Analysis on statistic characteristics of magnetotelluric signal

For magnetotelluric sounding (MT), many processing methods based on power spectrum have put forward lots of hypotheses, such as MT signals are Gaussian, linear and minimum-phase. If practical signals do not satisfy these requirements, the results will have a few problems as follows. Firstly, when signals are non-linear and non-Gaussian, the information of the earth contained in the MT signals cannot be sufficiently extracted; Secondly, when signals are non-Gaussian and non-minimum phase, the processed results cannot reflect the minimum phase characteristics of the signals. Hence, it is necessary for us to do further research on characteristics of MT signals (YAO, SUN, 1999; LI, CHENG, 2002; Nikias, Petropulu, 1993; ZHANG, 1996). Otherwise, we cannot judge the reliability of the processed results based on power spectrum.     

基于压缩感知重构算法的大地电磁强干扰分离

为压制大地电磁信号中的强人文干扰,提出一种基于压缩感知重构算法的大地电磁信号去噪方法.通过构建与常见典型强干扰相匹配而对有用信号不敏感的冗余字典原子,利用改进的正交匹配追踪算法,分离出大地电磁信号中的强干扰成分.为了验证所述方法的强干扰分离效果,首先通过在实测大地电磁信号中加入理想的强干扰信号进行了仿真分离实验,然后从大量实测数据中选取三种含有不同类型强干扰的时间域片段,用所述方法对实测数据中的强干扰进行分离,最后将所述方法应用于青海试验点以及庐枞矿集区某测点实测数据的综合处理.仿真实验结果表明,该方法在分离出强干扰的同时,能够较好地保留有用信号.实测数据处理结果表明,该方法能够有效压制强干扰,改善强干扰区大地电磁数据的质量.

     

基于循环神经网络的大地电磁信号噪声压制研究

由于天然电磁场源信号微弱,观测数据极易受到噪声干扰,严重影响反演和解释结果.传统去噪方法依赖于人工对时间序列和功率谱的筛选,去噪效率低,主观性强.本文提出利用循环神经网络对大地电磁时域信号进行特征噪声的识别和提取,进而重构出去噪后的大地电磁信号.在对大地电磁时域信号进行大量分析的基础上,对噪声进行分类并搭建含噪信号数据库,利用该数据库训练了两个循环神经网络,并选取长短时记忆单元优化循环神经网络结构,分别实现含噪数据段筛选和噪声形态提取.对仿真和实测数据分别进行了测试,循环神经网络均能准确筛选出大地电磁信号中的噪声段,本方法在避免人为操作主观性的同时提高了工作效率,视电阻率和相位曲线质量得到明显改善.

     

基于递归分析和聚类的大地电磁信噪辨识及分离

为了剖析大地电磁信号和强干扰的本质特征,进一步精细分离出微弱的大地电磁有用信号,提出基于递归分析和聚类的大地电磁信噪辨识及分离方法.首先,运用递归分析法扩展大地电磁一维时间序列的维数,分析了嵌入维数、延迟时间和判别阈值对递归图的性能,并研究了不同长度的序列对递归定量分析参数的影响情况.然后,构建典型的大地电磁强干扰类型和微弱的大地电磁有用信号样本库,针对样本库讨论了强干扰和微弱大地电磁信号之间的递归定量分析参数,分析了K均值聚类和模糊C均值聚类的信噪辨识效果.最后,对实测大地电磁数据进行信噪辨识处理,并仅对辨识为强干扰的时间段采用数学形态滤波进行噪声压制.实验结果表明,递归分析能定性及定量地描述大地电磁信号时间序列的非线性特征和原动力系统的本质规律,与聚类算法相结合能对矿集区实测大地电磁信号进行信噪辨识;处理后的卡尼亚电阻率-相位曲线更为光滑、连续,其结果更为精细地保留了大地电磁信号低频段的缓变化信息,整个低频段的大地电磁数据质量得到了明显改善.

