Subsurface Model Regression for Hydrocarbon Decision Accuracy
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current subsurface modeling techniques for hydrocarbon operations are inefficient and cumbersome, particularly in history matching, which relies on noisy production data and often results in underdetermined reservoir models, limiting decision-making capabilities and increasing the time required for hydrocarbon exploration, development, and production.
Innovation Solution
A method and system that utilize regression and classification techniques in subsurface models to evaluate hydrocarbon operations by creating multiple reservoir models based on initial data sets, simulating scenarios with and without specific operations, and transforming data into a feature space to determine the desirability of hydrocarbon operations such as adding new wells, without the need for extensive history matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If history matching is used to condition reservoir models to production data, then the models are adjusted to match measured data, but the process becomes computationally intensive and time-consuming
Solution Approach 1:
The patent pre-computes and stores response data from multiple reservoir models in a database before actual history matching is needed. This preliminary action creates a library of pre-simulated responses that can be quickly compared against actual production data, eliminating the need for time-consuming iterative simulations during the actual history matching process.
Solution Approach 2:
The patent creates synthetic copies of reservoir model responses through pre-simulation and stores them in a database. These synthetic response copies represent various possible reservoir behaviors and can be rapidly matched against actual data without re-running complex simulations, significantly reducing computational time while maintaining accuracy.
2Loss of information
If extensive data assimilation is performed through history matching, then production data is integrated into the model, but the process becomes cumbersome and limits decision-making capabilities
Solution Approach 1:
The patent extracts only the essential response characteristics from complex reservoir simulations and stores them in a simplified database format. By taking out only the critical response data needed for history matching rather than managing complete complex models, the system reduces process complexity while maintaining the ability to integrate production data effectively.
Solution Approach 2:
The patent introduces a response database as an intermediary layer between reservoir models and production data. This intermediary structure organizes and pre-processes model responses, making data integration more systematic and less cumbersome while preserving comprehensive information for decision-making.
3Measurement precision
If multiple reservoir models are created and simulated to evaluate hydrocarbon operations, then decision accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent pre-simulates and stores responses from multiple reservoir models in advance, creating a comprehensive library of possible outcomes. This preliminary computation of multiple scenarios allows for accurate decision evaluation when needed, while the actual decision-making process benefits from having pre-computed results rather than performing complex simulations in real-time.
Data Source
Figure 1
Figure 2
Figure 3A~3B
AI summary
A method and system are described for hydrocarbon exploration, development and production. The method relates to performing regression and/or classification in subsurface models to support decision making for hydrocarbon operations. The evaluation may then be used in performing hydrocarbon operations, such as hydrocarbon exploration, hydrocarbon development and/or hydrocarbon production.