Adaptive Coupling of Fluid-Flow and Geomechanical Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current fluid-flow and geomechanical models for petroleum systems require more efficient coupling strategies to accurately simulate reservoir behavior, especially in the presence of pre-existing fracture networks and stress changes, which existing numerical convergence criteria fail to address effectively, leading to high computational costs and inaccurate results.
Innovation Solution
A novel coupling method that analyzes historical data to identify triggering events causing abrupt stress changes, allowing for tighter coupling at specific time steps and automatic iteration calculation, while maintaining loose coupling during gradual changes, thereby reducing computational costs and improving accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If fully coupling or sequentially iterative coupling schemes are used to accurately simulate reservoir behavior, then simulation accuracy is improved, but computational cost increases significantly
Solution Approach 1:
The patent applies dynamic coupling by adjusting the coupling frequency and tightness adaptively based on triggering events. The system transitions between loosely coupled and tightly coupled modes depending on whether significant stress changes or pressure changes are detected, rather than maintaining a fixed coupling strategy throughout the simulation.
Solution Approach 2:
The patent changes the coupling parameter (coupling tightness/frequency) based on detected triggering events. When events such as significant stress changes, pressure changes, or compaction are detected, the system increases coupling tightness; otherwise, it maintains looser coupling to reduce computational burden.
2Measurement precision
If tighter coupling is applied at all time steps to capture abrupt stress changes, then simulation accuracy is improved, but computational time increases
Solution Approach 1:
The patent implements periodic tight coupling only at specific intervals determined by triggering events rather than continuously. The system monitors for events such as significant stress changes, pressure changes, and compaction, and applies tight coupling only when these events occur, otherwise using looser coupling schemes.
Solution Approach 2:
The patent performs preliminary identification of triggering events using analytical criteria before executing the full simulation. By pre-identifying when significant changes are likely to occur based on operational data and thresholds, the system can prepare and apply appropriate coupling strategies only when needed.
3Extent of automation
If numerical convergence criteria are used to determine coupling points, then coupling automation is improved, but ability to detect abrupt stress changes deteriorates
Solution Approach 1:
The patent uses feedback from multiple sources including operational data, detected triggering events, and analytical criteria to dynamically adjust coupling strategy. The system continuously monitors for events such as significant stress changes, pressure changes, and compiction, and uses this feedback to determine when tight coupling is necessary.
Solution Approach 2:
The patent segments the simulation time into different coupling regimes based on triggering events. Rather than using a uniform coupling approach, the system divides the simulation into periods of loose coupling and tight coupling based on the occurrence of significant events, allowing optimized computational effort at each segment.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Coupling fluid-flow model and geomechanical model for integrated petroleum systems, in one aspect, may comprise analyzing historical data associated with a reservoir to determine one or more triggering events that trigger abrupt changes in the state of stress of the reservoir solid framework and in the pore pressure. One or more time steps are defined based on the determined triggering events. The fluid-flow model and the geomechanical model are coupled at the one or more defined time steps, e.g., one-way or two-way. Number of iterations may be calculated automatically for the two-way coupling to converge.