| Abstract | This dissertation is on the practice and experience in deep excavations in soft soil of central Ho Chi Minh City (HCMC), Vietnam through reviews on the typical case histories, and on the estimation of the geotechnical parameters of the clays located in central HCMC for numerical analysis of deep excavations using Finite Element Method with Hardening Soil constitutive model.
The first part of the research, main methods of excavation and typical types of retaining wall and support that may be used in practice of deep excavation in urban zone are reviewed. The review then focuses on another major issue in deep excavation in soft ground which is stability of earth retaining wall corresponding with different modes of failures: overall failures involving push-in failure and basal failure, hydraulic failure involving sand boiling and upheaval failures. On one hand, some successful cases of stability in the literature are summarized on the design practice; on the other hand, some typical cases of failure of deep excavation in soft ground of are reviewed with the lessons learnt. The review, however, gives focus on the Finite element method in deep excavation analyses. Characteristics of main soil models implemented in the PLAXIS software are summarized for review. The Mohr Coulomb soil model (MCM), an elastic perfectly plastic soil model and Hardening Soil model (HSM), an advanced soil model are reviewed in this part with emphasis on the prediction of wall and ground movements induced by deep excavation.
The second part presents the geotechnical design considerations of the deep excavation performance in central HCMC soft clay through the typical case histories. The review of the current practice is on analyses of 18 cases of deep excavations in 4–16-m-thick soft clay in HCMC Central, three of which experienced severe instability. Geotechnical characteristics of the soft ground of HCMC Central lowland are reviewed and reported. The subsoils consist of a very soft to soft clay layer and saturated sand layers existing within and below the deep excavation depth range. The case studies identify that, the majority of failures in the deep excavation cases in HCMC Central soft ground are due to ground water and structural failures which may cause risk of sand boiling failure and excessive ground movements. The main contribution and findings of this part are through the review of the case histories, providing results of characterization of retaining walls with excavation depth and maximum wall lateral displacement. The ranges of wall displacement normalized with excavation depth and with different types of wall and different wall/support system stiffness, wall bending stiffness and factor of safety against basal heave are provided for current local design practice. Some of deep excavation improvements in design practice in HCMC soft ground are also suggested.
ii
Part three presents a well-controlled and thorough laboratory testing program to investigate the strength and deformation characteristics of the central HCMC subsoils. The main tests consist of oedometer, drained and undrained triaxial tests on the soft clay, Lower soft clay and stiff clay. The soil parameters of HCMC clays needed for Finite Element analysis (FEA) of deep excavations using HSM are determined from the test results. The consolidation parameters including both primary loading and unloading/reloading oedometer modulus are investigated and determined from the oedometer test results. The drained and undrained triaxial deformation parameters are also analyzed and obtained from the stress-strain hyperbolic relationships of the drained and undrained triaxial test results. The variation of the oedometer modulus with consolidation pressure, the variation of both the drained and undrained triaxial modulus with confining pressure are then established and investigated to estimate the reference oedometer modulus and the reference drained/undrained triaxial modulus, respectively, together with the modulus power. Each set of soil parameters directly interpreted from the laboratory testing for FEA using HSM are preliminarily selected, compared with the corresponding soil parameters of clays from other research in the literature and suggested to be used for FEA of deep excavation in HCMC Central soft ground.
The fourth part describes a detailed numerical analysis of two case histories of deep excavation in HCMC Central soft ground by 2D-FEM using PLAXIS program for prediction of wall and ground movements. Initially, based on evaluation of the sensibility of the model parameters, each pair of alternative sets of soil parameters interpreted from the laboratory testing for HSM are finally selected for the soft clay, Lower soft clay and stiff clay. The two case histories used in the FEA consist of a top down construction project and a bottom up one. The model predictions of wall and ground movements with the parameters resulting from the testing program are evaluated through comparisons with the measured data. The model sensitivity in the prediction of the lateral wall movements and the surface ground movements are compared between the FEA using HSM computed with both drained and consolidation analyses, FEA using MCM with unloading/reloading stiffness of soils and Undrained B type of material behavior for the clays (effective stress analysis with undrained shear strength, effective strength and stiffness). Hsieh and Ou (1998)‘s simplified method are adopted to predict the ground surface ground settlement for comparison. Appropriate ranges of realistic soil parameters of the HCMC clays for FEA using HSM are concluded and suggested to be used for prediction of wall and ground movements induced by deep excavation in HCMC Central soft ground. |