基于卷积完全匹配层的非规则网格时域有限元探地雷达数值模拟

复频移完全匹配层(Complex Frequency-Shifted PML,CFS-PML)在长时间时域计算中对凋落波、倏失波具有好的吸收效果,并被广泛应用于时域有限差分模拟中.而本文采用卷积方法将CFS-PML应用于时域有限元求解GPR波动方程的数值模拟中.论文以TM波为例,推导了基于CPML(Convolutional PML)边界的时域有限元GPR波动方程求解公式,采用Newmark-β方法对时间导数进行离散,有效改善了时域有限元GPR数值计算程序的稳定性.并以狭长模型为例,开展了CPML边界中关键参数m、R和κ的选取实验,通过对比反射误差大小确定了综合最优参数组合.相同时刻UPML与CPML波场快照、3个检测点的反射误差比较,说明CPML较UPML具有更好的吸收效果.最后,采用非规则四边形网格对1个复杂GPR模型进行剖分,应用加载CPML边界条件的FETD程序对该模型进行了正演,得到了二维剖面法、宽角法正演GPR剖面图,说明非规则四边形对复杂模型的良好适应性,基于CPML边界条件的FETD可有效减少边界反射误差,能实现对任意复杂不规则模型的正演模拟.     

基于PML边界条件的二阶电磁波动方程GPR时域有限元模拟

完全匹配层(PML)作为一种稳定高效的吸收边界条件,广泛应用于基于一阶电磁波动方程的探地雷达(GPR)数值模拟中.为解决基于二阶电磁波动方程的GPR数值模拟的吸收边界问题,本文借鉴二阶弹性波动方程的PML边界条件构建思想,提出了一种适合二阶电磁波动方程GPR时域有限元模拟的PML边界条件.从二阶电磁波动方程出发,基于复拉伸坐标变换,推导了PML算法的频域表达式;通过合理构造辅助微分方程,得到了PML算法的时域表达式,并以变分形式(弱形式)加载到GPR时域有限元方程中,实现了PML边界条件在二阶电磁波动方程GPR时域有限元模拟中的应用.在此基础上,对比了无边界条件、Sarma边界条件和PML边界条件下均匀模型的波场快照、单道波形、时域反射误差和能量衰减曲线,结果表明:PML边界条件的吸收效果要远优于Sarma边界条件,具有近似零反射系数.一个复杂介质模型的正演模拟验证了PML边界条件在非均匀地电结构中电磁波传播模拟的良好吸收效果.     

基于UPML边界条件的交替方向隐式有限差分法GPR全波场数值模拟

交替方向隐式差分(ADI-FDTD)法突破了Courand-Friedrich-Levy(CFL)条件的约束,具有无条件稳定的特点;而单轴各向异性完全匹配层(UPML)边界条件具有宽频带吸收特性,不需要对电场和磁场进行分裂,迭代公式简单,便于编程的特点.综合两者优势,本文提出了基于UPML边界条件的ADI-FDTD探地雷达数值模拟算法,通过对3个二维Maxwell方程进行离散化,推导了GPR波的ADI-FDTD及其UPML边界条件的两个子时间步的迭代差分公式,并分别给出了详细计算步骤.在此基础上,开发了相应的模拟程序,应用该程序对两个GPR模型进行了正演模拟,得到了两个正演模型的wiggle图、扫描图与全波场快照.通过分析这些雷达剖面图与波场快照,可以了解雷达波形在空间中的传播过程及变化规律,有助于雷达资料更可靠、更准确的解释.模拟结果表明,基于UPML边界条件的ADI-FDTD算法可取较大的时间步长,消除了截断边界处的强反射,能对简单与复杂GPR模型进行快速、高效模拟.     

