Resistivity Image Neural Networks for Real-Time Horizon Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing drilling systems face inefficiencies and inaccuracies in determining subsurface geological feature horizons, relying heavily on expert interpretation and struggling to effectively utilize seismic and resistivity data, leading to inconsistent and delayed boundary detection.

Innovation Solution

A horizon mapping system using a resistivity image mapping neural network generates horizon maps and augmented resistivity images in real-time, leveraging deep-learning methods and synthetic training data to predict accurate reservoir boundaries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If expert interpretation methods are used to determine subsurface horizons, then measurement precision may be improved, but productivity decreases due to labor-intensive processes

Engineering Contradiction:

Improvehorizon boundary detection accuracyVSAvoidboundary detection speed

Core Design Contradiction:

Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual expert interpretation with an automated deep learning system that processes resistivity images to identify horizon boundaries. The neural network model automatically detects subsurface features without requiring human experts to manually analyze seismic data, thereby maintaining measurement precision while dramatically improving productivity through automation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The system enables self-service by allowing the deep learning model to independently perform horizon identification without continuous human intervention. The automated pipeline processes resistivity images, generates horizon maps, and identifies boundaries autonomously, freeing experts from routine analysis tasks while maintaining consistent detection quality.

Inventive Principle:

Principle #25Self-service

2Reliability

If traditional seismic data analysis methods are used, then reliability of horizon identification may be maintained, but loss of time increases due to delayed boundary detection

Engineering Contradiction:

Improvehorizon identification consistencyVSAvoidboundary detection delay

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The system performs preliminary action by pre-training deep learning models on extensive seismic data before deployment. The models are prepared in advance with learned features and patterns, enabling them to quickly and reliably identify horizons in real-time applications without requiring time-consuming manual analysis during actual subsurface exploration.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The automated deep learning system enables continuous processing of resistivity images to generate horizon maps without interruption. Unlike batch processing by experts, the system maintains continuous operation, constantly analyzing incoming data and updating horizon boundaries in real-time, thereby eliminating detection delays while maintaining reliable identification.

Inventive Principle:

Principle #20Continuity of useful action

3Adaptability or versatility

If manual interpretation methods are used, then adaptability to complex geological features may be improved, but device complexity decreases due to lack of automated processing systems

Engineering Contradiction:

Improvehandling of complex geological featuresVSAvoidautomated processing system complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The system adapts to complex geological features by adjusting parameters within the deep learning model, such as training data composition, network architecture configurations, and loss function weights. By modifying these parameters, the system can handle diverse subsurface conditions including fault zones, folded structures, and varying rock types, while the underlying processing framework remains consistent.

Inventive Principle:

Principle #35Parameter changes

Solution Approach 2:

The deep learning system achieves universality by designing a single automated processing framework that can handle multiple types of geological features and resistivity image formats. The same neural network architecture processes various subsurface conditions through learned feature extraction, eliminating the need for separate manual interpretation methods for different geological scenarios while maintaining high adaptability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS20250225300A1Identifying subsurface horizons of geosphere sections automatically using deep-learning models

Publication Date: 2025.07.10 SCHLUMBERGER TECH CORP

AI summary

A system and method for determining a subsurface horizon in a drilling system that include generating a resistivity change interface using a resistivity image mapping neural network that determines horizon boundaries of geological features by encoding resistivity images of subsurface feature sections into feature vectors and decoding the feature vectors into the horizon boundaries. The system and method also include generating an augmented resistivity image based on the resistivity change interface and a resistivity image. The system and method further include providing the augmented resistivity image for display on a computing device to indicate a horizon boundary.