摘要
对于微型桩水平承载力,在目前的研究中尚未有统一的并被工程界普遍接受的表达式,本文就此问题进行了研究.首先选用西安地区黄土作为试验模型的滑坡介质,在假定干密度保持不变的前提下,选择土体饱和度、桩径和桩间距作为影响因素,每个因素选取三个试验水平,按照正交试验方案,对微型桩进行了室内试验,得到了不同影响因素和试验水平下的水平承载力大小关系,并借助MATLAB数据拟合功能,对水平承载力进行了一次线性和二次非线性函数的回归分析,对回归函数系数分别运用假设检验的相关原理进行了评价,最后将二次非线性回归函数应用于工程实例中和原设计方案进行了对比.试验结果表明微型桩的水平承载力与饱和度呈反比、与桩径呈正比关系,且三个因素对水平承载力影响大小的排列顺序依次为饱和度>桩径>桩间距;回归结果展示了二次非线性回归函数相比一次线性能较好地吻合试验结果,且具有较高的拟合精度;实例对比结果显示了二次非线性回归函数能够满足设计抗滑力的要求.
To start with, this article assumes that loess in Xi＇an is regarded as the slide mass and soil dry density is kept constant. Soil saturation, pile diameter and pile spacing are chosen as three experimental factors and there are three levels for every factor, so orthogonal experimental plan is made; and the micropile horizontal bearing capacity was obtained by indoor test under this plan. Secondly, with the aid of MATLAB, the linear and quadratic nonlinear function of regression analysis are obtained, and the coefficients of two regression functions are evaluated by the application of relevant principle of hypothesis test; Finally, the quadratic non-linear regression function is applied to an engineering example and compared with the original design scheme. The experimental results indicate that the horizontal bearing capacity has an inverse relationship with soil saturation and a positive relationship with the pile diameter, and the descending order is soil saturation, pile diameter and pile spacing according to the significant effect of these factors on horizontal bearing capacity. Furthermore, the analysis result shows that the quadratic polynomial fitting trend can well fit the test results compared with linear polynomial and the quadratic nonlinear polynomial can have higher fitting accuracy, However the scheme manifests that it can meet the requirements of the design of sliding resistance.
出处
《西安建筑科技大学学报（自然科学版）》
CSCD
北大核心
2015年第5期694-700,共7页
Journal of Xi'an University of Architecture & Technology(Natural Science Edition)
关键词
微型桩
水平承载力
室内试验
回归分析
假设检验
micropile
horizontal bearing capacity
indoor experiment
regression analysis
hypothesis testing
