地盤反應與土壤液化之數值模擬研究

Numerical Modeling of Site Response and Soil Liquefaction

張暐承(Wei-Chen Chang)

指導教授 ： 郭安妮

國立臺灣大學/工學院/土木工程學研究所/碩士(2014年)

https://doi.org/10.6342/NTU.2014.02002

摘要

當地震發生時，不同的地址狀況會使傳達的地震波產生改變，並產生放大或縮小的作用，而當地震波經過土層及岩盤時，波的性質會因為經過不同的土層而有所改變，可能改變的性質有頻率、振幅等等。在地震工程中，我們可以利用地盤反應分析理論來求得不同土層在震波之影響下會有甚麼樣的改變及反應。
地盤反應分析可分為時間域及頻率域來求解，在頻率域中之土壤行為可假設為線性及擬線性，而在頻率域中只能進行總應力分析，也就是無法考慮孔隙水壓之狀況；在時間域的土壤行為可假設為線性及非線性，並可做有效應力之分析，得到孔隙水壓之激發狀況。頻率域之分析是較為常見的分析方式，因為其所需之參數及步驟相較簡單，然而近年來有效應力的分析越來越多人關注，因為已發展了有效應力分析的基本參數選取及步驟，其相關研究也越來越多，但是有效應力分析在實際案例上的應用還是很少，對於參數上的選定及孔隙水壓之模型的適用性還有待釐清，本研究的目標是回顧近期有效應力動態分析之數值模擬並把總應力及有效應力之結果進行比對及討論，希望能釐清不同的參數對於有效應力分析結果之影響，並建立有效應力分析參數選取之準則或建議。本研究中進行的數值模擬包括一系列的簡易假想地址、動態三軸試驗及大型振動台試驗。而使用的數值模擬軟體有DEEPSOIL及OpenSees。
為了比較總應力及有效應力分析之結果差異，首先會進行不同假想地址之模擬比較，其中的土壤材料包含了砂土及黏土；接下來是對砂土進行動態三軸試驗之數值模擬比較，而最後是模擬大型振動台試驗；所有的數值分析都會將有效應力分析及總應力分析之結果做比較及討論。
模擬結果得到：當較小振幅時，任何分析分法的結果都十分接近。當振幅較大時，有效應力分析之結果開始與總應力分析有差異。在黏土層中所激發之孔隙水壓比砂土層來的小。用單一元素作模擬時，其結果和實驗室的數據相近。在DEEPSOIL中有模擬到類似液化的情形發生，而其孔隙水壓比也有達到１。

關鍵字

DEEPSOIL ； OpenSees ； 有限元素分析 ； 地盤反應分析 ； 有效應力分析

並列摘要

In the discipline of geotechnical earthquake engineering, theoretical site response analyses can be performed to evaluate how geologic deposit responds, in terms of particle motion and pore water pressure generation, when it is subjected to earthquake shaking.
Site response analyses can be classified according to their solution domain, the type of soil model employed, and whether pore water pressure response is considered. In common practice, frequency-domain total-stress site response analyses are often performed because parameter selection and code usage are relatively simple. Time-domain total stress site response analyses have become more popular because benchmarking studies had been performed to set up the proper parameter selection procedures and evaluate the differences between the ground motions predicted from frequency-domain and time-domain total stress analyses. On the other hand, effective-stress time-domain site response analyses are rarely performed because the parameter selection protocols for the soil material model and pore water pressure generation scheme are not available. The objective of this research is to review the currently available computer programs for effective-stress dynamic analyses and to compare the solutions from total stress and effective stress dynamic analyses. In this research, a series of numerical simulations had been run for simple hypothetical site conditions, cyclic triaxial tests and shaking table tests. From the simulation results, it is observed that the acceleration and pore pressure response predictions from different effective stress models are generally similar when the input motion level is low. However, at large input motion, the pore pressure response predictions from different effective stress models can be very different, even if the acceleration response predictions are similar. In addition, the pore pressure prediction model from OpenSees (compared to that in DEEPSOIL) seems to have a better performance as it is able to reproduce the shaking table test data on liquefied sand.

並列關鍵字

DEEPSOIL ； OpenSees ； effective stress ； site response analysis ； site effect

參考文獻

[36]. 李哲安, (2012).「一維與二維振動對砂土液化行為之影響」,國立台灣大學工學院土木工程學系碩士論文

[40]. 任芬頤, (2006). 「盤剪方式影響疏鬆砂土相對密度之研究」, 私立中原大學土木工程學系碩士學位論文

[1]. Abate G., Massimino M.R., Maugeri M. and Wood D.M. (2010). “Numerical Modelling of a Shaking Table Test for Soil-Foundation-Superstructure Interaction by Means of a Soil Consitutive Model Implemented in a FEM Code” Geotech Geol Eng (2010) 28:37-59.

[5]. Bernard R. W., Jason T. D., and Thomas S. (2012). “Guidelines for Estimation of Shear Wave Velocity Profiles” Pacific Earthquake Engineering Research Center.

[7]. Chang D., Travasarou T. and Chacko J. (2008). “Numerical Evaluation of Liquefaction-Induced Uplift for An Immersed Tunnel” The 14th World Conference on Earthquake Engineering October 12-17, 2008.