样本非一致性条件下洪水的分析与计算

发布时间：2018-05-29 16:46

本文选题：非一致性 + 下垫面变化　；参考：《天津大学》2014年硕士论文

【摘要】：大清河流域由于兴建大量防洪和水保工程，导致流域下垫面发生显著变化，实测水文序列（洪水、径流等）不再满足一致性假定，基于传统的水文频率分析法在下垫面变化条件下得到的设计成果应用到工程水文设计中存在一定的风险。为研究下垫面变化对设计洪水及水库调洪的影响，探讨下垫面变化条件下水库设计洪水的推求及调洪计算等问题，本文以大清河南支沙河上游王快水库为例，采用概率论与数理统计方法直接对水库非一致性洪水序列进行频率分析计算；基于非一致性洪水序列设计值推求了下垫面变化条件下水库的设计洪水，并进行水库调洪计算，给出了下垫面变化对水库设计洪水及水库调洪的定量影响，以期为下垫面变化条件下大清河流域设计洪水的修订及防洪调度提供技术支撑。主要研究内容与成果如下：（1）基于传统的水文序列变异诊断理论，提出了非一致性水文序列变异点综合诊断方法，分为初步诊断、详细诊断、综合诊断三个步骤。其中，在详细诊断中，提出了基于MWP检定的水文序列变异区间诊断方法，该法结合其他方法的诊断结论，可方便确认变异点的具体位置。通过对王快水库各洪水序列进行变异点综合诊断，表明各序列均发生变异，变异点均为1979年。（2）采用混合分布及条件概率分布直接对王快水库各非一致性洪水序列进行分布拟合，以K-S检验及AIC准则选择各序列的最优分布。基于最优分布拟合结果，计算王快水库非一致性序列单变量分布洪水设计值，运用同频率放大法对水库1963年典型洪水过程进行放大，得到基于非一致性序列单变量分布法的设计洪水过程线。与水库不考虑下垫面变化（不考虑变异）的原序列P-Ⅲ曲线拟合设计成果对比表明，洪峰、年最大1日洪量、年最大6日洪量设计值减小幅度均小于10%，年最大3日洪量设计值减小幅度为1%-15%左右。（3）考虑峰量相关性，基于水库年洪峰序列、年最大6日洪量序列的边缘分布结果，运用Copula函数构建二者的联合分布，对两变量联合重现期、两变量同现重现期及特定条件下的洪峰、洪量条件频率进行了分析，计算了非一致性洪水序列两变量联合分布设计值，采用变倍比放大法，推求了基于非一致性序列两变量联合分布的设计洪水过程线。与非一致性序列单变量分布对应的设计成果及不考虑变异的原序列P-Ⅲ曲线拟合设计成果对比表明，非一致性序列两变量联合分布洪水设计值均大于非一致性序列单变量分布洪水设计值；基于非一致性洪水序列两变量联合分布法得到最大6日洪量设计值比原序列P-Ⅲ曲线拟合得到最大6日洪量设计值大，但洪峰设计值与原序列P-Ⅲ曲线拟合得到的洪峰设计值相比，依不同的重现期对比结果有所不同。当重现期大于等于200年时，非一致性洪水序列两变量联合分布法得到的洪峰设计值小于原序列对应的洪峰设计值，当重现期小于200年时，非一致性洪水序列两变量联合分布法得到的洪峰设计值大于原序列对应的洪峰设计值。（4）将基于非一致性洪水序列单变量分布法得到的设计洪水过程和基于非一致性洪水序列两变量联合分布法得到的设计洪水过程分别作为水库入库洪水过程，按给定的调洪规则，对王快水库进行调洪计算，将得到的调洪成果分别与传统方法对应的调洪成果进行对比，结果表明：非一致性序列单变量分布设计洪水过程对应的最高库水位及最大泄量均小于传统方法对应的最高库水位及最大泄量，其中，最高库水位减小值在1m之内，最大泄量减小幅度不超过5%；非一致性洪水序列两变量联合分布设计洪水过程对应的调洪成果与传统方法的调洪成果相比，依不同的重现期对比结果有所不同，，当重现期小于或等于200年时，最高库水位及最大泄量有所增加，最高库水位增加值在2m以内，最大泄量增加幅度在1%以内；当重现期大于200年时，最高库水位及最大泄量有所减小，最高库水位减小值在1m以内，最大泄量减小幅度在10%以内。
[Abstract]:Due to a large number of flood control and water conservation projects in the Daqing River Basin, significant changes have occurred in the underlying surface of the basin, and the measured hydrological sequence (flood, runoff, etc.) no longer satisfies the conformance hypothesis. The design results obtained under the condition of the change of the underlying surface under the traditional hydrologic frequency analysis method have some risks. The influence of the change of the underlying surface on the design flood and the flood diversion of the reservoir is studied. The calculation of the flood control and the flood control of the reservoir under the condition of the change of the underlying surface are discussed. This paper takes the Wang quick reservoir of the upper reaches of the Henan branch of the Great Qing Dynasty as an example, and uses the probability theory and the mathematical statistics method to calculate the frequency of the non consistent flood sequence of the reservoir directly. Based on the design value of uncongruent flood sequence, the design flood of the reservoir under the condition of the underlying surface change is calculated, and the calculation of reservoir flood regulation is carried out. The quantitative influence of the change of the underlying surface on the reservoir design flood and the reservoir flood regulation is given in order to provide the technology for the revision of the design flood and the flood control dispatching in the Daqing River Basin under the condition of the underlying surface change. Support. The main research content and results are as follows:
(1) based on the traditional hydrological sequence variation diagnosis theory, a comprehensive diagnosis method of non consistent hydrological sequence variation points is proposed, which is divided into three steps: preliminary diagnosis, detailed diagnosis and comprehensive diagnosis. In the detailed diagnosis, a method of variable interval diagnosis for hydrologic sequence based on MWP verification is proposed, which combines the diagnostic conclusions of other methods, It is convenient to confirm the specific location of the variation point. Through the comprehensive diagnosis of the variation points of the flood sequence of the Wang fast reservoir, it is shown that all the sequences have been mutated and the variation points are all in 1979.
