A Fast Method for Imaging Potential Field Sources with Improved Generalized Linear Inversion
Several imaging methods for potential field data have been proposed that yield images of quantities related or proportional to the 3D magnetization/density distributions.Because all the methods were noniterative, computationally fast, they did provide a quick picture of physical property distributions.However, such source images are often too divergent.Aiming at this problem, we proposed a new method with improved generalized linear inversion.Firstly, we use generalized linear inversion approach to produce an image of a quantity proportional to the source distribution, and then the transformed field is transformed into source-density units by determining a constant with adequate physical dimension by a linear regression of the observed field versus the field computed from the generalized linear inversion image.Finally, we introduce a compacting function, which expressed as a matrix to make the source distribution models more convergent.The method is tested on synthetic gravity data, and the final imaged source is a good reconstruction of the geometry.
Generalized linear inversion fast imaging gravity gradient tensor 3D inversion
Qi Zhang Ping Yu
College of Geoexploration Science and Technology, Jilin University, Changchun, PRC
国际会议
杭州
英文
567-571
2018-06-10（万方平台首次上网日期，不代表论文的发表时间）