论文部分内容阅读
概率设计方法与有限元相结合分析桩土相互作用问题,可利用已有的有限元程序而直接调用,使桩土相互作用问题计算结果更加可靠。基于拉丁超立抽样和直接抽样方法对桩基沉降的蒙特卡洛概率有限元法进行应用研究。计算结果表明:拉丁超立抽样仅需要较少的抽样点即可达到较高的结果精度要求,结果要优于直接抽样方法;桩顶及桩端沉降服从三参数对数Logistic分布,桩端阻力服从正态对数分布,而桩身变形服从三参数Gamma分布;内摩擦角、膨胀角、土层弹性模量是影响桩顶沉降的主要因素,占70%左右;内摩擦角和膨胀角是影响桩端阻力的主要因素;桩顶沉降、桩端沉降与土层参数取值负相关,与桩身材料参数表现出很弱的相关性。采用拉丁超立抽样进行蒙特卡洛概率有限元分析抽样点分布,试验次数少,离散性小,计算结果精度高,可以节省大量的计算时间。可靠度分析计算结果可为桩基设计提供设计依据。
Probability design method combined with finite element analysis of pile-soil interaction problems, can be directly called using the existing finite element program, the pile-soil interaction problem calculation results more reliable. Application of Monte Carlo finite element method to settlement of pile foundation based on Latin superstandard sampling and direct sampling method. The calculation results show that the Latin super-discrete sampling method requires fewer sampling points to achieve higher precision results, and the result is better than the direct sampling method. The pile top and pile tip settlement follow logistic distribution of three parameters, Obey the normal logarithm distribution, while the pile deformation obeys the three-parameter Gamma distribution. The internal friction angle, expansion angle and soil elastic modulus are the main factors that affect the settlement of pile top, accounting for about 70%; the internal friction angle and the expansion angle are The main factors affecting pile tip resistance are the negative correlation between the settlement of top of pile and settlement of pile head and the parameters of soil layer, which shows weak correlation with pile material parameters. Monte Carlo probability finite element analysis using Latin super-discrete sampling sampling point distribution, the number of tests less discreteness, high accuracy of the results, you can save a lot of computing time. The results of reliability analysis can provide design basis for pile foundation design.