自回归总径流线性响应模型在洪水预报中的应用

采用自回归总径流线性响应(ATLR)模型建立淮河王家坝站洪水预报方案和海河流域南运河水系洪水预报调度方案.自回归TLR模型与TLR模型加误差自回归实时校正模型的结构相同,但前者计算简捷,预报精度也有所提高.     

洪水预报自回归实时校正多步外延方法研究

自回归模型在洪水预报实时校正中应用广泛。针对自回归模型进行连续多时段校正时中间误差系列缺失问题,提出一种基于历史洪水预报误差系列的样本重组自回归外延方法,以淮河流域王家坝断面为背景,选用洪量相对误差、洪峰相对误差、峰滞时间和确定性系数四个指标开展校正效果评估,并与时程递推外延方法对比。结果表明:样本重组外延方法可以提升洪水预报精度,延长洪水预报有效预见期,特别在降低洪量误差和提高洪水过程的拟合精度上优势更为显著。同时,该方法泛化能力较强,具有实用价值。     

实时洪水预报综合修正方法初探

利用误差的相似性扩大实时修正信息量,结合神经网络异联想记忆技术,提出了综合实时修正方法。并在七里街流域用11场洪水分别采用自回归(AR)模型和综合修正方法进行实时修正。分析结果表明,采用综合修正方法既能获得比AR模型更好的精度,又不损失洪水预报的预见期。进一步对闽江的6个支流流域近50场洪水进行综合修正方法实时修正检验,都获得了较好的结果。     

双校正模式下的大清河流域陆气耦合洪水预报研究

基于数值大气模式WRF、三维变分数据同化WRF-3DVar、河北雨洪模型以及实时校正模型ARMA,在北方半湿润半干旱地区的大清河流域构建了陆气耦合洪水预报系统,并利用2012、2013年发生的3场降雨洪水,对系统的降雨洪水预报结果进行分析。结果表明:雷达反射率与GTS数据的同时同化,可有效改善数值大气模式对中小尺度流域的降雨预报效果,从而降低系统的洪水预报误差,ARMA模型的应用,能够进一步提升系统的洪水预报精度,随着预见期的延长,系统的预报精度下降,但系统在6h预见期内仍表现出较好的应用效果。因此,在数据同化和实时校正的"双校正"模式下,陆气耦合洪水预报系统在延长洪水预报预见期的同时,具有较高的洪水预报精度,具有一定的应用前景。     

长江中下游实时洪水预报模拟

以长江中下游防洪系统为对象,概述了在大型复杂防洪系统水沙运动数值模拟基础上,成功地将面向长江中下游防洪规划论证需求的水沙数学模型转化为面向长江防洪系统防汛方案评估需求的长江中下游实时洪水预报数学模型.为适应实时预报调度快速、准确评估的要求,提出了基于水动力学的循环滚动计算模式和实时校正模式.实现了水文学实时校正方法与水动力学数学模型的耦合,建立了基于水动力学的实时校正模式和分洪溃口洪水预报模式.通过长江中下游防汛期间的试运行,较好地解决了洪水预报误差校正和分洪溃口后洪水预报等关键难题,为防汛方案的制定和实时洪水调度方案优化提供了技术支撑,主要成果已应用于长江中下游防汛调度方案中.     

河道洪水实时概率预报模型与应用

通过数据同化方法合理地将实时水文观测数据融入到洪水预报模型中,可提高洪水预报模型的实时性和精确度。选取沿程断面流量、水位和糙率系数作为代表水流状态的基本粒子,以监测断面实测水位数据作为观测信息,建立了基于粒子滤波数据同化算法的河道洪水实时概率预报模型。模型应用于黄河中下游河道洪水预报计算的结果表明,采用粒子滤波方法同化观测水位后,不仅可以直接校正水位,同时也可以有效地校正流量和糙率,为未来时刻模型预报计算提供更准确的水流初始条件和糙率取值区间,进而有效地提高模型预报结果的精度,给出合理的概率预报区间。不同预报期的预报结果表明,随着预报期的增长,同化效果减弱,模型预报结果的精度会有所降低,水位概率预报结果受粒子间糙率不同的影响不确定性增加,而流量概率预报结果受给定模型边界条件的影响不确定性降低。所提出模型可以有效同化真实水位观测数据,适合应用于实际的洪水预报工作中。     

人工神经网络模型与新安江模型的应用比较

新安江模型是我国研制的经典的概念性流域水文模型,其参数往往通过人工试错法进行率定,在洪水的实时预报中需建立误差自回归模型来修正预报值.人工神经网络模型是一种数据驱动模型,它可以通过算法调节权值和偏王值来模拟信息,实现了模型参数的自动率定.在实时预报的应用中,人工神经网络模型可以根据计算误差调节权值和偏置值,反映水文过程的时变性,模型结构显得更加简洁.本文将两种模型应用于潢川流域并作比较.它们的预报结果都达到了作业预报要求.在实际应用中可以根据资料情况选择模型进行洪水预报.     

