Graph-Based Light Hydrocarbon Correlations for Reservoir Complexity
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional geochemical analysis of hydrocarbon samples for reservoir characterization is qualitative and inconsistent, leading to poor reservoir characterization and inefficient wellbore planning due to reliance on subjective expert comparisons.
Innovation Solution
A method and system that constructs a graph representation of physical samples based on spatial-temporal data, applies graph-based algorithms to determine clusters, and uses these clusters to plan wellbore paths for hydrocarbon reservoirs, incorporating automated and quantitative reservoir complexity analysis.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional qualitative geochemical analysis is used, then expert subjectivity is involved, but consistency and reliability of reservoir characterization deteriorate
Solution Approach 1:
The patent replaces the mechanical system of human expert comparison with an automated computational system. Graph-based algorithms automatically process geochemical data, construct graphs representing sample relationships, and identify clusters without human intervention, thereby eliminating subjectivity and improving consistency.
Solution Approach 2:
The system performs self-service by automatically analyzing geochemical data and generating reservoir characterization results. The graph-based clustering algorithm independently processes samples, determines relationships between them, and produces objective conclusions without requiring external expert judgment.
2Measurement precision
If subjective expert comparisons are used, then flexibility in analysis is maintained, but measurement precision and objectivity deteriorate
Solution Approach 1:
The patent segments the complex geochemical data into discrete graph nodes representing individual samples. Relationships between samples are represented as edges, allowing the system to process complex multivariate geochemical data through systematic graph construction and clustering algorithms, thereby achieving precise measurements.
Solution Approach 2:
The graph representation serves as an intermediary structure between raw geochemical data and reservoir characterization conclusions. This intermediate graph model organizes complex sample relationships in a way that enables precise, objective analysis while managing system complexity through structured data representation.
3Productivity
If qualitative analysis methods are used, then ease of operation is maintained, but productivity and efficiency of wellbore planning deteriorate
Solution Approach 1:
The patent replaces manual expert analysis with automated graph-based algorithms that efficiently process geochemical data and generate wellbore planning recommendations. This substitution dramatically improves productivity by eliminating time-consuming subjective comparisons while maintaining operational simplicity through user-friendly interfaces.
Data Source
AI summary
A method for, at least, forming a graph representation of physical samples and identifying clusters of the physical samples. The method includes obtaining a plurality of physical samples from one or more wells, where each sample is associated with spatial-temporal data, and determining geochemical data for each sample. The method further includes constructing a graph representation of the plurality of physical samples, where in the graph representation each sample is represented as a node and a presence of an edge between any two nodes is based on the spatial-temporal data of the plurality of physical samples. The method further includes determining a weight for every edge in the graph representation and processing the graph representation with a graph-based algorithm to determine one or more clusters. The method may further include determining a wellbore plan that comprises a planned wellbore path to penetrate a hydrocarbon reservoir based on the determined clusters.