Fault Grouping ML for Scalable Subsurface Modeling Precision

Solution Overview

Problem

Modeling subsurface volumes with faults and fault interactions is challenging due to the complexity of fault geometry and requires significant manual effort, making automated modeling difficult.

Innovation Solution

A computer-implemented machine learning method that automatically partitions faults into groups based on descriptors like azimuth, dip, area, and orientation, applies a trained model to generate modeling parameter values, and models fault-related properties using these values.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Manufacturing precision

If manual fault modeling is performed, then modeling precision is improved, but productivity deteriorates

Engineering Contradiction:

Improvemodeling precisionVSAvoidproductivity

Core Design Contradiction:

Manufacturing precisionVSProductivity

Solution Approach 1:

The system performs automated fault modeling by having the computational system itself execute the modeling tasks without requiring manual intervention for each fault. The machine learning model automatically generates fault models based on input data, enabling the system to serve itself in the modeling process while maintaining precision through algorithmic consistency.

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The patent replaces manual mechanical modeling operations with an automated machine learning-based system. The machine learning model substitutes the manual mechanical process of fault modeling, allowing automated generation of fault models while maintaining the precision that previously required human expertise through systematic algorithmic approaches.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Productivity

If automated fault modeling is implemented, then productivity is improved, but modeling precision deteriorates

Engineering Contradiction:

ImproveproductivityVSAvoidmodeling precision

Core Design Contradiction:

ProductivityVSManufacturing precision

Solution Approach 1:

The patent replaces manual mechanical modeling operations with an automated machine learning-based system. The machine learning model substitutes the manual mechanical process of fault modeling, allowing automated generation of fault models while maintaining the precision that previously required human expertise through systematic algorithmic approaches.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system changes the parameters and approach by using machine learning models with multiple descriptors (azimuth, dip, area, orientation, eigenvalue) to characterize faults. This parametric approach allows automated processing while maintaining precision through comprehensive characterization of fault geometry and properties.

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If fault grouping is applied, then device complexity is reduced, but measurement precision deteriorates

Engineering Contradiction:

Improvedevice complexityVSAvoidmeasurement precision

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments faults into distinct groups based on their geometric and structural characteristics such as azimuth, dip, area, orientation, and eigenvalue. This segmentation organizes the complex fault system into manageable categories while preserving the unique characteristics of each fault group through multiple descriptors, reducing overall system complexity without sacrificing measurement precision.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The system changes the parameters and approach by using machine learning models with multiple descriptors (azimuth, dip, area, orientation, eigenvalue) to characterize faults. This parametric approach allows automated processing while maintaining precision through comprehensive characterization of fault geometry and properties.

Inventive Principle:

Principle #35Parameter changes

Data Source

PatentUS20260086257A1Automated machine learning fault modeling with grouping

Publication Date: 2026.03.26 SCHLUMBERGER TECH CORP

AI summary

Methods, computing systems, and computer-readable media for a machine learning method of modeling fault-related properties of a geological region are presented. The techniques include: obtaining seismic geological data for a geological region; obtaining from a user identifications of a plurality of faults in the geological region; automatically generating values for descriptors of respective faults of the plurality of faults; automatically partitioning faults of the plurality of faults into a plurality of groups according to the values for the descriptors; obtaining a mapping of respective groups of the plurality of groups to modeling parameter values; applying the mapping to a fault in the geological region outside of the plurality of faults to obtain a modeling parameter value for the fault outside of the plurality of faults; and modeling a fault-related property of the geological region based on the modeling parameter value for the fault outside of the plurality of faults.