Reservoir
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 paper presents a novel reservoir engineering/reservoir simulation approach—a data-driven interwell-connectivity model augmented as a digital twin—to predict reservoir dynamics and optimize operations in the Changqing oil field of China.
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.
-
The past couple of years has been a rollercoaster for energy professionals, but it did not stop the incredible achievements in machine-learning (ML) techniques, particularly neural networks to improve seismic imaging. -
SponsoredImprove decision making and reduce the uncertainty of CCS projects with accurate, detailed subsurface insights that help you estimate storage capacity, run play chance mapping, improve your injection strategy, and simulate carbon plumes over time.
-
Drilling is set to take place near the location where the Toyko-based energy company said it encountered a gas reservoir in a 2016 project.
-
You know about upstream, midstream, and downstream. Now, get ready for the coming age of "counterstream" which will leverage many of the industry's existing engineering skillsets.
-
The paper highlights two case studies to illustrate successful development of marginal prospects to unlock remaining potential.
-
This paper introduces methods to fully couple reservoir simulation with wellbore flow models in fractured injection wells.
-
A rigless chemical frac-hit remediation process was designed to address the damage mechanisms of capillary phase trapping, reduced hydrocarbon relative permeability, paraffin deposition, and minor scale deposition.
-
The disruptive nature of unconventional reservoirs has spurred the need for new ideas. But that doesn’t always mean that we need to toss out the traditional methods that got us here. Such might be the case for the recent developments unfolding around the diagnostic fracture injection test (DFIT).
-
This paper provides an alternative solution to identifying, classifying, and vertically distributing fractures and a lateral distribution method for reservoir modeling.
-
The paper demonstrates the ability of deep-learning generative models to enable new shale-characterization methods.
