基于DIRECT算法的微震震源快速网格搜索定位方法研究

微震监测定位技术的精度及效率是微震监测中的核心技术问题,震源扫描算法作为一种稳定、可靠的方法被广泛用于微震监测定位中,但是在多事件、多参数反演中,其反演效率严重制约其应用.本文将震源扫描算法与DIRECT算法思想结合,提出基于DIRECT算法的微震快速网格震源搜索定位方法,即在求解震源扫描算法目标函数最优解时,采用DIRECT搜索策略,无需划分网格大小,直接搜索目标函数最优解.通过地面及井中微震监测模型测试表明,基于DIRECT算法的微震快速网格震源搜索定位方法比传统的震源扫描算法搜索次数少、计算耗时短、计算精度高,尤其适合大空间内多事件、多参数的震源反演,为水力压裂微震监测过程中大量数据的实时、快速、准确定位提供了可能.     

煤矿井下微震震源高精度定位研究

煤矿井下微震震源准确定位,对于动力灾害监测预警具有重要意义.由于微震震源需要通过井下传感器接收信息反演确定,传感器的安装位置限制于煤矿井下巷道周围,传感器沿巷道近平面的不合理布置将大大降低震源定位精度.针对由传感器信息反演震源位置引起的病态问题,本文提出了基于微震监测测点优化布置的震源高精度定位算法.首先通过计算系数矩阵条件数,判定病态问题;然后利用中心化法和行平衡法联合进行病态矩阵预处理.对预处理后的矩阵A、b利用L曲线法计算正则参数,结合Tikhonov正则化算法计算得到震源坐标正则解.研究结果表明,中心化法有效降低了矩阵数量级,行平衡预处理降低了病态条件数,预处理后Tikhonov正则解的震源坐标误差最小可以达到3.09m,与预处理前的高斯消去解相比误差大大降低.通过上述优化处理,实现了井下受限空间微震监测震源高精度定位.     

利用模拟退火法反演大陆地震位置

为了精确得到大陆地震震源位置,利用模拟退火法对大陆地震参数进行反演,并对改变算法中相关参数后的反演结果进行分析,表明反演精度受降温速度和收敛阈值的影响较大,初始温度只影响迭代次数,模拟退火法在对模型参数进行全局搜索情况下能获得质量较好的最优解。     

利用模拟退火法反演大陆地震位置

为了精确得到大陆地震震源位置,利用模拟退火法对大陆地震参数进行反演,并对改变算法中相关参数后的反演结果进行分析,表明反演精度受降温速度和收敛阈值的影响较大,初始温度只影响迭代次数,模拟退火法在对模型参数进行全局搜索情况下能获得质量较好的最优解。     

微震震源定位研究现状及展望

微震震源定位研究是微震监测技术研究的重点之一,定位精度的高低直接影响着微震分析的结果.随着微震监测技术的广泛应用,出现了大量新的微震震源定位方法,针对不同的工程研究,这些方法各有千秋.首先,通过收集并分析近20年来国内外比较盛行的微震震源定位方法,其包括混合优化法、震源扫描算法、波速模型法和无预先测速法.然后,讨论这些方法的优缺点及适用条件,并主要分析微震检波器分布、首次到时拾取、波速模型和定位算法对微震震源定位的影响.最后,讨论地球物理学、声学、军事学等领域已有的先进定位方法在微震震源定位研究的意义.结果表明:(1)优化传感器的布置,必须协调自动到时拾取、波速模型和定位算法之间的关系;(2)减小波速模型和首次到达时间对震源位置的影响是微震监测未来发展的必然要求;(3)采用射线追踪理论、逆时偏移成像和被动时间反转镜技术可对速度结构模型进行优化;(4)利用双波定位、信息融合技术可进一步提高定位精度;(5)其他领域的先进定位方法在微震震源定位研究上具有潜在价值,可以达到微震震源定位更快、更准、更稳定的目的.     

