I presented in Amsterdam during the Practical Reservoir Monitoring Workshop in Amsterdam. This is the accompanying video.
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.
📝 Preprint: https://doi.org/10.31223/osf.io/zytp2
📝 Paper: https://doi.org/10.3997/2214-4609.201900028
📝 Presentation: https://doi.org/10.6084/m9.figshare.7963775
Here at The Way of the Geophysicists, we have written about social justice before. Today… Read More
This Friday we're looking at a machine learning state-of-the-art Dashboard and also a new way… Read More
It sure is an interesting time. Apologies I kept you waiting with more Friday Faves,… Read More
Aaaand it's gone. It's starting out with one of my new projects and then a… Read More
I'm starting a new project, where I take concepts from machine learning for science and… Read More
It's the holiday season, so let's keep this Friday Fave short, with a fave that… Read More