Search

A three-dimensional numerical analysis is conducted on a group pile adjacent to a deep multi-strutted excavation in a naturally occurring soil profile comprising both cohesive and non-cohesive layers. The contrasting soil layers affect the soil stiffness, strength, and deformation characteristics, thereby impacting the load transfer mechanism within a pile group. The primary aim of this study is to specifically understand how the presence of a non-cohesive soil layer within a predominantly cohesive soil profile, affects the behavior of a pile group during excavation. By systematically varying the position and thickness of the soil layers, the analysis highlights the significance of considering the actual soil profile's inherent heterogeneity rather than assuming a homogeneous clay layer. 19.2% and 17% deviations respectively are observed for pile group settlement and lateral deflection when a homogeneous clay layer is considered instead of the natural soil profile. The findings also show that the position of the non-cohesive soil layer within a predominantly clay soil profile is more critical than its thickness in determining the pile response to excavation-induced changes. The non-cohesive soil layers within the excavation depth, significantly influence the lateral and axial pile behavior, while those below the final excavation depth have a limited impact. A theoretical simplified framework is proposed for the preliminary assessment of the excavation-induced effect on an adjacent pile group. This simplification helps minimize the initial calculation time and effort while capturing the significant factors affecting the pile group behavior near the excavation. [ABSTRACT FROM AUTHOR]

Considering the non-linear creep characteristics of clays, the modified unified hardening (UH) relation is employed to reflect the visco-elastic-plastic deformation of soft soils. Simultaneously, Swartzendruber's flow in the form of a continuous function is captured to describe pore water flow through pores during consolidation. Thus, an upgraded one-dimensional (1-D) non-linear rheological consolidation (NRC) system of clays has been established, and its solutions are conducted using the finite volume method (FVM). Through comparisons against the theoretical solutions and laboratory test results, the effectiveness of the numerical algorithm and applicability of the modified UH relation is verified. Then, the effects of model parameters on non-linear rheological consolidation behavior were studied. The results indicate that ignoring the time effect of secondary consolidation coefficient overestimates the overall consolidation and settlement proceeding, and considering the viscous effect of the soils led to an unanticipated increase in excess pore-water pressure (EPP) at the initial loading period. Meanwhile, the rising EPP becomes more pronounced with the increase of the initial over-consolidated parameter, flow parameter, and soil thickness or the decrease of overlying load. Additionally, FVM and finite difference method (FDM) solutions, as well as the application scope of Darcy's law in the NRC system, are discussed. [ABSTRACT FROM AUTHOR]

This paper proposes a three-dimensional moisture migration model, which can simulate the moisture migration process in soil. A simple moisture transport example with analytic solutions is given to verify the correctness of the 3D moisture migration model. Combining the moisture migration model with the combined finite discrete element method (FDEM), a moisture migration-fracture model that simulates soil shrinkage and cracking is constructed and implemented in an in-house GPU parallel multiphysics finite element-discrete element software, namely MultiFracs. The calculation process of the three-dimensional moisture migration-fracture model is: first obtain the moisture field distribution of the solution domain through the moisture migration model; then obtain the shrinkage stress in all tetrahedral elements according to the change of the moisture field; finally, apply the shrinkage stress as a load to perform the deformation and fracture calculations of FDEM. The 3D moisture migration-fracture model is used to simulate an experiment of clayey soil shrinkage cracking. Image recognition technology is used to carry out quantitative statistics and analysis of desiccation crack patterns. The simulation results show high consistency with the laboratory experiment. Moreover, the influences of shrinkage parameter, water evaporation rate, and layer thickness on clayey soil desiccation cracking are studied comprehensively. [ABSTRACT FROM AUTHOR]

Full-flow penetrometers (ball and T-bar) are widely used for measurements of strength characteristics of clayey soils in offshore field investigations and laboratory tests. However, the accuracy of the estimated undrained shear strength is hampered by: (1) the drastic variation of the bearing factor for the penetrometer at shallow and transitional penetration depths where full-flow mechanisms are not achieved; and (2) the strain-softening of the soil, which results in a degradation of the estimated strength. This study conducted LDFE analyses to simulate the entire penetration process of full-flow penetrometers in strain-softening soils from shallow to deep embedment depths. Parametric studies were performed to investigate the effects of varied factors on the bearing factor and failure mechanism during the penetrometer penetration. Based on the LDFE analyses, empirical formulas are established to represent the critical penetration depths in the strain-softening soil, which are used to correct the bearing factor at shallow penetration depths. The discrepancies of the transitional failure mechanism and the trapped cavity bearing factor between ball and T-bar are emphasized. Finally, a back-analyses framework is proposed to estimate the strength parameters (including the original undrained shear strength prior to any strain-softening, and the strain-softening parameters) based on the measurements of the penetrometer during the first penetration from shallow to deep embedment depths. Through a series of model tests, the reliability of the back-analyses framework is validated. [ABSTRACT FROM AUTHOR]