瞬变电磁测深资料的ABC算法反演研究

目前瞬变电磁测深资料的反演基本局限于一维,且多数是基于最小二乘法,实际上是将非线性问题做线性化处理,再利用迭代的方法逐步逼近最优解,所以在一定程度上会丢失许多细节信息,同时极度依赖初始模型,易陷入局部最优解,而非全局最优解.为提高瞬变电磁测深法的反演解释精度,本文以瞬变电磁测深的一维正演理论为基础,同时针对野外实测资料的特点,建立非线性反演问题的目标函数,考虑到现有非线性反演算法存在的易陷入局部优化和过学习等问题,首次引入了一种当前优化领域的十分新颖且性能优良的智能优化算法——ABC(人工蜂群)算法,来求解反演问题.ABC算法对求解多极值非线性问题优势明显,它不需要了解待求解问题特殊信息,只需要对问题进行优劣的比较,通过单独人工蜂的局部寻优的行为,最终将全局最优值突显出来,具有很快的收敛速度.通过在Matlab平台下对算法参数的详细试算研究,最终确定了一套最佳反演参数,并以二、三、四层的地电模型为例进行了验证.在野外实测资料的反演解释中,经过与实地钻孔资料及其他反演方法的对比分析,表明利用ABC算法建立的瞬变电磁测深一维反演系统具有明显的理论优势和实用价值.     

基于粒子群算法的瑞雷波频散曲线反演研究

瑞雷波频散曲线反演的本质是对目标函数求极值的过程.传统的线性局部反演算法容易陷入局部极小值,增加了反演结果的不确定性.粒子群算法作为一种全局非线性优化手段,能够保证各搜索空间内局部寻优的同时,逼近全局最优,保证迭代反演的收敛性.建立含软夹层型地层模型,正演计算得到理论频散曲线,采用粒子群算法反演得到的横波速度与模型高度吻合.同时,在理论频散曲线中加入10%的高斯白噪声,粒子群算法的反演结果仍然具有很好的可靠程度.通过对实测频散曲线的反演表明:粒子群算法在含软夹层型地层的横波速度探测中相对于最小二乘法优势明显,具有很高的应用价值.     

基于L1范数的瞬变电磁非线性反演

瞬变电磁反演存在高度的非线性特征,常用的最小二乘等线性反演方法往往对初始模型高度依赖,并且极易陷入局部最优解.本文基于观测数据与模拟数据的L1范数建立目标函数,采用模拟退火非线性全局最优化方法实现瞬变电磁一维反演.初始模型完全随机产生,通过指数函数退温机制模拟系统能量最小实现迭代,通过接收概率函数评价当前模型,实现局部最优解的跳出,最终实现全局最优化求解.通过数值算例发现,无论给定的反演层数等于还是大于设计模型,都可以获得较好的反演效果,因而可以在反演初始就设计较多的层数,实现反演模型的自动拟合;同时,利用含噪声数据反演进一步验证算法的稳定性.最后,对实测数据进行了反演测试,结果与钻孔编录基本一致,表明提出的基于L1范数的模拟退火反演可用于实测数据处理.     

布谷鸟马尔科夫链蒙特卡洛混合高斯地质统计学随机反演

地质统计学随机反演可以获得比常规反演更高分辨率的结果,目前已成为储层高分辨率预测的主流方法.地下不同岩相储层参数存在明显差异,本文在地质统计学反演框架下构建了岩相和储层参数同步反演目标函数,实现不同岩相条件下储层参数分布精细描述.在求解该高维数据多参数同步反演问题时,本文将可以动态调节搜索步长的布谷鸟算法与马尔科夫链蒙特卡洛方法融合,采用多条马尔科夫链进行Levy飞行产生新解的策略扩大解的空间范围,通过适应度最佳选择输出最优解实现全局优化迭代,有效提升了反演方法的稳定性和全局最优性,避免了传统马尔科夫链蒙特卡洛方法因抽样随机性而陷入局部最优的问题.通过含噪声模型和实际数据分析验证了本文方法的有效性.     

