Loading...

Discrete Element Method (DEM)Modeling of the Behavior of Uncemented and Cemented Coarse-Grained Soils in Triaxial Test Using Non-Spherical Particles

Rabinezhad, Mohammad Reza | 2024

0 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 57477 (09)
  4. University: Sharif University of Technology
  5. Department: Civil Engineering
  6. Advisor(s): Pak, Ali
  7. Abstract:
  8. The present research focuses on the numerical modeling of the behavior of uncemented and 1% and 2% cemented coarse-grained materials commonly used in the base and subbase of roads and railways. Portland cement is used as a stabilizing material to improve the performance of aggregates in road/railroad construction, which requires an understanding of the mechanical properties of these materials for optimal design. Numerous triaxial tests under monotonic and cyclic loading have been conducted to investigate the mechanical properties of these materials. Conducted laboratory experiments have several limitations, such as high costs, membrane rupture under high confining pressures, and the complexity of conducting dynamic tests, which have encouraged researchers to simulate these tests using the Discrete Element Method (DEM). DEM, due to its inherent assumption that considers soil as a discrete medium, which aligns well with the granular nature of soils, is preferred over other numerical methods that are developed based on continuum mechanics. In the present research, triaxial tests on uncemented and cemented aggregates under monotonic loading were simulated using DEM and employing non-spherical dyad particles. In dyad particles, the ratio of the largest to the smallest dimension is 1.5. In this study, the flexible Latex membrane of the triaxial test was precisely modeled to sufficiently replicate the performance of the latex membrane in laboratory experiments. The strength results obtained in both uncemented and cemented samples showed very good match with laboratory results, and by using clumps instead of spherical particles, the error in the pre-peak behavior of the samples was significantly reduced. For example, the error in determining the axial strain corresponding to peak strength in uncemented samples was reduced from about 30% to zero. The results of the dilatancy behavior of the samples also showed good qualitative agreement with laboratory observations. In the present research, some relationships were proposed and validated for considering changes in microparameters under different confining pressures. Furthermore, using concepts such as force chains, coordination number, and failure mechanisms, the microscopic behavior of gravelly soils was examined, and it was found that the applied loads were transmitted through vertical force chains, and the confining pressure and the presence of surrounding particles played a stabilizing role in preventing the buckling of these chains. In fact, the strength of coarse-grained soils against failure depends on the resistance to buckling of contact chains. It was also observed that in uncemented samples, the coordination number had a decreasing trend and by using clumps instead of spherical particles, the coordination number at 10% axial strain improved from an unrealistic value of 0.8 to a reasonable value of 3.0. In contrast, in cemented samples, the coordination number first increased and then decreased. In the present research, a sensitivity analysis was also conducted on the microparameters, revealing that an increase in sliding friction coefficient, rolling friction coefficient, effective Young’s modulus between particles, the ratio of the largest to the smallest dimension of particles, and a decrease in the membrane particle radius led to an increase in the peak strength of the samples. It is noteworthy that the sensitivity of peak strength to the sliding friction coefficient was greater than that to the rolling friction coefficient.
  9. Keywords:
  10. Aggregates ; Triaxial Test ; Discrete Element Method ; Road Base ; Cemented Sample ; Gravel ; Coarse Aggregates ; Numerical Modeling

 Digital Object List

 Bookmark

No TOC