(2) the mixed distribution and the conditional probability distribution are used to direct the distribution of the uncongruent flood sequence of the Wang fast reservoir, and the optimal distribution of each sequence is selected by K-S test and AIC criterion. Based on the optimal distribution fitting result, the flood design value of the single variable distribution of the uncongruent sequence of the Wang fast reservoir is calculated, and the reservoir 1963 is applied to the reservoir by the same frequency amplification method. The typical flood process is magnified in the year, and the design flood process line based on the uncongruent sequence single variable distribution method is obtained. The comparison of the original sequence P- III curve fitting design results with the reservoir without considering the change of the underlying surface (without considering the variation) shows that the maximum 1 day flood volume of the flood peak is less than 10%, and the annual maximum 6 day flood volume is less than 10%. The decrease of the design value of the flood in 3 days is about 1%-15%.
(3) considering the peak quantity correlation, based on the annual flood peak sequence of the reservoir and the edge distribution of the annual maximum 6 day flood sequence, the joint distribution of the two persons is constructed by Copula function. The combined recurrence period of the two variable, the two variable recurrence period and the flood peak and the flood condition frequency under the specific conditions are analyzed, and the two variation of the non consistent flood sequence is calculated. A design flood process line based on the two variable joint distribution of non conformance sequence is deduced by using the variable double distribution design value. The comparison of the design results corresponding to the single variable distribution of non consistent sequence and the results of the original sequence P- III curve fitting design without considering the variation show that the uncongruent sequence of two variables joint distribution The design value of flood is greater than that of uncongruent sequence single variable distribution flood design value. The maximum 6 day flood design value based on the two variable joint distribution method based on the non consistent flood sequence two variables is larger than the original sequence of 6 days, but the design value of the flood peak is compared with the original sequence of P- III curve fitting the flood peak design value. When the recurrence period is more than 200 years, the design value of the flood peak is less than the original sequence corresponding to the original sequence when the two variable joint distribution method of the non consistent flood sequence is less than 200 years, and the flood peak design value obtained by the uncongruent flood sequence two variable joint distribution method is larger than that of the uncongruent flood sequence. The design value of the flood peak corresponding to the original sequence.
(4) the design flood process based on the uncongruent flood sequence single variable distribution method and the design flood process based on the two variable joint distribution method based on the non consistent flood sequence are respectively used as the reservoir flood process. According to the given flood regulation rules, the flood diversion calculation of the Wang fast reservoir will be calculated, and the results of the flood diversion will be transmitted respectively. The results show that the maximum reservoir water level and maximum discharge of the uncongruent sequence single variable distribution design flood process are less than the maximum reservoir water level and the maximum discharge amount corresponding to the traditional method. The maximum reservoir water level reduction value is within 1m, and the maximum discharge reduction range is not more than 5%. Compared with the traditional method, the result of flood regulation corresponding to the two variable joint distribution design flood process of the sexual flood sequence is different from the traditional method. When the recurrence period is less than or equal to 200 years, the maximum reservoir water level and maximum discharge increase, the maximum reservoir water level is within 2m, and the maximum amount of discharge increases. Within 1%, when the recurrence period is more than 200 years, the maximum reservoir water level and maximum discharge decrease, the maximum reservoir water level decreases within 1m, and the maximum discharge decrease is less than 10%.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TV122

【参考文献】