We have developed spatio-temporal neural-network-based models that can produce high-fidelity interpolated or extrapolated seismic images effectively and efficiently. Specifically, our models are built on an autoencoder, and incorporate the long short-term memory (LSTM) structure with a new loss function regularized by optical flow. We validate the performance of our models in monitoring and forecasting the CO2 storage using real 4D post-stack seismic imaging data acquired at the Sleipner CO2 sequestration field.
Keywords:machine learning, subsurface storage, interpolation network, enhanced recovery, deep learning, applied geoscience, linear extrapolation, artificial intelligence, upstream oil & gas, autoencoder
This content is only available via PDF.
Copyright 2021, Society of Exploration Geophysicists