梯级水库洪灾风险分析理论方法研究

发布时间：2018-06-22 06:08

  本文选题：梯级水库群 + 最大可能降雨　； 参考：《大连理工大学》2015年博士论文

【摘要】：梯级水库群受到不同风险源的综合影响,其风险源复杂,灾害链长,影响程度大。从流域整体安全性的角度出发,研究分析梯级水库群系统的洪灾风险问题显得尤为重要。一个流域梯级式开发的水利工程,是空间中一系列单元组成的系统,各单元之间相互联系,上游水库失效将会对相邻的下游水库带来一定程度的影响,从而对整个系统的安全性造成影响。本文以此为切入点,以大渡河流域梯级水库群系统为例,研究梯级水库群系统的洪灾风险分析问题。主要研究内容及成果如下：(1)构建梯级水库群多源风险源分析的理论框架,依据风险分析的主要流程,建立了包括风险源的识别、风险的量化和估计及风险评价的理论体系。在风险识别方面,通过分析梯级水库群的结构特征,认为梯级水库群在运行中主要受到自然、工程两类风险源的影响。本文主要分析在这两种风险源的影响下,系统整体的洪灾风险问题。分别从物理成因和概率统计的角度识别系统中的薄弱性工程。并在分析溃坝洪水的基础上,对系统中水库原设计参数提出合理的建议。在风险量化和估计方面,采用水库失效风险率模型分析单元水库的失效风险率,建立以贝叶斯网络为理论基础的梯级水库群系统失效风险率模型,量化系统中各水库及其整体在原设计参数和建议设计参数情况下失效风险的后验概率。在风险评价方面,通过计算两种情况下系统的生命风险损失,评价系统的风险是否在可接受范围内。通过以上理论体系的构建,对全文的研究提供支撑。(2)分析系统中各水库工程等级、防洪标准等设计参数,以及各水库设计洪水的计算方法,发现设计洪水均是根据其控制水文站的设计洪水成果利用面积内插法求得。仅根据各水库的设计参数,无法指定系统中的薄弱性工程。为了从流域整体的角度考虑各水库的状态,以及在上游水库发生溃坝风险时,下游各水库的安全性。通过分析大渡河自然地理特征和气候特性,依据流域中主雨区的不同,分为以泸定到瀑布沟区间降雨为主(A7)和以泸定以上降雨为主(B7)两种情景。利用水文气象学的方法计算出大渡河梯级系统中各研究水库的最大可能降水(PMP),并转化为相应的最大可能洪水(PMF)。(3)识别系统的薄弱性工程,从物理成因的角度,分别在两种降雨情境下识别系统中结构性和功能性失效状态的水库。从数理统计的角度,在计算得到各单元水库失效风险率的基础上,以现行规范和特级水库风险标准判定系统中结构性和功能性两种失效状态的水库。将两个角度的识别结果对比分析,最终确定系统中的薄弱性工程。考虑到土石坝漫顶易溃的原因,利用溃坝分析软件,计算系统中水库漫顶溃坝,溃坝洪水演进至下游后,若造成水库连溃的情况,此时系统中各水库的状态。并以流域安全为目标,对系统中水库的设计参数提出合理的建议。在梯级水库群系统中,薄弱性工程是触发风险源,一旦系统中发生连溃事件,必须保证控制性工程的安全性,使其能够分担和消纳由于上游水库失效对系统造成的风险,起到阻断风险的作用。(4)针对梯级水库群系统中单元水库失效风险量化评价问题,本文分别分析三个方面的内容：单元水库超标准洪水洪峰序列的随机分布特征,调洪最高水位序列的随机分布特征,以及单元水库的失效风险率。基于统计学中极值理论,分别以校核洪峰流量、校核洪水位、坝顶高程为阈值,研究大渡河流域梯级水库的超标准(超校核)洪峰流量随机分布特征和各水库的调洪最高水位的随机分布特征。结果表明：大渡河流域梯级水库的超标准(超校核)洪峰流量与P-III型曲线拟合较好,符合我国大部分地区洪水要素的分布规律；各水库的调洪最高水位与对数函数拟合较好。(5)在梯级水库系统中单元水库的失效概率分析方面,依据梯级水库群中各梯级的开发次序,将大渡河干流的梯级水库根据不同的建设时期分为三个阶段,利用蒙特卡洛模拟方法分别计算各单元水库的失效风险率。将三个阶段中各单元水库的失效风险率和已识别出的流域中的薄弱性工程对比分析,结果表明,为了保证流域长期的安全,需将系统中各单元水库依据其对流域安全的影响,分批次建设于流域中。同时,所得到的各单元水库失效风险率将为下一章搭建贝叶斯网络计算系统失效概率提供数据支撑。(6)由于梯级系统的物理结构特性,系统中各单元水库之间也存在着一定的相关性,系统中各单元水库的失效是一个条件概率事件。鉴于贝叶斯网络在不确定知识的表达、因果推理等方面有突出的优点,本文建立基于贝叶斯网络的梯级系统失效风险率分析模型。根据流域中的控制性工程,将系统划分为巴拉-双江口、猴子岩-瀑布沟两段(D1、D2)。利用贝叶斯网络分别分析D1、D2段系统在原设计参数和建议设计参数两种情况下的失效风险率。同时,建立各单元水库的状态对系统产生影响的贝叶斯网络模型。结果表明,在不同情景下,各段系统中,建议设计参数的情况下与原设计参数相比,失效风险率明显降低。各段系统的失效概率和其控制性工程的失效概率较为接近。本文分别从特级水库的风险控制标准和生命风险损失的可接受程度两个方面评价梯级水库群系统失效风险,结果表明,大渡河流域梯级水库系统在建议设计参数情况下,系统失效风险有显著降低并且在可接受的范围内。
[Abstract]:The cascade reservoir group is affected by different risk sources, its risk source is complex, the disaster chain is long and the influence degree is great. It is particularly important to study and analyze the flood risk of cascade reservoir group system from the point of view of the overall safety of the basin. The water profit project developed by a cascade cascade is a system of a series of units in the space. Each unit is connected with each other. The upstream reservoir failure will bring a certain degree of influence on the adjacent downstream reservoirs, thus affecting the safety of the whole system. This paper takes this as a breakthrough point and takes the cascade reservoir group system of Dadu River Basin as an example to study the flood risk analysis of the cascade reservoir groups. The results are as follows: (1) the theoretical framework of multi source risk source analysis of cascade reservoirs is constructed. According to the main flow of risk analysis, a theoretical system including the identification of risk sources, quantification and estimation of risk and risk assessment is established. In the aspect of risk identification, the structure characteristics of cascade reservoirs are analyzed, and the cascade