弹性波正演模拟中改进的非分裂式PML实现方法(英文)

在弹性波有限差分正演模拟中,吸收边界条件常用来吸收截断边界处引入的不期望边界反射,其中完全匹配层(PML)吸收边界条件被认为是目前最理想的吸收边界条件。但是PML吸收边界条件的传统实现却存在着很大不足:全局分裂式PML吸收边界条件实现简单但是需要占用太多内存;局部分裂式PML吸收边界条件需要考虑多个边界和角点区域,编程实现非常复杂;非分裂式PML吸收边界条件由于涉及卷积运算,计算量很大。本文基于非分裂式PML吸收边界条件,结合复频移伸展函数,提出了一种新的数值实现方法,其计算方程简单、占用内存小、编程实现容易,是对PML介质理论数值实现的改进和完善。     

基于递归积分的复频移PML弹性波方程交错网格高阶差分法

常规完全匹配吸收边界(PML)对以近掠射角入射到界面上的波以及低频波、损耗波都会产生虚假边界反射.基于递归积分的不分裂复频移PML算法,利用复频移拉伸函数,极大地改善了PML边界条件的性能,我们进一步推导出基于递归积分的不分裂复频移PML弹性波方程交错网格高阶差分法,对长条形介质模型进行数值模拟,与常规PML算法进行比较说明该算法对以掠射角入射到PML界面的波以及PML层内损耗波的吸收效果.     

基于混合边界条件的有限单元法GPR正演模拟

从Maxwell方程组出发,推导了探地雷达(GPR)有限元波动方程.阐述了透射边界条件和Sarma边界条件的原理,推导了这两种边界条件的理论公式;通过在衰减层内加入过渡带优化了Sarma边界条件的加载方法,压制了介质区和衰减层交界面处的人为反射.考虑到透射边界条件与Sarma边界条件不同的理论机制,提出了一种结合透射边界条件和Sarma边界条件的混合边界条件,它利用Sarma边界条件对到达边界区域的GPR波能量衰减功能和透射边界对GPR波能量的透射功能,使GPR波经过Sarma边界条件的衰减吸收后,再通过透射边界条件将剩余能量透射出去,集成了二者的优势.并以二维均匀模型中的中心脉冲激励源方式为例,通过Matlab程序实现,以GPR的全波场快照的直观方式,对比了有、无边界条件及不同边界条件对人工截断边界的处理效果,说明了该混合边界条件对到达截断边界处的GPR波的处理优于单一边界条件.最后,以基于混合边界条件的有限单元法对两个典型的GPR地电模型进行了正演模拟,指导了GPR数据处理与工程实践.     

不分裂卷积完全匹配层与旋转交错网格有限差分在孔隙弹性介质模拟中的应用

完全匹配层吸收边界在地震波模拟中已广泛使用,但常用的场分裂格式完全匹配层吸收边界(SPML)和传统的不分裂完全匹配层吸收边界(NPML)对极低频入射波或大角度入射波的边界吸收效果不好.一种无需分裂和显式卷积计算的完全匹配层吸收边界(CPML)不仅能够解决常规PML吸收边界的不足,而且具有存储量小、计算效率高、易于编程实现的特点.本文将这种完全匹配层(CPML)吸收边界引入到孔隙弹性介质速度-应力格式的旋转交错网格有限差分算法中,对完全匹配层吸收边界参数进行数值分析,得到一组优化的参数.孔隙弹性介质数值模拟结果表明这种不分裂卷积完全匹配层的吸收效果优于常规完全匹配层.     

衰减夹层GPR模拟的时频域全局反射误差

全局反射误差分析是深入研究探地雷达(GPR)吸收边界条件吸收效率的有力工具.基于常规完全匹配层(PML)的标准交错网格有限差分算法必须满足严格的CFL条件限制,即在单位时间步长内,不容许电磁波传播的距离超过单元网格尺寸.为了提高主区域所有网格节点的计算效率,并有效地吸收传播后期出现的低频隐失波,提出基于非分裂递归卷积完全匹配层(UCPML)的旋转交错网格(RSG)GPR正演算法.该算法采用不同的网格交错策略,并在边界条件中引入了吸收低频隐失波的自由可变因子,使得该算法允许选取较大的时间步长,提高了计算效率,并且实现了对低频隐失波的高效吸收.本文首先给出了RSG差分格式,推导了基于UCPML的RSG差分更新方程,探讨了数值色散的稳定性条件,然后以绕射现象严重的衰减夹层数值模拟为例,分别从波场快照、单道波记录、时间域反射误差(TDRE)、频率域反射误差(FDRE)四个方面分析了UCPML与常规PML的全局反射误差,说明了基于UCPML和RSG的GPR正演算法能更有效地吸收低频隐失波.     

