A micromechanical study of the standard penetration test
- Autores: Ningning Zhang
- Directores de la Tesis: Antonio Gesn Solé (dir. tes.), Marcos Arroyo Álvarez de Toledo (dir. tes.), Matteo Ciantia (codir. tes.)
- Lectura: En la Universitat Politècnica de Catalunya (UPC) ( España ) en 2020
- Idioma: español
- Tribunal Calificador de la Tesis: Mingjing Jiang (presid.), José Antonio Gili Ripoll (secret.), Pierre Breul (voc.)
- Programa de doctorado: Programa de Doctorado en Ingeniería del Terreno por la Universidad Politécnica de Catalunya
- Materias:
- Texto completo no disponible (Saber más ...)
- Resumen
This thesis explores the potential of models based on the discrete element method (DEM) to study dynamic probing of granular materials, considering realistic particle-scale properties.
The virtual calibration chamber technique, based on the discrete element method, is applied to study the standard penetration test (SPT). A macro-element approach is used to represent a rod driven with an impact like those applied to perform SPT. The rod is driven into a chamber filled with a scaled discrete analogue of a quartz sand. The contact properties of the discrete analogue are calibrated simulating two low-pressure triaxial tests. The rod is driven changing input energy and controlling initial density and confinement stress. Energy-based blowcount normalization is shown to be effective. Results obtained are in good quantitative agreement with well-accepted experimentally-based relations between blowcount, density and overburden.
A comprehensive energetic balance of the virtual calibration chamber is conducted. Energy balance is applied separately to the driven rod and the chamber system, giving a detailed account of all the different energy terms. The characterization of the evolution and distribution of each energy component is investigated. It appears that the SPT test input energy is mainly dissipated in friction. The energy-based interpretation of SPT dynamic response proposed by Schnaid et al. (2017) is then validated in comparisons between static and dynamic penetration results. Moreover, microscale investigation provides important information on energy dissipation mechanisms.
A well-established DEM crushing contact model and a rough Hertzian contact model are combined to incorporate both effects in a single contact model. The efficient user defined contact model (UDCM) technique is used for the contact model implementation. Parametric studies explore the effect of particle roughness on single particle crushing event. The model is then used to recalibrate the contact properties of the quartz sand, being able to use realistic contact properties and then correctly capture both load-unload behaviour and particle size distribution evolution. The calibration chamber results are exploited to investigate the relation between static and dynamic penetration test. This is done first for unbreakable materials and later for crushable and rough-crushable ones. It is shown that the tip resistance measured under impact dynamic penetration conditions is very close to that under constant velocity conditions, hence supporting recent proposals to relate CPT and SPT results. It is also shown that penetration resistance reduces if particles are allowed to break, particularly when roughness is also considered.
