Efficient finite-difference modelling of seismic waves using locally adjustable time steps

Conventional finite-difference modelling algorithms for seismic forward modelling are based on a time-stepping scheme with a constant (global) time step. Large contrasts in the velocity model or in the spatial sampling rate cause oversampling in time for some regions of the model. The use of locally adjustable time steps can save large amounts of computation time for certain modelling configurations. The computation of spatial derivatives across the transition zone between regions of the model with different temporal sampling requires the definition of the wavefield at corresponding time levels on both sides of the transition zone. This condition can be obtained by extrapolation in time, which is inaccurate, or by multiple time integration in the transition zone. The error in the latter solution is of the same order as the conventional time-stepping scheme because both methods are based on the same iteration formula. The technique of multiple time integration simply requires the use of different sizes of time step. It is applicable only for certain factors of variation of the time step.     

基于深度学习UNet++网络的初至波拾取方法

初至波拾取是地震资料处理中一项基础而重要的工作.为解决我国西部沙漠、黄土塬、戈壁等地区地震资料信噪比低,致使初至波拾取准确率不高的难题.本文创新提出一种基于图像分割技术——UNet++神经网络应用于初至波智能拾取.输入原始地震数据及少量初至时间的标签数据进行监督学习,并建立UNet++模型,应用西部某工区地震数据测试,实验证明,UNet++模型性能稳定,炸药震源初至波拾取准确率达到98％,可控震源初至波拾取准确率达到98％.此外,本方法与商业软件、U-net网络的初至拾取对比表明,UNet++优势明显,具有准确率高,抗噪能力强,性能稳定、高效等特点.     

A real time processing system of seismic waves using personal computers-Function and characteristics

Ａｒｅａｌｔｉｍｅｐｒｏｃｅｓｉｎｇｓｙｓｔｅｍｏｆｓｅｉｓｍｉｃｗａｖｅｓｕｓｉｎｇｐｅｒｓｏｎａｌｃｏｍｐｕｔｅｒｓ—ＦｕｎｃｔｉｏｎａｎｄｃｈａｒａｃｔｅｒｉｓｔｉｃｓＪＵＮＦＡＮ１）（范军）ＴＩＡＮＣＨＡＮＧＣＨＥＮ１）（陈天长）ＷＥＩ?..     

对人造地震动反应谱求解及拟合的几个相关问题探讨

为更高效合理地生成与设计反应谱相符的人造地震动信号,本文基于自主开发的地震动信号处理软件EQSignal对人造地震动生成过程中涉及的几个问题进行了探讨,并通过对单自由度简谐共振体系的响应分析,提出了具体的解决建议:① 反应谱求解时应该对高频段和低频段分别采用频域传递函数法和逐步积分法求解;② 反应谱周期控制点的分布模式也应分段处理,短周期段宜采用对数平均分布,长周期段宜采用线性平均分布,反应谱总周期控制点不宜少于120个;③ 人造地震动反应谱与设计反应谱拟合的过程中,使用频域方法与时域方法相结合可兼顾效率与收敛性。      

一种地震P波和S波初至时间自动拾取的新方法

地震P波、S波初至时间的拾取是地震波分析的一项基础性工作.本文提出了一种新的地震波初至时间自动拾取的方法:首先,把地震波的三分量时程曲线变换为一组空间向的能量变化率时程曲线;然后对能量变化率时程曲线进行STA/LTA(Short Time Average/Long Time Average,短时间的均值/长时间的均值)处理,拾取地震P波和S波的大致初至时间;最后提出采用一种二次方自回归模型对初至附近的能量变化率曲线进行二次方自回归处理,精确拾取出P波和S波的初至时间.本文采用了10组芦山地震的记录数据和150组汶川地震的记录数据对此方法的可靠性进行了检验.以人工拾取结果为参考,此方法具有很高的准确率和稳定性,同时,相比于常用的STA/LTA方法和AIC(Akaike Information Criterion,Akaike信息准则)方法,此方法在计算时间效率方面稍微逊色,但是对S波初至时间的拾取精度和可靠性更高.此方法丰富了地震P波、S波初至时间的自动拾取方法.     

