Rapid assessment of flood loss based on neural network ensemble

Considering the defects of low accuracy and slow speed existing in traditional flood loss assessment, firstly, the technical route of flood loss assessment was presented based on the neural network ensemble. Secondly, through the study of certain country of Poyang Lake district, the flood loss assessment indicators of the test area were analyzed and extracted by utilizing analytic hierarchy process (AHP), and the weights of the impact factors were assigned. Subsequently, the approaches to generate individuals and conclusions of neural network ensemble model were also investigated. In the platform of C# language and neural network library under AForge.NET open source, a flood loss assessment program which could rapidly build neural network ensemble models was developed. Finally, the proposed method was tested and verified. The comparison results between the assessment results of the proposed method and the actual statistical flood loss proved the feasibility of this method, thus a new approach for flood loss assessment was provided.

Flood Risk Analysis and Flood Potential Losses Assessment

The heavy floods in the Taihu Basin showed increasing trend in recent years. In this work, a typical area in the northern Taihu Basin was selected for flood risk analysis and potential flood losses assessment. Human activities have strong impact on the study area’s flood situation (as affected by the polders built, deforestation, population increase, urbanization, etc.), and have made water level higher, flood duration shorter, and flood peaks sharper. Five years of different flood return periods [(1970), 5 (1962), 10 (1987), 20 (1954), 50 (1991)] were used to calculate the potential flood risk area and its losses. The potential flood risk map, economic losses, and flood-impacted population were also calculated. The study’s main conclusions are: 1) Human activities have strongly changed the natural flood situation in the study area, increasing runoff and flooding; 2) The flood risk area is closely related with the precipitation center; 3) Polder construction has successfully protected land from flood, shortened the flood duration, and elevated water level in rivers outside the polders; 4) Economic and social development have caused flood losses to increase in recent years.

A High Precision Comprehensive Evaluation Method for Flood Disaster Loss Based on Improved Genetic Programming

Precise comprehensive evaluation of flood disaster loss is significant for the prevention and mitigation of flood disasters. Here, one of the difficulties involved is how to establish a model capable of describing the complex relation between the input and output data of the system of flood disaster loss. Genetic programming (GP) solves problems by using ideas from genetic algorithm and generates computer programs automatically. In this study a new method named the evaluation of the grade of flood disaster loss (EGFD) on the basis of improved genetic programming (IGP) is presented (IGP-EGFD). The flood disaster area and the direct economic loss are taken as the evaluation indexes of flood disaster loss. Obviously that the larger the evaluation index value, the larger the corresponding value of the grade of flood disaster loss is. Consequently the IGP code is designed to make the value of the grade of flood disaster be an increasing function of the index value. The result of the application of the IGP-EGFD model to Henan Province shows that a good function expression can be obtained within a bigger searched function space; and the model is of high precision and considerable practical significance. Thus, IGP-EGFD can be widely used in automatic modeling and other evaluation systems.

Assessing Land Loss from Flooding in the Lake St. Martin Basin in Manitoba, Canada

Floodwaters from the Assiniboine River are typically diverted to Lake Manitoba and then Lake St. Martin to save the City of Winnipeg and Portage la Prairie from flooding. The four Indigenous communities living Lake St. Martin basin experienced many negative impacts, including long-term displacement after the 2011 flood. Data analysis of historical water levels of Lake Manitoba and Lake St. Martin were conducted to understand the impact of control structures and water levels. Satellite imagery allowed geographic information system (GIS) raster analysis of the shoreline change of Lake St. Martin before and after the 2011 super flood. From 1986 to 2010, the Lake St. Martin area increased slightly by approximately 0.63% but in 2011 the lake area increased by 13%. 11,000 acres were lost mainly around Lake St. Martin First Nation (LSMFN) reserve and the Little Saskatchewan First Nation, as a result and many houses and other buildings (churches, band offices, etc.) flooded in these communities. The shoreline change analysis showed that, in particular, the west and north shoreline of Lake St. Martin lost its beaches and lakefront properties, to become swampland after the 2011 flood. Thousands of Indigenous people were displaced for eight years, as the community could only start to rebuild its infrastructure and houses on higher ground after a government settlement was reached.

