Application of seismic cone for characterization of ground improved by vibro-replacement

The objective of this thesis was to gain a better understanding of the physical process of ground improvement by vibro-replacement and of how the induced changes in ground conditions affect the interpretation of seismic cone penetration testing used to assess its effectiveness. This was achieved by a combination of field testing and monitoring supported by numerical modelling of both the vibro-replacement process and of in situ testing. Field measurements were made of the input motions created by the vibrator and the induced response of the ground. The measured vibrator motions were input to a numerical model of the soil mass and the results of the analysis were compared to the measured data. The results of seismic cone testing before and after treatment at 15 sites and existing chamber test data were analyzed and additional numerical modelling of seismic cone testing in the post-treatment ground conditions was carried out. The study showed that the ground response to the vibrator attenuated with distance due to geometrical spreading and material damping. The direction of the principal horizontal acceleration changed with distance from the vibrator. It was shown that vibro-replacement creates a young, heterogeneous deposit comprising the native soil with increases in density and horizontal stress varying with distance from stiffer stone columns. If the induced heterogeneity is neglected when interpreting in situ test results, there will be some over-estimation of the soil properties close to the cone hole and considerable under-estimation of the average properties of the composite mass. The stiffer stone columns also change the wave propagation regime during down-hole testing and prevent reliable interpretation of the shear wave velocity o f the improved native soil. Analysis of field test data showed that vibro-replacement causes an apparent shift in soil behaviour type classification. The combined effects of changes in density, horizontal stress and time dependent soil stiffness were shown to be important when interpreting seismic cone results. Friction ratio was found not diagnostic of changes in soil conditions. For the soils studied, a correlation was derived between achievable tip resistance and pre-treatment soil behaviour type for several stone column spacings.Applied Science, Faculty ofCivil Engineering, Department ofGraduat