基于改进蚁群算法在面波频散曲线反演中的应用

为快速准确的反演得到近地表地层结构,将一种新颖而强大的非线性算法——蚁群算法引入到瑞雷波频散曲线领域,并对其进行相应的改进,改进蚁群算法的优点是运算效率快、精度高、算法简单、灵活易于实现,需要调节控制参数也较少。文中分别在无噪声\,含噪声以及实测数据进行反演测试,通过模型数据和实测数据表明,应用于瑞雷波反演中的改进蚁群算法在收敛速度与收敛精度之间能达到良好的平衡,所得解具有较高可信度。而且算法为促进所得解快速收敛到全局最优,在搜索中分全局搜索与局部搜索两个方式进行,能够有效地避免局部最优解产生。借助人工合成的瑞雷波数据以及真实观测数据,验证了改进蚁群算法在反演近地表剪切波速度时的有效性和通用性。此外,文中与遗传算法进行比较,得出改进蚁群算法具有高效性和高精度性的优点。     

Cross double‐difference inversion for simultaneous velocity model update and microseismic event location

Microseismic monitoring is an approach for mapping hydraulic fracturing. Detecting the accurate locations of microseismic events relies on an accurate velocity model. The one‐dimensional layered velocity model is generally obtained by model calibration from inverting perforation data. However, perforation shots may only illuminate the layers between the perforation shots and the recording receivers with limited raypath coverage in a downhole monitoring problem. Some of the microseismic events may occur outside of the depth range of these layers. To derive an accurate velocity model covering all of the microseismic events and locating events at the same time, we apply the cross double‐difference method for the simultaneous inversion of a velocity model and event locations using both perforation shots and microseismic data. The cross double‐difference method could provide accurate locations in both the relative and absolute sense, utilizing cross traveltime differences between P and S phases over different events. At the downhole monitoring scale, the number of cross traveltime differences is sufficiently large to constrain events locations and velocity model as well. In this study, we assume that the layer thickness is known, and velocities of P‐ and S‐wave are inverted. Different simultaneous inversion methods based on the Geiger's, double‐difference, and cross double‐difference algorithms have been compared with the same input data. Synthetic and field data experiments suggest that combining both perforation shots and microseismic data for the simultaneous cross double‐difference inversion of the velocity model and event locations is available for overcoming the trade‐offs in solutions and producing reliable results.     

Feasibility of joint 1D velocity model and event location inversion by the Neighbourhood algorithm

Using a set of synthetic P‐ and S‐wave onsets, computed in a 1D medium model from sources that mimic a distribution of microseismic events induced by hydrofrac treatment to a monitoring geophone array(s), we test the possibility to invert back jointly the model and events location. We use the Neighbourhood algorithm for data inversion to account for non‐linear effects of velocity model and grid search for event location. The velocity model used is composed of homogeneous layers, derived from sonic logging. Results for the case of one and two monitoring wells are compared. These results show that the velocity model can be obtained in the case of two monitoring wells, if they have optimal relative position. The use of one monitoring well fails due to the trade‐off between the velocity model and event locations.     

三维地震与地面微地震联合校正方法

由于地面微地震监测台站布设在地表,会受到地表起伏、低降速带厚度和速度变化的影响,降低了微地震事件的识别准确度和定位精度,限制了地面微地震监测技术在复杂地表地区的应用.因此,将三维地震勘探技术的思路引入到地面微地震监测中,提出了三维地震与地面微地震联合校正方法,将油气勘探和开发技术更加紧密地结合在一起.根据三维地震数据和低降速带测量数据,通过约束层析反演方法建立精确的近地表速度模型,将地面微地震台站从起伏地表校正到高速层中的平滑基准面上,有效消除复杂近地表的影响.其次,根据射孔数据和声波测井速度信息,通过非线性反演方法建立最优速度模型,由于已经消除复杂近地表的影响,在进行速度模型优化时不需要考虑近地表的影响,因而建立的速度模型更加准确.最后,在精确速度模型的基础上,通过互相关方法求取剩余静校正量,进一步消除了复杂近地表和速度模型近似误差的影响.三维地震与地面微地震联合校正方法采用逐步校正的思路,能够有效消除复杂近地表的影响,提高微地震数据的品质和速度模型的精确度,保证了微地震事件的定位精度,具有良好的应用前景.     