基于遗忘因子的BP神经网络水文实时预报方法

在应用神经网络进行洪水预报时,因洪水系统随着河道上游来流、区间降雨、河床演变等因素的动态变化,其特性并不总是按照基本相同的规律变化,对这类系统的参数辨识,要求算法具有较强的实时跟踪能力,以适应模拟或预测洪水运动变化过程的要求。在BP神经网络模型的基础上,运用最小二乘递推算法,引入时变遗忘因子实时跟踪模型中时变参数的变化,建立了神经网络在非线性系统中动态系统输入、输出数据间的映射关系。计算实例表明:该法对参数的快速时变具有较快的跟踪能力和较高的辨识精度,是一种非常实用的水文实时预报方法。     

全过程联合校正方法在饶河流域洪水预报中的应用研究

为提高饶河流域洪水预报精度,将全过程联合校正(EPJC)方法与三水源新安江(XAJ)模型结合,按一定比例划分洪水预报总误差为各过程误差,基于系统响应理论反演得到面雨量计算误差和模型参数误差,重新输入流域水文模型正演得出校正后的洪水过程.选取洪峰流量相对误差、峰现时间绝对误差、径流深相对误差和确定性系数作为模型评价指标,...     

新安江模型前期土湿的修正（校正）算法

概念性流域水文模型的产流量，尤其是产流总量对前期土湿十分敏感。本文研究新安江模型前期土湿的修正（校正）算法。这种算法可直接应用于模型率定中次洪资料的前期土湿修正，从而保证了总水量平衡。在实时预报中，采用该法对上次已发生洪水的前期土湿进行修正，则洪水末期土湿得到校正，为下场洪水预报提供了更为精确的前期土湿，从而提高了预报精度。该算法在实践中已取得了较好的效果。     

中国洪水预报系统设计建设研究

分析了我国洪水预报系统的现状,阐述了中国洪水预报系统的建设的必要性、目标及开发原则,介绍了中国洪水预报系统的软硬件环境、预报模型和方法、预报方案、模型率定、实时作业预报、实用模块、系统管理7部分的设计建设。     

The sensitivity to the microphysical schemes on the skill of forecasting the track and intensity of tropical cyclones using WRF-ARW model

The Advanced Research WRF (ARW) model is used to simulate Very Severe Cyclonic Storms (VSCS) Hudhud (7–13 October, 2014), Phailin (8–14 October, 2013) and Lehar (24–29 November, 2013) to investigate the sensitivity to microphysical schemes on the skill of forecasting track and intensity of the tropical cyclones for high-resolution (9 and 3 km) 120-hr model integration. For cloud resolving grid scale (<5 km) cloud microphysics plays an important role. The performance of the Goddard, Thompson, LIN and NSSL schemes are evaluated and compared with observations and a CONTROL forecast. This study is aimed to investigate the sensitivity to microphysics on the track and intensity with explicitly resolved convection scheme. It shows that the Goddard one-moment bulk liquid-ice microphysical scheme provided the highest skill on the track whereas for intensity both Thompson and Goddard microphysical schemes perform better. The Thompson scheme indicates the highest skill in intensity at 48, 96 and 120 hr, whereas at 24 and 72 hr, the Goddard scheme provides the highest skill in intensity. It is known that higher resolution domain produces better intensity and structure of the cyclones and it is desirable to resolve the convection with sufficiently high resolution and with the use of explicit cloud physics. This study suggests that the Goddard cumulus ensemble microphysical scheme is suitable for high resolution ARW simulation for TC’s track and intensity over the BoB. Although the present study is based on only three cyclones, it could be useful for planning real-time predictions using ARW modelling system.     

一个稳健的河网水情数据同化耦合模型

为解决河网水情动态系统的数据同化问题,将水情量测数据作为内边界条件,建立了一个基于误差最小二乘原理的河网水情数据同化耦合模型.同时,为了有效控制水情状态变量的校正程度,避免因校正过度而破坏河网系统的水量平衡关系,构建了一个能有效控制及平衡各量测变量校正程度的权重矩阵.通过模拟算例和实例应用,系统检验了模型在河网水情仿真与预报中的实时校正能力,并得到了理想的结果.结果表明:模型理论完备、构建简便、数据同化可控性好,能够有效地进行河网水情动态系统的数据同化,提高河网水情仿真与预报的精度.     

