| The research focuses on the consolidation behaviour and the bearing capacity of soft soil ground improved by Deep Cement Mixing (DCM) method. It includes two major parts: physical modelling and numerical modelling. Firstly, three physical model tests were conducted to investigate the consolidation behaviour of soil ground with a single DCM column, and the vertical bearing capacity, and the failure mode of the soil ground improved by a DCM column group. Secondly, a series of numerical analyses were performed to improve understanding of the findings obtained in the model tests.; Physical modelling studies reveal that the DCM column behaved as a vertical drain partially, similar to a partial vertical drain. The stress concentration ratio was found to be dependent on the external pressure and consolidation of the surrounding soil in an axisymmetric condition. Excess pore water pressures in the soil around a DCM column appeared to dissipate faster than that in the soil around a vertical drain. Under a plain-strain condition, the softening of bearing capacity of the DCM column group improved soil ground was observed and a wedge-shaped failure pattern was explored. A three-dimensional Elastic Visco-Plastic (EVP) constitutive model developed by Yin and Graham (1999) was incorporated into a Finite Element (FE) code, ABAQUS through a User MATerial (UMAT) subroutine. Several single element tests were simulated to evaluate the performance of the UMAT subroutine. The FE package with UMAT was employed to conduct analyses of two model tests and two field cases. In general, good agreement was obtained between measurements and predictions. In the analyses of two model tests, the DCM column was modelled using Mohr-Coulomb model with a reducing cohesion. The vertical bearing capacity of the DCM treated soil model ground was predicted nicely. The permeability of DCM column and soil viscosity was found to have an influence on the excess pore pressure dissipation and the bearing capacity. Numerical modelling results also show good agreement between predictions and observed data for two field cases. Compared to Elastic Plastic (EP) numerical modelling, it seems that EVP numerical modelling is capable of providing better predictions. |
