Shale Gas Production Forecasting via Surrogate Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The predictability of shale gas production is hindered by a lack of understanding of controlling factors and limitations in characterizing shale gas formations and available tools, making it difficult to assess the commercial value of projects, especially in areas with limited data availability.
Innovation Solution
A method involving statistical analysis and simulation modeling to generate history-matched reservoir models, which can forecast production in new exploration areas by using data from well-characterized formations, incorporating various decline curves and sensitivity analysis to identify key parameters affecting production.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional characterization tools and methods are used for shale gas formations, then the understanding of controlling factors is limited, but the predictability of production is improved
Solution Approach 1:
The patent creates surrogate models that copy the production behavior of well-characterized shale gas formations. These surrogate models are developed by matching simulation models to historical production data from analogous formations, then using them to predict production in new formations with limited data. This allows the system to leverage knowledge from well-understood formations to improve predictability in data-scarce environments.
Solution Approach 2:
The patent performs sensitivity analysis to identify key parameters that control production behavior in shale gas formations. By determining which parameters have the greatest impact on production (such as permeability, porosity, fracture characteristics), the system focuses characterization efforts on measuring these critical parameters accurately, thereby improving predictability without requiring complete understanding of all formation properties.
2Measurement precision
If statistical analysis and simulation modeling are performed to generate history-matched models, then the forecasting accuracy is improved, but the complexity of the method increases
Solution Approach 1:
The patent performs preliminary statistical analysis and sensitivity studies on well-characterized formations to develop surrogate models before applying them to new exploration areas. By pre-processing the data and establishing the relationships between parameters and production behavior in advance, the system reduces the complexity of analysis needed when forecasting for new formations, as the surrogate models can be directly applied with minimal additional processing.
3Reliability
If data from well-characterized formations is used to forecast production in new exploration areas, then the commercial value assessment is improved, but the data availability in new areas remains limited
Solution Approach 1:
The patent introduces surrogate models as intermediaries between well-characterized formations and new exploration areas. These surrogate models act as mediators that translate production behavior and parameter relationships from data-rich formations into predictive tools for data-poor formations. By using these intermediary models, the system can assess commercial value in new areas without requiring extensive local data, as the surrogate models bridge the information gap.
Data Source
AI summary
A method can include providing data for at least one shale gas formation; performing a statistical analysis on the data for each of the at least one shale gas formation; providing a simulation model; history matching the simulation model for each of the at least one shale gas formation based at least in part on the performed statistical analysis to generate a history matched model for each of the at least one shale gas formation; and forecasting production for another shale gas formation by plugging in data for the other shale gas formation into each generated history matched model. Various other apparatuses, systems, methods, etc., are also disclosed.