高铁地震数据干涉成像技术初探

高铁运行会引起铁轨的震动,从而产生地震波向地下介质中传播,通过研究该地震波可对高铁沿线的地质情况进行持续监测.与常规地震勘探中的震源相比,高铁地震中的震源较为复杂,为移动震源,而地震干涉技术可以通过地震记录间的相互干涉,消除震源的影响,因此可利用地震干涉技术对高铁地震信号进行处理并成像.本文通过分析研究,总结出地震干涉方法在处理高铁地震数据时的关键技术问题：不同于常规地震干涉中先干涉后叠加的干涉成像方式,高铁地震移动源的特点使得干涉顺序变为先叠加后干涉,由此带入了大量震源串扰噪声;初步提出两种解决高铁地震干涉成像的思路：通过对高铁地震信号的处理,使高铁变相"提速"或"降速",给出了"提速"或"降速"后各自的成像思路,并给出了数据处理的技术设想.     

红山地震台震相到时准确测定

利用工具软件对选取的红山地震台波形数据进行分析处理,从波形包含的频率成分特征上可以区分地脉动、干扰、地震;利用Fourier变换、滤波、复时谱分析、高斯小波分析、莫莱小波分析、DB小波分析等手段,综合分析处理可准确标注地震震相.     

高铁地震信号直达波波形反演建模

高铁地震信号由高速行驶的高铁列车与铁轨挤压形变后产生,具有频带宽、低频强、重复性好等优点.如何利用高铁地震信号进行地下介质重构,尚处于前期探索阶段.全波形反演建模技术是目前地球物理领域建模精度较高的方法之一,但其需要低频数据才能保证结果的稳定收敛,因此利用全波形反演对高铁地震信号进行近地表建模及属性变化监测和灾害预警具有独特优势.本文通过对高铁地震信号的分析,研究了高铁地震的理论震源信号和基于桥墩的离散震源信号,并通过模型测试实现了利用高铁地震信号的全波形反演建模及属性变化监测.

     

Modal control of seismic structures using augmented state matrix

In conventional methods of modal control, the number of controllable structural modes is usually restrained by the number of sensors that feedback the structural signals. In this paper a modal control scheme where the feedback gain is formulated in an augmented state space is proposed. The advantage of the proposed method is that it increases the number of the controllable modes without adding extra sensors. The method is verified experimentally by an earthquake simulation test with a full‐scale building model. The proposed modal control was also compared with the conventional ones in the test. For the building model tested, the performance of the proposed control with only one feedback signal can be as efficient as that of modal control with full state feedback. Copyright © 2001 John Wiley & Sons, Ltd.     

Unsupervised separation of seismic waves using the watershed algorithm on time-scale images

This paper illustrates the use of image processing techniques for separating seismic waves. Because of the non‐stationarity of seismic signals, the continuous wavelet transform is more suitable than the conventional Fourier transforms for the representation, and thus the analysis, of seismic processes. It provides a 2D representation, called a scalogram, of a 1D signal where the seismic events are well localized and isolated. Supervised methods based on this time‐scale representation have already been used to separate seismic events, but they require strong interactions with the geophysicist. This paper focuses on the use of the watershed algorithm to segment time‐scale representations of seismic signals, which leads to an automatic estimation of the wavelet representation of each wave separately. The computation of the inverse wavelet transform then leads to the reconstruction of the different waves. This segmentation, tracked over the different traces of the seismic profile, enables an accurate separation of the different wavefields. This method has been successfully validated on several real data sets.     

