基于贝叶斯理论的土性参数空间变异性量化方法

发布时间：2018-05-25 22:28

本文选题：空间变异性 + 有效内摩擦角　；参考：《岩土力学》2017年11期

【摘要】：岩土工程可靠度分析和设计中,合理地选取随机场参数和相关函数,并准确地描述土性参数空间变异性十分困难。基于贝叶斯理论,本文提出了一套量化砂土有效内摩擦角空间变异性的方法。该方法根据先验信息和静力触探试验锥尖阻力数据,确定砂土有效内摩擦角的随机场参数和相关函数。该方法合理地考虑了砂土有效内摩擦角与锥尖阻力间经验回归方程的不确定性。采用马尔科夫链蒙特卡洛模拟(Markov Chain Monte Carlo Simulation,MCMCS)获取服从后验分布的随机场参数样本。利用MCMCS样本构建随机场参数的Gaussian Copula函数求解后验分布。估计备选相关函数的概率,选择概率最大的为最可能的相关函数。最后,采用美国德州农工大学国家岩土工程砂土试验场的CPT数据算例验证了文中所提方法的有效性。结果表明:文中所提方法可以正确、合理地利用间接测量的锥尖阻力数据确定砂土有效内摩擦角的随机场参数和相关函数,准确量化其空间变异性。对于美国德州农工大学国家岩土工程砂土试验场的砂土有效内摩擦角,建议选用二阶自回归函数作为其最可能的相关函数。
[Abstract]:In the reliability analysis and design of geotechnical engineering, it is very difficult to reasonably select random field parameters and correlation functions, and accurately describe the spatial variability of soil parameters. Based on Bayesian theory, a method for quantifying the spatial variability of effective angle of internal friction of sand is presented in this paper. Based on the prior information and the cone tip resistance data of static penetration test, the random field parameters and correlation functions of the effective internal friction angle of sand are determined. The uncertainty of the empirical regression equation between the effective internal friction angle of sand and the resistance of cone tip is reasonably considered in this method. Markov Chain Monte Carlo simulation MCMCSs are used to obtain random field parameter samples of the posteriori distribution of Markov chain Monte Carlo. The Gaussian Copula function of random field parameters is constructed by using MCMCS samples to solve the posterior distribution. Estimate the probability of alternative correlation function and select the most probable correlation function. Finally, the effectiveness of the proposed method is verified by using CPT data from the National Geotechnical Engineering Sand Test site of Texas University of Agriculture and Technology, USA. The results show that the proposed method can be used to determine the random field parameters and correlation functions of the effective angle of internal friction of sand, and to quantify the spatial variability of sand accurately. The second order autoregressive function is suggested as the most likely correlation function for the effective internal friction angle of sand in the National Geotechnical Engineering Sand Test site of Texas University of Agriculture and Technology.
【作者单位】：武汉大学水资源与水电工程科学国家重点实验室;武汉大学水工岩石力学教育部重点实验室;香港城市大学土木与建筑工程系;
【基金】：国家重点研发计划课题(No.2016YFC0800208) 国家自然科学基金项目(No.51329901,No.51409196,No.51579190)~~
【分类号】：TU41

【相似文献】