3D inversion modeling of joint gravity and magnetic data based on a sinusoidal correlation constraint

Gao, Xiaojing; Xiong, Shaoxiang; Zeng, Zhaofa; Cao, Yu; Zhang, Guibin; Sun, Shasha

doi:10.1007/s11770-019-0792-z

Cited by 6 publications

(2 citation statements)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Because of the strong constraint ability of the correlation-analysis constraints [e.g., Gao et al 2019], The regularization method of elastic nets [Zhang and Li., 2019] can avoid excessive sharp boundaries of the recovered models obtained by L1-norm regularization technology and improve excessive smooth boundaries of the recovered models obtained by L2-norm regularization technology. For these reasons we adopt the elastic-nets regularization method in this paper.…”

Section: Objective Functions For Joint Inversionmentioning

confidence: 99%

“…The second type involves joint inversion methods based on statistical or empirical petrophysical relationships [e.g., Gardner et al, 1974;Lines et al, 1988;Li, 2015, 2016]. Other joint inversion methods are based on different physical property models with similar spatial distribution structures, such as cross-gradient joint inversion [e.g., Haber and Oldenburg, 1997;Gallardo and Meju, 2003;Fregoso and Gallardo, 2009;Vatankhah et al, 2022], correlation-analysis joint inversion [e.g., Oldenburg and Li, 1999;Lelièvre et al, 2012;Yin et al, 2018;Gao et al, 2019], and the joint inversion based on Gramian constraint [e.g., Zhdanov et al, 2012;Lin and Zhdanov, 2018]. In this case, the non-uniqueness can be effectively reduced by properly combining the correlation-analysis constraint with the joint inversion [e.g., Yin et al, 2018;Gao et al, 2019].…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

3-D data-space joint inversion of gravity and magnetic data using a correlation-analysis constraint

Liu

Jin²,

Xuan³

et al. 2022

Annals of Geophysics

View full text Add to dashboard Cite

Non-uniqueness, low computational efficiency and large memory requirements are main issues for geophysical data inversion. In this paper, we propose an efficient algorithm for 3D correlationanalysis joint inversion of gravity and magnetic data with high accuracy and low computation effort. Firstly, since the number of the observed field data is smaller than the number of inverted parameters, the calculations of the correlation-analysis for joint inversion of gravity and magnetic data in model space (MS) are transformed into the equivalent calculations in data space (DS), which can reduce the dimensions of the calculation domain, improve the computation efficiency and reduce the non-uniqueness. Then, an improved conjugate gradient (ICG) method is employed for the optimization algorithm, which can facilitate the use of stable functions with sparse factors and improve the accuracy of the inversion. The inversion performed by the combined DS-ICG method for synthetic data tests shows the calculation effort can be effectively reduced, and the issues with non-uniqueness are improved. Finally, the test by real field data can delineate the distribution of underground geological bodies, which illustrates the strong stability and good applicability of our extended method.

show abstract

Section: Objective Functions For Joint Inversionmentioning

confidence: 99%