考虑时间变化的洪涝灾害损失评估

洪涝灾害是对我国影响最大的自然灾害之一。聚焦灾害损失评估问题,本文构建了考虑时间变化的洪涝灾害损失函数。一方面,表征灾害损失随着致灾因子强度先递增、而后增幅趋缓的静态S型曲线关系;另一方面,通过量级、形状和位置三个参数综合考虑灾害损失随时间的动态变化情况。基于《中国水旱灾害防御公报》提供的受灾人口、直接经济损失数据,对于各省级行政区2006—2021年洪涝灾害损失进行评估分析。结果表明:通过拟合洪涝灾害损失函数,洪涝灾害损失关于致灾因子强度和时间的三维空间散点图可有效地拓展成为连续的三维空间曲面及二维空间热力图。各省级行政区受灾人口整体上呈现出逐年下降的特征,但随致灾因子强度的增加而趋于增加。各省级行政区直接经济损失则表现出明显的差异,随时间呈现出大幅下降、缓慢下降和略有增加等不同特征。考虑时间变化的洪涝灾害损失函数是变化环境下灾害评估的有效工具,可为区域洪涝灾情评估提供借鉴和参考。

Life Loss Estimation Based on Dam-Break Flood Uncertainties and Lack of Information in Mountainous Regions of Western China

Compared with urban floods, dam-break floods are associated with greater uncertainties, including variable dam-break modes and hydrological characteristics, so conventional flood estimation methods cannot be directly applied in the estimation of dam-break flood loss. In particular, there is scant information regarding the conditions of affected area and hydrological characteristics in southwest China. In this paper, we introduce an integrated model for estimating flood loss that is adapted to the mountainous regions of southwestern China in light of the relative lack of available information. This model has three major components: a basic information model, a dam-routed flood propagation simulation model, and a loss estimation model. We established the basic information model despite the relative lack of available information using 3S technology [remote sensing (RS); geographical information system (GIS); global positioning system (GPS)], data mining technology, and statistical analysis techniques. Our dam-routed flood propagation simulation model consists of major hydrologic processes and their governing equations for flow propagation, which we solve using finite-difference schemes. In this model, the flood propagation area is divided into grids and each grid is determined by the characteristic parameters obtained from the propagation simulation. We present a case study of the Lianghekou hydropower station in Sichuan Province, China to illustrate the practical application of this integrated model for life loss estimation. © 2017, Tianjin University and Springer-Verlag Berlin Heidelberg.

The Analysis on the Temporal and Spatial Variation of Strong Precipitation Caused Flood and Agricultural Disaster Loss in Huaihe River Basin of Anhui Province during Meiyu Period of 2007

[Objective] The research aimed to analyze temporal and spatial variation of strong precipitation caused flood and agricultural disaster loss in Huaihe River basin of Anhui Province during Meiyu period of 2007.[Method] On the basis of rainfalls of each station in Huaihe River basin of Anhui,rainfall data during Meiyu period of 2007 and flood disaster data in the same period,the temporal and spatial distribution characteristics of strong precipitation caused flood during Meiyu period of 2007 and its harm on agriculture were analyzed.The variation rule,distribution characteristics of strong precipitation during Meiyu period in Huaihe River basin of Anhui and its relationship with agricultural disaster loss were discussed.[Result] During Meiyu period of 2007 in Huaihe River basin of Anhui,the rainstorm was more,and the rainfall was large.The precipitation variation showed 'three-peak' trend.Rainfall in Huaihe River basin during Meiyu period of 2007 was greatly more than that homochronously in Yangtze River basin.The rain area over 400.0 mm during Meiyu period mainly located in Huaihe River basin,and the rain area over 600.0 mm mainly located from area along Huaihe River to central Huaibei.The rainfall during Meiyu period gradually decreased toward south and north by the north bank of Huaihe River as the symmetry axis.The rainfall in area along Huaihe River showed wavy distribution in east-west direction.The flood disaster loss index and disaster area of crops in Huaihe River basin of Anhui both increased as rainfall in Meiyu period.[Conclusion] The research provided theoretical basis for flood prevention,disaster reduction and agricultural flood-avoiding development in Huaihe River basin.

