Methods for adaptive optimization of enhanced oil recovery performance under uncertainty

Patent metadata

Field	Value
Publication number	US-11598185-B2
Application number	US-201514949032-A
Country	US
Kind code	B2
Filing date	Nov 23, 2015
Priority date	Nov 24, 2014
Publication date	Mar 7, 2023
Grant date	Mar 7, 2023

Abstract

Methods are provided for adaptive optimization of enhanced oil recovery project performance under uncertainty. Predictive physics-based reservoir simulation is used to estimate performance of the project. Input parameters of the model are divided into control variables and uncertain variables. The reservoir model is optimized to obtain values of control variables maximizing mean value of a chosen performance metric under initial uncertainty of formation and fluid properties. An efficient frontier can characterize dependence between the optimized mean value of the performance metric and its uncertainty expressed by the standard deviation. Global sensitivity analysis (GSA) is then applied to quantify and rank contributions from uncertain input parameters to the standard deviation of the optimized values of the performance metric. Additional measurements can be performed to reduce uncertainty in the high-ranking parameters. Constrained optimization of the model with reduced ranges of uncertain parameters is performed and a new efficient frontier is obtained.

First claim

What is claimed is:

1. A method of performing an enhanced oil recovery (EOR) project for a formation containing fluids, the method comprising:
defining a performance metric, a plurality of control variables, and a plurality of uncertain variables of the EOR project;
applying the plurality of control variables and the plurality of uncertain variables as input parameters into a predictive physics-based reservoir model;
optimizing the performance metric under initial uncertainty of the plurality of uncertain variables to obtain a set of optimized values of the control variables that provide an initial optimum of an objective function based on the performance metric;
conducting a global sensitivity analysis for the plurality of uncertain variables using the set of optimized values of the control variables, wherein the sensitivity analysis comprises a first order sensitivity index;
performing a measurement, using at least one of a core analysis surface tool or a seismic survey surface tool, on at least one of the plurality of uncertain variables to reduce uncertainty in at least one of the plurality of uncertain variables;
optimizing the performance metric with reduced uncertainty in the at least one of the plurality of uncertain variables to obtain an updated set of optimized values of the control variables that provide an updated optimum of the objective function based on the performance metric;
running the reservoir model with the updated set of optimized values of the control variables in order to determine performance of the EOR project; and
performing the EOR project on a formation according to the updated set of optimized values of the control variables, wherein performing the EOR project includes altering an operation of a production well or an injection well based upon, at least in part, the updated set of optimized values of the control variables.

2. The method according to claim 1 , further comprising: repeating the conducting, the performing, the optimizing with reduced uncertainty, and the running a plurality of times.

3. The method according to claim 1 , further comprising: determining contributions of the plurality of uncertain variables to total uncertainty of the performance metric.

4. The method according to claim 3 , further comprising: ranking the plurality of uncertain variables based on their contribution to total uncertainty of the performance metric, wherein the reducing comprises selecting an uncertain variable from the plurality of uncertain variables contributing to the total uncertainty.

5. The method according to claim 4 , wherein the selecting comprises selecting a highest ranked uncertain variable, and wherein the sensitivity analysis comprises a total sensitivity index.

6. The method according to claim 3 , further comprising: comparing the contribution of the at least one uncertain variable to a threshold value.

7. The method according to claim 6 , further comprising: setting a value of at least one uncertain variable to a fixed value prior to the repeating based on whether a contribution of the at least one uncertain variable is less than the threshold value.

8. The method according to claim 7 , wherein the setting a value comprises setting the value to a midpoint value for the uncertain variable.

9. The method according to claim 7 , wherein the objective function is F(α,β)=μ(α, β)−λσ(α, β) where a are the plurality of control variables, β are the plurality of uncertain variables, μ and σ are the mean and standard deviation of the performance metric computed based on an output of the reservoir model respectively, and λ is a non-negative parameter defining a tolerance to risk.

10. The method according to claim 9 , wherein the plurality of uncertain variables include at least two of:
viscosity of water as a function of an EOR agent concentration,
surface tension of water-oil interface as a function of an EOR agent concentration,
saturated concentration of an EOR agent adsorbed by the formation, and
miscibility of oil and water with an EOR agent as a function of the capillary number.

11. The method according to claim 9 , wherein the plurality of control variables include at least one of:
target rate of a production well,
target rate of an injection well, and
an EOR agent concentration in the injected fluid in the injection well corresponding to at least one period of injection.

12. The method according to claim 1 , wherein the defining comprises defining probability distribution functions (pdfs) for the uncertain variables.

13. The method according to claim 1 , wherein the objective function is F(α, β)=μ(α, β)−λσ(α, β) where α are the plurality of control variables, β are the plurality of uncertain variables, μ and σ are the mean and standard deviation of the performance metric computed based on an output of the reservoir model respectively, and λ is a non-negative parameter defining a tolerance to risk.

14. The method according to claim 1 , wherein the plurality of uncertain variables include at least two of:
water relative permeability at residual oil saturation,
gas relative permeability at residual oil saturation,
viscosity of water as a function of an EOR agent concentration,
surface tension of water-oil interface as a function of an EOR agent concentration, saturated concentration of an EOR agent adsorbed by the formation, and
miscibility of oil and water with an EOR agent as a function of the capillary number.

15. The method according to claim 1 , wherein the plurality of control variables include at least one of:
target rate of a production well,
target rate of an injection well, and
an EOR agent concentration in the injected fluid in the injection well corresponding to at least one period of injection.

16. The method according to claim 1 , wherein the performance metric include at least one of: oil recovery efficiency, incremental oil production, total oil production, and financial indicator of project profitability.

17. The method according to claim 1 , wherein performing the measurement on at least one of the plurality of uncertain variables comprises:
performing a measurement of at least one of the formation and the fluids within the formation.

18. The method according to claim 1 , further comprising: performing the EOR project using the updated set of optimized values of the control variables.

19. The method according to claim 1 , wherein:
the plurality of uncertain variables include an aqueous phase viscosity at a non-zero shear rate, and
the objective function is F(α, β)=μ(α, β)−λσ(α, β) where α are the plurality of control variables, β are the plurality of uncertain variables, μ and σ are the mean and standard deviation of the performance metric computed based on an output of a the reservoir model respectively, and λ is a non-negative parameter defining a tolerance to risk.

20. A method for performing an enhanced oil recovery (EOR) project under uncertainty, the method comprising:
utilizing a predictive physics-based reservoir model to estimate the performance of the EOR project;
identifying a plurality of input parameters into predictive physics-based reservoir the model as control variables and uncertain variables;
optimizing the predictive physics-based reservoir model to obtain values of control variables maximizing a mean value of a chosen performance metric under initial uncertainty of formation and fluid properties;
applying a global sensitivity analysis to quantify and rank contributions from uncertain input parameters to the stand

Assignees

• Schlumberger Technology Corp

Inventors

• Chugunov Nikita
• Ramakrishnan Terizhandur S
• Barrat-Charlaix Pierre

Classifications

• E21B43/16 Primary

  Enhanced recovery methods for obtaining hydrocarbons · CPC title

• G06F17/11

  for solving equations {, e.g. nonlinear equations, general mathematical optimization problems (optimization specially adapted for a specific administrative, business or logistic context G06Q10/04)} · CPC title