数据空间磁异常模量三维反演

强剩磁的存在通常导致了总磁化强度方向未知,进而影响了磁异常的反演和解释.磁异常模量是一种受磁化方向影响小的转换量,可以在强剩磁条件下通过反演三维磁化强度大小分布来推测场源分布状态.我们提出了一种数据空间磁异常模量反演算法来减少剩磁的影响.与标准的模型空间L2范数正则化反演方法相比,我们的方法有两个优点:一是无需搜索正则化参数(需要反复求解非线性反演问题),因而可以减少计算时间;二是反演结果更加聚焦,深度分辨率更高,我们对此进行了原因分析.通过模型和实测数据测试证明了该算法的有效性和更好的反演效果.     

基于交叉梯度约束的时间域激发极化法二维同步反演

时间域激发极化法(Time-domain induced polarization method,简称为TDIP)已有的反演算法采用的是分步反演的思路,即先由视电阻率资料反演电阻率,固定电阻率再由视极化率资料反演极化率,这样就存在极化率结果严重依赖于电阻率反演结果的问题.为了有效解决这一问题,本文实现了TDIP二维数据空间分步反演算法,提出了基于交叉梯度约束的TDIP二维同步反演策略,实现了交叉梯度约束的电阻率和极化率二维同步反演算法.分别用电阻率和极化率结构一致和不一致的二维模型合成数据进行了分步和同步反演试算,对不同模型试算结果进行了对比分析.结果表明:对于电阻率和极化率结构一致和不一致模型,同步反演结果比分步反演结果能更好地确定异常体的空间分布范围,反演得到的电阻率和极化率值更接近真值.理论模型算例表明本文提出的同步反演算法有效解决了分步反演的问题,优于分步反演算法,具有更好的实用性.     

大地电磁三维数据空间反演并行算法研究

目前大地电磁三维反演实际应用的主要问题是计算效率低.在对大地电磁三维数据空间反演算法进行深入分析的基础上,本文提出了基于频点和矩阵划分的大粒度并行反演方案和具体实现步骤,并在曙光TC5000A高性能计算平台上实现了基于MPI的大地电磁三维数据空间反演并行算法.该算法实现了包括三维正演、灵敏度矩阵、叉积矩阵以及模型改正量的并行执行,不仅计算效率高,而且每个节点机上灵敏度矩阵的存储空间只需原来微机上的2/N(N是参加并行计算的节点机个数),大大地减少了内存开销.通过两个理论模型合成的数据对实现的三维数据空间反演并行算法进行试算,对比分析了多个节点机下程序的执行效率.测试结果表明,所实现的三维数据空间反演并行算法是可行的、高效的,与单机相比,不仅可以提高运行速度,缩短计算时间,而且还可以扩大计算规模,极大地推动了大地电磁三维反演的实用化.     

Two-dimensional direct current (DC) resistivity inversion: Data space Occam's approach

A data space Occam's inversion algorithm for 2D DC resistivity data has been developed to seek the smoothest structure subject to an appropriate fit to the data. For traditional model space Gauss–Newton (GN) type inversion, the system of equations has the dimensions of M × M, where M is the number of model parameter, resulting in extensive computing time and memory storage. However, the system of equations can be mathematically transformed to the data space, resulting in a dramatic drop in its dimensions to N × N, where N is the number of data parameter, which is usually less than M. The transformation has helped to significantly reduce both computing time and memory storage. Numerical experiments with synthetic data and field data show that applying the data space technique to 2D DC resistivity data for various configurations is robust and accurate when compared with the results from the model space method and the commercial software RES2DINV.     

Two-dimensional data-space joint inversion of magnetotelluric,gravity, magnetic and seismic data with cross-gradient constraints

In recent years, joint inversion has been widely used for integrated geological interpretation. We extended a data-space joint inversion algorithm of magnetotelluric, gravity and magnetic data to include first-arrival seismic travel-time and normalized cross-gradient constraints. We describe the main features of the algorithm and apply it to synthetic data generated for hypothetical models. For the synthetic data, we find that the joint inversion with multiple parameters is superior to the joint inversion with two or even three parameters, which can reduce the multisolution of inversion results more effectively. Furthermore, data-space joint inversion involves fewer memory requirements and better calculation speeds than traditional model-space joint inversion. The normalized cross-gradient constraints can better couple model parameters of different magnitudes compared with traditional unnormalized cross-gradient constraints, resulting in higher levels of structural similarity among resistivity, density, magnetic susceptibility and velocity models.