考虑超标准洪水作用的沥青心墙坝溃决生命损失评估模型

抗冲刷能力更强的沥青心墙坝一旦遭遇超标准洪水,将面临溃决风险,对下游人民造成严重威胁。现有针对沥青心墙坝的溃决生命损失研究尚属匮乏,亟需开展生命损失评估。本文分析超标准洪水作用下沥青心墙坝的溃决特征,考虑生命损失影响因素之间的相互作用过程,基于溃决洪水致灾机制构建生命损失定量评估模型,实现溃决生命损失的快速定量分析;通过方法对比和敏感性分析,明晰不同影响因素对溃决生命损失的作用规律。将模型应用于新疆射月沟溃决案例发现:溃决洪峰流量为3320 m^(3)/s,溃决历时3.35 h,与实际结果相符;生命损失定量模型分析结果为31人,与实际误差在11%以内,优于其他损失模型;风险人口的敏感性指数I_(max)和I_(min)值最大,分别为1.436、0.486,是射月沟溃决生命损失最重要的影响因素,其他依次为洪水严重性、是否撤离、警报时间、淹没范围。

基于场次洪涝灾害的损失函数构建方法

构建洪涝灾害损失函数对于认识洪涝灾害损失的发展规律、提高灾害损失的评估时效性具有重要意义。提出了耦合洪涝致灾因子空间分布的动态模拟、社会经济数据空间化处理、灾害直接和间接损失计算以及曲线拟合的场次洪涝灾害损失函数构建方法,构建了洪涝灾害损失—时间函数,并建立了函数参数的调整策略,在上海市开展了实例研究。研究结果表明:场次洪涝灾害直接经济损失的发展过程呈“S型”曲线的变化特征,可将weibull分布函数作为其损失函数;减停产损失的日变化过程呈“L型”曲线的变化特征,可将双曲递减函数作为其损失函数;产业关联损失的日变化过程呈“倒U型”曲线的变化特征,可将Hoerl函数作为其损失函数;累计产业关联损失约为停产或减产损失的1.77倍,间接经济损失约为直接经济损失的1.29倍;损失函数参数与灾害发生概率多呈非线性的单调递增或递减关系,基于拟合参数的洪涝灾害损失评估相对误差在5%以内。本研究的方法被证实具有较高的模拟精度,为认识洪涝灾害损失的发展规律和提高洪涝灾害损失评估时效性提供了新途径。

上海市洪涝灾害损失对长江下游地区的辐射影响

随着区域经济一体化程度的加深,跨区域产业关联损失已成为灾害损失评估不容忽视的一部分,准确量化洪涝灾害损失的辐射影响对于制定防灾减灾措施、减少洪涝灾害损失具有重要作用。在系统评估了不同暴雨重现期情景下上海市减停产损失的基础上,采用两区域投入产出模型,测算和分析了长江下游地区不同省(直辖市)的区域内乘数效应、区域间溢出效应和区域间反馈效应,量化了上海市洪涝灾害损失对长江下游地区的辐射影响。结果表明,暴雨重现期10 a、20 a、30 a、50 a、100 a时,上海市的减停产损失分别为14.01亿元、19.83亿元、25.58亿元、33.82亿元和46.24亿元;在长江下游地区,上海市与浙江省的经济相互影响强度最大,且跨区域产业关联损失以溢出效应为主导;上海市发生洪涝灾害时,跨区域波及效应将导致长江下游地区损失增加26.24%。因此,在城市之间联系日益紧密的情况下,不仅要关注洪涝灾害对自身的影响,同时也要采取有效措施减少地域之间的波及影响,以便更有效地开展防灾减灾工作。

