Coupled Reservoir and Wellbore Flow Simulation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for modeling fluid flows in underground porous media around wells face challenges such as high computation time, convergence issues, and mass balance deterioration, particularly when simulating both large-scale reservoir flows and detailed near-well phenomena simultaneously.
Innovation Solution
A computer-implemented method that uses two flow simulators: one for the reservoir and another for the well surroundings, with linear interpolation of boundary conditions between time intervals to iteratively update numerical productivity indices, allowing for efficient modeling of fluid flows and optimizing well operations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If hybrid meshes combining reservoir meshes and locally refined well meshes are used in a single flow simulator, then the behavior of flows in the vicinity of the well is better taken into account, but computation time increases significantly
Solution Approach 1:
The patent divides the simulation domain into two separate models: a reservoir model with coarse meshes for the bulk reservoir, and a well model with fine meshes for the near-wellbore region. This segmentation allows each model to use appropriate mesh densities, avoiding the computational burden of a fully refined hybrid mesh while maintaining accuracy where needed.
Solution Approach 2:
The patent introduces an intermediary coupling mechanism that exchanges boundary conditions and flow rates between the reservoir model and well model at the wellbore interface. This mediator enables the two separate models to work together as a unified system, achieving hybrid mesh accuracy without the computational cost of a single integrated hybrid mesh model.
2Loss of time
If domain decomposition techniques are used to separate reservoir and well simulations, then computation time is reduced, but convergence and stability problems arise
Solution Approach 1:
The patent implements a feedback mechanism where the reservoir model provides boundary conditions to the well model, and the well model returns updated flow rates and pressures to the reservoir model. This iterative feedback loop ensures convergence by continuously adjusting the interface conditions until consistency is achieved between the two models.
Solution Approach 2:
The patent performs preliminary initialization of the well model using boundary conditions from the reservoir model before running the coupled simulation. This preliminary action ensures that both models start from consistent initial states, improving numerical stability and convergence behavior during the iterative coupling process.
3Device complexity
If a single flow simulator is used for both reservoir and well meshes, then model integration is simplified, but mass balance deteriorates in domain decomposition approaches
Solution Approach 1:
The patent uses an intermediary coupling interface that explicitly tracks and balances mass transfer between the reservoir model and well model. The coupling mechanism ensures that flow rates and pressures exchanged at the wellbore interface satisfy mass conservation, preventing the mass balance deterioration that occurs in domain decomposition methods without proper balancing.
4Measurement precision
If very small meshes are used for well discretization, then near-wellbore phenomena are resolved accurately, but the number of meshes and computation time increase
Solution Approach 1:
The patent segments the computational domain so that fine meshes are used only in the well model for the near-wellbore region, while coarse meshes are used in the reservoir model for the bulk reservoir. This segmentation concentrates computational resources where they are most needed, achieving high accuracy near the well without the prohibitive cost of refining the entire reservoir grid.
Solution Approach 2:
The patent extracts the wellbore region from the full reservoir model and treats it as a separate, independent well model. This extraction allows the use of very fine meshes only in the extracted well region, rather than requiring fine meshes throughout the entire reservoir, thereby maintaining accuracy where it matters while preserving overall calculation efficiency.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
The method involves simulating flow of fluids within a porous medium using a simulator over a time interval defined between times, and deducing conditions at updated limits for another simulator. The flow of fluids is simulated near a well using the latter simulator over the same time interval by using the conditions at the updated limits, and updated digital productivity indices are deduced for the former simulator. The flow of fluids is modeled within the porous medium during a time period between the times by repeating simulations for successive time intervals between the times. An independent claim is also included for a method for exploiting a porous underground reservoir using a well.