Reservoir Simulation Optimization Using Statistical Moments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current reservoir simulation methods are computationally expensive and time-consuming due to the need to model a large number of geological realizations to achieve accurate results, especially when optimizing operations like well liquid flow rates and oil flow rates for objectives such as maximizing net present value (NPV).
Innovation Solution
A method that selects a minimal number of realizations based on statistical moments of the simulated reservoir data, optimizing control variables through partial and full simulations, and iteratively refining the number of realizations until convergence is achieved, reducing the computational burden while maintaining accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a large number of geological realizations are modeled to achieve accurate reservoir simulation results, then measurement precision is improved, but productivity deteriorates due to computational expense and time consumption
Solution Approach 1:
The patent extracts only the essential realizations needed to capture statistical moments (mean, variance, skewness, kurtosis) from the full set of geological realizations. By identifying and removing redundant realizations that do not contribute significantly to statistical accuracy, the method maintains measurement precision while reducing the computational burden proportionally.
Solution Approach 2:
The patent transforms the problem from modeling all realizations to modeling a reduced set characterized by statistical moments. By changing the parameter representation from individual realization details to aggregated statistical parameters (mean, variance, skewness, kurtosis), the method achieves the same informational content with fewer computational elements.
2Productivity
If the number of realizations is reduced to improve productivity, then computational efficiency is improved, but measurement precision deteriorates due to insufficient sampling
Solution Approach 1:
The patent performs preliminary analysis to identify which realizations contribute most to each statistical moment before conducting full optimization. By pre-selecting realizations that capture the essential statistical characteristics (mean, variance, skewness, kurtosis), the method ensures adequate sampling coverage is maintained while reducing the total number of realizations needed for subsequent detailed simulation.
Solution Approach 2:
The patent uses a minimal sufficient set of realizations rather than the complete set. By determining the minimum number of realizations required to accurately represent each statistical moment, the method applies just enough sampling action to achieve statistical accuracy without the excess computational cost of modeling all possible realizations.
3Manufacturing precision
If iterative optimization with multiple simulation rounds is performed to achieve convergence, then manufacturing precision is improved, but loss of time increases due to repeated full and partial simulations
Solution Approach 1:
The patent segments the optimization process into distinct phases: initial partial simulations on reduced realizations to obtain preliminary control variables, followed by full simulations on the complete realization set for final optimization. This segmentation allows each phase to focus on specific computational tasks, reducing total iteration time while maintaining convergence accuracy through the coordinated use of both simulation types.
Solution Approach 2:
The patent performs preliminary optimization using partial simulations on a reduced set of realizations before conducting full simulations. By obtaining preliminary control variable estimates from the computationally lighter partial simulations, the method prepares initial conditions that guide the subsequent full simulations, reducing the number of iteration rounds needed to achieve convergence and thereby reducing time loss.
Data Source
AI summary
A method, computer program product, and computing system are provided for receiving reservoir data associated with the reservoir. A simulation may be performed on the reservoir data to generate simulated reservoir data. A subset of realizations including a minimal number of realizations from a plurality of realizations may be determined based upon, at least in part, one or more statistical moments of the simulated reservoir data. An optimized reservoir model associated with an objective may be generated based upon, at least in part, the subset of realizations including the minimal number of realizations.