模拟地震波传播的优化质量矩阵Legendre谱元法

波动方程的数值求解是地震波正反演的重要环节,而数值算法的计算精度直接关系到地震波的模拟结果和成像质量.当前,谱元法由于同时具备有限元法的网格灵活性与谱方法的高精度性已被成功应用于不同尺度模型中的地震波模拟.然而,常见的Legendre谱元法在求解地震波运动方程时采用Gauss-Lobatto-Legendre（GLL）数值积分计算质量矩阵所包含的积分项,由于GLL数值求积无法对积分项精确估计,从而造成谱元法精度损失.针对谱元法精度上的不足,本文提出一种优化算法用于提升其精度.首先构造关于GLL数值求积积分权与质量矩阵对角线元素精确值的最小二乘目标函数,然后利用共轭梯度法求解目标函数得到优化权系数,该权系数能减小质量矩阵的离散误差最终提高谱元法的计算精度.通过数值频散分析、数值算例证实了本文给出的优化算法用于提升谱元法数值模拟精度的可行性和有效性.

     

Source spectral phases of surface waves recovered from observed spectral amplitudes by using Hilbert transforms

Spectral phases of surface waves radiated at the seismic source are recovered by means of the Hilbert transform of spectral amplitudes. Through numerical simulations as well as the study of an earthquake with well known focal mechanism, it is shown that the recovered phases satisfy a criterion of coherency, so that they can be used in an inversion process for the retrieval of the seismic moment tensor.     

用初至P震相联合深度震相sPL测定海南中小地震深度

两种测定中小地震震源深度的方法:一是基于走时测定震源深度,二是基于波形测定震源深度。以海南地区4次中小地震为例,采用初至P震相联合深度震相sPL测定震源深度的方法,测定其结果为11—12 km,结果一致性较好,表明两种方法适合用于测定中小地震的震源深度。      

Quantitative detection of fluid distribution using time-lapse seismic

Although previous seismic monitoring studies have revealed several relationships between seismic responses and changes in reservoir rock properties, the quantitative evaluation of time‐lapse seismic data remains a challenge. In most cases of time‐lapse seismic analysis, fluid and/or pressure changes are detected qualitatively by changes in amplitude strength, traveltime and/or Poisson's ratio. We present the steps for time‐lapse seismic analysis, considering the pressure effect and the saturation scale of fluids. We then demonstrate a deterministic workflow for computing the fluid saturation in a reservoir in order to evaluate time‐lapse seismic data. In this approach, we derive the physical properties of the water‐saturated sandstone reservoir, based on the following inputs: V_P, V_S, ρ and the shale volume from seismic analysis, the average properties of sand grains, and formation‐water properties. Next, by comparing the in‐situ fluid‐saturated properties with the 100% formation‐water‐saturated reservoir properties, we determine the bulk modulus and density of the in‐situ fluid. Solving three simultaneous equations (relating the saturations of water, oil and gas in terms of the bulk modulus, density and the total saturation), we compute the saturation of each fluid. We use a real time‐lapse seismic data set from an oilfield in the North Sea for a case study.     

基于支持向量机的地震体波震相自动识别及到时自动拾取

面对海量地震资料,自动准确地拾取震相并确定其到时的需求非常迫切.基于支持向量机技术,本文提出了使用两个分类器SSD和SPS自动识别地震体波震相并自动拾取其到时的方法.相比于传统的自动拾取方法,本文方法能够更准确地识别震相并区分P波和S波.进一步地,我们提出了利用台阵资料辅助识别震相的方案,有效地提高了地震震相拾取的准确率.

     

On the possibility of using the tidal modulation of seismic waves for forecasting earthquakes

The earthquake forecasting technique based on the effect of the modulation of high frequency seismic noise (HFSN) by the Earth’s tides is discussed and the experience of its practical application in Kamchatka is described based on the extensive measurement data for 1996–2013. The empirical development of the lower magnitude threshold on the epicentral distance is refined. The reliability and validity of the precursor are assessed. The efficiency of the precursor is estimated in two ways. It is shown that the results of applying the prognostic procedure statistically significantly differ from a random guess. The presented estimates are based on the catalog containing 68 earthquakes of 1996–2013 in the predicted magnitude interval M ≥ 5.0 at the epicentral distances that are linked to the magnitude by functional dependence.     

