Comparisons of Model Responses and Input Parameter Calibrations of PM4Sand (Version 3.1) Model

抄録

Liquefaction hazards can cause damage to structures and loss of life and property. The simplified prediction equation, numerical simulation and model test are three commonly used approaches to evaluate the damage of the liquefaction hazards or the effectiveness of the mitigation plan. Because the numerical simulation can include the site-specific conditions into the evaluation and perform a parametric study under a reasonable time and affordable budget, many research and engineering projects choose this approach as an evaluation tool. In order to obtain rational and comparable simulation results, appropriate constitutive models and model inputs are essential. Many liquefaction constitutive models (Finn model, UBCSAND model, SANISAND model, PM4Sand model etc.) have been introduced and coded for numerical programs (PLAXIS, FLAC, MIDAS etc.) to study liquefaction hazards and many related topics. Among these constitutive models, PM4Sand model has been frequently used in recent projects to model the soil liquefaction. PM4Sand model is modified from the bounding surface plasticity model for geotechnical earthquake engineering applications and coded as a user-written constitutive model in FLAC. In this study, inputs of PM4Sand (version 3.1) model are calibrated against the triggering curve of simplified liquefaction analysis procedures suggested by Seismic Design Specifications and Commentary of Buildings of Taiwan for future liquefaction evaluation projects. In addition, model responses of PM4Sand model are discussed and compared with laboratory tests