Continuous Grain Size Logs from Thin Section Images
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating grain size in hydrocarbon reservoirs are non-representative and lack accuracy due to their reliance on samples from specific locations, failing to provide a comprehensive understanding of the subsurface material.
Innovation Solution
A method involving the extraction of core samples from multiple depths, creation of structured data sets combining wireline logs and discrete grain sizes, and the use of machine learning models, such as artificial neural networks, to generate continuous grain size logs across a well, enabling more accurate and representative grain size determination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If grain size is estimated from thin section samples of core plugs, then discrete grain size measurements can be obtained, but the results are non-representative of the reservoir and lack accuracy
Solution Approach 1:
The patent segments the reservoir into multiple depth intervals and obtains core samples from multiple locations (first well and second well) at different depths. By dividing the reservoir assessment into discrete depth segments and combining results, the method achieves both localized measurement precision and overall reservoir representativeness.
Solution Approach 2:
The patent introduces grain size logs as an intermediary between discrete thin section analysis and continuous reservoir characterization. The grain size logs serve as a bridge that translates pore-scale thin section measurements into well-scale continuous data, enabling representative reservoir assessment while maintaining measurement accuracy.
2Ease of manufacture
If core samples are extracted from specific locations in a well, then grain size analysis can be performed, but the results cannot be extended to represent the entire reservoir
Solution Approach 1:
The patent creates a universal grain size log that can be applied across multiple wells and depth intervals. By developing a methodology that works across different locations (first well and second well) and depths, the approach achieves universality, allowing grain size characterization to be extended from specific sample locations to the entire reservoir.
Solution Approach 2:
The patent uses grain size logs as a copy mechanism that replicates thin section analysis results across continuous depth intervals and multiple wells. Instead of performing physical thin section analysis at every location, the methodology creates continuous grain size logs that copy and extend the discrete measurements, making the results applicable to the entire reservoir.
3Reliability
If continuous grain size logs are generated using machine learning models, then representative reservoir characterization is achieved, but complex data processing and modeling are required
Solution Approach 1:
The patent employs machine learning models that automatically process and integrate data from multiple sources (wireline logs, grain size logs, core samples) without requiring manual intervention. The system self-services by autonomously generating continuous grain size logs from the input data, reducing the complexity burden on operators while maintaining high reliability.
Solution Approach 2:
The patent replaces manual data processing and interpretation methods with automated machine learning models. Instead of mechanically processing data through multiple manual steps, the system uses computational algorithms to automatically integrate wireline logs, core sample data, and generate continuous grain size logs, significantly reducing processing complexity.
Data Source
AI summary
Systems and methods for determining a continuous grain size log from a collection of petrographic thin section images are provided. Thin section images from core samples from one or more wells may be obtained and analyzed to estimate grain sizes. Using wireline logs from the one or more wells and the estimated grain sizes, a data structure (for example, a database) of grain sizes and wireline logs at depths may be constructed. The data structure may be used to train a machine learning model. Next, a wireline tool may be used to obtain wireline logs in a new well, and a continuous grain size log may be determined from the wireline logs of using the machine learning model. Computer-readable media for determining reservoir rock grain sizes from a collection of petrographic thin section images is also provided.