考虑多维地震动及重力作用的土层场地地震反应分析模型

合理有效的工程场地自由场分析是开展地下结构地震反应计算、发展地下结构实用抗震分析方法的前提。目前常用的自由场分析方法多局限于单向地震动输入，难以考虑多向地震动共同作用时的非线性耦合效应。鉴于此，提出一种考虑水平和竖向地震动及重力荷载共同作用的土层地震反应分析力学模型，该模型采用捆绑约束条件作为侧面边界条件，采用基于黏性边界的地震动输入方法将输入地震动转化为等效地震荷载，可实现自由场计算中水平和竖向地震动以及重力荷载的同时施加，简化了计算步骤，减少了计算成本。此外，该方法考虑了水平和竖向地震荷载的叠加效应，对于考虑材料非线性及大变形影响的土层场地地震反应分析具有更为良好的计算精度与适用性。     

基于自由场局部变形的地下结构抗震Pushover分析方法

对地下结构抗震Pushover分析方法进行了改进,采用自由场局部变形峰值作为目标位移,局部变形峰值时刻对应的土层水平加速度作为等效惯性加速度输入。给出了局部变形峰值和等效惯性加速度确定方式,详细介绍了基于自由场局部变形的地下结构抗震Pushover分析方法实施步骤、使用方法和功能特点。该方法更有针对性地考虑了强地震作用下不同埋深地下结构与土体的非线性特征以及两者之间的相互作用,通过分析变形和受力情况可以得到完整的能力曲线,更好地评估地下结构抗震性能。使用本文方法对3种埋深的地下结构进行计算,并与动力非线性分析结果进行对比研究。结果表明:本方法在计算稳定性和模拟精度方面优于基于自由场整体变形的Pushover方法;对于不同的输入地震波,能力曲线的吻合程度更高;在强震和罕遇地震情况下,对于深埋地下结构,计算结果略大于动力非线性结果,对实际工程而言更加安全。     

地下结构地震反应分析拟静力法与动力非线性时程法的比较

以典型的两层三跨地铁地下车站结构为研究对象,采用基于自由场位移场频域等效线性分析结果的反应位移法和反应加速度法分析地下结构的地震弯矩,并与基于ABAQUS软件的地下结构地震反应非线性时程分析的计算结果进行了比较,对比了不同地震动强度、结构刚度及顶板埋深对地下结构地震弯矩的影响。结果表明:输入地震动强度越大,拟静力法与动力非线性时程法计算结果之间的差异越大;地下结构刚度越大,两者之间的差异也越大;中等强度的地震动作用时,相对于动力非线性时程法的计算结果,浅埋结构的拟静力法结果偏大,深埋结构的拟静力法计算结果偏小;反应位移法和反应加速度法的计算结果差异很小。该研究结果对地下结构抗震设计方法的比选具有一定的指导意义。     

基于ABAQUS的粘弹性边界单元及在重力坝抗震分析中的应用

考虑了2种不同的地震输入模型,即无质量地基模型和考虑辐射阻尼的粘弹性边界模型。以大型非线性有限元程序ABAQUS为平台,采用FORTRAN语言开发了用户单元子程序VSB_UEL.for,将粘弹性边界有效地嵌入到ABAQUS中。在尽量避免求解自由场的前提下,将离散的地震荷载转化为等效结点荷载,编写的计算程序可以方便地将等效荷载施加到人工边界结点上,并通过数值算例验证了粘弹性边界单元及波动输入程序的正确性,其求解效率和计算精度均令人满意,最后将该程序应用到在建的某水电站厂房坝段的抗震分析中。结果表明,与传统的固定边界无质量地基相比,考虑了粘弹性人工边界后,坝体的动力响应峰值减小了20%~40%;在进行结构动力响应分析时,考虑无限域地基的辐射阻尼影响是很有必要的。文中的用户子程序及波动输入程序很容易扩展至三维,为同类工程的抗震分析提供了简洁、合理的计算模式。     

