DocumentCode :
83297
Title :
An Adjoint-Based Hybrid Computational Method for Crosswell Seismic Inversion
Author :
Danping Cao ; Wenyuan Liao
Author_Institution :
China Univ. of Pet., Huadong, China
Volume :
16
Issue :
6
fYear :
2014
fDate :
Nov.-Dec. 2014
Firstpage :
60
Lastpage :
67
Abstract :
Waveform tomography, or full waveform inversion (FWI), is in general better than ray tomography; however, a reliable initial model is usually required to ensure success. Here, the authors designed a cascade-like hybrid tomography technology to solve the crosswell seismic inversion problem. The new method is a combination of several widely used tomography technologies, implemented in a sequence, where the cheaper method is used as the initial model feeder for the more expensive method. They start with the Radon-transform-based back-projection method; then the Linear Iterative Reconstruction method, such as selective internal radiation therapy, is adopted to update this initial model; which is further improved by the nonlinear, Gradient-based Eikonal Equation Tomography method. Finally, the velocity model reconstructed from the previous multiple ray tomography methods is used as the initial model for the FWI.
Keywords :
Radon transforms; conjugate gradient methods; geophysical techniques; geophysics computing; nonlinear equations; seismology; FWI; Radon-transform-based back-projection method; adjoint-based hybrid computational method; cascade-like hybrid tomography technology; crosswell seismic inversion problem; full-waveform inversion; linear iterative reconstruction method; nonlinear-gradient-based Eikonal equation tomography method; ray tomography methods; selective internal radiation therapy; velocity model reconstruction; waveform tomography; Computational modeling; Iterative methods; Mathematical model; Numerical models; Seismic measurements; Subspace constraints; Tomography; FWI; GEET; LIR; algorithm design; back projection; inverse problem; numerical algorithm; ray tomography; scientific computing; waveform tomography;
fLanguage :
English
Journal_Title :
Computing in Science & Engineering
Publisher :
ieee
ISSN :
1521-9615
Type :
jour
DOI :
10.1109/MCSE.2014.1
Filename :
6728939
Link To Document :
بازگشت