地震绕射波弱信号U-net网络提取方法

绕射波携带大量小尺度非均匀地质体信息,对于提高地震勘探分辨率具有重要意义.绕射波能量远小于反射波,在地震记录中常被强反射波掩盖,因此分离并单独成像绕射波,为探测小尺度非均匀地质体的关键问题.传统绕射波分离方法受限于理论模型假设,对陡倾角反射波去除效果不佳,且易对绕射波造成损伤.基于经典编码-解码框架下的U-net网络和注意力机制,本文提出了一种绕射波智能分离方法,通过编码器自动提取地震数据中的绕射波特征,再由解码器恢复绕射波,从而隐性去除反射波.该方法作为端到端的机器学习,训练后的U-net网络可自适应地分离绕射波.本文通过数值模拟数据与实际数据构建训练数据集,利用训练后的U-net网络分离绕射波,并将结果偏移成像.数值模型测试和实际资料应用表明,融合了注意力机制的U-net网络能够有效压制反射波能量,保留绕射波动力学特征,克服了传统绕射波分离方法难以去除陡倾角反射的局限性,其提取的绕射波弱信号特征较为完整,能够进一步提高地震成像分辨率,在小尺度断裂刻画上具有优势.

     

逐次线性化衰减层析成像方法研究

在时间域衰减成像的基础上,建立了一种与当前走时层析成像类似的衰减成像方法——逐次线性化衰减层析成像方法,这是一种基于射线理论的方法.文中推导了逐次线性化过程的一系列公式,给出了相应的迭代算法及实现技术,并通过数值实验验证了方法的可行性.作为走时层析成像的补充方法,实现多参数成像.工程勘察实践表明,走时与衰减联合成像方法的应用效果良好.     

基于模拟退火法的青藏高原东缘噪声层析成像研究

青藏高原是全球造山带研究的热点地区,此前在青藏高原开展的三维层析成像研究大多基于线性反演方法.本文利用青藏高原东缘及邻区布设的127个宽频带固定地震台站记录的连续波形资料,首先通过噪声互相关提取了3~50 s Rayleigh波群速度频散曲线并反演得到群速度分布,再进一步采用模拟退火法反演了研究区的三维S波速度及泊松比结构.结果显示：（1）松潘—甘孜地块的中下地壳低速异常主要分布在龙日坝断裂带、鲜水河断裂带、龙门山断裂带和岷山隆起所围限的区域,而该区域的中下地壳仅具有中等泊松比值,推测松潘—甘孜地块中下地壳的低速物质可能是青藏高原与扬子块体长期相互作用产生的塑性低速滑脱层;上地壳脆性物质在板块作用下沿中地壳低速滑脱层顶界面发生逆冲增厚,造成龙门山的持续抬升和地形起伏,并在构造边界带形成了应变积累和应力集中;而龙门山断裂带的上地壳低速软弱物质为地壳发生破裂提供了有利条件,从而在某种程度上促进了汶川地震和芦山地震的发生.（2）岷山隆起一带中下地壳的高泊松比异常呈"凸起"形态,结合前人研究发现的较高热流和岩石快速抬升现象,推测岷山隆起一带可能存在岩石圈的拆沉,导致地幔热物质上涌而形成下地壳高泊松比物质.（3）川滇地块的北部和南部具有不同的S波速度和泊松比分布特征.30 km深度下川滇地块北部具有明显的低速异常,而该深度下并不具有明显的高泊松比值特征;此外剖面成像结果也显示川滇地块内的低速异常与高泊松比的分布不一致,因此川滇地块的研究结果不支持下地壳流模型.综合其他地震学证据,本文认为川滇地块的变形模式为上地壳纯剪切增厚,块体变形主要受块体内部的走滑断裂及活动边界断裂控制.     

基于HAFMM的无射线追踪跨孔雷达走时层析成像

本文使用最小二乘线性迭代反演方法对跨孔雷达直达波初至时数据进行反演,每次迭代过程中,用有限差分法求解走时程函方程,并用高精度快速推进方法(HAFMM)进行波前扩展,通过追踪波前避免了进行射线追踪.为了验证该方案,我们对三组合成数据进行了测试,分析了单位矩阵算子、一阶差分算子和拉普拉斯算子等三种不同模型参数加权算子对模型的约束和平滑效果;讨论了FMM和HAFMM对反演精度的影响;测试了LSQR,GMRES和BICGSTAB等三种矩阵反演算法的反演效果.此外,我们还对一组野外实测数据进行了反演,对比了基于本方案以及基于平直射线追踪和弯曲射线追踪的走时层析成像反演效果.对比分析结果表明,使用拉普拉斯算子和HAFMM进行反演能较好地进行目标体重建,而三种矩阵反演方法对反演效果的影响差别不大;并且通过对波前等时线图的分析可以定性地判断异常体的性质和位置;而在对实测数据目标体的重建上,本方案能达到甚至优于弯曲射线算法的重建效果.     