基于DEEPSOIL的软土场地地震反应研究

软土场地地震反应分析是目前工程场地地震安全性评价中的重要组成部分,对场地设计地震动参数的确定具有重要意义。利用一维场地地震反应分析软件DEEPSOIL,可进行场地线性、等效线性化和时域非线性等多种分析,并可考虑孔隙水压的影响。笔者根据土层计算参数,编制了DEEPSOIL软件场地模型输入文件的自动生成程序,可高效、快速地完成对场地的建模。通过数值算例验证了DEEPSOIL软件的精度。同时通过对某典型Ⅲ类软土场地的地震反应分析,研究了拟合参数的敏感性以及等效线性化方法和时域非线性方法对峰值加速度和地表加速度反应谱的影响,并指出了等效线性化方法在分析软土场地地震反应中的不足。对于软土场地建议采用DEEPSOIL软件进行时域非线性分析,因为其参数简单并容易确定,适合建模快速和使用方便的要求。     

桩-土-结构体系随机地震响应的实用计算方法及参数分析

考虑地震的随机性和土体的非线性,引入粘弹性边界,提出了一种在谐响应分析中直接输入加速度的加载方式,建立了基于虚拟激励法、等效线性化法和确定性动力有限元法相结合的非线性桩-土-结构体系随机地震响应的实用计算方法.采用该法对单桩支承的剪切型结构进行了较为全面的参数分析.结果表明基岩地震动输入水平,桩长度和土体模量都对结构反应有较大影响.该方法可方便的用于随机地震下桩-土-结构体系的参数分析中.     

海洋平台结构地震反应的近似计算

本文对海洋平台结构地震反应运动方程中的非线性流体阻力项,按振动一周力作功相等的原则,等效为线性的粘滞阻尼力。在等效的过程中,作者提出了所谓“拟静态位移”的概念,用它代替等效粘滞阻尼系数中与频率有关的位移幅值,从而使计算大为简化。文中还以一简单的海洋工程结构为例,按该近似方法计算了它在某地震记录输入下的反应,并与“Fourier变换-时域迭代混合法”计算的结果进行了对比。结果表明:利用本文的近似方法计算海洋平台结构的地震反应不仅速度快、费用少,而且具有较高的精度。 该法还可以推广应用于同时受地震荷载及波浪荷载作用的海洋平台结构的反应分析。     

土-结构动力相互作用问题分析中地震动输入的一种新方法

为实现地震作用下土-结构动力相互作用问题的有限元模拟,需要在人工边界上完成地震动的有效输入,目前工程和科研中常用的地震动输入方法有两种:波动输入方法和振动输入方法。波动输入方法的模拟精度高,但实施上相对复杂且耗时,而振动输入方法处理简单,但模拟精度较低。针对应力型人工边界提出一种在人工边界上实现地震动输入的新方法,该方法通过对土-结构有限元模型中由人工边界节点及相邻节点组成的局部子结构施加自由波场位移时程并进行动力分析,从而直接获得可实现地震波动有效输入的等效地震荷载,然后在土-结构有限元模型的人工边界节点上施加等效输入地震荷载并完成动力计算,由此完成土-结构动力相互作用问题的地震动输入和地震反应计算。与原有波动输入方法相比,新方法避免了原方法需分别计算人工边界上自由场应力和由引入人工边界条件引起的附加力,以及需要根据不同人工边界面的外法线方向确定荷载作用方向等较为复杂的处理过程,具有等效地震荷载计算简便、地震动输入过程更易于实施的特点。采用均匀弹性半空间和成层弹性半空间一维地震反应算例初步验证新方法的正确性和可靠性。     

考虑土-结构相互作用的上海中心大厦地震反应

采用ANSYS有限元软件建立土-桩-上海中心大厦相互作用简化模型.其中,桩土区采用等效模型,近域土体定为塑性区,用DP模型模拟;外围的土域定为弹性区,用超单元来模拟.对考虑土,桩-结构相互作用的整体结构和以刚性地基为假定的上部结构分别进行地震反应分析,并完成了比较.最后,在整体结构中提取上部结构与下部结构处的加速度反应与原地震波叠加,形成修正地震波,为输入修正地震波能考虑相互作用因素来分析相同结构的精细模型地震反应提供了条件.     

