1887

Abstract

Summary

We investigate a simple Deep Learning Neural Network architecture to invert High Resolution seismic data for reflectivity and acoustic impedance. We generate synthetic reflectivity model and corresponding seismic traces by a simple convolution with a wavelet for training the machine learning network. Synthetic reflectivity models are correctly recovered when the reflectors separation is in the range of the train set. Noise in the data is correctly handled if the network is trained with adding noise on the data. Finally, we investigate the prediction of the acoustic impedance using the Marmousi model.

Loading

Article metrics loading...

/content/papers/10.3997/2214-4609.202020169
2020-12-07
2024-04-16
Loading full text...

Full text loading...

References

  1. Kim, Y., & Nakata, N. [2018] Geophysical inversion versus machine learning in inverse problems. The Leading Edge, 37(12), 894–901.
    [Google Scholar]
  2. Kingma, D. P., & Ba, J. [2014] Adam: A method for stochastic optimization. arXiv preprint arXiv: 1412.6980.
    [Google Scholar]
  3. Li, S., Liu, B., Ren, Y., Chen, Y., Yang, S., Wang, Y., & Jiang, P. [2019] Deep learning inversion of seismic data. arXiv preprint arXiv: 1901.07733.
    [Google Scholar]
  4. Puzyrev, V., Egorov, A., Pirogova, A., Elders, C., & Otto, C. [2019] Seismic Inversion with Deep Neural Networks: a Feasibility Analysis. In: 81st EAGE Conference and Exhibition2019 (Vol. 2019, No. 1, pp. 1–5).
    [Google Scholar]
  5. Santos, C.G.P., Evsukoff, A.G., & Mansur, W.J. [2009]. Post-stacking Acoustic Seismic Inversion using Recurrent Neural Network. In: 11th International Congress of the Brazilian Geophysical Society (pp. cp-195).
    [Google Scholar]
http://instance.metastore.ingenta.com/content/papers/10.3997/2214-4609.202020169
Loading
/content/papers/10.3997/2214-4609.202020169
Loading

Data & Media loading...

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error