Machine Learning Applications to Solving Inverse Problems in Fractured Layer Seismic Surveys
PDF (Russian)


machine learning
convolutional neural networks
grid-characteristic method
seismic survey
inverse problems
fractured media

How to Cite

Muratov M.V., Petrov D.I., Ryazanov V.V., Biryukov V.A. Machine Learning Applications to Solving Inverse Problems in Fractured Layer Seismic Surveys // Russian Journal of Cybernetics. 2022. Vol. 3, № 1. P. 8-13. DOI: 10.51790/2712-9942-2022-3-1-1.


In this paper, we solve inverse problems of exploration seismology in rocks with uniformly oriented fractured inclusions using convolutional neural networks. This type of neural network was chosen due to the large dataset. We used simulation to build a neural network training sample from the direct problem results. For the numerical solution of direct problems, a grid-characteristic method was applied to unstructured meshes. This numerical method was used since the studied dynamic processes are of wave nature, which is very suitable for the grid-characteristic method. This approach is proven in building correct computational algorithms for boundary and interface conditions, in particular, for defining discrete fracture arrays. The problem statement is to determine the characteristics of a single fracture and layers of such fractures. The inverse exploration seismology problem for a fractured layer with six unknown parameters was successfully solved. These parameters are the height and angle of inclination of the fractures, the density of the fractures, the horizontal extent of the formation, and its 2D position. The vibration velocities in the seismic data array and the frequency spectra were inputs for the neural network training and validation sample recognition.
PDF (Russian)


Claerbout J. F. Coarse Grid Calculations of Waves in Inhomogeneous Media with Application to Delineation of Complicated Seismic Structure. Geophysics. 1970;36(3):407-418.

Etgen J., Gray S., Zhang Y. An Overview of Depth Imaging in Exploration Geophysics. Geophysics. 2009;74:WCA5-WCA17.

Jiao K., Huang W., Vigh D., Kapoor J., Coates R., Starr E. W., Cheng X. Elastic Migration for Improving Salt and Subsalt Imaging and Inversion. SEG Technical Program Expanded Abstracts. 2012;1-5.

Luo Y., Tromp J., Denel B., Calandra H. 3D Coupled Acoustic-Elastic Migration with Topography and Bathymetry Based on Spectral-Element and Adjoint Methods. Geophysics. 2013;78(4):S193-S202.

Zhang C., Frogner C., Araya-Polo M., Hohl D. Machine-learning Based Automated Fault Detection in Seismic Traces. EAGE Conference and Exhibition. 2014.

Araya-Polo M., Dahlke T., Frogner C., Zhang C., Poggio T., Hohl D. Automated Fault Detection without Seismic Processing. The Leading Edge. 2017;36(3):194-280.

Nikitin I. S., Golubev V. I., Golubeva Y. A., Miryakha V. A. Numerical Comparison of Different Approaches for the Fractured Medium Simulation. Smart Innovation, Systems and Technologies. 2021;214:87-99.

Новацкий В. К. Теория упругости. М.: Мир; 1975. 872 с.

Golubev V. I., Shevchenko A. V., Khokhlov N. I., Nikitin I. S. Numerical Investigation of Compact Grid-Characteristic Schemes for Acoustic Problems. Journal of Physics: Conference Series. 2021;1902(1):012110.

Coates R. T., Shoenberg M. Finite-Difference Modeling of Faults and Fractures. Geophysics. 1995;60:1514-1523.


Download data is not yet available.