饱和砂土透镜体液化对建筑物地震反应的影响

采用一种能分析有结构物存在的场地地震液化问题的二维有效应力有限元分析方法,研究饱和砂土透镜体液化对建筑物地震反应的影响。计算中采用了更为合理的迭代方式处理土的非线性,考虑了Ｋｃ对孔压的作用,引入了透射边界。取建筑物为短周期结构。考虑了透镜化宽度、厚度、埋深以及输入地震动类型幅值对结构加速度反应的影响。计算结果表明：（１）所采用的方法与已有的模型实验结果有很好的对应关系,可用于招考莪存在下的场地砂土     

土-结构相互作用分析中几类地震输入方式的比较

地震作用下土-结构动力相互作用的整体有限元分析需要在人工边界处输入地震动。目前可能采用的地震输入方法包括黏弹性边界自由场输入方法、自由场应力方法、自由场位移方法以及侧边界自由方法。由于采用近似人工边界条件或者未完全考虑地震自由场,上述地震输入方法均为近似方法。本文以大开地铁车站二维有限元分析为例,根据规范建议的边界位置,研究了上述地震输入方法的精度,研究成果可为土-结构相互作用分析的合理地震输入提供一定参考。     

On the effective seismic input for non-linear soil-structure interaction systems

Two equivalent semi-discrete formulations are presented for the problem of the transient response of soil-structure interaction systems to seismic excitation, considering linear behaviour of the soil material and arbitrary non-linear structural properties. One formulation results in a direct method of analysis in which the motion in the structure and the entire soil medium, rendered finite by an artificial absorbing boundary, is determined simultaneously. The other represents a substructuring technique in which the structure and the soil are analysed separately. The forces induced in the discretized system by the incident seismic motion are obtained as part of the general formulation by using the free-field motion of the unaltered soil as the earthquake input. It is shown that these forces act within the soil region in the direct method, but only on the soil-structure interface in the substructure formulation. Both sets of forces, however, involve only the displacements and tractions acting on the fictitious surface in the unaltered (linear) soil which coincides with the soil-structure interface of the complete system. It is shown, further, that the free-field displacements alone define a minimal set of data for evaluating the seismic response of the structure, since the tractions and displacements on that surface are interrelated. In practice, the minimal set must be obtained by extrapolating the available information, as the free-field ground motion at a site is usually specified at a single reference point.     

Correlation of predicted seismic response using hybrid modelling with EPRI/TPC lotung experimental data

The hybrid modelling method is presented herein along with the equivalent linearization method to take account of the strain-dependent non-linearity of soils in a soil-structure interaction (SSI) seismic analysis. A refined substructuring of the soil-structure system is utilized and two separate analyses are made to determine the soil free-field and SSI motions induced by earthquake excitation. This method is used to predict the seismic response of a 1/4-scale containment model built in the seismically active area of Lotung, Taiwan. The results obtained show excellent correlation with the field test results.     

Longitudinal integral response deformation method for the seismic analysis of a tunnel structure

For the longitudinal seismic response analysis of a tunnel structure under asynchronous earthquake excitations, a longitudinal integral response deformation method classified as a practical approach is proposed in this paper. The determinations of the structural critical moments when maximal deformations and internal forces in the longitudinal direction occur are deduced as well. When applying the proposed method, the static analysis of the free-field computation model subjected to the least favorable free-field deformation at the tunnel buried depth is performed first to calculate the equivalent input seismic loads. Then, the equivalent input seismic loads are imposed on the integral tunnel-foundation computation model to conduct the static calculation. Afterwards, the critical longitudinal seismic responses of the tunnel are obtained. The applicability of the new method is verified by comparing the seismic responses of a shield tunnel structure in Beijing, determined by the proposed procedure and by a dynamic time-history analysis under a series of obliquely incident out-ofplane and in-plane waves. The results show that the proposed method has a clear concept with high accuracy and simple progress. Meanwhile, this method provides a feasible way to determine the critical moments of the longitudinal seismic responses of a tunnel structure. Therefore, the proposed method can be effectively applied to analyze the seismic response of a long-line underground structure subjected to non-uniform excitations.     