Lebedev网格改进差分系数TTI介质正演模拟方法研究

本文采用一种新的交错网格-Lebedev网格(LG)进行TTI介质的正演模拟研究,避免了Virieux标准交错网格(SSG)算法在处理TTI、单斜等各向异性介质时波场插值引入的数值误差,提高了模拟精度.在方法实现过程中,本文针对有限差分正演模拟面临的网格频散与边界反射两个关键性问题分别做了优化,并通过模型试算验证了它们的有效性与可行性:(1)结合最小二乘思想推导出新的频散改进差分系数(DIC),该系数比Taylor系数更能有效地压制粗网格引起的数值频散,可以节约内存,提高计算效率;(2)将分裂的多轴完全匹配层(M-PML)吸收边界条件引入到LG算法中,解决了传统PML边界条件在某些各向异性介质中的不稳定现象并且具有较好的边界吸收效果.     

黏弹TTI介质中旋转交错网格高阶有限差分数值模拟

以Carcione黏弹各向异性理论为基础,给出了适用于黏弹性具有任意倾斜对称轴横向各向同性介质(黏弹TTI介质)的二维三分量一阶速度-应力方程,采用旋转交错网格任意偶数阶精度有限差分格式求解该方程,并推导出了二维黏弹TTI介质完全匹配层(PML)吸收边界条件公式和相应的旋转交错网格任意偶数阶精度有限差分格式,实现了该类介质的地震波场数值模拟.数值模拟结果表明:该方法模拟精度高,边界吸收效果好,可以得到高精度的波场快照和合成记录;并且波场快照和合成记录能较好地反映地下介质的各向异性特征和黏弹性特征.     

探地雷达三维高阶时域有限差分法模拟研究

探地雷达数值模拟中,时域有限差分法在时间和空间上一般采用二阶精度的中心差分近似(FDTD(2,2)),其形式简单,但数值色散误差较大,在复杂模型模拟时不能很好地反映模型的精细变化.高阶时域有限差分法能很好地改善数值色散带来的误差,提高模拟精度.本文基于三维高阶时域有限差分法的基本原理实现了探地雷达正演模拟,采用单轴各向异性完全匹配层(UPML)作为吸收边界条件,可以有效地吸收外向传播的电磁波,在大大地提高计算效率的同时,也能很好地改善边界的吸收效果.分析对比正演模拟结果,通过三维高阶时域有限差分正演能获得目标体准确电磁响应信息,并能很好的提高模拟精度.     

Comparison of perfectly matched layer and multi-transmitting formula artificial boundary condition based on hybrid finite element formulation

The theory of perfectly matched layer (PML) artificial boundary condition (ABC), which is characterized by absorption any wave motions with arbitrary frequency and arbitrarily incident angle, is introduced. The construc- tion process of PML boundary based on elastodynamic partial differential equation (PDE) system is developed. Combining with velocity-stress hybrid finite element formulation, the applicability of PML boundary is investi- gated and the numerical reflection of PML boundary is estimated. The reflectivity of PML and multi-transmitting formula (MTF) boundary is then compared based on body wave and surface wave simulations. The results show that although PML boundary yields some reflection, its absorption performance is superior to MTF boundary in the numerical simulations of near-fault wave propagation, especially in corner and large angle grazing incidence situations. The PML boundary does not arise any unstable phenomenon and the stability of PML boundary is better than MTF boundary in hybrid finite element method. For a specified problem and analysis tolerance, the computa- tional efficiency of PML boundary is only a little lower than MTF boundary.     

