Reservoir simulation
The authors present an open-source framework for the development and evaluation of machine-learning-assisted data-driven models of CO₂ enhanced oil recovery processes to predict oil production and CO₂ retention.
The authors of this paper propose hybrid models, combining machine learning and a physics-based approach, for rapid production forecasting and reservoir-connectivity characterization using routine injection or production and pressure data.
Virtual reality and related visualization technologies are helping reshape how the industry views 3D data, makes decisions, and trains personnel.
-
With the easy conventional oil in Argentina having been produced, one remaining way to find new oil in existing fields is to convert fields from primary or secondary production to secondary or tertiary production, respectively. -
Computational advances in reservoir simulation have made possible the simulation of thousands of reservoir cases in a practical time frame. This enables exhaustive exploration of subsurface uncertainty and development/depletion options.
-
This paper describes how seismic reservoir integration, advanced production analysis, and accurate nanoscale and 3D full-field simulations may address profitability issues and help oil companies to be more efficient in developing unconventional portfolios.
-
Reservoir-simulation-model inputs are numerous, and uncertainty is pervasive—before, during, and after development. With the pressure to deliver results quickly, how do we find the right balance?
-
In upstream oil and gas, cloud computing is very immature because the industry has always been challenged by storage and computational capability. However, high-performance cloud computing may create an opportunity for smaller companies lacking infrastructure for scientific applications.
-
With the recent drop in oil prices, operators are shifting to optimization of existing assets with minimal costs. For mature floods (water, chemical, and CO2), one low-cost optimization strategy is the intelligent adjustment of well-rate targets.
-
Permanent downhole gauges (PDGs) can provide a continuous record of flow rate and pressure, which provides extensive information about the reservoir. In this work, a machine-learning framework based on PDG data was extended to two applications: multiwell testing and flow-rate reconstruction.
-
In this paper, the authors introduce a novel semianalytic approach to compute the sensitivity of the bottomhole pressure (BHP) data with respect to gridblock properties.
-
This paper critically investigates the impact of using realistic, inaccurate simulation models. In particular, it demonstrates the risk of underestimating uncertainty when conditioning real-life models to large numbers of field data.
-
History matching is only one part of something more comprehensive—reservoir modeling.
