Machine learning and geophysical inversion — A numerical study

Машинное обучение и геофизическая инверсия — численное исследование
Brian Russell
2019-07-01

feedforward neural networkgeophysical inversionnonlinear regressionreflection coefficientsseismic deconvolution
Abstract As geophysicists, we are trained to conceptualize geophysical problems in detail. However, machine learning algorithms are more difficult to understand and are often thought of as simply “black boxes.” A numerical example is used here to illustrate the difference between geophysical inversion and inversion by machine learning. In doing so, an attempt is made to demystify machine learning algorithms and show that, like inverse problems, they have a definite mathematical structure that can be written down and understood. The example used is the extraction of the underlying reflection coefficients from a synthetic seismic response that was created by convolving a reflection coefficient dipole with a symmetric wavelet. Because the dipole is below the seismic tuning frequency, the overlapping wavelets create both an amplitude increase and extra nonphysical reflection coefficients in the synthetic seismic data. This is a common problem in real seismic data. In discussing the solution to this problem, the topics of deconvolution, recursive inversion, linear regression, and nonlinear regression using a feedforward neural network are covered. It is shown that if the inputs to the deconvolution problem are fully understood, this is the optimal way to extract the true reflection coefficients. However, if the geophysics is not fully understood and large amounts of data are available, machine learning can provide a viable alternative to geophysical inversion.
1
A reflection-coefficient dipole below seismic tuning frequency produces overlapping wavelets, causing amplitude increases and spurious nonphysical reflection coefficients.
2
The analysis compares deconvolution, recursive inversion, linear regression, and nonlinear feedforward-neural-network regression for recovering true reflection coefficients.
3
The study uses a synthetic seismic example to clarify the mathematical structure shared by geophysical inversion and machine-learning methods.
4
When geophysical understanding is incomplete but large datasets are available, machine learning offers a viable alternative to conventional geophysical inversion.
5
When the deconvolution inputs are fully known, deconvolution is optimal for extracting the true reflection coefficients.

Synthetic seismic response generated by convolving a below-tuning-frequency reflection-coefficient dipole with a symmetric wavelet

Extraction of the underlying true reflection coefficients and comparison of geophysical inversion with machine-learning inversion, including the effects of overlapping wavelets and nonphysical coefficients

Publication Details
Publication Date
2019-07-01
Journal
Publisher
ISSN
Cited by
54
Access Type
Author Information
Authors
Brian Russell
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%