Summary Reservoir simulation is the industry standard for prediction and characterization of processes in the subsurface. However, large gridblock counts simulation is computationally expensive and time-consuming. This study explores data-driven reduced-order models (ROMs) as an alternative to detailed physics-based simulations. ROMs that use neural networks (NNs) effectively capture nonlinear dependencies and only require available operational data as inputs. NNs are usually labeled black-box tools that are difficult to interpret. On the other hand, physics-informed NNs (PINNs) provide a potential solution to these shortcomings, but they have not yet been applied extensively in petroleum engineering. In this study, a black-oil reservoir simulation model from Volve public data release was used to generate training data for an ROM leveraging long short-term memory (LSTM) NNs’ temporal modeling capacity. Network configurations were explored for their optimal configuration. Monthly oil production was forecast at the individual wells and full-field levels, and then validated against real field data for production history to compare its predictive accuracy against the simulation results. The governing equations for a capacitance resistance model (CRM) were then added to the reservoir-scale NN model as a physics-based constraint and to analyze parameter solutions for efficacy in characterization of the flow field. Data-driven ROM results indicated that a stateless LSTM, with single time lag as input, generated the most accurate predictions. Using a walk-forward validation strategy, the single well ROM increased prediction accuracy by about 95% average when compared with the reservoir simulation and did so with much less computational resources in short time duration. Physical realism of reservoir-scale predictions was improved by the addition of CRM constraint, demonstrated by the removal of negative flow rates. Parameter solutions to the governing equation showed good agreement with the field-scale streamline plots and demonstrated the ROM ability to detect spatial irregularities. These results clearly demonstrate the ease with which ROMs can be built and used to meet or exceed the predictive capabilities of certain time-history production data using the reservoir simulation.
Offshore petroleum production operations pose a unique set of challenges. A common undesirable phenomenon that occurs in these multiphase flow systems is known as slug flow. Slug flow is an oscillatory flow regime that creates large bullet shaped bubbles (also known as Taylor Bubbles) followed by large slugs of liquid. This high-rate alternation of liquid and gas production volumes in the surface facilities causes severe pressure oscillations. These oscillations adversely affect the structural health and individual components. A bench-scale closed flow loop was built with capabilities of measuring pressure and flow rates at different relevant sections. PID control strategy to mitigate the harmful effects of slug flow regime showed promise, although the tests were performed in the low pressure conditions of bench scale setup. The sensors and valve were programmed with MATLAB® to provide real time analysis, and a PID controller was utilized to adjust the back pressure. Initial experimental data and visual observation provided better understanding of slug flow regime and some quantitative data was obtained through image processing. Theoretical estimates of Taylor bubble velocities were found to be in agreement with presented observations. Further experiments are being carried out to gather data and showcase this model to develop better multiphase flow control strategies.
Slug flow is a major problem to the structural integrity and production equipment in offshore production platforms. Pressure oscillations due to the alternation of liquid and gas phases in slug flow regime can cause fatigue on the structural components of the platform. Also, the intermittent high flow rates can cause adverse effects on the production equipment. A 28-foot pilot scale model was constructed to simulate the riser on offshore platforms. Three pressure sensors were attached to the model to monitor and record pressures in the riser during operations. A PID control strategy was utilized to regulate the pressure oscillations in the system by use of a linear actuated valve. Similarity between the pressure signals in the pilot scale model is qualitative when compared to actual pressures observed in an offshore riser system. A MATLAB® GUI was designed to allow for manipulation of the valve and allow for instant graphing of data for real time visualization of the pressure signals. Pressure oscillations during slug flow with “no control” vary greatly and result in natural vibrations of the designed system. By pinching down on the choke valve to a designated opening, the back pressure in the riser increased, thereby slowing down the liquid slugs. However, an increase in the magnitude of the higher frequency oscillations can have adverse effects on the system. With the implementation of an active control, such as a linear actuated valve, a better control of back pressure on the riser and reduction in the magnitude of the higher frequency oscillations on the system is achieved.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.