Том: 21st International Conference on Computational Science and Its Applications, ICCSA 2021. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Cagliari, Italy, September 13-16, 2021)
Том: 12950
, Год издания: 2021
Многотомное издание: 21st International Conference on Computational Science and Its Applications, ICCSA 2021. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Cagliari, Italy, September 13-16, 2021)
Страницы: 282-295
Аннотация
We present a robust nonlinear traveltime inversion procedure that enables fast iterative 3D depth velocity model reconstruction. It is based on automatic grid reflection traveltime tomography. To obtain input data necessary for inversion (traveltimes of reflected waves), we utilize conventional seismic time processing products such as normal-moveout velocities and a set of time-migrated reflection horizons picked in the final prestack time migration image. Ray-based tomographic inversion fits prestack traveltimes. They are approximated by hyper-bolae using the developed effective engine of grid reflection tomography. We prove the effectiveness of the proposed approach numerically using a real-life example.
