基于解耦传播的波场分解方法在VTI介质弹性波逆时偏移中的应用

由于弹性波逆时偏移更符合实际情况,而且转换波的成像结果有更高的分辨率,因此弹性波逆时偏移越来越受到重视。然而,弹性波逆时偏移需要多波多分量数据,为了减少成像结果中的串扰假象,在逆时偏移过程中进行P波和S波分解就变得非常必要。结合基于向量的激发振幅成像条件,我们把基于解耦传播的波场分解方法应用到弹性波逆时偏移中,并对比了其在各向同性介质和横向各向同性（VTI）介质中的应用效果。结果说明,该方法可以在各向同性介质中完全分解P波和S波,并保留波场的向量信息。虽然在VTI介质中有较小的分解残余,但是该分解残余不会在逆时偏移结果中产生明显的串扰;因此,这种波场分解方法可以应用于各向同性介质和VTI介质弹性波逆时偏移。该方法是在时间空间域实现的,可以在波场传播过程中直接对P波和S波进行分离,应用方便,计算效率高。与利用Helmholtz分解的弹性波逆时偏移相比,该方法避免了在PS波成像结果中的极性反转问题。在复杂Hess VTI模型的逆时偏移结果中,高速岩体和断层的成像清晰,甚至是两个低速薄夹层也能较好成像;这说明该方法对复杂介质具有较好的适应性。PS波成像结果中的各向异性体成像清晰,说明各向异性介质弹性波逆时偏移可以对传统逆时偏移不能很好成像的构造进行成像。     

A Study of the Stability Properties in Simulation of Wave Propagation with SPH Method

When ordinary Smoothed Particle Hydrodynamics (SPH) method is used to simulate wave propagation in a wave tank, it is usually observed that the wave height decays and the wave length elongates along the direction of wave propagation. Accompanied with this phenomenon, the pressure under water decays either and shows a big oscillation simultaneously. The reason is the natural potential tensile instability of modeling water motion with ordinary SPH which is caused by particle negative stress in the computation. To deal with the problems, a new sextic kernel function is proposed to reduce this instability. An appropriate smooth length is given and its computation criterion is also suggested. At the same time, a new kind dynamic boundary condition is introduced. Based on these improvements, the new SPH method named stability improved SPH (SISPH) can simulate the wave propagation well. Both the water surface and pressure can be well expressed and the oscillation of pressure is nearly eliminated. Compared with other improved methods, SISPH can truly reveal the physical reality without bringing some new problems in a simple way.     

Aftershock seismic assessment taking into account postmainshock residual drifts

This paper introduces and evaluates a methodology for the aftershock seismic assessment of buildings taking explicitly into account residual drift demands after the mainshock (i.e., postmainshock residual interstory drifts, RIDR_o). The methodology is applied to a testbed four‐story steel moment‐resisting building designed with modern seismic design provisions when subjected to a set of near‐fault mainshock–aftershock seismic sequences that induce five levels of RIDR_o. Once the postmainshock residual drift is induced to the building model, a postmainshock incremental dynamic analysis is performed under each aftershock to obtain its collapse capacity and its capacity associated to demolition (i.e., the capacity to reach or exceed a 2% residual drift). The effect of additional sources of stiffness and strength (i.e., interior gravity frames and slab contribution) and the polarity of the aftershocks are examined in this study. Results of this investigation show that the collapse potential under aftershocks strongly depends on the modeling approach (i.e., the aftershock collapse potential is modified when additional sources of lateral stiffness and strength are included in the analytical model). Furthermore, it is demonstrated that the aftershock capacity associated to demolition (i.e., the aftershock collapse capacity associated to a residual interstory drift that leads to an imminent demolition) is lower than that of the aftershock collapse capacity, which mean that this parameter should be a better measure of the building residual capacity against aftershocks. Copyright © 2014 John Wiley & Sons, Ltd.     

黏弹性节理岩体的位移不连续模型研究

选用三参数标准线性固体作为岩石本构,提出了一种考虑岩体黏弹性的位移不连续模型;根据一维黏弹性波的特征线法,推导了节理处质点速度、应力和应变递推公式。首先,基于分离式霍普金森压杆（SHPB）对砂层进行试验,得到其应力-应变关系,并换算出砂层节理的法向刚度;接着,通过一维强间断黏弹性波的波速公式、高频波衰减系数以及任一频率下的衰减系数,确定出数值算法中的三参数。最后,基于自制的摆锤装置,探讨了一维应力波在节理岩体中的传播规律,试验中以两根长1 000 mm、直径为68.50 m的岩杆作为入射和透射杆,以3 mm砂层模拟节理。试验和数值结果吻合度良好,进一步验证了该方法的可靠性。     

岩石单轴压缩下能量与损伤演化规律研究

能量变化是材料各参数变化的本质特征。通过颗粒流模拟岩石单轴压缩试验,从细观力学角度分析能量转化、裂纹扩展及损伤演化规律。颗粒流利用细观颗粒的运动克服了宏观力学理论不易实现试件多裂纹破坏形态的缺点,以此研究能量变化更加合理。应力-应变各阶段能量与微裂纹及损伤存在着相互对应的发展关系。通过3种工况论证了宏-细观力学参数对应关系,采用最小二乘法拟合得出微裂纹与轴向应变呈幂次函数关系。采用割线模量定义损伤变量,取模量加速下降处（对应裂纹加速扩展）为损伤门槛,对应损伤阈值为0.158。3种工况下微裂纹数量与损伤呈线性发展关系,为损伤演化的深入研究提供参考。     

