三维光滑磁化率成像反演及在大冶铁矿的应用

3D smooth inversion of magnetic data and its application to the Daye iron mine

为使磁测数据的解释结果能够刻画复杂的磁性体,探讨了应用三维磁化率成像技术来反演磁测数据.本文应用基于Occam反演的模型粗糙度约束的光滑成像算法来求解纯欠定的磁化率成像问题.在求解大规模线性方程组时,本文使用预优共轭梯度法并将深度加权函数作为预优因子,既提高了反演的收敛速度,又使磁化率分布在合理的深度上.采用绝对约束的方法对反演的模型值的取值范围进行了约束,取得了较好的约束效果.二维和三维典型模型的数值实验表明,本文给出的磁化率成像方法可以恢复真实模型的主要特征信息.将本文的磁化率成像反演技术应用于大冶铁矿狮子山矿段的实测数据的反演解释,获得了与地质信息具有很好对应关系的成像结果.

I present a method for inverting the magnetic data to describe the complicated susceptibility source model. In this paper, I use the smooth inversion method which is based on the Occam＇s model roughness constraint to solve the underdetermined susceptibility imaging problem. I use the preconditioned conjugate gradient method to solve the large-scale equations and the depth weighting function as the pre-conditioner which accelerate the convergence speed and make the susceptibility distribute at the right depth. A positive constraint is imposed by the absolute bound method and a good result is obtained. Testing by the synthetic examples of 2D and 3D representative models, the imaging method presented in this paper recovers the major information of the real model. Applying the 3D susceptibility imaging method to interpret the field data of the Shizishan ore section of the Daye iron mine, a good result is yielded which matches very well with the geology of the work area.