     

Multi-station analysis of surface wave dispersion

Methods based on multi-station recordings are presented for constructing the experimental dispersion curve of Rayleigh waves. Multi-station recording permits a single survey of a broad depth range, high levels of redundancy with a single field configuration, and the ability to adjust the offset, effectively reducing near field and far field effects. A method based on the linear regression of phase angles measured at multiple stations is introduced for determining data quality and filtering criteria. This method becomes a powerful tool for on site quality control in real time. The effects of multiple modes and survey line parameters, such as near offset, receiver spacing, and offset range, are investigated. Parametric studies result in general guidelines for the field data acquisition. A case study demonstrates how to easily deploy commonplace seismic refraction equipment to simultaneously record data for P-wave tomographic interpretation and multi-station analysis of surface wave.     

Processing of magnetotelluric data

The processing of magnetotelluric data involves concepts from electromagnetic theory, time series analysis and linear systems theory for reducing natural electric and magnetic field variations recorded at the earth's surface to forms suitable for studying the electrical properties of the earth's interior.The electromagnetic field relations lead to either a scalar transfer impedance which couples an electric component to an orthogonal magnetic component at the surface of a plane-layered earth, or a tensor transfer impedance which couples each electric component to both magnetic components in the vicinity of a lateral inhomogeneity.A number of time series spectral analysis methods can be used for estimating the complex spectral coefficients of the various field quantities. These in turn are used for estimating the nature of the transfer function or tensor impedance. For two dimensional situations, the tensor impedance can be rotated to determine the principal directions of the electrical structure.In general for real data, estimates of the apparent resistivity are more stable when calculated from the tensor elements rather than from simple orthogonal field ratios (Cagniard estimates), even when the fields are measured in the principal coordinates.     

The magnetotelluric inverse problem

The magnetotelluric inverse problem is reviewed, addressing the following mathematical questions: (a)Existence of solutions: A satisfactory theory is now available to determine whether or not a given finite collection of response data is consistent with any one-dimensional conductivity profile. (b)Uniqueness: With practical data, consisting of a finite set of imprecise observations, infinitely many solutions exist if one does. (c)Construction: Several numerically stable procedures have been given which it can be proved will construct a conductivity profile in accord with incomplete data, whenever a solution exists. (d)Inference: No sound mathematical theory has yet been developed enabling us to draw firm, geophysically useful conclusions about the complete class of satisfactory models. Examples illustrating these ideas are given, based in the main on the COPROD data series.     

Quasi-one-dimensional method for solving the inverse magnetotelluric problem

We consider the iterative numerical method for solving two-dimensional (2D) inverse problems of magnetotelluric sounding, which significantly reduces the computational burden of the inverse problem solution in the class of quasi-layered models. The idea of the method is to replace the operator of the direct 2D problem of calculating the low-frequency electromagnetic field in a quasi-layered medium by a quasi-one dimensional operator at each observation point. The method is applicable for solving the inverse problems of magnetotellurics with either the E- and H-polarized fields and in the case when the inverse problem is simultaneously solved using the impedance values for the fields with both polarizations. We describe the numerical method and present the examples of its application to the numerical solution of a number of model inverse problems of magnetotelluric sounding.     

大地电磁自适应正则化反演算法

针对大地电磁正则化反演中正则化因子的选取困难问题提出了自适应正则化反演算法（Adaptive Regularized Inversion Algorithm, ARIA）. 在该算法中, ①提出了一种新的数据方差处理方法：数据方差规范化,使得数据方差的大小只对数据的拟合发生影响,不对数据目标函数和模型约束目标函数的权重产生影响,从而减少了正则化因子取值的影响因素;②提出了粗糙度核矩阵的概念,并给出了由基本结构插值基函数计算粗糙度核矩阵的公式,使得模型目标函数的构建更为简便、直接;③根据数据目标函数、模型约束目标函数和正则化因子之间的关系,提出了两种正则化因子自适应调节方法. 本文详细阐述了最平缓模型约束下的大地电磁一维连续介质反演的ARIA实现,以几个算例的分析比较来说明ARIA的有效性.     

误差对大地电磁测深反演的影响

本文讨论了几种不同分布的误差对大地电磁测深反演的影响.首先基于给定的正演模型计算得到相应的理论合成数据,然后在合成数据中分别加入满足不同均值和方差的高斯分布、均匀分布和Gamma分布的误差,最后使用非线性共轭梯度法进行反演.反演结果表明, 较简单的模型在考虑的几种误差分布下可以较好地给出模型的基本信息;但复杂的模型随着误差方差的增大,与已知模型的偏差越明显; 在均值和方差相同时,上述三种不同分布的误差的反演结果相差很小.     