考虑人体与水流相互作用的溃坝洪水生命损失评估模型

溃坝洪水会给下游人民群众造成巨大的生命损失,开展溃坝洪水造成的生命损失风险定量评估对应急抢险救灾具有重要的现实意义。本文在已有生命损失贝叶斯网络HURAM1.0模型基础上,引入了人体稳定性物理模型,考虑人体与水流相互作用关系,对处于洪水中的人先进行稳定性判定,并进行溺水判定,运用蒙特卡洛模拟方法,综合了水深和水流速对生命损失的影响,建立HURAM2.0模型;并将该模型应用于唐家山堰塞坝溃坝洪水生命损失分析。结果表明:HURAM2.0模型建立了水流流速对生命损失影响的定量关系,更精确地刻画了人体在水流中的稳定性和求生能力,相比HURAM1.0模型对较强洪水强度条件下的生命损失预测结果更准确。同时,在本文建立的模型中,除水深度、洪水严重程度变化不大,其余变量的敏感性均有所上升,其中居民区住宅层数、在建筑物中庇护情况和溃坝时长等变量对模型计算结果的最大影响程度分别增加142%、95%和93%,加强了模型在低、中、高3类洪水强度下的解释性,与HURAM1.0相比在贝叶斯反演分析中更占优势。在唐家山堰塞坝溃坝风险分析中,HURAM2.0能区分出不同水流速条件下的生命损失,更符合实际情况,即开挖泄流槽前风险大、死亡率高,在现场勘测和开挖泄流槽后风险及死亡人数大大降低,建议结合预警疏散以降低生命损失风险。

周期性淹水-落干和施用秸秆对白云石改良棕壤磷素流失风险的模拟研究

农田土壤磷累积及潜在的磷素流失风险在中国受到广泛的关注。施用白云石是降低土壤磷流失风险的有效措施之一,但常见农田管理措施(秸秆还田、交替灌溉以及它们的组合措施等)如何影响白云石改良后土壤的磷素流失风险及其相关机制目前还不清楚。采用培养试验,探究施用秸秆、周期性淹水-落干、周期性淹水-落干和施用秸秆联合处理对白云石改良土壤活性磷含量、磷吸附/解吸特性、连续浸提磷组分和土壤理化性质的影响。结果表明:与仅施用白云石的对照(CK)相比,施用秸秆(S)和周期性淹水-落干处理(FW)使土壤不稳定磷(CaCl_(2)-P)分别增加62.3%和15.9%;并同时降低土壤磷最大吸附容量(6.26%-12.5%)和Na HCO_(3)-P_(i)(6.97%-19.2%)。秸秆本身含有的磷和易解吸态磷(Na HCO_(3)-P_(i))的释放分别是上述处理土壤活性磷增加的主要原因。周期性淹水-落干和施用秸秆联合处理(FWS)使土壤CaCl_(2)-P和Bray-P分别降低43.4%和23.2%。这一处理也显著提高了土壤非晶质铁质量分数(21.6%),增强土壤对磷的吸附,并抑制吸附态磷的解吸,进而降低土壤活性磷质量分数。此外,周期性淹水-落干和施用秸秆联合处理也显著提高了土壤Fe^(2+)(423%)和铁活化度(18.8%),促进了土壤Fe/Al-P向稳定态磷转化,使HCl-P增加26.9%。土壤磷吸附能力的提高、非晶质铁对活性磷的固定和稳定态磷(HCl-P)质量分数的增加是周期性淹水-落干和施用秸秆联合处理中磷流失风险降低的重要原因。结构方程模型表明,土壤非晶质铁和有机质是影响磷素流失的关键因素。研究表明,适当的淹水-落干措施配合秸秆还田能增强白云石改良土壤中磷素的固持,降低磷素流失风险,可为胶东半岛高磷土壤的管理提供科学依据。

