This project focuses on discrete element modelling of fracturing of concrete material at meso-scale, and particularly on calibration of the particle assembly parameters to reproduce phenomenological properties of concrete, and on applying the discrete element method to analyze the failure mechanisms in a three-point bending test and debonding between the FRP sheet and the concrete. The particle flow code PFC2D and PFC3D are employed to carry out the parametric study but only PFC2D is used in the case studies. The calibration of properties of the numerical samples is conducted to determine the effects of the particle level input parameters on the elastic constants, the uniaxial compressive strengths and failure mode of particle assembly. The input parameters are divided into two groups, model constitutive parameters (e.g., particle and bond stiffness, bond shear and normal strengths and friction coefficient) and geometric and physical parameters (e.g., particle and specimen size, particle distribution and loading velocity.). The analysis is constructed using dimensional analysis and numerical uniaxial tests. A random aggregate generation algorithm is incorporated in the DEM code to reproduce the aggregate structure in real concrete material. The aggregate generation algorithm utilizes polygon and polyhedron as the basic shapes of aggregate and is capable of producing multi-graded concrete specimens with aggregate content up to 80% and 60% for two-dimensional and three-dimensional samples respectively. The mode I fracture behavior of three-phased concrete is then simulated by performing a virtual three-point bending test. The mortar matrix phase is simulated with the linear elastic-pure-brittle and softening bond model to ensure a fair comparison. The dynamic debonding process between the FRP sheet and the concrete is simulated with a particle assembly by a regular hexagonal packing arrangement where the heterogeneity of concrete is taken into account by incorporating the Weibull distribution. Based on the analysis of the modelling results, it is conclude that the fracture behavior of concrete can be satisfactorily captured by meso-scale DEM model and comprehensive parameter study allows more confidently implementation of particle flow code.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:678449 |
| Date | January 2013 |
| Creators | Tang, Lingwei |
| Publisher | Swansea University |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | https://cronfa.swan.ac.uk/Record/cronfa42301 |
Page generated in 0.0017 seconds