随机加权法辅助小样本岩土参数概率分布的边坡稳定性分析

Analysis of Slope Stability Based on Probability Distribution of Small Sample Geotechnical Parameters by Random Weighting Method

根据小样本测试数据的实际情况,提出随机加权多项式拟合法,并通过K—S检验法,检验所求概率密度函数正确性。并以吉林某河段库区土工试验、地基工程勘察获取的土工试验数据为基础,采用Bayes推断方法确定岩土参数,通过对岩土参数,包括孔隙比、液限、压缩系数等结合工程类比数据,对其进行了统计分析,通过K—S检验法,对整体分布特征进行了拟合,对岩土边坡稳定性进行了研究。结果表明,通过优化先验信息,提高样本可信度,得到符合实际岩土参数。利用蒙特卡洛模拟法(Monte Carlo),统计边坡体安全系数,获得边坡体失稳概率,进而对边坡体稳定性进行分析。

In this paper,according to the actual situation of small sample test data,a random weighting polynomial fitting method is proposed,and the probability density function is verified by K — S test. And in Jilin a river reservoir engineering soil test,foundation soil test data acquisition as the basis,using Bayes inference method to determine the parameters of rock and soil,the parameters of rock and soil,such as void ratio,liquid limit,compression coefficient combined with engineering analogy data,carries on the statistical analysis by K — S test method on the whole,fitting the distribution characteristics of rock and soil slope stability were studied. The results show that the reliability of the sample is improved by optimizing the prior information. By analyzing the safety factor of the slope,the probability of failure of the slope is obtained,and then the stability of the slope is analyzed. By using the Monte Carlo simulation（ Monte Carlo）,the safety factor of the slope is calculated,and the probability of failure of the slope is obtained.