Temporal sampling and role of flux measurements for subsurface heterogeneous characterization in groundwater basins using hydraulic tomography

Accurate characterization of heterogeneity in groundwater basins is crucial to the sustainable management of groundwater resources. This study explores the temporal sampling issues and the role of flux measurements in the characterization of heterogeneity in groundwater basins using numerical experiments. The experiments involve a digital basin imitating the groundwater basin of the North China Plain (NCP), where the groundwater exploitation reduction program is ongoing. Using the experiments, we champion that the reduction program could collect groundwater level information induced by operational variations of existing pumping wells at different locations in the basin. Such a dataset could serve as a basin-scale hydraulic tomography (HT) to characterize the basin-scale heterogeneity cost-effectively. Both steady-state and transient-state inversion experiments demonstrate the advantage of HT surveys in characterizing basin-scale heterogeneity over conventional pumping tests at fixed well locations. Additionally, head data at the early, intermediate, and late time from well hydrographs should be selected for the HT analysis to maximize HT's power and save computational costs. When accurate geological zones are incorporated in prior information, flux measurements significantly improve parameter estimates based on conventional pumping tests. However, their effects are less noticeable for long-term HT surveys in such basin-scale aquifers without fissures or fractures. This basin-scale tomographic survey example serves a guide for field data collection and optimization of the analysis of future basin-scale HT.     

A hybrid inversion method of damped least squares with simulated annealing used for Rayleigh wave dispersion curve inversion

Surface wave methods are becoming increasingly popular in many geotechnical applications and in earthquake seismology due to their noninvasive characteristics.Inverse surface wave dispersion curves are a crucial step in most surface wave methods.Many inversion methods have been applied to surface wave dispersion curve inversion,including linearized inversion and nonlinearized inversion methods.In this study,a hybrid inversion method of Damped Least Squares(DLS) with Very Fast Simulated Annealing(VFSA) is developed for multi-mode Rayleigh wave dispersion curve inversion.Both synthetic and in situ fi eld data were used to verify the validity of the proposed method.The results show that the proposed method is superior to the conventional VFSA method in aiming at global minimum,especially when parameter searching space is adjacent to real values of the parameters.The advantage of the new method is that it retains both the merits of VFSA for global search and DLS for local search.At high temperatures,the global search dominates the runs,while at a low temperatures,the local search dominates the runs.Thus,at low temperatures,the proposed method can almost directly approach the actual model.     

An iterative method for finding small magnetic objects in the subsurface by linear and non-linear inversion1

A method to determine the position and magnetization vector of buried objects producing a magnetic anomaly is described. The data used were collected in boreholes. Since the anomaly is due to a number of objects, a ‘stripping’ procedure is employed for finding them, and therefore the process of inversion for finding all objects causing the anomaly consists of a few inversion steps. In each inversion step, two dipoles are considered as a model which approximates an object. The position and magnetic moments of the dipoles are the unknown parameters. The initial parameters are optimized by minimization of an objective function. The optimization procedure consists of a combination of linear and non-linear inversion. The solution of the linear inversion is obtained by singular value decomposition and that of the non-linear inversion by a six-dimensional simplex method (polytope algorithm). After finding one object, its effect is subtracted (‘stripped’) from the data and a new inversion step is started with new initial models and with a reduced data set. The inversion steps for finding different objects are continued until the absolute norm of the data becomes less than some adjustable value. The data will also be inverted assuming a three-dipole model in order to find the effect of using a more complex model in the inversion. The efficiency of the method is demonstrated using synthetic and real borehole data.     

Optimal selection of number and location of rainfall gauges for areal rainfall estimation using geostatistics and simulated annealing

This paper presents a method for establishing an optimal network design for the estimation of areal averages of rainfall events. The problem consists of minimising an objective function which includes both the accuracy of the areal mean estimation (as expressed by the kriging variance of estimation) and the economic cost of the data collection. The well known geostatistical variance-reduction method is used in combination with simulated annealing as an algorithm of minimisation. This methodology has several advantages which will be demonstrated in this paper. Several synthetic examples are shown in order to illustrate the performance of the methodology in two different optimisation problems: the optimal selection of a subset from a set of stations that already exist and the optimal augmentation of a previously existing network.     

