Heavy Oil Reservoir Simulation Using Capillary Number-Dependent Relative Permeability
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional reservoir simulators fail to accurately account for heavy oil solution gas drive mechanisms, leading to underestimated heavy oil production forecasts due to static relative permeability curves independent of fluid flow rates.
Innovation Solution
Developing capillary number dependent correlations for gas relative permeability, which adjust baseline correlations based on local fluid velocities and depletion rates, to capture the effects of heavy oil solution gas drive and enhance reservoir simulation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If static relative permeability curves are used in reservoir simulation, then the simulation model is simple and easy to implement, but the heavy oil production forecast is underestimated and inaccurate
Solution Approach 1:
The patent applies the dynamics principle by transforming static relative permeability curves into dynamic curves that adapt to changing flow conditions. The relative permeability is made velocity-dependent through the introduction of a velocity correction factor that adjusts the baseline relative permeability based on local fluid velocities and capillary numbers, allowing the simulation model to capture the dynamic nature of heavy oil solution gas drive mechanisms
Solution Approach 2:
The patent applies parameter changes by modifying the relative permeability parameters from fixed static values to dynamic values that change with flow conditions. The key parameter change is the introduction of velocity-dependent relative permeability, where the correction factor is derived from capillary number correlations that relate relative permeability to fluid velocity, thereby improving forecast accuracy without requiring completely new model structures
2Reliability
If velocity-dependent relative permeability is implemented, then heavy oil production forecast accuracy is improved, but computational complexity and data requirements increase
Solution Approach 1:
The patent applies preliminary action by pre-calculating and storing velocity correction factors in lookup tables before the main simulation runs. The correction factors are computed based on capillary number correlations and stored as functions of velocity and saturation, allowing the simulation to efficiently retrieve and apply these factors during runtime without performing complex calculations at each time step, thus balancing accuracy with computational efficiency
Solution Approach 2:
The patent introduces the capillary number as an intermediary parameter that mediates between fluid velocity and relative permeability. Rather than directly coupling velocity to relative permeability in a complex manner, the capillary number serves as an intermediate variable that simplifies the relationship through established correlations, making the velocity-dependent model more computationally tractable while maintaining physical accuracy
Data Source
AI summary
A method, a system and a program storage device for predicting a property of a fluid, such as fluid production from a subterranean reservoir containing heavy oil entrained with gas is described. The method includes developing a baseline correlation of gas relative permeability krg versus gas saturation Sg. A capillary number dependent correlation is determined capturing the relationship between at least one of critical gas saturation Sgc and capillary number Nca and gas relative permeability krgro and capillary number Nca phased upon a plurality of depletion rates. Capillary numbers Nc are calculated for a plurality of cells in a reservoir model representative of the subterranean reservoir. The baseline correlation is then adjusted to comport with at least one of Sgc and krgro selected from the capillary number dependent correlation to produce a plurality of corresponding adjusted baseline, correlations. Gas relative permeabilities krg for the plurality of cells are selected from the corresponding adjusted baseline correlations. A reservoir simulation is then run utilizing the selected relative permeabilities krg to predict a property of at least one fluid in a subterranean reservoir containing heavy oil entrained with gas.