【作者】 李玲；

【导师】 黄博；

【作者基本信息】 浙江大学 ， 岩土工程， 2013， 硕士

【摘要】 小应变剪切模量Gmax是场地地震响应分析中的重要参数。在以往的场地地震响应有效应力分析法中,通常假定循环振动作用过程中土体的Gmax等同于相同有效应力静力状态下的Gmax,采用Hardin公式计算Gmax。而Hardin公式是根据土体静力状态下的试验所得到的,只考虑了有效应力和孔隙比等参数对Gmax的影响,未考虑循环振动历史这一重要因素。但已有的研究表明,在地震动中大应变幅振动将导致Gmax附加衰减,即地震过程中Gmax要比相应有效应力下Hardin公式计算得到的Gmax还要低,这部分由于地震振动导致的Gmax与静力状态下Gmax的区别,被称为附加衰减。目前,尚未有人探讨这种由于地震振动作用而产生的Gmax附加衰减给场地地震响应带来的影响。本文在回顾前人对各类土体Gmax受大振幅振动影响研究的基础上,得到了对应于强结构性土和弱结构性土的两种不同Gmax附加衰减模式。将附加衰减模量值GmaxⅡ与相同有效应力下静力状态的小应变剪切模量值GmaxⅠ之比作为参量进行分析发现,弱结构性土和强结构性土的模量比GmaxⅡ/GmaxⅠ与有效应力比σm’/σm0’。的关系可分别用两段式和三段式来表示,文中拟合得到了不同衰减模式对应的数学表达式。通过编制可考虑Gmax附加衰减影响的一维场地地震响应分析程序,对典型场地地震响应进行了分析对比。结果发现：Gmax附加衰减对场地加速度和剪应力响应均有较大影响,并将加速场地液化、扩大场地最终液化范围。地表加速度谱分析表明,考虑Gmax附加衰减后土体更加软化,对地震波短周期部分有减震作用,对长周期部分有加震作用。此外,文中还考察了场地条件和输入波卓越周期的影响,发现Gmax附加衰减对场地地震响应的影响程度随地下水位下降而增强,随场地土层厚度的增加无明显变化；强结构性土场地受到的影响比弱结构性土场地更大；不同卓越周期输入下均表现出由于Gmax附加衰减加快场地液化的现象。为进一步考察Gmax附加衰减对场地地震响应的影响,采用该程序模拟分析了饱和砂土场地的离心机振动台模型试验。为获得离心机振动台试验所采用福建砂的孔压模型参数,进行了排水条件下等剪应变幅扭剪试验和压缩回弹试验,得到应变型孔压模型中计算体应变增量Δεvd和土体回弹模量Er的试验参数。通过对不同幅值地震加速度输入下模型场地动力响应的数值模拟,并和试验结果比较,两者所得加速度响应吻合得很好,孔压响应存在数值模拟中孔压增长慢于试验实测结果的现象,认为可能是由于离心机振动台试验模型边界和确定孔压参数的试验中空心圆柱试样边界不同造成的；两者所得土层液化深度及各层达到液化的时间均较为相符。考虑Gmax附加衰减对模型场地的影响,得出：当输入峰值加速度较小时,Gmax附加衰减对场地地震响应有显著影响；当场地输入峰值加速度较大时,由于场地迅速发生液化,Gmax附加衰减对场地地震响应影响减弱甚至无影响。更多还原

【Abstract】 Gmax is an important parameter in site seismic response analysis, and it was often obtained by Hardin Formula in previous effective stress analysis method. However, Hardin formula is based on static tests, considering only the influence of effective stress and void ratio, but not cyclic stress history. Soil’s Gmaxunder cyclic stress was assumed to be equivalent to the same effective stress static state Gmax. However, many studies have proven the attenuation of Gmax under large dynamic strain, namely the value of Gmax under earthquake is lower than the corresponding one calculated by Hardin Formula with the same effective stress. The attenuation caused by earthquake was named additional attenuation. Currently, few studies had been done on additional attenuation’s influence on site response of earthquake.Previous studies on Gmax under large amplitude cyclic load are reviewed, and two modes of additional attenuation of Gmax are summarized corresponding to strong and weak structural soils. It’s found that the relation between modulus ratio Gmax11/Gmax1(Gmax1refers small strain modulus under static stress state, Gmax11refers additional attenuation modulus under cyclic dynamic loading) and effective stress ratio σm/σm0can be represented by two curves or three curves. Different decay modes are presented in this paper to facilitate for numerical calculations.A refined one-dimensional equivalent linearization method has been composed on the basis of effective stress principle, to account for the influence of additional attenuation of Gmax.It can be observed that the additional attenuation of Gmax has influence both on the response of accelerations and shear stresses. Besides, the additional attenuation of Gmax can speed up the liquefaction of site and expand the final liquefied range. The influences of different site conditions and input predominant period on seismic site response are also investigated. The results indicate that the influence of the additional attenuation of Gmax on site increases as the underground water level decline, but slight changes occur as the thickness of site increases. Moreover, the influence of the additional attenuation of Gmax on strong structural soil site is greater than weak structural soil site. Faster liquefaction due to additional attenuation of Gmax happens in different input predominant periods.To further investigate the influence of Gmax additional attenuation on the seismic response of sites, the program mentioned above was used in the simulation and analysis of dynamic centrifuge model test of saturated sands. Strain control shear test and compression and resilience test were carried out to obtain the pore pressure generation model’s parameters of Fujian sand used in dynamic centrifuge model test, and parameters useful to calculate the body strain increment△εvd and soil resilient modulus Er are obtained. Numerical calculations were carried out to simulate the responses of site model under different seismic acceleration amplitudes, comparison between numerical calculations and tests showed that the responses of both methods agreed with each other perfectly. The growth rate of pore pressure of numerical calculations were slower than that of dynamic tests, which may be due to the initial condition differences between dynamic centrifuge model tests and hollow cylinder tests which determine the pore pressure parameters. No obvious differences of depth of liquefied soil and beginning time of each layer were found in both methods. Conclusions can be drawn as follows, additional attenuation of Gmax makes obvious influence on the seismic response of site when subjected to slight earthquake, but it makes little influence or even no impact on the seismic response of site when subjected to severe earthquake because of soil’s rapid liquefaction.更多还原