Effective anisotropic velocity model from surface monitoring of microseismic events

We develop a methodology to obtain a consistent velocity model from calibration shots or microseismicity observed on a buried array. Using a layered 1D isotropic model derived from checkshots as an initial velocity model, we invert P‐wave arrival times to obtain effective anisotropic parameters with a vertical axis of symmetry (VTI). The nonlinear inversion uses iteration between linearized inversion for anisotropic parameters and origin times or depths, which is specific to microseismic monitoring. We apply this technique to multiple microseismic events from several treatments within a buried array. The joint inversion of selected events shows a largely reduced RMS error indicating that we can obtain robust estimates of anisotropic parameters, however we do not show improved source locations. For joint inversion of multiple microseismic events we obtained Thomsen anisotropic parameters ε of 0.15 and δ of 0.05, which are consistent with values observed in active seismic surveys. These values allow us to locate microseismic events from multiple hydraulic fracture treatments separated across thousands of metres with a single velocity model. As a result, we invert the effective anisotropy for the buried array region and are able to provide a more consistent microseismicity mapping for past and future hydraulic fracture stimulations.     

Understanding acquisition and processing error in microseismic data: An example from Pouce Coupe Field,Canada

A challenge in microseismic monitoring is quantification of survey acquisition and processing errors, and how these errors jointly affect estimated locations. Quantifying acquisition and processing errors and uncertainty has multiple benefits, such as more accurate and precise estimation of locations, anisotropy, moment tensor inversion and, potentially, allowing for detection of 4D reservoir changes. Here, we quantify uncertainty due to acquisition, receiver orientation error, and hodogram analysis. Additionally, we illustrate the effects of signal to noise ratio variances upon event detection. We apply processing steps to a downhole microseismic dataset from Pouce Coupe, Alberta, Canada. We use a probabilistic location approach to identify the optimal bottom well location based upon known source locations. Probability density functions are utilized to quantify uncertainty and propagate it through processing, including in source location inversion to describe the three-dimensional event location likelihood. Event locations are calculated and an amplitude stacking approach is used to reduce the error associated with first break picking and the minimization with modelled travel times. Changes in the early processing steps have allowed for understanding of location uncertainty of the mapped microseismic events.     

Approximate traveltime inversion in downhole microseismic monitoring

In downhole microseismic monitoring, accurate event location relies on the accuracy of the velocity model. The model can be estimated along with event locations. Anisotropic models are important to get accurate event locations. Taking anisotropy into account makes it possible to use additional data – two S-wave arrivals generated due to shear-wave splitting. However, anisotropic ray tracing requires iterative procedures for computing group velocities, which may become unstable around caustics. As a result, anisotropic kinematic inversion may become time consuming. In this paper, we explore the idea of using simplified ray tracing to locate events and estimate medium parameters. In the simplified ray-tracing algorithm, the group velocity is assumed to be equal to phase velocity in both magnitude and direction. This assumption makes the ray-tracing algorithm five times faster compared to ray tracing based on exact equations. We present a set of tests showing that given perforation-shot data, one can use inversion based on simplified ray-tracing even for moderate-to-strong anisotropic models. When there are no perforation shots, event-location errors may become too large for moderately anisotropic media.     

