Deep Learning Application for 4D Pressure Saturation Inversion Compared to Bayesian Inversion on North Sea Data

This is a Preprint and has not been peer reviewed. The published version of this Preprint is available: This is version 2 of this Preprint.

Add a Comment

You must log in to post a comment.


There are no comments or no comments have been made public for this article.


Download Preprint

Supplementary Files

Jesper Sören Dramsch , Gustavo Corte, Hamed Amini, Mikael Lüthje, Colin MacBeth


In this work we present a deep neural network inversion on map-based 4D seismic data for pressure and saturation. We present a novel neural network architecture that trains on synthetic data and provides insights into observed field seismic. The network explicitly includes AVO gradient calculation within the network as physical knowledge to stabilize pressure and saturation changes separation. We apply the method to Schiehallion field data and go on to compare the results to Bayesian inversion results. Despite not using convolutional neural networks for spatial information, we produce maps with good signal to noise ratio and coherency.



Applied Statistics, Earth Sciences, Geophysics and Seismology, Physical Sciences and Mathematics, Statistics and Probability


Deep learning, Geophysics, Seismic Inversion, North Sea, Neural Network, 4D, 4D seismic, time-lapse seismic, pressure, reservoir, saturation


Published: 2019-02-21 00:28

Last Updated: 2019-02-22 23:16

Older Versions

CC BY Attribution 4.0 International