Subsurface Structure Identification With CNN Seismic Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Interpretation of large seismic volumes is a time-consuming and expensive task, especially with increasing data collection and improved acquisition techniques, necessitating automated seismic interpretation systems.

Innovation Solution

A seismic interpretation system using machine learning techniques, specifically convolutional neural networks, to automate geo-body segmentation by training on labeled seismic data and accelerating the training process through parallel processing and hyperparameter variation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional manual seismic interpretation methods are used, then interpretation accuracy can be maintained, but the time required for geo-body segmentation increases significantly

Engineering Contradiction:
Improveinterpretation accuracyVSAvoidtime required for geo-body segmentation
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent replaces manual mechanical interpretation processes with an automated computer-based system that uses machine learning algorithms (specifically convolutional neural networks) to perform seismic data analysis and geo-body segmentation, thereby eliminating the time-consuming manual review process while maintaining interpretation quality

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service interpretation by training the neural network on labeled seismic data, allowing the computer to automatically learn and perform segmentation tasks without requiring continuous human intervention or manual review of each interpretation result

Inventive Principle:
Principle #25Self-service

2Measurement precision

If more comprehensive seismic data is collected, then subsurface analysis quality improves, but processing time and computational resources increase

Engineering Contradiction:
Improvesubsurface analysis qualityVSAvoidprocessing speed
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The system performs preliminary actions by pre-training the neural network on large volumes of labeled seismic data before actual interpretation tasks, enabling the model to quickly process new comprehensive seismic datasets without requiring proportional increases in processing time during actual operations

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent changes the processing parameters by transitioning from traditional sequential processing methods to parallel neural network processing, fundamentally altering how computational resources are utilized to handle large seismic datasets more efficiently

Inventive Principle:
Principle #35Parameter changes

3Productivity

If automated interpretation systems are implemented, then processing speed increases, but system complexity increases

Engineering Contradiction:
Improveprocessing speedVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent applies universality by designing a multi-functional neural network system that can handle various seismic interpretation tasks (segmentation, feature extraction, analysis) through a single trained model, reducing the need for multiple specialized systems and thereby managing complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

Data Source

PatentEP3966599B1System and method for identifying subsurface structures
Publication Date: 2025.10.01 SERVICES PETROLIERS SCHLUMBERGER SA
  • EP3966599B1 patent drawingFigure 1
  • EP3966599B1 patent drawingFigure 2
  • EP3966599B1 patent drawingFigure 3

AI summary

A subsurface structure identification system includes one or more processors and a memory coupled to the one or more processors. The memory is encoded with instructions that when executed by the one or more processors cause the one or more processors to provide a convolutional neural network trained to identify a subsurface structure in an input migrated seismic volume, and to partition the input migrated seismic volume into multi-dimensional sub-volumes of seismic data. The instructions also cause the one or more processors to process each of the multi-dimensional sub-volumes of seismic data in the convolutional neural network, and identify the subsurface structure in the input migrated seismic volume based on a probability map of the input migrated seismic volume generated by the convolutional neural network.