Decarbonization

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.
This work uses a novel pseudosteady-state-based simulation to reduce training-data-generation cost while maintaining high-performance predictions of data-driven proxy models for carbon-sequestration projects.
Sustainable energy continues to grow as a focus for reliable, affordable, and secure energy as seen from the past year of papers reviewed for this feature. Three primary areas are being reported on heavily: carbon use for enhanced oil recovery, geological hydrogen discovery, and critical minerals from the subsurface.

Page 14 of 60