元胞自动机洪水淹没分析模型的有效性。结果表明:相比于无源淹没分析方法和基于MIKE SHE的坡面流模型,本文建立的模型可以充分考虑流域下垫面条件以及降雨产汇流过程的影响,更加合理地模拟洪水淹没过程,并能够动态输出不同时刻洪水淹没数据。所建模型具有应用推广性强、实际物理意义明确、计算过程较为简便,并且可准确地显示实时降雨下洪水淹没过程等特点。研究成果可为强降雨作用下洪水灾害风险评估、防汛抢险决策方案的制定提供理论参考。