The study of effective shear wave velocity measurement in composite soil by application of three-dimensional simulation of surface wave

• Main Authors: Chang, Yu-Cheng, 張友誠
• Other Authors: Lin, Chih-Ping
• Format: Others
• Language: zh-TW
• Published: 2013
• Online Access: http://ndltd.ncl.edu.tw/handle/48429507136056420467

碩士 === 國立交通大學 === 土木工程系所 === 101 === Soft ground is often improved to increase strength and stiffness by methods such as jet grouting and stone column which result in heterogeneous ground with improved columns. Conventional methods used to assess such ground improvement are subjected to several limitations such as small sampling volume, time-consuming, and cost ineffectiveness. It’s difficult to assess the average property of the improved ground and the actual replacement ratio of ground improvement. The use of seismic surface wave method (i.e. multi-station analysis of surface wave, MASW) for such a purpose seems to be a good candidate. But the surface wave method is essentially a 1-D method assuming horizontally-layered medium. What MASW measures in the highly heterogeneous improved ground remains to be investigated. The feasibility of MASW in characterizing the heterogeneous ground with improved columns has been demonstrated and the relationship between replacement ratio and velocity increase has been studied by 2D model. This study aims to investigate the homogenization of shear wave velocity measured by MASW in real 3D condition. Testing results may also be affected by the survey line location relative to the improved columns. The lateral sampling space of the surface wave testing was also investigated by 3D modeling. 3D spectral element method was utilized for the numerical investigation. Numerical results show that the effective shear velocity can be derived based on lower bound of equivalent velocity model when the heterogeneous ground is a 2D wall-like structure. However, in real 3D column-embedded structure, the relation between the effective shear wave velocity and replacement ratio falls between upper bound and lower bound of the equivalent modulus model. A predictive model is derived based on the numerical simulations. The lateral sampling range is about 1.3 wavelength. In the same spatial range, the measured velocity is affected more by offline heterogeneous material for higher velocity contrast. At constant velocity contrast, lower velocity material along the survey line is affected more by offline heterogeneity.