Reservoir simulation

Digital space with glowing lines, 3d rendering.

Geothermal energy

Digital Twin Initiative Could Reduce Geothermal Production Uncertainties

The DOE-backed EGS-Twin project aims to simulate geothermal production systems, helping operators better predict performance and maximize output.

Carbon dioxide CO2 model on a bad smog background.

Carbon capture and storage

A Sub-Grid Model for Convective Mixing Applied to the 11th SPE Comparative Solution Project

Modeling solubility trapping at reservoir scale remains a key challenge for carbon storage projects. This study introduces a novel sub-grid convective mixing model, implemented in the Open Porous Media Flow simulator and applied to the SPE11 benchmark, that enables accurate prediction of carbon dioxide dissolution trapping in geological storage using computationally e…

Fig. 1—Accelerated GNN training with multi-GPU parallelism implementation.

Reservoir simulation

Fast, Scalable Graph Neural Networks Used for Subsurface in Porous-Media Simulations

This work describes a study in which distributed data parallel training, paired with a node-local caching pipeline, enabled efficient multigraphics-processing-unit scaling for a CO₂-storage graph-neural-network surrogate while maintaining generalization.