岩土参数的联合分布识别与边坡稳定性动力可靠度分析

发布时间：2018-04-29 00:12

本文选题：边坡稳定性 + 可靠度分析　；参考：《西安理工大学》2017年硕士论文

【摘要】：边坡的稳定性评价是岩土工程中的研究热点。目前,边坡稳定性评价方法主要包括安全系数法与可靠度方法,但这两种方法都存在一些缺点与不足:前者无法考虑岩土参数以及所受动荷载中存在的不确定性和随机性,使得边坡工程的安全水平含混不清,不能提供说明边坡安全度的评价指标;而后者虽可考虑诸多因素的不确定性,但其在岩土参数相关性、小样本条件下岩土参数的统计特性及动力可靠度的评判指标等方面尚不完善,导致该法的准确性和实用性较差。本文针对以上几种问题,提出了相对的解决方法,并在此基础上编写了边坡稳定可靠度分析的计算程序,主要研究成果如下:(1)简单介绍了边坡稳定性分析和可靠度分析的基本理论,并探讨了不同可靠度分析方法的优劣;提出了基于响应面法的边坡概率临界滑面搜索方法,使用二次多项式响应面函数代替难以求偏导的安全系数隐式表达式,从而得到边坡最小可靠度指标及其对应的滑面;通过改变边坡稳定性分析方法和参数的变异系数、相关系数及其所服从的概率分布,进行了边坡稳定性静力可靠度的敏感性分析。(2)利用DA-检验对3个实际工程中的数据进行分析,发现常用的正态分布并不一定是c、?所服从的概率分布,对数正态、极值I型与Weibull分布都可能是c、?服从的分布,直接使用正态分布模拟其概率分布将导致较大的误差。(3)为准确描述参数间的相关性引入Copula理论,利用Copula函数建立参数的联合概率分布并进行边坡可靠度分析,发现:抗剪强度参数c、?的负相关性对可靠度结果具有很大的影响,不考虑相关性的边坡可靠度分析计算结果偏于保守,带来不必要的浪费。(4)针对小样本数据条件下的边坡可靠度分析,首次引入随机加权法进行岩土参数的统计分析,模拟不同备选Copula函数AIC值及边坡失效概率的变异性,再利用百分位法得到边坡失效概率的置信区间。通过一简单算例和工程实例说明了上述方法能够解决小样本数据条件下的可靠度分析。(5)提出以边坡永久位移作为边坡稳定性动力可靠度的评判指标,在考虑地震动随机性的基础上通过Newmark有限位移理论得到永久位移,并通过永久位移的累计概率分布图得到任意可能发生的永久变形对应的失效概率,使得工程人员能够根据边坡的重要性以及边坡修复的难易程度选择不同的边坡允许变形作为控制标准,全面协调安全与经济的矛盾。(6)在考虑到地震烈度自身随机性的基础上,通过地震场地危险性分析得到烈度的概率分布,随后利用全概法则求得任意边坡永久位移对应的概率;在考虑了岩土参数不确定性及相关性的基础上,研究了不同因素如参数变异性、参数分布概型等对动力可靠度结果的影响。
[Abstract]:Slope stability evaluation is a hot topic in geotechnical engineering. At present, the evaluation methods of slope stability mainly include safety factor method and reliability method, but these two methods have some shortcomings and shortcomings: the former can not take into account the rock and soil parameters and the uncertainty and randomness existing in the dynamic load. The safety level of slope engineering is ambiguous and can not provide an evaluation index to explain the slope safety degree, and the latter can take into account the uncertainty of many factors, but its correlation in geotechnical parameters, The statistical characteristics of geotechnical parameters and the evaluation index of dynamic reliability are not perfect under the condition of small sample, which leads to the poor accuracy and practicability of the method. In view of the above problems, this paper puts forward a relative solution method, and on the basis of this, compiles a calculation program for the reliability analysis of slope stability. The main research results are as follows: (1) the basic theories of slope stability analysis and reliability analysis are briefly introduced, and the advantages and disadvantages of different reliability analysis methods are discussed, and the slope probabilistic critical sliding surface search method based on response surface method is proposed. The quadratic polynomial response surface function is used to replace the implicit expression of safety factor which is difficult to obtain the slope reliability index and its corresponding sliding surface, and by changing the slope stability analysis method and the variation coefficient of the parameters, the minimum reliability index and the corresponding slip surface of the slope are obtained. The correlation coefficient and the probability distribution of the reliability of slope stability are analyzed by using DA-test. It is found that the usual normal distribution is not necessarily cu? The probabilistic distribution, logarithmic normal distribution, extreme type I and Weibull distribution may all be cu? The direct use of normal distribution to simulate its probability distribution will lead to a large error. The Copula theory is introduced to accurately describe the correlation between parameters. The joint probability distribution of parameters is established by using Copula function and the slope reliability is analyzed. Find out: shear strength parameter cu? The negative correlation has a great influence on the reliability results. The calculated results of slope reliability analysis without considering the correlation are conservative and bring unnecessary waste.) for the slope reliability analysis under the condition of small sample data, For the first time, the statistical analysis of geotechnical parameters is carried out by introducing the stochastic weighting method to simulate the variability of AIC value and slope failure probability of different Copula functions, and then the confidence interval of slope failure probability is obtained by using the percentile method. A simple example and an engineering example show that the above method can solve the reliability analysis under the condition of small sample data. The permanent displacement of the slope is proposed as the evaluation index of the dynamic reliability of the slope stability. On the basis of considering the randomness of ground motion, the permanent displacement is obtained by Newmark finite displacement theory, and the failure probability corresponding to any permanent deformation is obtained by the cumulative probability distribution of permanent displacement. According to the importance of slope and the difficulty of slope restoration, engineers can choose different allowable deformation of slope as control standard, and fully coordinate the contradiction between safety and economy. On the basis of considering the randomness of seismic intensity itself, The probability distribution of intensity is obtained by seismic site risk analysis, and then the probability corresponding to permanent displacement of any slope is obtained by using the full probability rule, and the uncertainty and correlation of geotechnical parameters are taken into account. The influence of different factors, such as parameter variability and parameter distribution probability type, on the dynamic reliability results is studied.
【学位授予单位】：西安理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU435

【参考文献】