球坐标系下多震相走时三参数同时反演成像

球坐标系下多震相走时三参数(速度、震源位置和反射界面)同时反演需要解决两个关键问题:(1)球坐标系下3D速度模型中多次透射、反射(折射)及转换波精确、快速的射线追踪;(2)同时反演时三种不同参数间的强耦合问题.为此,我们将直角坐标系下分区多步不规则最短路径算法推广至球坐标系中,进行区域或者全球尺度的多震相射线追踪.然后将其与适合多参数同时反演的子空间算法相结合,形成一种球坐标系下联合多震相走时三参数同时反演的方法技术.与双参数(速度和反射界面或速度和震源位置)同时反演的数值模拟对比分析显示:三参数与双参数的同时反演结果大体接近,并且它们对到时数据中可容许的随机噪声不太敏感.结果说明本文中的同时反演成像为一种提高成像分辨率,同时反演速度、震源位置和反射界面的有效方法.     

Development and Applications of Double-difference Seismic Tomography

Double-difference (DD) tomography is a generalization of DD location; it simultaneously solves for the three-dimensional velocity structure and seismic event locations. DD tomography uses a combination of absolute and more accurate differential arrival times and hierarchically determines the velocity structure from larger scale to smaller scale. This method is able to produce more accurate event locations and velocity structure near the source region than standard tomography, which uses only absolute arrival times. We conduct a stability and uncertainty analysis of DD tomography based on a synthetic data set. Currently three versions of the DD tomography algorithms exist: tomoDD, tomoFDD and tomoADD. TomoDD assumes a flat earth model and uses a pseudo-bending ray-tracing algorithm to find rays between events and stations while tomoFDD uses a finite-difference travel-time algorithm and the curvature of the Earth is considered. Both codes are based on a regularly distributed inversion grid, with the former for a local scale and the latter for a regional scale. In contrast, tomoADD adapts the inversion mesh to match with the data distribution based on tetrahedral and Voronoi diagrams. We discuss examples of applying DD tomography to characterize fault zone structure, image high-resolution structure of subduction zones, and determine the velocity structure of volcanoes.     

基于局部一维模型与走时标定的区域 三维速度模型构建技术研究

构建区域介质三维速度模型并以之获得准确的区域震相走时, 是提高区域地震定位精度的重要手段之一. 为充分利用已有的一维模型、 GT事件、 地质资料等实现三维模型构建, 尝试基于目标区域内已有的部分局部一维模型, 通过克里金空间插值建立初始三维模型, 然后利用GT事件走时数据并参考其它地震地质资料对其不断进行修正, 使得其走时偏差图与GT事件走时偏差图一致, 进而获得能够提高区域地震定位精度的三维模型. 使用不同模型进行的地震定位实验表明, 以此方法建立的三维模型的定位偏差较初始模型减少约20%, 较好地起到了减小区域震相走时残差, 提高区域地震定位精度的作用.      

Comparison of surface and borehole locations of induced seismicity

Monitoring of induced microseismic events has become an important tool in hydraulic fracture diagnostics and understanding fractured reservoirs in general. We compare microseismic event and their uncertainties using data sets obtained with surface and downhole arrays of receivers. We first model the uncertainties to understand the effect of different acquisition geometries on location accuracy. For a vertical array of receivers in a single monitoring borehole, we find that the largest part of the final location uncertainty is related to estimation of the backazimuth. This is followed by uncertainty in the vertical position and radial distance from the receivers. For surface monitoring, the largest uncertainty lies in the vertical position due to the use of only a single phase (usually P‐wave) in the estimation of the event location. In surface monitoring results, lateral positions are estimated robustly and are not sensitive to the velocity model. In this case study, we compare event location solutions from two catalogues of microseismic events; one from a downhole array and the second from a surface array of 1C geophone. Our results show that origin time can be reliably used to find matching events between the downhole and surface catalogues. The locations of the corresponding events display a systematic shift consistent with a poorly calibrated velocity model for downhole dataset. For this case study, locations derived from surface monitoring have less scatter in both vertical and horizontal directions.