论文部分内容阅读
地质统计学是用于模拟土体固有空间变异性的方法之一,以变差函数为工具,采用Kriging插值提供未采样点处土工参数值的最优线性无偏估计。将地质统计学方法应用于宿-新(宿迁至新沂)高速公路某试验段内孔压静力触探(piezocone penetration test,CPTU)锥尖阻力qt空间变异性研究中,采用回归分析移除数据中的趋势项,从而获得具有弱平稳性的残差数据。指数型理论变差函数能够准确描述试验段内土体的连续空间变异性特征。根据估计结果,试验段内锥尖阻力qt残差的变程具有显著各向异性,在水平方向和竖直方向分别为4.05 m和1.2 m。采用普通Kriging插值结合趋势分析,绘制了qt在试验段的空间分布图和平面投影图,用于指导工程实践。结果表明,普通Kriging插值的估计结果能够与试验段内实测资料形成较好的对比,仅仅在部分极值变化和远离采样点的位置处估计值可靠性会降低。
Geostatistics is one of the methods used to simulate the inherent spatial variability of soils. Using the variogram as a tool, Kriging interpolation is used to provide the optimal linear unbiased estimate of geotechnical parameters at the unsampled points. Geostatistics applied to the sink - Study qt tip resistance spatial variability of pore pressure at a Test Static CPT new (Suqian to Xinyi) Expressway (piezocone penetration test, CPTU), the regression analysis is removed Trend items in the data to obtain the residual data with weak stationary. The exponential theoretical variation function can accurately describe the continuous spatial variability of soil in the test section. According to the estimation results, the variation range of the cone tip resistance qt residuals in the experimental section has significant anisotropy, which is 4.05 m and 1.2 m respectively in the horizontal direction and the vertical direction. Using ordinary Kriging interpolation and trend analysis, the spatial distribution and plane projection of qt in the experimental section were plotted to guide the engineering practice. The results show that the estimation results of ordinary Kriging interpolation can be in good contrast with the measured data in the experimental section, and the reliability of the estimation value will only decrease at partial extremum changes and positions away from the sampling points.