Font Size: a A A

A Study Of Three-dimensional Magnetotelluric Conjugate Gradient Inversion Based On Data Space

Posted on:2017-03-18Degree:MasterType:Thesis
Country:ChinaCandidate:G N LiFull Text:PDF
GTID:2180330503479263Subject:Geological Resources and Geological Engineering
Abstract/Summary:PDF Full Text Request
Magnetotelluric sounding method(MT), which observe the natural alternating electromagnetic field worldwide, is a frequency domain electromagnetic method study the electrical property of the underground stratum. However It’s hard to delineate the targets by traditional 2D profile survey because the ore targets always present three dimensional electrical property. Thus, developing the 3D inversion is the key point in the world.Geophysicists never stop to improving the inversion methods because a good method can reduce the computer memory and so much more.We accomplished the three-dimensional magnetotelluric forward by using the staggered-grid finite-difference method and then verified the accuracy and applicability of this algorithm using the typical geoelectric models. One this basis, The three-dimensional magnetotelluric inversion, which based on the data space conjugate gradient algorithm, has been accomplished. This algorithm updates the operator by using conjugate gradient algorithm model which based on the Occam algorithm. It can calculate the sensitivity from the data space instead of the model space, greatly reduces the computational running memory, On the other hand, this algorithm calculates the sensitivity matrix or the matrix transpose and vector product, avoiding the accessing and computing the sensitivity matrix directly.The validity and precision have been proved by several typical geoelectrical model that DCG algorithm can recover underground terrestrial electricity correctly, the inversion result shows that the rate of convergence is faster and has a smaller memory usage than the Nonlinear Conjugate Gradient(NLCG)algorithm.Finally this paper analyze the impact of noise immunity, the initial background resistivity, the selection of frequency, the influence of the selection of data sets, the paper showed that: DCG inversion algorithm has a good anti-noise property; As the initial background resistivity close to the target volume resistivity value, It will has a better inversion effect; the frequency should be appropriately increased near the target body and It has a better effect which select the total impedance tensor for inversion.
Keywords/Search Tags:Data space Ocaam algorithm, conjugate gradient(CG)algorithm, 3D inversion, Magnetolluric method
PDF Full Text Request
Related items