Granular Material Shear Behaviior using Elliptic Particles by Five Circle Method

Autor: Her-Min Hu, 胡賀銘
Rok vydání: 2009
Druh dokumentu: 學位論文 ; thesis
Popis: 97
This study proposes an Alternative Elliptic Particle, based on overlapping five circles with various diameters called Five Circles Method, to simulate mechanical analyses of gravel deposits. The Five Circles Method inherits the simplicity of circle, due to being combination of circles, and avoids the complication and instability of intersection calculation, argument of real contact point and conflict of tangencies between real elliptic particles. Additionally, in comparison with similar type model, Four Arcs Method, it is found that the procedure of Five Circles Method is of brevity and faster than Four Arcs Method by five times. Obviously, Five Circles Method has the advantage in algorithm. Since the normal contact force of circular particle is always through the center of mass, it is helpless for resistance of rotation. Then, the tangent friction force becomes the only source of resistance of rotation moment. That is why circular particles are easily to rotate and have poor interlocking effect. The Five Circles Method is implemented in a program with (DEM) Distinct Element Method and the advantage of shapes for numerical simulations, laboratory direct shear test of gravel deposits is simulated to research the factors of affecting shear strength by the advantage of Alternative Elliptic Particle shapes. The numerical study reveals that e (aspect ratio)is bigger and the shear strength is higher, but e=0.25 is the highest after that the shear strength becoming lower situation. Finally, a comparison with field direct shear test of San-Yi west Taiwan gravel deposits is made using circular particles and Alternative Elliptic Particles. The results of numerical simulation reveal that the Alternative Elliptic Particles could be closer the real experiment and truly more effective than circular particles to simulate shear behavior of granular materials at the same conditions.
Databáze: Networked Digital Library of Theses & Dissertations