AI/machine learning
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 downtime of manufacturing machinery, engines, or industrial equipment can cause an immediate loss of revenue. Reliable prediction of such failures using multivariate sensor data can prevent or minimize the downtime. With the availability of real-time sensor data, machine-learning and deep-learning algorithms can learn the normal behavior of the sensor systems, dis… -
Privacy concerns about AI systems are growing. So researchers are testing whether they can remove sensitive data without retraining the system from scratch.
-
This paper describes an artificial intelligence deep Q network for field-development plan optimization.
-
In this paper, the authors describe a model that uses augmented artificial intelligence to optimize well spacing by use of data sculpting, domain and feature engineering, and machine learning.
-
SponsoredRod lift failure frequency in horizontal wells drives significant operating expenses. Understanding why these failures occur leads to a solution — production optimization with automated setpoint changes, which can extend the life of this equipment and reduce downtime.
-
Cube drilling was an exciting idea several years ago. Since then, the luster seems to have faded. Now, production software company Novi Labs says machine learning may bring life back to the concept.
-
Four CEOs describe what goes into turning a world of data into a data-driven world.
-
The authors write that even simple deep-learning architectures can identify a leak using pressure data.
-
The authors demonstrate how artificial intelligence and machine learning can help build a purely data-driven reservoir simulation model that successfully history matches dynamic variables for wells in a complex offshore field and that can be used for production forecasting.
-
One of the major characteristics of petroleum data analytics is its incorporation of explainable artificial intelligence (XAI). Predictive models of petroleum data analytics are not represented through unexplainable black-box behavior. Predictive models of petroleum data analytics are reasonably explainable. This second part of a two-part series presents the use of XA…
