Machine Learning Model for Seismic Data Geological Body Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current hydrocarbon exploration methods face challenges in accurately interpreting seismic data to identify geological bodies and assess stability for safe hydrocarbon extraction, leading to potential wellbore failures and reduced production.

Innovation Solution

A machine-learning model is applied to seismic data to classify geological bodies such as salt, oil, and gas bodies, using a supervised training method with labeled datasets, and generates outputs that assist in determining the suitability of a target area for hydrocarbon extraction, with additional processing operations like image analysis and 3D simulations to visualize and predict extraction challenges.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If traditional seismic data interpretation methods are used, then the process is simpler and more straightforward, but the accuracy of identifying geological bodies and assessing stability is insufficient

Engineering Contradiction:

Improveaccuracy of identifying geological bodiesVSAvoidcomplexity of interpretation system

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent replaces traditional mechanical/manual seismic data interpretation methods with a machine learning-based automated system. The machine learning model processes seismic data to identify geological bodies and assess stability, substituting human expert analysis with an automated computational approach that provides higher accuracy and consistency.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent introduces a machine learning model as an intermediary between raw seismic data and geological interpretation. This intermediary layer processes the seismic data through trained algorithms, extracting features and patterns that are not easily detectable through traditional methods, thereby improving identification accuracy of geological bodies.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If traditional interpretation methods are used, then the system is easier to operate, but the reliability of hydrocarbon extraction operations is reduced due to potential wellbore failures

Engineering Contradiction:

Improvereliability of hydrocarbon extraction operationsVSAvoidease of seismic data interpretation

Core Design Contradiction:

ReliabilityVSEase of operation

Solution Approach 1:

The patent replaces manual interpretation processes with an automated machine learning system that consistently applies trained models to seismic data. This substitution eliminates human error and variability, providing more reliable and consistent results for identifying geological bodies and assessing operational stability, thereby reducing wellbore failures.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The machine learning model incorporates feedback mechanisms where it processes seismic data, generates predictions about geological bodies and stability, and can be retrained or refined based on actual outcomes. This feedback loop continuously improves the reliability of the interpretation system, making it more accurate over time while maintaining ease of operation.

Inventive Principle:

Principle #23Feedback

3Productivity

If more detailed analysis and processing operations are applied, then the productivity of hydrocarbon exploration improves, but the loss of time for data processing increases

Engineering Contradiction:

Improveproductivity of hydrocarbon explorationVSAvoidtime for seismic data processing

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent applies preliminary action by pre-training the machine learning model on extensive seismic data before actual exploration. The model learns patterns and features in advance, so when real seismic data needs to be analyzed, the processing is much faster. This preliminary training phase enables rapid, accurate analysis during actual hydrocarbon exploration operations.

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

The patent replaces time-consuming manual analysis methods with automated machine learning processing. The machine learning system can analyze large volumes of seismic data much faster than human experts, significantly reducing processing time while maintaining or improving analysis quality, thereby enhancing overall exploration productivity.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11428835B2Facilitating hydrocarbon exploration and extraction by applying a machine-learning model to seismic data

Publication Date: 2022.08.30 LANDMARK GRAPHICS CORP

AI summary

Hydrocarbon exploration and extraction can be facilitated using machine-learning models. For example, a system described herein can receive seismic data indicating locations of geological bodies in a target area of a subterranean formation. The system can provide the seismic data as input to a trained machine-learning model for determining whether the target area of the subterranean formation includes one or more types of geological bodies. The system can receive an output from the trained machine-learning model indicating whether or not the target area of the subterranean formation includes the one or more types of geological bodies. The system can then execute one or more processing operations for facilitating hydrocarbon exploration or extraction based on the seismic data and the output from the trained machine-learning model.