基于地震成像数据稀疏反演的不连续及非均质地质体检测方法

本文针对油气藏储层预测中的不连续及非均质地质信息识别问题,研究基于地震成像数据的稀疏反演方法.由于该类地质体的地震响应特征为弱信号,因此利用平面波破坏滤波器由地震成像数据中去除强反射同相轴.在此基础上,对剩余的地震数据进行非线性加强滤波,并构建L1稀疏反演模型.为有效求解L1模型,采用非光滑泛函L1范数逼近和拟牛顿求解算法.该方法考虑稀疏先验信息,能够提高反演结果信噪比.缝洞模型测试验证该方法在检测断点、微断裂、散射点等小尺度地质体上的有效性,塔北缝洞型碳酸盐岩储层预测的应用效果进一步证实该方法的实用性.     

Boundary integral equation method for the diffraction of elastic waves using simplified green's functions

An alternate formulation of the ‘substructure deletion method’ suggested by Dasgupta in 1979¹ has been successfully implemented. The idea is to utilize simple Green's functions developed for a surface problem to replace the more complicated Green's functions required for embedded problems while still being able to generate an accurate solution. Since the exterior medium is usually represented by a continuum model, the interior medium in the present approach will also be represented by a continuum model rather than a finite element model as suggested originally, thereby eliminating the incompatibility between the solutions of the interior and exterior media. Detailed studies of the method's accuracy and limitations were performed using two-dimensional examples in wave scattering of canyons and alluvial valleys, problems which are more suitable for this method than the embedded foundation problem. The results obtained indicate that the alternate formulation gives accurate results only when the vertical dimension of the scattering object is not too large; if the aspect ratio (vertical over lateral) exceeds a certain limit, the results will not approach the known results given by boundary integral equation solutions or indirect boundary integral equations no matter what the refinement of the model may be. The greatest advantage of the present method is that the task of calculating Green's functions is reduced significantly; computational time using this new formulation is approximately five times less than for conventional boundary integral equation methods.     

用地震波波形拟合方法研究中小地震的震源机制

采用网格搜索方法,利用区域地震三分量全波形资料,对中小地震震源机制的反演方法进行了研究。反演采用初动约束、分段求相关系数,以及幅值加权的混合方法计算误差函数,通过格点搜索法确定震源深度。理论模型试验表明,反演结果具有较好的稳定性和可靠性。云南地区区域数字地震台网的观测资料反演表明,该方法可以合理地拟合观测波形资料。将2001年7月9日(24.93°N,101.47°E)发生的MS 5.3地震的震源机制解与哈佛大学CMT进行了对比,结果表明,Z分量波形拟合略比CMT拟合的好,R分量和T分量的波形拟合明显好于CMT的对应分量。 我们对2001年4月5日在(25.77°N,102.30°E)发生的mb 4.8地震进行了震源机制反演。结果表明,该地震主要表现为走滑断层,震源机制解为(180°,80°,11°)。根据这两个地震震源机制推断的主应力轴,与该地区现代构造应力场特征基本一致。     

基于信号整体与局部特征的地震数据自动处理方法研究

提出一种基于信号整体与局部特征的地震数据自动处理新方法, 该方法不同于以往仅采用包络线互相关来直接检测事件. 新方法依然按照检测、 识别、 关联和定位等4个步骤进行处理, 但在进行单个震相信号检测的同时, 也检测信号波列并利用其包络线特征来识别和关联震相. 文中详细阐述了数据处理过程中如何定义一个波列及抽取和应用其特征. 相关的数据处理技术目前已成功应用于区域台网的日常数据处理分析中. 作为例子, 给出了对新疆区域台网连续16天数据进行测试处理的结果. 实际应用结果表明, 这种新方法可以大幅度降低自动处理结果的误检、 漏检率, 在实际应用中具有很好的前景.     

Seismic hazard and risk scenarios for Barcelona,Spain, using the Risk-UE vulnerability index method

The vulnerability index method, in its version developed in the framework of the European project Risk-UE, has been adapted and applied in this article, to evaluate the seismic risk for the city of Barcelona (Spain) through a GIS based tool. According to this method, which defines five damage states, the action is expressed in terms of the macroseismic intensity and the seismic quality of the buildings by means of a vulnerability index. The probabilities of damage states are obtained considering a binomial or beta-equivalent probability distribution. The most relevant seismic risk evaluation results obtained, for current buildings and monuments of Barcelona, are given in the article as scenarios of expected losses.     

