可控源音频大地电磁法张量数据二维反演研究

发布时间：2018-06-18 16:48

本文选题：可控源电磁法 + 张量　；参考：《中国地质大学(北京)》2015年硕士论文

【摘要】：可控源音频大地电磁法(CSAMT)最大特点就是采用了人工场源,大大增加了信号强度,从而弥补了天然场源电磁信号微弱,不易观测等缺点,已经被广泛应用于油气勘探、矿产普查、水文环境等方面,并且发挥了巨大的作用。此外,它还具有工作效率高,高阻屏蔽作用小,水平和垂直分辨率高等优点。但是,传统的可控源音频大地电磁法采用的多为单一偶极装置的标量测量,其缺点一是这种标量测量只适合于简单的地质环境,不适于复杂地质的勘查;二是观测范围会受到发射装置的限制,在区域观测或者长剖面观测中,发射源的位置往往需要调整,如果发射源正好布置在断裂构造上,那么由于受到场源效应的影响就会导致多个发射源的观测数据无法处理。所以可控源张量数据的反演研究是有实际应用价值并且非常有必要的。在正演中,本文首先分别将两个互相垂直的与测线斜交45°的场源分解为平行X轴和平行Y轴的子场源,然后分别计算了X方向子场源和Y方向子场源作用下产生的电磁场分量,通过张量阻抗公式计算出张量阻抗元素,最后通过卡尼亚视电阻率公式计算出视电阻率和相位。在计算单个电偶极子源作用下产生的电磁场分量时,采用了二次场的方法来减小源对计算结果的影响,提高计算精度。通过有限元推导形成正演矩阵方程,求解矩阵方程得到波数域的结果,最后通过反傅氏变换得到空间域的结果。在反演中,工作重点是灵敏度矩阵的求取,本文在计算各电磁场分量灵敏度矩阵时采用了Mc Gillivray等人提出的伴随方程法,选择不同的伴随场源即可求得不同电磁场分量的灵敏度矩阵;在计算视电阻率和相位的灵敏度矩阵时,采用了Rodi(2001)的方法。采用在Occam反演基础上改进的数据空间反演方法来反演合成数据,设计了不同的模型进行反演,得到了比较好的结果,验证了算法的正确性和可靠性,实测数据的反演算例表明了该反演算法可用于实测资料的处理解释。
[Abstract]:The most important feature of controllable source audio magnetotelluric method (CSAMT) is the use of artificial field sources, which greatly increases the signal intensity, thus making up for the weakness of electromagnetic signals from natural field sources, such as weak and difficult to observe, and has been widely used in oil and gas exploration and mineral survey. Hydrological environment and other aspects, and played a huge role. In addition, it has the advantages of high efficiency, low resistance shielding and high horizontal and vertical resolution. However, the traditional controlled source audio magnetotelluric method is mostly a scalar measurement of a single dipole device. One of its disadvantages is that the scalar measurement is only suitable for simple geological environment and is not suitable for the exploration of complex geology. Second, the observation range will be limited by the launcher. In regional observation or long profile observation, the position of the emitter often needs to be adjusted, if the emitter is located on the fault structure, The observation data of multiple emitters can not be processed because of the effect of field source effect. Therefore, the inversion of controllable source Zhang Liang data has practical application value and is very necessary. In the forward modeling, the two field sources which are perpendicular to each other and oblique to the measuring line 45 掳are decomposed into parallel X axis and parallel Y axis subfield sources respectively, and then the electromagnetic field components generated by the action of X direction subfield source and Y direction subfield source are calculated, respectively. The Zhang Liang impedance element is calculated by Zhang Liang impedance formula, and the apparent resistivity and phase are calculated by Kania apparent resistivity formula. In the calculation of the electromagnetic field components produced by a single electric dipole source, the quadratic field method is used to reduce the influence of the source on the calculation results and to improve the accuracy of the calculation. The forward matrix equation is derived by finite element method, the results of wavenumber domain are obtained by solving the matrix equation, and the results of space domain are obtained by inverse Fourier transform. In the inversion, the emphasis is on the calculation of sensitivity matrix. In this paper, the adjoint equation method proposed by MC Gillivray et al is used to calculate the sensitivity matrix of each electromagnetic field component. The sensitivity matrix of different electromagnetic field components can be obtained by choosing different adjoint field sources, and the method of Rodigen 2001 is used to calculate the sensitivity matrix of apparent resistivity and phase. The improved data space inversion method based on Occam inversion is used to inverse the synthetic data. Different models are designed for inversion. The results show that the algorithm is correct and reliable. An example of the inversion of measured data shows that the algorithm can be used to interpret the measured data.
【学位授予单位】：中国地质大学(北京)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P631.325

【相似文献】