FREQUENCY DOMAIN ELASTIC WAVEFORM INVERSION USING THE GAUSS-NEWTON METHOD

被引:0
|
作者
Chung, Wookeen [1 ]
Shin, Jungkyun [2 ]
Bae, Ho Seuk [2 ]
Yang, Dongwoo [3 ]
Shin, Changsoo [2 ]
机构
[1] Korea Maritime Univ, Dept Energy & Resources Engn, Pusan 606791, South Korea
[2] Seoul Natl Univ, Dept Energy Syst Engn, Seoul 151742, South Korea
[3] Western Australian Energy Res Alliance, Kensington, WA 6151, Australia
来源
JOURNAL OF SEISMIC EXPLORATION | 2012年 / 21卷 / 01期
关键词
Gauss-Newton method; conjugate gradient; CGLS; elastic; frequency domain; waveform inversion; SEISMIC DATA; FIELD;
D O I
暂无
中图分类号
P3 [地球物理学]; P59 [地球化学];
学科分类号
0708 ; 070902 ;
摘要
Chung, W., Shin, J., Bae, H.S., Yang, D. and Shin, CS., 2012. Frequency domain elastic waveform inversion using the Gauss-Newton method. Journal of Seismic Exploration, 21: 29-48. In spite of several advantages of using the Gauss-Newton method for waveform inversion, calculating a Hessian-matrix is problematic in the current computing environment. Consequently, many researchers have suggested a Conjugate Gradient Least Square (CGLS) algorithm to circumvent the Hessian matrix problem in the Gauss-Newton method. There has been no attempt, however, at using the CGLS method to perform elastic media waveform inversion. It is necessary to consider the characteristics of elastic media because the acoustic wave equation does not accurately simulate waveforms propagated through elastic media. Our objective is to develop a frequency domain waveform inversion algorithm using the CGLS method for elastic media. Numerical experiments with the Marmousi-2 model and the SEG/EAGE salt model verify enhancement of the inversion results. We confirm that a greatly improved waveform inversion can be carried out over a complicated layer structure with the Marmousi-2 model. We also verify that structures containing a high velocity salt can be reproduced correctly through an inversion experiment using the SEG/EAGE salt model. However, the improved inversion result of the SEG/EAGE salt model is limited to use with the long wavelength initial velocity, necessitating a method for obtaining the long wavelength velocity model.
引用
收藏
页码:29 / 48
页数:20
相关论文
共 50 条
  • [31] Truncated Gauss-Newton full-waveform inversion of pure quasi-P waves in vertical transverse isotropic media
    Ren, Zhi-Ming
    Wang, Lei
    Bao, Qian-Zong
    PETROLEUM SCIENCE, 2024, 21 (05) : 3102 - 3124
  • [32] LEAST-SQUARES METHOD FOR THE BUBBLE STABILIZATION BY THE GAUSS-NEWTON METHOD
    Kim, Seung Soo
    Lee, Yong Hun
    Oh, Eun Jung
    HONAM MATHEMATICAL JOURNAL, 2016, 38 (01): : 47 - 57
  • [33] Gauss-Newton method for convex composite optimizations on Riemannian manifolds
    Wang, Jin-Hua
    Yao, Jen-Chih
    Li, Chong
    JOURNAL OF GLOBAL OPTIMIZATION, 2012, 53 (01) : 5 - 28
  • [34] Multi-scale elastic full waveform inversion in frequency domain
    Li, Yuanyuan
    Li, Zhenchun
    Zhang, Kai
    Geophysical Prospecting for Petroleum, 2015, 54 (03) : 317 - 323
  • [35] A distributed Gauss-Newton method for distribution system state estimation
    Li, Keqiang
    Han, Xueshan
    INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS, 2022, 136
  • [36] Adaptive Gauss-Newton Method for Solving Systems of Nonlinear Equations
    Yudin, N. E.
    DOKLADY MATHEMATICS, 2021, 104 (02) : 293 - 296
  • [37] A nonmonotone damped Gauss-Newton method for nonlinear complementarity problems
    Dong, Li
    ITALIAN JOURNAL OF PURE AND APPLIED MATHEMATICS, 2023, (49): : 206 - 215
  • [38] ON THE CONVERGENCE OF INEXACT GAUSS-NEWTON METHOD FOR SOLVING SINGULAR EQUATIONS
    Argyros, Ioannis K.
    George, Santhosh
    JOURNAL OF NONLINEAR FUNCTIONAL ANALYSIS, 2016,
  • [39] Comparison of TE and TM Inversions in the Framework of the Gauss-Newton Method
    Mojabi, Puyan
    LoVetri, Joe
    IEEE TRANSACTIONS ON ANTENNAS AND PROPAGATION, 2010, 58 (04) : 1336 - 1348
  • [40] Nonlinear Compensation Using the Gauss-Newton Method for Noise-Robust Speech Recognition
    Zhao, Yong
    Juang, Biing-Hwang
    IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2012, 20 (08): : 2191 - 2206