reservoirs are considered to be in operation. It is mainly influenced by the two types of risk sources of nature and engineering. This paper mainly analyzes the overall flood risk of the system under the influence of the two types of risk sources, and identifies the weak projects in the system from the perspective of physical causes and probability statistics. And on the basis of the analysis of the dam break flood, the original design parameters of the reservoir are put forward reasonable. In the aspect of risk quantification and estimation, the reservoir failure risk rate model is used to analyze the failure risk rate of the unit reservoir, and the failure risk rate model of cascade reservoir group system based on Bayesian network is established to quantify the failure risk of each reservoir and its whole in the original design parameters and the proposed design parameters. In the aspect of risk assessment, by calculating the life risk loss of the system in two cases, the risk of the system is evaluated in the acceptable range. Through the construction of the above theoretical system, the research provides support for the full text. (2) analysis of the design parameters of various reservoir engineering grades, flood control standards, and the design flood of each reservoir. It is found that the design flood is calculated according to the area interpolation method based on the design flood results of the hydrologic station. Only according to the design parameters of each reservoir, the weak engineering in the system can not be specified. In order to consider the state of the reservoir from the point of view of the whole basin and the risk of dam break in the upstream reservoir, the downstream reservoir By analyzing the natural geographical features and climatic characteristics of the Dadu River, according to the difference of the main rain area in the valley, two scenarios are divided from Luding to the waterfall gully (A7) and the rainfall dominated by Luding (B7). The maximum possible precipitation of each study reservoir in the Dadu River cascade system is calculated by hydrographic Meteorology (P MP), and converted to the corresponding maximum possible flood (PMF). (3) the weak engineering of the identification system, from the point of view of physical origin, to identify the structural and functional failure states of the system in the two rainfall situations respectively. Two reservoirs with structural and functional failure in the evaluation system of the special reservoir risk standard, the results of the identification of the two angles are compared and analyzed, and the weak engineering in the system is finally determined. Considering the cause of the overtopping of the earth dam, the dam break dam in the calculation system is calculated by the dam break analysis software, and the dam break flood will evolve to the lower reaches. After that, if the reservoir collapse is caused and the state of each reservoir in the system is at this time, the design parameters of the reservoir in the system are put forward reasonably. In the cascade reservoir group system, the weak project is the trigger of the risk source. Once the system occurs, it must ensure the safety of the controlled engineering and make it possible. Enough to share and eliminate the risk caused by the upstream reservoir failure to the system. (4) according to the quantitative evaluation problem of