Machine Learning Formation Property Volume Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining subsurface formation properties, such as porosity and permeability, face challenges in areas where logging tool measurements are unavailable, particularly at interwell locations outside the vicinity of drilled wells.
Innovation Solution
A method using a computer processor to obtain well log data from various wells, assign subsets based on geological attributes, and apply machine-learning algorithms for intrawell interpolation and interwell extrapolation to generate formation property volumes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If logging tool measurements are used to determine subsurface formation properties, then measurement precision is improved, but coverage is limited to areas where wells are drilled
Solution Approach 1:
The patent uses machine learning models as intermediaries to transfer formation property information from logged wells to unlogged locations. The ML models are trained on available logging data and then applied to predict properties in areas without direct measurements, effectively mediating between measured and unmeasured spaces
Solution Approach 2:
The patent creates copies of formation property patterns learned from logged wells and applies them to unlogged locations through machine learning. The models learn typical formation property relationships from measured data and replicate these patterns in areas without direct measurements
2Area of stationary object
If machine learning algorithms are applied to interpolate and extrapolate well log data, then coverage area is improved, but device complexity increases
Solution Approach 1:
The patent segments the problem into distinct processing stages: data acquisition from multiple wells, grouping similar well logs together, applying machine learning models to grouped data, and generating formation property volumes. This segmentation makes the complex overall task more manageable and implementable
Solution Approach 2:
The patent develops universal machine learning models that can handle multiple logging types and be applied across different wells and locations. These multi-functional models reduce the need for separate specialized systems for each logging type or location
Data Source
AI summary
A method may include obtaining well log data for various wells regarding a geological region of interest. The well log data may correspond to various well logs with different logging types. The method may include assigning, using a grouping algorithm, subsets of the well log data to various groups based on one or more geological attributes. The method may include determining, using the groups and a machine-learning algorithm, various well zones for different portions of a respective well among the wells. The method may include determining interpolated log data using the well log data, the well zones, and an intrawell interpolation process. The method may include generating a formation property volume based on the interpolated log data and the well log data.