A compact representation of magnetotelluric responses for two-layer models

In this paper the locations where ρ_app = ρ₁ and ? = π/4 and where these parameters reach an extreme value in two-layer magnetotelluric (MT) sounding curves are summarized in an extremely compact form. The key parameters over two-layer models with conductivities σ₁, σ₂ and upper layer thickness h are the real S and α, where S is the conductivity contrast and α is the distance between the observation site and the conductivity interface, normalized to the half skindepth in the first layer. If the impedance components, various resistivity definitions ( ρ_{Re Z}, ρ_{Im Z} and ρ_|Z|, based on different parts of the complex impedance Z ) and the magnetotelluric phase ? are derived as a function of S and α, then the conditions for the apparent resistivity ρ_app and the phase ? are that they either satisfy ρ_app = ρ₁ and ? = π/4 or attain extreme values which can be given in terms of simple algebraic equations between S and α. All equations are valid for observation sites at any depth 0 ≤ z ≤ h in the first layer. The set of equations, presented in a tabular form, may make it possible to determine a layer boundary from the short period part of the sounding curves, in particular the ρ_{Re Z} and the ?_MT curves.     

Possibilities of magnetotelluric methods in geophysical exploration for ore minerals

In the past decade, the applications of magnetotelluric method in the electric prospecting for ore bodies have been rapidly progressing. In the present work, we summarize the first results on this way. We discuss the specificity of the geoelectrical models in the problems of mining prospecting for ore bodies. The state-of-the-art capabilities of the method, which rely on the synchronous observation systems and the procedure of joint inversion of magnetotelluric and magnetovariational responses, are considered in the context of ore mineral exploration. The results of modeling a typical mining audio-magnetotelluric survey for ore minerals are presented. On the basis of these simulations and the data provided by in-situ soundings, the efficient approaches to the processing, analysis, and inversion of these data are discussed and illustrated. The future trends in magnetotellurics as applied to the mining prospecting are analyzed.     

一种新的三维大地电磁积分方程正演方法

采用规则六面体单元和并矢Green函数奇异积分等效积分技术,已有的大地电磁积分正演方法具有不能有效模拟地下复杂地质体和计算精度偏低的缺点.本文提出了一种新的三维大地电磁积分方程正演技术,即采用四面体单元、解析的并矢Green函数奇异积分表达式,达到既能模拟地下复杂异常体,又能有效提高已有积分方程法计算精度的目的.首先,采用四面体网格技术离散地下复杂异常体,获得四面体单元上的大地电磁积分方程.然后,利用针对四面体单元开发的新的奇异值积分的解析表达式,准确计算线性方程中的并矢Green函数的奇异积分,从而获得精确的线性方程.借助于PARDISO高性能并行直接求解器,实现了三维大地电磁问题的高精度求解.最后,基于国际标准3D-1模型和六棱柱模型,通过与其他方法结果的对比分析,验证了本文方法的正确性、处理高电导率对比度的能力（1000：1）和处理复杂模型的能力.

     

Some aspects of magnetotelluric field procedures

Progress in magnetotelluric field procedures that has taken place over the past few years is reported upon. These include calibration procedure of equipment, misorientation effects, recording characteristics, in-field processing and trend towards future developments.     