A Probability-Based Renewal Rainfall Model for Flow Forecasting

In real-time flood warning systems, sufficient lead-time is important for people to take suitable actions. Rainfall forecasting is one of the ways commonly used to extend the lead-time for catchments with short response time. However, an accurate forecast of rainfall is still difficult for hydrologists using the present deterministic model. Therefore, a probability-based rainfall forecasting model, based on Markov chain, was proposed in this study. The rainfall can be forecast one to three hours in advance for a specified nonexceeding probability using the transition probability matrix of rainfall state. In this study, the nonexceeding probability, which was hourly updated on the basis of development or decay of rainfall processes, was taken as a dominant variable parameter. The accuracy of rainfall forecasting one to three hours in advance is concluded from the application of this model to four recording rain gauges. A lumped rainfall-runoff forecasting model derived from a transfer function was further applied in unison with this rainfall forecasting model to forecast flows one to four hours in advance. The results of combination of these two models show good performance with agreement between the observed and forecast hydrographs.     

Development of roughness updating based on artificial neural network in a river hydraulic model for flash flood forecasting

Flood is the worst weather-related hazard in Taiwan because of steep terrain and storm. The tropical storm often results in disastrous flash flood. To provide reliable forecast of water stages in rivers is indispensable for proper actions in the emergency response during flood. The river hydraulic model based on dynamic wave theory using an implicit finite-difference method is developed with river roughness updating for flash flood forecast. The artificial neural network (ANN) is employed to update the roughness of rivers in accordance with the observed river stages at each time-step of the flood routing process. Several typhoon events at Tamsui River are utilized to evaluate the accuracy of flood forecasting. The results present the adaptive n-values of roughness for river hydraulic model that can provide a better flow state for subsequent forecasting at significant locations and longitudinal profiles along rivers.     

Real-time flood forecasting of the Tiber river in Rome

An adaptive, conceptual model for real-time flood forecasting of the Tiber river in Rome is proposed. This model simulates both rainfall-runoff transformations, to reproduce the contributions of 37 ungauged sub-basins that covered about 30% of the catchment area, and flood routing processes in the hydrographic network. The adaptive component of the model concerns the rainfall-runoff analysis: at any time step the whole set of the model parameters is recalibrated by minimizing the objective function constituted by the sum of the squares of the differences between observed and computed water surface elevations (or discharges). The proposed model was tested through application under real-time forecasting conditions for three historical flood events. To assess the forecasting accuracy, to support the decision maker and to reduce the possibility of false or missed warnings, confidence intervals of the forecasted water surface elevations (or discharges), computed according to a Monte Carlo procedure, are provided. The evaluation of errors in the prediction of peak values, of coefficients of persistence and of the amplitude of confidence intervals of prediction shows the possibility to develop a flood forecast model with a lead time of 12 h, which is useful for civil protection actions.     

延河流域雨洪特性及洪水预报

延河是黄河的一级支流，流经安塞、延安、延长等陕北南部地区的重要城镇，是引发该地区城镇洪水的主要河流，因此，延河洪水预报对这些城镇的防洪安全具有重要意义。在分析延河流域雨洪特性的基础上，以提高大洪水的预报精度为主、适当兼顾中小洪水的原则，采用具有成因概念的系统模型和相关图法，建立了以延安市为重点、包括安塞及延长县城的延河区段洪水预报方案，可供作业预报试用。     

Prévision des crues éclair

Flash-flood events resulting from paroxystic meteorological events concentrated in time and space are insufficiently documented as they produce destructive effects. They are hardly measurable and present single features that are not transposable to another event. In the South of France, the flash flood of November 1999 gives a perfect illustration of these characteristics. The physical complexity of the process and consequently the volume and the variety of the data to take into account are incompatible with the real time constraint allocated to the forecasters confronted to the occurrence of such phenomena. So, we have to make choices to afford acceptable simplifications to the complete mechanical model. MARINE (‘Modélisation de l'Anticipation du Ruissellement et des Inondations pour des évéNements Extrêmes’) is the operational and robust tool we developed for flash-flood forecasting. This model complies with the criterions of real-time simulation. It is a physically based distributed model composed of two parts: first the flood runoff process simulation in the upstream part of the basin modelled from a rainfall–runoff approach, then the flood propagation in the main rivers described by the Saint-Venant equations. It integrates remote sensed data – Digital Elevation Model, land-use map, hydrographic network for the observations from satellites and the rainfall evolution from meteorological radar. The main goal of MARINE is to supply real time pertinent information to the forecasters. Results obtained on the Orbieu River (Aude, France) show that this model is able to supply pertinent flood hydrograph with a sufficient precision for the forecasting service to take the appropriate safety decisions. Furthermore, MARINE has already been tested in the French National Flood Forecasting Service of Haute-Garonne in real conditions. To cite this article: V. Estupina Borrell et al., C. R. Geoscience 337 (2005).     

辽河福—铁区间洪水预报模型的建立与应用

根据辽河福德店～铁岭区间（水库以下）集水面积产、汇流以及干流各预报站资料,建立多种模型结构、多种预报方法比较、综合滚动式实时的洪水预报模式,利用这些模型对流域自上而下在预报和预警条件下实现对各站点多种预报方案所得的结果进行比较和优选,结果表明,该预报模型在实时状态下能够对全流域各种条件、各种调度手段及其组合快速地生成多个可行的调度方案。