Subsurface Formation Declustering for Unbiased Reservoir Statistics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for characterizing subsurface geological formations are biased due to nonuniform and unrepresentative measurement data collection, leading to skewed global statistics and inaccurate modeling of reservoir properties.
Innovation Solution
A method that defines a distance of influence based on the property and heterogeneity of the formation to identify clusters, calculates an average value for each cluster, and replaces multiple values with this average, thereby reducing bias and improving summary statistics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If measurements are collected at multiple locations to improve representation, then measurement precision improves, but data clustering occurs causing bias in global statistics
Solution Approach 1:
The patent applies local quality by defining a distance of influence that varies with location and property type. Each measurement location has a customized influence radius based on local heterogeneity, allowing the system to account for spatial variability while avoiding overrepresentation of clustered measurements. This resolves the contradiction by making the representation accuracy location-dependent rather than uniform.
Solution Approach 2:
The patent changes the parameter of distance of influence based on the measured property and local heterogeneity. By dynamically adjusting the distance parameter rather than using a fixed value, the system can accommodate different spatial scales of geological variability. This allows measurements to be weighted appropriately based on their spatial context, improving global statistics without requiring uniform sampling density.
2Measurement precision
If the distance of influence is increased to reduce clustering bias, then summary statistics become more accurate, but locations that are actually distinct may be incorrectly grouped
Solution Approach 1:
The patent uses local quality by defining distance of influence that varies with local heterogeneity characteristics. In homogeneous regions, a larger distance can be used to group measurements, while in heterogeneous regions, a smaller distance prevents incorrect grouping. This location-specific approach resolves the contradiction between reducing clustering bias and maintaining accurate cluster boundaries.
Solution Approach 2:
The patent applies dynamics by making the distance of influence a variable parameter rather than a fixed value. The distance adapts based on the measured property type and local geological heterogeneity, allowing the system to optimize cluster identification for each specific context. This dynamic adjustment resolves the contradiction by enabling the distance to increase where appropriate while remaining constrained where geological variability requires finer resolution.
Data Source
AI summary
Systems and methods for characterizing a subsurface formation include: measuring a property of the subsurface formation at a plurality of locations; storing values of the property of the subsurface formation at the plurality of locations; defining a distance of influence for the property in the subsurface formation based on the property and on a degree of heterogeneity of the subsurface formation; identifying clusters of locations that are located within the distance of influence of each other; for each cluster identified, calculating an average value of the property of the subsurface formation in the cluster; for each cluster identified, replacing multiple values of the property and locations of the identified cluster with the average value of the property of the subsurface formation in the cluster at a single location; and calculating summary statistics of the reduced data set.