初至波菲涅尔体地震层析成像

根据地震波传播的有限频理论,对于某个特定震相的观测信息,不仅射线路径上的点对该信息具有影响,射线领域上的其他点对接收信息也具有影响,这种影响可以用核函数来表达.本文基于波动方程的Born近似与Rytov近似,给出了非均匀介质情况下初至地震波振幅与走时菲涅尔体层析成像单频、带限层析核函数的计算方法.通过对均匀介质情况下初至地震波菲涅尔体层析成像核函数解析表达式的理论模型实验与分析,给出了不同维度振幅、走时单频菲涅尔体的空间分布范围,进而给出了带限菲涅尔体边界的确定方法.将本文的走时菲涅尔体层析成像理论应用于表层速度结构反演中,理论模型试验与实际资料处理结果表明,初至波菲涅尔体地震层析成像方法比传统的初至波射线层析成像理论具有更高的反演精度.     

多点输入下大跨结构地震反应的频域精细传递矩阵法

地震输入问题一直是工程结构抗震研究所关注的焦点.在对大跨度结构进行抗震设计时,需要考虑多点地震输入的影响已成为国内外学术界和工程界的共识.本文根据傅氏变换,运用精细传递矩阵法推导出了大跨结构在多点地震输入下的频域精细传递矩阵,并在频域内对大跨度结构进行了动力分析.这种方法公式简单,能够快速、高精度地进行结构的地震反应分析.算例显示了本文方法的有效性.     

Amplitudes of explosion-generated seismic waves

Summary Records of underwater explosions in Iceland in 1959 and 1960, mainly performed for crustal studies, have been investigated with regard to amplitudes. The amplitudes ofP2 in the 1959 explosions and of the first arrivingP waves in the 1960 explosions were found to be proportional to the first power of the charge weight, whereas amplitudes ofP1 andS1 in the 1959 explosions were proportional to the 3/4-power of the charge up to 200 kg, but increased much slower for larger charges. The influence of water depth of shot point on the charge weights, required to obtain a certain amplitude, has been determined and it was found that the logarithm of the charge weight has a linear relation to water depth down to about 8 meters. The amplitudes of first arrivingP waves decrease as the inverse 2.2-power of the distance up to about 30 km. For greater distances an exponential decrease of the form (const./) exp (–) is valid. ForP2 waves with a frequency of 10 cps we found =0.027±0.003 km^–1 for a profile across central Iceland and =0.009±0.004 km^–1 for a profile in the western part of Iceland.

---

Zusammenfassung In den Jahren 1959 und 1960 wurden auf Island Unterwasser-Explosionen ausgelöst, deren Hauptziel Untersuchung der Erdkruste waren. Die Aufzeichnungen der Explosionen werden hier in Hinsicht auf die Amplituden der Wellen untersucht. DieP2-Amplituden in den Explosionen von 1959 und die Amplituden der zuerst ankommendenP-Wellen in den Explosionen von 1960 waren proportional der ersten Potenz des Ladungsgewichtes, während diePl-undSl-Amplituden in den Explosionen von 1959 proportional der 3/4-Potenz des Ladungsgewichtes bis 200 kg waren. Für höhere Ladungsgewichte wuchsen die Amplituden langsamer an. Es wurde der Einfluss der Tiefe des Schusspunktes unterhalb des Wasserspiegels auf das für eine gewisse Amplitude benötigte Ladungsgewicht bestimmt. Es ergab sich dabei, dass der Logarithmus des Ladungsgewichtes in einer linearen Beziehung zur Tiefe, bis etwa 8 m, steht. Die Amplituden der zuerst ankommendenP-Wellen sind umgekehrt proportional der 2.2-Potenz der Entfernung, bis etwa 30 km. Für grössere Entfernungen ist die Abnahme der Amplituden exponential und kann durch (Konst./) exp (–) dargestellt werden. FürP2 mit einer Frequenz von 10 cps haben wir =0.027±0.003 km^–1 für ein Profil in Zentral-Island und =0.009 ± 0.004 km^–1 für ein Profil in West-Island gefunden.