the failure risk of the unit reservoir in the cascade reservoir group system, this paper analyzes the contents of three aspects respectively: the random distribution characteristics of the super standard flood peak sequence of the unit reservoir and the maximum water level sequence of the flood control The random distribution characteristics of the reservoir and the failure risk rate of the unit reservoir. Based on the statistical extreme value theory, the random distribution characteristics of the flood peak flow random distribution characteristics of the cascade reservoirs in Dadu River Basin and the maximum flood water level of each reservoir are studied by checking the flood peak flow, checking the flood water level and the height of the dam as the threshold. The results show that the super standard (super calibration) flood peak flow of the Dadu River Basin and the P-III type curve fit well, which is in line with the distribution law of flood factors in most areas of our country; the maximum water level of the reservoir is fitted well with the logarithm function. (5) the failure probability analysis of the unit reservoir in the cascade reservoir system is based on the ladder. The cascade reservoirs in the reservoir group are divided into three stages according to the different construction period. The failure risk rate of each unit reservoir is calculated by Monte Carlo simulation method. The failure risk rate of each unit in the three stages is compared with the weak project in the identified basin. The results show that, in order to ensure the long-term safety of the river basin, each unit reservoir in the system should be built in the basin according to its impact on the safety of the basin. At the same time, the failure risk rate of each unit reservoir will provide data support for the failure probability of the next chapter of the Bayesian network computing system. (6) due to the cascade system There is also a certain correlation between the various units of the reservoir in the system, and the failure of the various units in the system is a conditional probability event. In view of the outstanding advantages of the Bayesian network in the uncertain knowledge expression and causality reasoning, the failure risk rate analysis of the cascade system based on Bayesian network is established in this paper. The system is divided into Bala double Jiangkou, monkey rock and waterfall gully two sections (D1, D2) according to the control engineering in the basin. The failure risk rate of D1, D2 segment system in the original design parameters and the proposed design parameters is analyzed by Bias network respectively. At the same time, the influence of the state of each unit reservoir on the system is built. The results show that, under different scenarios, the failure probability of the proposed design parameters is significantly lower than that of the original design parameters in the various segments of the system. The failure probability of each section of the system and the failure probability of its controlled engineering are close. The failure risk of cascade reservoirs system is evaluated by two aspects. The results show that the system failure risk of cascade reservoir system in Dadu River Basin is significantly reduced and within acceptable range under the proposed design parameters.
【学位授予单位】：大连理工大学
【学位级别】：博士
【学位授予年份】：2015
【分类号】：TV697.13
，

本文编号：2051887