GEOPHYSICS-BASED GROUND CONTROL AND EXCAVATION DESIGN METHODOLOGY BASED ON ACCEPTABLE DAMAGE

This dissertation is structured around a novel conceptual framework for designing deep excavation retaining systems and using geophysical data to estimate the mechanical response of soft soil deposits. It begins with presenting an approach to design excavation retaining walls based on limiting damage to adjacent infrastructure. In this approach, the damage is defined based on critical distortions of an idealized laminate beam model used for representing the adjacent building deformations. The wall and support elements of the support system are then designed such that the system yields the limiting ground deformations. The resulting excavation support system limits damage to adjacent structures below an acceptable level and automatically satisfies the structural stability requirements. More significantly, the design of the excavation support system does not require an iterative process. Also, this dissertation presents an experimental study of geophysical measurements, shear wave velocities, for soft soils under isotropic consolidation. The presented results show that the variation of effective stress during consolidation can be determined based on shear wave measurements. An approach to estimate consolidation processes based on a 1D hypoplastic model with three parameters is presented. In addition, the proposed method showed favorable results for ko-consolidation conditions. Based on the defined parameters of the model, it is possible to estimate the complete stress history of the soil. Finally, this study proposed a shear wave-based approach to predicting the triaxial behavior of cohesive soils. Laboratory tests with bender elements were performed for silt-predominant samples from the state of Kentucky. A function to relate mean effective stresses and shear wave velocities was adapted from the measured behavior to predict undrained and drained triaxial behavior. Using the previous function in conjunction with a hypoplastic model for soft soils expressed in stress invariants, the deviatoric strains, volumetric strains, and excess porewater pressures developed during shearing were predicted. The proposed methodology performed very well in simulating the various soils under undrained and drained conditions