甘肃省庆城县极端暴雨下水保措施抵御暴洪灾害的能力

[目的]最大洪峰流量是导致暴洪灾害的决定性因素。研究小流域降水频率、治理程度与洪峰流量、暴洪灾害损失的关联性,为水土保持减灾防灾工作提供科学依据。[方法]采用回归分析、对比分析等方法,对甘肃省庆阳市庆城县蔡家庙、野狐沟两条相邻流域1981—2022年水保措施的调水效益以及降水因子、水保措施对洪峰流量的影响进行分析。[结果](1)小流域水保措施调水效益随治理程度的提高而增加,随降水频率的提高而减少。(2)年最大日降水量对洪峰流量变化具有极显著的正向影响(p<0.01),贡献率系数为0.707;治理程度对洪峰流量呈负影响,贡献率系数为-0.237。(3)30 a一遇洪峰流量(频率为3%)与50 a一遇洪峰流量(频率为2%)灾害损失强度分别达到1.74×10^(5),3.13×10^(6)元/km^(2),后者比前者增长了16.95倍。[结论]影响洪峰流量变化的主要因素是年最大日降水量。水保措施对削减洪峰流量具有显著的影响。流域治理程度越高,削减洪峰流量的效果越明显,减少暴洪灾害经济损失的能力越大。

HEC-HMS水文模型不同降雨损失方法对比研究

为对比分析HEC-HMS模型三种降雨损失方法在沁河流域的适用性。借助Morris筛选法识别降雨损失方法的关键参数,选用流域内5场雨洪资料进行参数率定和模拟精度分析。结果表明:(1)SCS CN值曲线法、Green-Ampt法、Initial and Uniform法主要敏感性参数分别为CN值、土壤饱和导水率、恒定损失率。(2)选取洪峰流量、洪水总量、峰现时刻误差以及Nash系数对模型模拟精度进行评价,SCS CN值曲线法和Initial and Uniform法模拟结果达到乙等精度,Green-Ampt法模拟结果达到丙等精度。研究成果可为半湿润地区中小流域降雨损失方法的选择提供参考。

基于机器学习的洪涝灾害损失评估模型预测精度优化研究

随着国家整体防灾减灾能力的提升,洪涝灾害损失呈逐年波动下降趋势,但单次灾害损失波动幅度巨大,在遭遇农业大灾风险时,保险业承担能力有限。财政补贴与保险费率厘定需基于预期损失的精准评估,而基于传统灾害损失的评估精度较低,运用风险序贯性链式结构理论,结合机器学习提高损失评估精度。以河北省玉米洪涝损失评估为例,检验发现,以决策树、BP神经网络、XGBoost回归模型预测区域降水量表征的风险分级函数对SVM模型加以改进,显著提高了灾害损失评估模型精度,而随着国家防灾减灾整体能力的提升,灾害损失呈现波动下降趋势。

海相砂岩薄层稠油油藏地热水驱效果评价

南海东部薄层边水稠油油藏注地热水开发效果好,鲜有针对地热水驱提高采收率机理及开采效果方面的研究。本文从稠油注热实验、考虑围岩热损失的数值模拟以及矿场实践三个维度研究地热水驱提高稠油采收率机理及效果。研究结果表明,注地热水可以大幅降低原油黏度,残余油饱和度降低7.24%~8.22%,驱油效率提高8.85%~9.11%;注地热水油藏早期水驱前沿与周围地层发生热交换,温度迅速降低,地热水驱影响范围在50~150 m,以低速均匀注水保压驱油为主;后期通过大液量提高温度的波及范围,提高注热采收率。E油藏注地热水后采油井产量快速上升,注水井吸水指数因近井地带黏度降低、水相渗透率提高而逐步增大。通过完善注采井网,油藏压力系数逐年上升,产量持续增加,预测采收率达到40.0%,注热提高采收率4.8%。

