Experimental Testing and Numerical Modeling of Geomaterials Interacting with Rigid Bodies Under Impact Loading
Open Access
- Author:
- Reese, Lynsey
- Graduate Program:
- Civil Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 16, 2016
- Committee Members:
- Tong Qiu, Dissertation Advisor/Co-Advisor
- Keywords:
- Anti-Ram Barriers
LS-DYNA
Smoothed Particle Hydrodynamics
Soil-Structure Interaction
Numerical Modeling
Expanded Polystyrene Concrete
Full-Scale Testing - Abstract:
- Experimental tests of geo-materials interacting with rigid bodies under impact loading were conducted to collect critical information about the behavior of such systems. Full-scale vehicular tests were conducted to assess the performance and behavior of single boulders embedded in AASHTO uniformly-graded coarse aggregate soil as an anti-ram vehicle barrier, while drop weight tests were conducted on expanded polystyrene crushable concrete to develop a material model capable of incorporating strain rate effects for use in vehicle barrier designs. Numerical models were developed for each experimental test to calibrate material constitutive models. Four full-scale vehicular tests were conducted to assess the performance and behavior of single boulders embedded in compacted AASHTO soil. LS-DYNA numerical models were developed and a soil constitutive model was calibrated based on a full-scale test with minimal soil movement. Each full-scale test progressed to larger boulder translation and rotation and soil deformation with the last test exhibiting boulder rotation out of the ground. Two modeling techniques were recommended based on the expected boulder and soil deformation. For small deformations, traditional finite element method (FEM) can be used for the boulder and soil domain. For large deformations, a hybrid approach combining Smoothed Particle Hydrodynamics (SPH) near areas of large deformation and FEM in areas of minimal deformation for soil was developed. Both numerical methods used the same calibrated constitutive model, Mohr-Coulomb failure criterion, for compacted AASHTO soil. A crushable concrete mix design was tested where expanded polystyrene was used as the primary cell material to take advantage of its energy dissipating properties. Quasi-static and dynamic drop weight tests were conducted on fully confined specimens to help determine material constitutive properties for numerical modeling. A LS-DYNA model, using Material Type 16, “Pseudo Tensor”, was developed to simulate both the quasi-static and dynamic drop weight tests. The numerical model was able to capture the overall responses of the specimens during the quasi-static testing and consecutive impacts from the dynamic drop weight tests. Several compressed specimens from the dynamic drop weight test, representing different strain levels, were studied using non-destructive X-ray CT imaging to ascertain damage levels. From the CT scans, the volume of concrete with respect to voids along the height of each specimen could be determined. As the level of dynamic compression increased, the volume of concrete increased and the voids decreased, showing that the polystyrene beads were being crushed. Also, the bottom of the specimen exhibited higher amounts of crushing of the polystyrene beads. This was likely due to reflective waves from the bottom steel platen used to support and confine the concrete specimen.