Analysis of electrojet-distorted magnetotelluric sounding curves

South America presents several unique geomagnetic features, one of which is the Equatorial Electrojet (EEJ), a current system which extends itself east to west in Northeastern Brazil for almost 3500 km. Considering the fact that the influence of this phenomenon at low magnetic latitudes can be detected at great distances from its central axis, it is important to simulate its effect in magnetotelluric (MT) exploration. To accomplish this and by the use of an integral equation scheme, we have modeled the MT response of confined three-dimensional (3D) bodies (dykes in an homogeneous half-space) and deep 3D structures (horst and graben located at Marajo Basin in Northern Brazil). As the inductive source, we replace the classical plane wave source by a common line of current, besides gaussian and undulatory electrojets. The results of the modelling indicate that the studied effect is prominent in both one-dimensional (1D) and 3D media for periods (T) greater than 10 s. It decreases with distance, but it is detectable as far as 3000 km from the center of the EEJ. It is also observed that for T greater than 10 s, the computed soundings can be strongly distorted, mainly by EEJ effects in the host medium which, in turn, cause changes in the final branch of the 3D soundings. For structures in the Marajo Basin, our results reveal that the 3D galvanic effect prevails in the interval 0.1<T<10 s, while source influence occurs mainly for T>10 s. On the other hand, automatic inversion of these 3D data shows that, in equatorial regions having complex geology, some errors arise in conventional 1D interpretation of the MT soundings. This is due to the superposition of the host medium response, the galvanic effect of 3D structures, and the EEJ influence.     

面向目标自适应三维大地电磁正演模拟

本文将面向目标的自适应算法应用于三维大地电磁数值模拟.使用基于非结构网格的矢量有限单元法对起伏地表大地电磁正演模拟问题进行求解.使用利用垂向电流密度在物性界面上的连续性对后验误差进行估算的算法指导网格优化.由于全局自适应算法针对观测点优化网格的能力较差,本文通过求解正演问题的对偶问题计算后验误差的加权系数,并对相关加权系数进行改进,从而实现了面向目标的自适应算法.与传统基于结构化网格的电磁正演算法相比,采用非结构网格能够更好地拟合起伏地表和地下不规则异常体.由于使用了面向目标的自适应算法,本文能够使用更少的网格达到较高的计算精度.通过对比本文模拟结果与半空间响应和全局自适应算法计算结果,并通过对比使用改进前和改进后加权系数得到的网格剖分结果验证了本文算法的有效性.

     

Impedance-admittance regression analysis of magnetotelluric fields

The paper is devoted to the development of the theory of magnetotelluric field processing. A new method is proposed for complete consistent robust impedance-admittance regression estimation. This approach eliminates divergences between results obtained by independent solution of linear systems for the determination of impedance and admittance matrices and, moreover, is a good stabilizer of solution. Formulas for statistical estimates of results obtained by this method are derived. The theory developed in the paper is based on a linear algebraic approach. The regression problem is analyzed in a linear complex space with a nonrigidly specified metric. The metric of the solution space is formed concurrently with the solution of the problem on the basis of currently processed time series. All statistical estimations of results are performed in the metric obtained as a consequence of the solution and, in this sense, are optimal. The full regression model is effective for diagnosing the presence of electromagnetic fields that cannot be completely reduced to the plane wave approximation. the possibilities of the regression estimation are considered in detail on the basis of correlation analysis applied to sets of spectral components of electromagnetic fields.     

Estimation of the magnetotelluric impedance functions

The successful interpretation of magnetotelluric data demands well presented, and accurate, estimates of the impedance and tipper functions.In this paper, a method based on the singular value decomposition of the local data matrix, using well known results from multivariate statistical analysis, is proposed as an alternative to current least-squares methods. The advantage obtained is that error on all data channels is recognised, so that impedance estimates are not biased by the choice of “independent noise-free” fields.An important result of the method is an extensive and structured capability for the analysis of the frequency-dependent noise which could be used to improve the resulting impedance estimates.     

大地电磁测深反演方法现状与评述

如何通过改进大地电磁测深(MT)反演方法来提高数据解释的精确度一直都是大地电磁测深研究领域的重要课题.本文以作者所在研究组从事的科研项目为依托,介绍了国内外主要的大地电磁反演方法.通过总结将反演方法分为定性近似反演方法、基于目标函数的线性或非线性选代反演方法和全局搜索最优反演方法三大类,并从目标函数构建、灵敏度矩阵计算...     

Damped vibration mode superposition method for dynamic response analysis

In the dynamic response analysis of extremely complex structural systems in which the damping characteristics of each element are independent, the damping matrix is not always diagonalized by the use of undamped free vibration mode shapes. In the present paper, a mode-superposition method by the use of damped free vibration mode shapes is developed for such structural systems. It is also shown that the Fast Fourier Transform (FFT) procedures, that are available for the dynamic response analysis of linear structural systems, are used effectively in this mode-superposition method with good accuracy.