基于混合有限元格式的完美匹配层与多次透射公式人工边界比较研究

介绍了完美匹配层(PML)人工边界可以吸收不同频率和任意角度入射波的原理以及PML人工边界的构造方法. 在此基础上，将PML人工边界应用于地震波动数值模拟的速度应力混合有限元格式中，探讨了PML应用的可行性，并通过数值试验研究了PML人工边界的反射率，比较了PML人工边界与多次透射公式(MTF)人工边界应用于体波和面波模拟中数值反射的差异，对两种边界的透射效果进行了分析. 结果表明， 尽管数值离散后PML人工边界不再保持完美匹配特性，但PML人工边界在近场波动数值模拟中可获得比MTF人工边界更为理想的吸收效果，在角点透射、大角度掠射情形下尤为明显；PML人工边界在混合有限元格式的数值算法中，未见失稳等不良反应，比MTF人工边界有更好的稳定性；在合理选择参数的情况下，PML人工边界的运算量可接受.      

Perfectly matched layer-absorbing boundary condition for finite-element time-domain modeling of elastic wave equations

The perfectly matched layer (PML) is a highly efficient absorbing boundary condition used for the numerical modeling of seismic wave equation. The article focuses on the application of this technique to finite-element time-domain numerical modeling of elastic wave equation. However, the finite-element time-domain scheme is based on the second-order wave equation in displacement formulation. Thus, the first-order PML in velocity-stress formulation cannot be directly applied to this scheme. In this article, we derive the finite-element matrix equations of second-order PML in displacement formulation, and accomplish the implementation of PML in finite-element time-domain modeling of elastic wave equation. The PML has an approximate zero reflection coefficients for bulk and surface waves in the finite-element modeling of P-SV and SH wave propagation in the 2D homogeneous elastic media. The numerical experiments using a two-layer model with irregular topography validate the efficiency of PML in the modeling of seismic wave propagation in geological models with complex structures and heterogeneous media.     

Theoretical and quantitative evaluation of hybrid PML-ABCs for seismic wave simulation

A good artificial boundary treatment in a seismic wave grid-based numerical simulation can reduce the size of the computational region and increase the computational efficiency, which is becoming increasingly important for seismic migration and waveform inversion tasks requiring hundreds or thousands of simulations. Two artificial boundary techniques are commonly used: perfectly matched layers (PMLs), which exhibit the excellent absorption performance but impose a greater computational burden by using finite layers to gradually reduce wave amplitudes; and absorbing boundary conditions (ABCs), which have the high computational efficiency but are less effective in absorption because they employ the one-way wave equation at the exterior boundary. Naturally, PMLs have been combined with ABCs to reduce the number of PMLs, thus improving the computational efficiency; many studies have proposed such hybrid PMLs. Depending on the equations from which the ABCs are derived, there are two hybrid PML variants: the PML+unstretched ABC (UABC), in which the ABC is derived from a physical equation; or the PML+stretched ABC (SABC), in which the ABC is derived from the PML equation. Even though all the previous studies concluded that hybrid PMLs can improve the absorption performance, none of them quantified how many PMLs can be removed by combining the PML with the ABC compared with the pure PML. In this paper, we systematically study the absorption performance of the two hybrid PML variants. We develop a method to distinguish the artificial reflections from the PML-interior interface and those caused by the PML exterior boundary to accurately approximate the additional absorption achieved by using the UABC and the SABC. The reflection coefficients based on a theoretical derivation and numerical tests both show that the UABC amplifies most reflections and is not recommended in any situation; conversely, the SABC can always diminish reflections, but the additional absorption achieved by the SABC is relatively poor and cannot effectively reduce the number of PMLs. In contrast, we find that simply increasing the damping parameter improves absorption better than the PML+SABC. Our results show that the improvement in absorption achieved by combining the PML with either the SABC or the UABC is not better than that obtained by simply adjusting the damping profile of the PML; thus, combining the PML with the ABC is not recommended in practice.     