New supervised learning classifiers for structural damage diagnosis using time series features from a new feature extraction technique

The motivation for this article is to propose new damage classifiers based on a supervised learning problem for locating and quantifying damage. A new feature extraction approach using time series analysis is introduced to extract damagesensitive features from auto-regressive models. This approach sets out to improve current feature extraction techniques in the context of time series modeling. The coefficients and residuals of the AR model obtained from the proposed approach are selected as the ...     

求解二次方程的二维VTI介质qSV波和qSH波走时快速扫描算法

地震波走时计算在观测系统设计、偏移成像、速度模型走时反演和地震定位等方面起到重要作用.各向异性广泛存在于地球介质中,影响地震波传播的振幅和走时,忽略各向异性的影响将对成像、反演以及地震定位等造成一定的误差.因此对于高分辨率成像和反演,走时计算中考虑各向异性十分重要.快速扫描法不需要存储和追踪波前面信息,在各向异性初至波走时计算方面应用广泛.传统的方法通过将慢度四次方程转换为走时四次方程并结合快速扫描法求解走时.该方法没有对程函方程做近似,适用于强各向异性介质,但存在计算效率低的问题.对于求解qSV波走时,本文发展了一种在局部解中将慢度四次方程简化为二次方程解析地快速求解走时的方法,极大地提高了计算效率.对于qSH波,慢度方程是二次的,可以直接解析求解.最后,本文用各向异性均匀模型和BP复杂模型进行测试,计算结果表明走时计算准确,验证了该方法的有效性.

     

Seismic response analysis using characteristic ground motion records for risk‐based decision‐making (3R method)

The concept of intensity‐based assessment for risk‐based decision‐making is introduced. It is realized by means of the so‐called 3R method (response analysis, record selection and risk‐based decision‐making), which can be used to check the adequacy of design of a new building or of the strengthening of an existing building by performing conventional pushover analysis and dynamic analysis for only a few ground motions, which are termed characteristic ground motions. Because the objective of the method is not a precise assessment of the seismic risk, a simple decision model for risk acceptability can be introduced. The engineer can decide that the reliability of a no‐collapse requirement is sufficient when collapse is observed in the case of less than half of, for example, seven characteristic ground motions. From the theoretical point of view, it is shown that the accuracy of the method is acceptable if the non‐linear response history analyses are performed at a low percentile of limit‐state intensity, which is also proven by means of several examples of multi‐storey reinforced concrete frame buildings. The 3R method represents a compromise between the exclusive use of either pushover analysis or dynamic analysis and can be easily introduced into building codes provided that its applicability is further investigated (e.g. asymmetric structures and other performance objectives) and that the procedure for the selection of characteristic ground motions is automated and readily available to engineers (www.smartengineering.si).     

Seismic velocity structure around the Hyuganada region, Southwest Japan, derived from seismic tomography using land and OBS data and its implications for interplate coupling and vertical crustal uplift

The Hyuganada region, a forearc region of Southwest Japan, is characterized by several interesting geological and geophysical features, i.e., significant aseismic crustal uplift of 120 m during the past 120 thousand years at the Miyazaki Plain, negative free-air gravity anomalies with the maximum magnitude of −130 mgal, and relatively less cohesive interplate coupling compared with that for off the Shikoku and Kii Peninsula. In order to examine the causes of these observations, we determined a detailed three-dimensional seismic velocity structure based on the seismic data observed by ocean bottom seismometers (OBS) and land stations. P- and S-wave tomographic velocity structures clearly indicate the subducting slab and also the zones of high Poisson's ratio at 25–35 km depth along the coastline of the northeastern part of the Hyuganada. The region with high Poisson's ratio may correspond to the serpentinized mantle wedge as suggested for other mantle wedges, and appears to be coincident with the zone for observed aseismic slips such as the slow-slip and after-slip events. Also, the detection may be related to a relatively weak interplate coupling in the Hyuganada region. The tomographic structures also indicate low velocity zones with a horizontal scale comparable to the Kyushu-Palau Ridge in and around the subducting slab. If we assume that the low velocity zones correspond to the subducted Kyushu-Palau Ridge, then the predicted gravity anomaly due to the density contrast between the low velocity zones and the surrounding region can explain about 60% of the gravity anomaly in the Hyuganada region. The buoyancy is probably an important factor for the crustal uplift observed in the Miyazaki Plain, the steep bending of the subducting slab and the normal fault-type earthquakes around the Hyuganada region.