起伏地表弹性波传播的间断Galerkin有限元数值模拟方法

间断Galerkin有限元法（DG-FEM）作为一种有效的高阶有限元法受到了国内外学者的广泛关注.本文基于任意高阶间断Galerkin有限元法对弹性波方程进行空间离散,并将离散后所得的非齐次线性常微分方程系统齐次化,最后结合针对齐次问题的强稳定性保持龙格库塔（SSP Runge-Kutta）算法,将DG-FEM推广至时间任意高阶精度.另外,借鉴近最佳匹配层（NPML）的思想,基于复频移（CFS）拉伸坐标变换推导了一种新的PML吸收边界条件（简称为CFS-NPML）,该CFS-NPML能够与DG-FEM算法很好地结合,形成有效的起伏地表地震波传播数值模拟技术.数值试验结果表明,DG-FEM具有高阶精度,可以适应任意复杂起伏地表和复杂构造情况下的弹性波传播数值模拟.同时,CFS-NPML对包括面波等震相的人为边界反射都具有良好的吸收效果.     

1D System identification of a 54‐story steel frame building by seismic interferometry

A 54‐story steel, perimeter‐frame building in downtown Los Angeles, California, is identified by a wave method using records of the Northridge earthquake of 1994 (M_L = 6.4, R = 32 km). The building is represented as a layered shear beam and a torsional shaft, characterized by the corresponding velocities of vertically propagating waves through the structure. The previously introduced waveform inversion algorithm is applied, which fits in the least squares sense pulses in low‐pass filtered impulse response functions computed at different stories. This paper demonstrates that layered shear beam and torsional shaft models are valid for this building, within bands that include the first five modes of vibration for each of the North–South (NS), East–West (EW), and torsional responses (0–1.7 Hz for NS and EW, and 0–3.5 Hz for the torsional response). The observed pulse travel time from ground floor to penthouse level is τ ≈1.5 s for NS and EW and τ ≈ 0.9 s for the torsional responses. The identified equivalent uniform shear beam wave velocities are β_eq ≈ 140 m/s for NS and EW responses, and 260 m/s for torsion, and the apparent Q ≈ 25 for the NS and torsional, and ≈14 for the EW response. Across the layers, the wave velocity varied 90–170 m/s for the NS, 80–180 m/s for the EW, and 170–350 m/s for the torsional responses. The identification method is intended for use in structural health monitoring. Copyright © 2013 John Wiley & Sons, Ltd.     

Series expansion of complex ground amplifications with a sequence of simple transfer functions

Amplification of earthquake ground motions at actual deposit sites is an important factor to consider when assessing the risk of an earthquake disaster. In order to identify the amplification properties, several preprocessings such as the Fourier transform are required. I propose a series expansion of the amplification with simple ground transfer functions as a new preprocessing. I define a sequence of transfer functions based on a two‐layered structure excluding an internal damping and a function space spanned by the set of the functions. I mathematically prove that the function space is equal to L² space. This indicates that all the functions belonging to L² space, that is, an arbitrary ground amplification, have a unique series expansion. This expansion is applied to the physics‐based decomposition of the amplification. Some numerical examples indicate that the similarity between a target complex structure and a simple model is measured by the absolute value of each coefficient in the series expansion. Copyright © 2014 John Wiley & Sons, Ltd.     

库木库里背斜活动性初步研究

库木库里盆地位于青藏高原北缘,与柴达木盆地一山之隔,是二者的过渡地带,也是高原主体部分向NE扩展的前缘地区;现今构造表现为被3条大型活动构造带(走滑的阿尔金断裂带、东昆仑断裂带和逆冲的祁漫塔格褶皱逆冲系)所夹持。因此,该盆地对于研究青藏高原北缘的构造活动性、活动历史,探讨高原的扩展模式具有十分重要的意义。虽然库木库里盆地南、北两侧均发育活动性很强的大型走滑断裂,但是在盆地中央发育1条大型背斜,走向NWW-SEE,与祁漫塔格褶皱逆冲系和柴达木盆地内的褶皱构造走向一致,说明盆地目前遭受NNE向的挤压。通过对盆地地形横、纵剖面和阶地展布形态的分析,得出背斜有自西向东扩展变形的特征;野外调查和测年结果显示,背斜东段冰川融水形成了大型冰水扇,形成年龄为(87.09±2.31)~(102.4±3.7)ka,进而获得背斜东段自晚更新世以来平均隆升速率的最大值为(2.78±0.28)~(3.28±0.28)mm/a。库木库里盆地整体的活动性很强,在构造上与其北边的柴达木盆地类似,都受控于阿尔金断裂南侧的NNE向的区域挤压作用。     

云南鲁甸M_S6.5地震余震重定位及其发震构造

整合了鲁甸震区周边的云南省地震台网、昭通市地震台网、巧家台阵,以及流动台站2个月的震相观测数据,对鲁甸地震序列进行了重新定位,得到了1 750个地震的震源参数。重定位结果显示,余震有2个优势分布方向,分别为SE向和SW向,具有不对称的共轭分布特征。2个余震条带的展布长度相当,约为16km,夹角约100°。余震分布显示鲁甸地震的发震断层为高倾角的走滑断层。主震位于2个余震条带中间略偏西南的位置,早期余震主要沿NW-SE向垂直于昭通-鲁甸断裂分布,主震西南侧的余震可能为后期触发的。根据余震分布与周边断层的关系、主震震源机制、烈度分布的长轴方位,以及滑坡分布等资料,认为鲁甸地震的发震断层为NW向的包谷垴-小河断裂。包谷垴-小河断裂南北两侧无论是在地震活动、深部速度结构,还是块体运动方向和速率方面都存在显著差异,断裂北侧的高速异常可能是阻止余震向北继续扩展的主要原因。