Reservoir simulation
The authors propose a deep-learning-based approach enabling near-real-time CO2-plume visualization and rapid data assimilation incorporating multiple geological realizations for predicting future CO2 plume evolution and area-of-review determination.
In this study, forward simulation is executed by a commercial reservoir simulator while external code is developed for backward calculations.
In this study, the authors propose the use of a deep-learning reduced-order surrogate model that can lower computational costs significantly while still maintaining high accuracy for data assimilation or history-matching problems.
-
This paper presents results from evaluating the rate of thermal energy that can be extracted under various completion scenarios in end-of-life oil and gas wells using a transient flow simulator. -
This paper presents a numerical simulation work flow, with emphasis on hydraulic fracture simulation, that optimizes well spacing and completion design simultaneously.
-
The authors discuss the development of a deep-learning model to identify errors in simulation-based performance prediction in unconventional reservoirs.
-
The authors of this paper compare case studies from the Bakken and the STACK plays to conclude that mineralogy, petrophysics, and reservoir-condition differences between basins cause differences in the effect of fracture-driven interactions.
-
The authors of this paper propose a novel approach to data-driven modeling for transient production of oil wells.
-
This paper describes a novel distributed quasi-Newton derivative-free optimization method for reservoir-performance-optimization problems.
-
The complete paper builds on existing tools in the literature to quantify the effect of changing well spacing on well productivity for a given completion design, using a new, simple, intuitive empirical equation.
-
The paper describes an approach to history matching and forecasting that does not require a reservoir simulation model, is data driven, and includes a physics model based on material balance.
-
This paper presents a physics-informed neural network technique able to use information from fluid-flow physics as well as observed data to model the Buckley-Leverett problem.
-
Whether it be Derrick Turk asking us to “resist the temptation to accrue vocabulary rather than understanding” or Mark Bentley telling us that “if you can sketch it you can model it,” there does seem to be a growing pushback against the notions that the modeling/simulation process can be successfully shrink-wrapped and that fundamental understanding is increasingly a …