基于损失函数的城市洪涝灾害弹性演化规律研究

为了提升城市防洪水平与洪涝灾害恢复能力以及推进以人为核心的新型城镇化战略的目标任务,以潍坊市城区为研究区域,借助基于元胞自动机的CAflood模型获取淹没水深数据,并通过数据挖掘与相似性分析方法直接移植引用城市损失率建立研究区淹没水深-损失率函数,构建基于损失函数的城市洪灾弹性模型方法并计算了潍坊城区在遭遇5、50、100及200 a重现期降雨时的内涝弹性指数分布。同时,分析研究区在2012、2017和2022年3种不同土地利用情况下,潍坊市城区洪涝灾害弹性的时空变化趋势,揭示潍坊城区城市演化与区域洪涝灾害弹性响应关系。结果表明:潍坊市城区洪涝灾害弹性随重现期增加总体呈下降的趋势,且下降剧烈区域主要集中在人造地表面积占比较大的中心城区、街道和东部开发区,以及河流周边平坦低洼区域。从空间分布上看,这种下降趋势呈现出以主城区为中心向外扩张的趋势。此外,随着城市化程度的增加,研究区各分区的平均弹性指数均有不同程度的降低,其中C7山丘区一直处于弹性较高水平,C4高新开发区在多种模拟情景下均处于较低水平,并且在遭遇相同重现期降雨后,城市化程度越高,各分区的平均弹性指数降幅越大。研究结果为潍坊市应对未来增长的洪涝风险,实施城市洪灾弹性提升策略提供了科学支撑。

湖泊型流域洪涝灾害经济损失多情景模拟——以南四湖流域为例

洪涝灾害情景模拟与损失预估对湖泊型流域防洪减灾和可持续发展具有重要意义。以我国典型的湖泊型流域南四湖流域为例,利用CA-Markov模型预测2030年三种用地情景,采用P-Ⅲ型概率曲线构建10、20、50、100 a四个重现期强降雨情景,得到12种组合情景;通过InVEST模型和等体积淹没法量化不同情景的径流和淹没深度;在各产业GDP空间预测的基础上计算12种情景的洪涝直接经济损失。结果表明:(1)CA-Markov模型在南四湖流域的土地利用模拟精度较高,三种用地情景的各地类变化显著;(2)用地变化对流域洪涝灾害的影响显著,其中城镇发展情景下洪涝灾害风险最大,而生态保护情景风险最低;(3)暴雨强度会明显增加各用地情景的经济损失,但增幅在50 a重现期后有所减缓;生态保护情景的经济损失最少,介于97.39—128.81亿元之间,与其他情景相比,该情景可在一定程度上减少洪灾损失。为此,在气候变化背景下湖泊型流域可持续发展应充分考虑土地利用变化对洪涝灾害风险的影响,合理扩张城镇建设用地,优化布局生态用地,充分发挥国土空间规划在防洪减灾方面的作用。

基于局部洪泛优化的低轨星座分布式路由算法

针对传统分布式路由算法应用于大型低轨星座时拥塞控制能力有限的问题,提出了基于局部洪泛优化的分布式路由算法。该算法通过局部洪泛的机制收集星间链路的拥塞信息,以应对局部通信需求密集所导致的常见拥塞现象;通过局部路径优化,以较低运行开销实现了流量疏导和网络负载均衡。并且该算法可适用于多层低轨卫星网络结构,不会产生路由环路,对卫星随机故障具有宽适应性。仿真实验表明:该算法在通信需求分布不均的高网络负载情况下能降低2%~10%的丢包率,并对网络随机故障具有鲁棒的表现。

氧化铟对铅酸蓄电池性能的影响

实验采用循环伏安法和线性伏安扫描法,在电化学工作站上研究了Pb电极和Pb-In电极。通过线性伏安扫描曲线图,分析了电极的析氢电位与析氢电流。得出初步结论后,将氧化铟作为添加剂加入负极铅膏中,做成实验富液式铅酸电池,并与常规富液电池进行对比测试。测试项目有初期性能、水损耗、充电接受等。实验结果表明,加氧化铟的电池整体性能优于常规电池,而且水损耗明显降低。