基于粗糙集与核主成分的震区泥石流风险评价研究

发布时间：2018-05-01 19:07

本文选题：泥石流 + 风险评价　；参考：《成都理工大学》2017年硕士论文

【摘要】：汶川地震诱发了大量崩塌、滑坡等次生地质灾害,为泥石流活动提供了丰富的松散固体物源,同时在地震后效应的影响下,岩体内部的剪切力作用使得岩体结构破坏、坡体失稳,在后期降雨条件下极易启动为泥石流提供间接补给。据统计,震后爆发的泥石流次数较震前明显增加,地震导致的灾害链效应会影响震后几十年甚至更长一段时间,因此对震区泥石流风险评价的研究十分重要。针对震区泥石流的特点,本文以清平镇绵远河流域泥石流为研究对象,通过现场调研并结合前人研究资料,分析了泥石流的地质环境和发育背景,将灾害链效应考虑到泥石流危险性评价当中。利用统计学方法和RUSLE模型对泥石流爆发时的物源动储量进行估算。通过文献调研的方式初步确立了20个危险性评价指标,采用粗糙集对初选指标进一步约简。利用核主成分分析对泥石流危险性进行评价,最后结合对泥石流区域范围内承灾体的易损性评价,得到震区泥石流风险评价结果。本文提出的震区泥石流风险评价的研究思路和计算方法,有较高的可行性和科学依据,取得的主要成果有:(1)基于清平镇泥石流爆发历史及现状。确定以绵远河流域11条泥石流沟为研究对象进行风险评价。详细调查了研究区域的气象水文、地形地貌、地层岩性、地质构造、植被条件等泥石流发育背景资料,分析震前震后的泥石流特征以及地震对泥石流物源储量和启动条件的影响。(2)建立泥石流物源动储量估算模型。利用样本数据,控制变量法找出单一因子与动储量的关系,并建立多元非线性回归模型,通过该模型估算出崩滑堆积类泥石流物源动储量,结合RUSLE模型估算出的坡面侵蚀类泥石流物源动储量,从而得到泥石流物源动储量。(3)结合前人对泥石流危险评价的研究成果和研究区现场调查结果,对泥石流危险性影响因子进行初选,再根据粗糙集理论采取差别矩阵对所选取的评价指标进行属性约简,得到泥石流危险性评价指标体系。(4)介绍了核函数及核主成分方法的基本理论,对各沟域原始数据进行标准化处理后,利用核主成分分析提取出最优主成分,通过对主成分的计算得到研究区各沟域的危险性评价结果。(5)在地质灾害风险评价理论的基础上,探讨了泥石流灾害风险评价中易损性评价的技术方法,论述了泥石流易损性的影响因素,即被破坏的可能性、被损坏的难易程度以及承灾体的价值,初步建立了适用于山区单沟泥石流灾害易损性评价的方法,并结合危险性评价得到泥石流风险评价结果。
[Abstract]:Wenchuan earthquake induced a large number of secondary geological disasters, such as collapse, landslide and other secondary geological disasters, which provided a rich source of loose solid for debris flow activities. At the same time, under the influence of post-earthquake effect, the internal shear force of rock mass caused the destruction of rock mass structure. The slope body is unstable and can easily start to provide indirect recharge for debris flow under the late rainfall condition. According to statistics, the frequency of debris flow after earthquake is obviously increased than that before earthquake, and the disaster chain effect caused by earthquake will affect the risk assessment of debris flow in earthquake area for decades or even longer after the earthquake, so it is very important to study the risk assessment of debris flow in earthquake area. According to the characteristics of debris flow in earthquake area, this paper takes debris flow in Mianyuan River basin of Qingping Town as the research object, analyzes the geological environment and development background of debris flow through field investigation and combining with previous research data. The hazard chain effect is taken into account in the risk assessment of debris flow. The dynamic reserves of debris flow were estimated by statistical method and RUSLE model. Through literature investigation, 20 risk evaluation indexes were preliminarily established, and rough sets were used to further reduce the primary indexes. The risk of debris flow is evaluated by nuclear principal component analysis. Finally, the risk assessment results of debris flow in seismic area are obtained by combining the vulnerability evaluation of debris flow in the area of debris flow. The research thought and calculation method of debris flow risk assessment in seismic area presented in this paper have high feasibility and scientific basis, and the main results obtained are: 1) based on the history and present situation of debris flow in Qingping Town. Take 11 debris flow gullies in Mianyuan River Basin as the research object to carry on the risk evaluation. The background data of debris flow development, such as meteorology, hydrology, topography, stratigraphic lithology, geological structure, vegetation conditions, etc., were investigated in detail. The characteristics of debris flow before and after the earthquake and the influence of the earthquake on the source reserves and start-up conditions of debris flow are analyzed. A model for estimating the dynamic reserves of debris flow source is established. Based on the sample data, the control variable method is used to find out the relationship between the single factor and the dynamic reserves, and a multivariate nonlinear regression model is established to estimate the source dynamic reserves of debris flows. Combined with the RUSLE model to estimate the slope erosion debris flow source reserves, the debris flow source reserves. 3) combined with the previous research results of debris flow risk assessment and site investigation results in the study area. The influence factors of debris flow risk are selected by primary selection, and then the attribute reduction of the selected evaluation index is carried out by using the difference matrix according to rough set theory. The basic theory of kernel function and kernel principal component method is introduced. After standardized processing of original data in each gully domain, the optimal principal component is extracted by kernel principal component analysis. Based on the theory of geological hazard risk assessment, the technical method of vulnerability assessment in debris flow hazard risk assessment is discussed. This paper discusses the influencing factors of debris flow vulnerability, that is, the possibility of being destroyed, the degree of damage and the value of disaster bearing body, and establishes a method for evaluating the vulnerability of debris flow in mountain area. The risk assessment results of debris flow are obtained based on the risk assessment.
【学位授予单位】：成都理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：P642.23

【参考文献】