| description abstract | This paper presents a numerical study of the dynamic compaction (DC) process, based on the finite element method, with the main attention on the role of water content on the soil response. Dynamic compaction is one of the most cost-effective techniques available for soil improvement, where the soil is compacted by repeatedly dropping free-falling heavy weights, often using cranes. Despite its simplicity in practice, finding a closed-form solution for the problem is a tedious task. This is because the efficiency of the procedure is dependent on the characteristics of dynamic wave propagation, which is predominantly determined by the soil’s hydromechanical properties. To simplify the problem, the soil has been often modeled as a single-phase material containing no water. The relationship between the compaction efficiency and the soil response has long been investigated experimentally. Despite this knowledge, modeling the moisture content in numerical analyses has been a difficult task due to the strong nonlinearity and challenges imposed by the presence of large deformations (and the associated mesh distortions), inertia effects, soil–structure interaction, spurious reflections of waves from truncated boundaries, and the simultaneous presence of three material phases (solid, gas, and liquid), together with their interactions. In this study, the problem is investigated within the framework of multiphase porous media and unsaturated soil mechanics. Some fundamental observations are revealed in solving this problem, particularly those that cannot be directly measured in practice due to the extreme energy release after the impact. A fully coupled finite element framework developed for immiscible flows is employed for this purpose. To reduce complexity, the dependency of the soil–water characteristic curve on volume changes and its hysteretic response (as shown in the literature)] is ignored. The paper also proposes a generalized version of the viscous boundary for the case of unsaturated soil dynamics and proposes an Arbitrary Lagrangian–Eulerian (ALE) approach to tackle mesh distortion; hence, providing a chance to model the problem at higher applied energies. Using the proposed numerical framework, a parametric study has also been conducted to reveal the role of the soil plasticity parameters in the dynamic compaction problem. | |
