Proxy Model for Fast Oil and Gas Field Production Model Updating
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current oil and gas field production management is hindered by the time-consuming process of physically evaluating and managing individual wells, leading to outdated data for real-time decision-making, and the complexity of physics-based models which are cumbersome and slow to execute, making it difficult to respond to production issues promptly.
Innovation Solution
Implementing a method that uses physical and proxy simulators to rapidly update oil and gas field production models by establishing a base model, defining uncertain parameters, and using a decision management system with a proxy model to automatically generate and validate parameter ranges for real-time optimization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If physics-based simulators are used to model oil and gas field production, then model accuracy and reliability are improved, but model execution time and complexity increase significantly
Solution Approach 1:
The patent creates a proxy model that replicates the behavior of complex physics-based simulators. The proxy model is trained using data from multiple simulator runs with different uncertain parameters, then uses this trained model to rapidly predict production outcomes without re-running the full physics-based simulations, thus maintaining accuracy while reducing execution time
Solution Approach 2:
The patent performs preliminary actions by pre-executing physics-based simulators with various uncertain parameter combinations to generate training data before actual production decisions are made. This pre-computed data is stored and used to train the proxy model, so that when real-time decisions are needed, the system can quickly query pre-computed results rather than running full simulations
2Reliability
If physics-based simulators are used for production optimization, then production decisions are based on accurate physical models, but the tuning and history matching process becomes complex and time-consuming
Solution Approach 1:
The patent extracts the complex tuning and history matching process from manual expert operations and transforms it into an automated computational workflow. The system automatically executes multiple simulator runs with varying uncertain parameters, collects results, trains a proxy model, and identifies optimal parameter ranges without requiring manual intervention from reservoir engineers
Solution Approach 2:
The system performs self-service by automatically conducting the history matching and parameter calibration processes that traditionally required expert engineers. The proxy model training and validation processes autonomously adjust uncertain parameters to match historical production data, eliminating the need for manual model tuning
3Measurement precision
If continuous monitoring sensors are installed to provide real-time data, then data availability and measurement frequency are improved, but data analysis complexity and response time requirements increase
Solution Approach 1:
The patent introduces the proxy model as an intermediary between continuous monitoring sensors and production decision-making. The proxy model processes and interprets the continuous data stream, translating raw sensor measurements into actionable production recommendations, thus simplifying the analysis complexity while maintaining real-time responsiveness
4Reliability
If multiple physics-based simulators are run to account for uncertain parameters, then production predictions become more comprehensive, but computational resources and execution time become impractical
Solution Approach 1:
The patent applies partial action by running physics-based simulators with a strategically selected subset of uncertain parameter combinations rather than exhaustively testing all possible parameter values. The experimental design process identifies the most influential parameters and their optimal ranges, allowing the system to achieve sufficient prediction accuracy with fewer simulator runs
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods, systems, and computer readable media are provided for fast updating of oil and gas field production optimization using physical and proxy simulators. A base model (30) of a reservoir (100), well (100), or a pipeline network (100) is established in one or more physical simulators (26). A decision management system (24) is used to define uncertain parameters for matching with observed data (114). A proxy model is used to fit the uncertain parameters to outputs of the physical simulators (26), determine sensitivities of the uncertain parameters, and compute correlations between the uncertain parameters and output data from the physical simulators (26). Parameters for which the sensitivities are below a threshold are eliminated. The decision management system (24) validates parameters which are output from the proxy model in the simulators (26). The validated parameters are used to make production decisions.