Reservoir Simulation Input Data Generation via Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for simulating hydrocarbon reservoirs with steady-state, spatially varying water-hydrocarbon interfaces are hindered by incomplete, erroneous, or unavailable data, often relying on trial-and-error or simplified modeling approaches that result in inaccurate simulations.
Innovation Solution
An automated method using optimization techniques to infer input data for simulating hydrocarbon reservoirs, where control parameters are determined to minimize the error between observed and simulated interface data, incorporating physical processes and heterogeneous/anisotropic reservoir properties.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If trial-and-error methods or simplified modeling methods are used to infer input data, then the data generation process is simpler, but the accuracy of the simulated interface is poor
Solution Approach 1:
The patent transforms the data generation process from trial-and-error parameter adjustment to an optimization-based parameter determination approach. By formulating an objective function that quantifies the match between simulated and observed interface data, the system automatically determines optimal input parameters (such as boundary conditions and initial conditions) that minimize the error, thereby achieving high accuracy without manual trial-and-error iterations.
Solution Approach 2:
The patent implements a feedback mechanism where the simulated interface results are continuously compared with observed interface data through an objective function. This feedback loop allows the system to evaluate the accuracy of input parameters and automatically adjust them to minimize the difference between simulated and observed data, ensuring high precision in the final simulation results.
2Measurement precision
If field measurements are conducted to obtain accurate interface data, then the accuracy of observed interface is improved, but the time and cost required increases
Solution Approach 1:
The patent performs preliminary simulations using available incomplete or approximate data to generate an initial interface model. This preliminary action allows the system to identify key parameters that need accurate measurement and prioritize field measurements only for those critical parameters, rather than requiring comprehensive field measurements for all input data, thereby reducing overall measurement time and cost.
Solution Approach 2:
The patent creates a computational copy of the physical reservoir system through numerical simulations. By using this virtual model to reproduce and analyze interface behavior, the system can infer missing or uncertain interface characteristics without requiring additional physical field measurements, thus saving time and resources while maintaining measurement precision for critical parameters.
3Manufacturing precision
If automated optimization methods are used to infer input data, then the accuracy of simulated interface is improved, but the computational complexity increases
Solution Approach 1:
The patent replaces manual trial-and-error adjustment mechanisms with an automated optimization algorithm. The objective function serves as a mathematical substitute for human judgment, automatically evaluating the quality of input parameters and guiding the optimization process. This substitution achieves high accuracy through systematic computational methods while maintaining reasonable automation levels by focusing optimization on key control parameters rather than all input data.
Data Source
AI summary
Methods and systems, including computer programs encoded on a computer storage medium are described for generating data used to simulate properties of a target area in a subterranean region. A system obtains data describing a water-hydrocarbon interface of the target area and selects control parameters for processing the data based on a formation mechanism corresponding to a pressure or temperature of the target area. An objective function is determined that measures a delta between first values observed in the data and second values from simulations of the target area. Based on outputs of the function, the system calculates values for the control parameters that minimizes the delta between the first values observed in the data and the second values from simulations of the target area. The input data is generated to simulate properties of the target area based on calculated values of the control parameters.