Using Discrete Element Method to Simulate Excavation Behavior in Gravel

• Main Authors: Guo-Wei Lin, 林國偉
• Other Authors: Jian-Zhong Jiang
• Format: Others
• Language: zh-TW
• Published: 2014
• Online Access: http://ndltd.ncl.edu.tw/handle/40426119153722969733

碩士 === 國立雲林科技大學 === 營建工程系 === 102 === If the strength of the gravel layer is not good enough, grouting will be performed every 2-3 meters deep to prevent collapses during the gravel excavations. Therefore, a discrete element program PFC2D (Itasca 2004) was employed to model the gravel layer on site and, according to the geological survey of certain areas in the western Taichung City, to simulate staged excavations. This research was aimed to control the material parameters of parallel-bond models of the grouting effect at the simulated excavation sites. With such simulation, we expected a series of parameter analyses to obtain safe and cost-efficient design principles of excavation as reference. If the excavation sites not given bond model to simulate the grouting, the excavation sites will be fully collapse. Cross validation of parameter analysis showed that, firstly, the parallel-bond shear stiffness (Pb_ks) had more influence on the horizontal displacements of excavated wall than the parallel-bond normal stiffness (Pb_kn). Secondly, the 50cm granular-bond thickness of simulated grouting effect on the excavation sites was better than 30cm one. The maximum horizontal displacement variation of the 50cm granular-bond thickness in the three stages were 43%, 55% and 58% respectively, compared with the 30cm granular-bond thickness.The maximum horizontal displacements occurred between 0 and 0.75 meters deep, and tended to increase while approaching the surface. There were heave at the bottom of excavation in the first two stages, while, without squeezing downward, the 3rd stage excavation had a less rugged bottom despite its wider settlement on the ground surface.