Physics-Aware Aquifer Selection Models for Carbon Storage
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current approaches for selecting aquifers for carbon storage are inefficient due to the lack of consideration for complex physics, such as two-phase flow and solubility of carbon dioxide in formation brine, and require extensive computing resources and time.
Innovation Solution
A system utilizing a computer system and model manager that generates simulation data and auxiliary quantities using numerical simulation models and physics-based equations, training a machine learning model to select aquifers that meet carbon storage requirements.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If numerical simulation models and physics-based equations are used to select aquifers, then measurement precision and reliability improve, but productivity deteriorates due to extensive computing resources and time required
Solution Approach 1:
The patent pre-generates simulation data using numerical simulation models and physics-based equations during an offline training phase. This preliminary action creates a trained machine learning model that can perform rapid aquifer selection without requiring real-time computational resources, thus resolving the contradiction between high measurement precision and productivity.
Solution Approach 2:
The patent creates a simplified copy of the complex physics-based selection process through machine learning model training. The trained model replicates the behavior of full numerical simulations but executes much faster, maintaining accuracy while dramatically improving computational efficiency for actual aquifer selection tasks.
2Measurement precision
If complex physics considerations are included in aquifer selection, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent replaces the complex mechanical/computational system of numerical simulation models and physics-based equations with a machine learning model. This substitution maintains the ability to consider complex physics (through the training data) while simplifying the actual selection process to a straightforward model inference, thus reducing system complexity.
3Measurement precision
If manual updates and computations are performed for aquifer selection, then measurement precision improves, but loss of time increases
Solution Approach 1:
The patent performs all time-consuming manual computations and simulations in advance during the model training phase. The trained machine learning model then provides rapid aquifer selection results without requiring manual updates or computations, significantly reducing time loss while maintaining precision.
Solution Approach 2:
The machine learning model serves itself by automatically performing aquifer selection based on input parameters without requiring manual intervention, updates, or computations. This self-service capability eliminates time loss associated with manual processes while maintaining measurement precision.
Data Source
AI summary
An aquifer management system is provided. The aquifer management system includes a computer system and a model manager. The model manager generates simulation data for simulation parameters for aquifers using a set of numerical simulation models with historical data for aquifer parameters for the aquifers as inputs to the set of numerical simulation models. The model manager generates auxiliary quantities for auxiliary parameters for the aquifers using a set of physics-based equations and the simulation data for a set of selected simulation parameters. The model manager creates a training dataset using the simulation data and the auxiliary quantities. The model manager trains a machine learning model using the training dataset to select a set of aquifers that meet a set of carbon storage requirements to store carbon dioxide.