THE APPLICATION OF PML BOUNDARY CODITIONS IN THE SIMULATION OF SEISMIC WAVE BY THE HIGH-ORDER STAGGERED-GRID FINITE DIFFERENCES AT THE TRANSVERSELY ISOTROPIC MEDIA

In the realm of the numerical simulation, finite difference method and finite element method are more intuitive and effective than other simulation methods. In the process of simulating seismic wave propagation, the finite differences method is widely used because of its high computational efficiency and the advantage of the algorithm is more efficient. With the demand of precision, more and more researchers have proposed more effective methods of finite differences, such as the high-order staggered-grid finite differences method, which can restore the actual process of wave propagation on the premise of ensuring accuracy and improving the efficiency of operation. In the past numerical simulation of seismic wave field, different models of isotropic medium are mostly used, but it is difficult to reflect the true layer situation. With the research demand of natural seismology and seismic exploration, the research on anisotropic media is more and more extensive. Transversely isotropic(TI)media can well simulate the seismic wave propagation in the formation medium, such as gas-bearing sandstone, mudstone, shale et al., the character of TI media is reflected by introducing the Thomsen parameters to reflect its weak anisotropy of vertical direction by using Thomson parameter. Therefore, studying the process of seismic wave propagation in TI media can restore the true information of the formation to the greatest extent, and provide a more reliable simulation basis for the numerical simulation of seismic wave propagation. In the geodynamic simulation and the numerical simulation of the seismic wave field, under the limited influence of the calculation area, if no boundary conditions are added, a strong artificial boundary reflection will be generated, which greatly reduces the validity of the simulation. In order to minimize the influence of model boundaries on the reflection of seismic waves, it is often necessary to introduce absorbing boundary conditions. At present, there are three types of absorption boundary conditions: one-way wave absorption boundary, attenuation absorption boundary, and perfectly matched layer(PML)absorption boundary. In terms of numerical simulation of seismic waves, the boundary absorption effect of PML is stronger than the first two, which is currently the most commonly used method, and it also represents the cutting-edge development direction of absorption boundary technology. The perfectly matched layer absorbing boundary is effectively applied to eliminating the reflective waves from model boundaries, but for transversely isotropic medium, the effect of the absorbing is not very well. For this reason, the elastic dynamic wave equations in transversely isotropic media are derived, and we describe a second-order accurate time, tenth-order accurate space, formulation of the Madariaga-Virieux staggered-grid finite difference methods with the perfectly matched layer(PML)are given. In addition, we have established vertical transversely isotropic(VTI)media and arbitrary inclined tilted transversely isotropic(TTI)media models, using a uniform half-space velocity model and a two-layer velocity model, respectively. By combining the actual geoscience background, we set the corresponding parameters and simulation conditions in order to make our model more research-oriented. When setting model parameters, different PML thickness, incident angle, source frequency and velocity layer models were transformed to verify the inhibition of boundary reflection effect by PML absorption boundary layer. The implementations of this simulation show that the formula is correct and for the transversely isotropic(TI)media of any angular symmetry axis, when the thickness of the PML layer reaches a certain value, the seismic wave reflection effect generated by the artificial boundary can be well suppressed, and the absorption effect of PML is not subject to changes in incident angle and wave frequency. Therefore, the results of our study indicate that our research method can be used to simulate the propagation process of seismic waves in the transversely isotropic(TI)media without being affected by the reflected waves at the model boundary to restore the actual formation information and more valuable geological research.     

考虑卷积完全匹配层数值色散的井间电磁三维正演

本文将以卷积完全匹配层为吸收边界条件的时域有限差分法应用到井间电磁的三维正演模拟中.证明了卷积完全匹配层中的数值色散会因为有效延伸因子而产生,列举常规有效延伸因子和网格间距对电磁波相速度各向异性的影响,并通过波场快照验证卷积完全匹配层中数值色散的存在;进而推广三维卷积完全匹配层中最大有效延伸因子、最大网格间距与激励源主频之间的约束,以此完善卷积完全匹配层的最优参数设置方案.在此基础上,通过正演结果中二次场的垂直磁场分量和走时来展示灵敏度分布,以此划定井间电磁勘探的优势区域,提出井间电磁正反演所需要的约束测井系列和最佳取井方案.