地下综合管廊土-结构接触面参数对地震动力响应的影响数值分析

为研究土-结构接触面参数对地下综合管廊地震动力响应特征的影响,建立动力有限元数值模型,模型边界采用激励侧固定边界、远离激励侧黏性边界、其余侧自由场边界的优化组合动力边界,土体本构采用HSS模型,接触面采用改进Goodman单元,动力荷载考虑三种情况(Rayleigh波的作用、底部激励了美国加利福尼亚Upland地震波以及前两者的共同作用),分别研究不同地震动输入、接触面折减系数的改变对综合管廊内力及加速度的影响。研究结果表明:在相同的折减系数条件下,与静力作用相比,动力作用下的结构内力明显增大,综合管廊设计时应考虑地震荷载作用下内力增大的情况;随着界面折减系数的增加,正弯矩极值减小,负弯矩极值增大,加速度峰值增大;在相同接触面折减系数条件下,底部地震波输入产生的结构内力极值显著高于仅有Rayleigh波输入的情况;考虑Rayleigh波和地震波共同作用条件下,引起的管廊结构内力极值与仅考虑底部地震波输入时的结构内力极值差异不大。研究成果可供地下综合管廊结构地震响应精细化数值模拟及抗震设计参考。     

Influence of soil to structure stiffness on the accuracy of the pushover method for underground structures

The pushover method for underground structures is a seismic analysis method featured by high calculation accuracy and a simple implementation process. The method has been widely used in seismic design and other related scientific research; however, the influence of different soil-structure flexibility ratios on the accuracy of this method is still not well understood. In this study, we select the cross-section structures beneath the Daikai subway station as the research object and establish 12 finite element analysis models with different soil-structure flexibility ratios using ABAQUS. All models are computed by the dynamic time-history method or the pushover method. Furthermore, the dynamic time-history solution result is taken as the standard solution, and the precision and application of the pushover analysis method are discussed based on the parameters of peak interlayer displacement and peak internal force of the middle column section. The results show that the soil-structure flexibility ratio has a significant influence on the calculation accuracy of the pushover method, and the calculation accuracy of this method is the most ideal when the soil-structure flexibility is equal to 1. The research results can provide significant references for the seismic design of underground structures or the improvement of simplified seismic analysis methods.     

极限安全地震动下考虑土-结构相互作用的核电站安全壳响应分析

根据黏弹性人工边界的基本原理,结合有限元分析软件ABAQUS和MATLAB辅助程序,在地基有限区域上添加黏弹性人工边界并实现极限安全地震动的输入。基于ABAQUS软件平台,对CPR1000安全壳构建了土-结构相互作用体系的数值模拟模型,分析其在极限地震动下的动力响应,并将计算结果与考虑刚性基础的安全壳结构响应数据进行对比。结果表明:核电站CPR1000安全壳结构在极限安全地震动下仍能保持良好的密闭性。考虑土-结构相互作用后分析所得安全壳结构受到的应力、加速度峰值和相对位移均有所增大,使用刚性地基模型要偏于危险。     

A simple boundary element formulation and its application to wavefield excited soil-structure interaction

A simple boundary element formulation which is based directly on the point load solutions for an elastic full-space is presented. It is integrated in a finite element program to calculate dynamic soil-structure interaction problems. The combined boundary and finite element method is applied to structures which are excited by horizontally propagating waves in the soil. For three different types of flexible structure-elastic beams, low and high (square) shear walls-and the corresponding rigid structures the vibration modes and the soil-structure transfer functions have been investigated. The flexible foundations display the same wave pattern as the exciting free-field of the soil, but the amplitudes are reduced with increasing frequency, depending on the stiffness or wave resistance of the structure. Rigid structures show, in part, quite different behaviour, giving free-field reductions caused by kinematic and inertial soil-structure interaction.     

Antiplane response of a dike with flexible soil-structure interface to incident SH waves

Studies of the effects of differential ground motions on structural response generally do not consider the effects of the soil-structure interaction. On the other end, studies of soil-structure interaction commonly assume that the foundation of the structure (surface or embedded) is rigid. The former ignore the scattering of waves from the foundation and radiation of energy from the structure back to the soil, while the latter ignore quasi-static forces in the foundations and lower part of the structure deforming due to the wave passage. This paper studies a simple model of a dike but considers both the soil-structure interaction and the flexibility of the foundation. The structure is represented by a wedge resting on a half-space and excited by incident plane SH-waves. The structural ‘foundation’ is a flexible surface that can deform during the passage of seismic waves. The wave function expansion method is used to solve for the motions in the half-pace and in the structure. The displacements and stresses in the structure are compared with those for a fixed-base model shaken by the free-field motion. The results show large displacements near the base of the structure due to the differential motion of the base caused by the wave passage.