Seismic Surface Classification via Machine Learning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for classifying seismic surfaces in geophysical prospecting are largely manual and lack automation, especially in differentiating between sequence boundaries and flooding surfaces, which are crucial for hydrocarbon exploration but require human interpretation and are time-consuming and prone to inconsistencies.
Innovation Solution
A method using a computer-based approach to classify seismic surfaces or surface patches by learning a classification model from labeled training data, applying seismic attributes, and quantifying confidence in classification, allowing for the differentiation between sequence boundaries, flooding surfaces, and their subtypes like maximum flooding and transgressive flooding surfaces.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual classification methods are used for seismic surfaces, then interpretation accuracy can be maintained through human expertise, but the process becomes time-consuming and prone to inconsistencies
Solution Approach 1:
The system performs preliminary classification of seismic surfaces using trained machine learning models before final interpretation. Training data is preprocessed and models are trained in advance, allowing rapid automated classification during actual operation while maintaining consistency through predefined classification criteria
Solution Approach 2:
A machine learning classification model acts as an intermediary between raw seismic data and human interpretation. The model processes seismic attributes and surface characteristics, providing consistent automated classifications that reduce human variability while maintaining accuracy through trained algorithms
2Productivity
If automated classification methods are implemented, then processing efficiency increases, but differentiation between sequence boundaries and flooding surfaces may lack the nuance of human interpretation
Solution Approach 1:
The system incorporates feedback mechanisms where classification results are evaluated and used to refine model performance. Confidence scores are generated for each classification, allowing the system to identify uncertain cases that may require human review, thereby maintaining reliability while improving efficiency
Solution Approach 2:
The classification model uses multiple seismic attributes and parameters (amplitude, frequency, continuity, geometry, termination patterns) to differentiate between surface types. By analyzing changes in these parameters across different contexts, the model achieves reliable differentiation between sequence boundaries and flooding surfaces
3Measurement precision
If comprehensive seismic attributes are analyzed for classification, then classification accuracy improves, but computational complexity and data processing requirements increase
Solution Approach 1:
The system extracts only the most relevant seismic attributes and features needed for classification from the full seismic dataset. By selecting key parameters (termination patterns, surface geometry, amplitude characteristics) rather than processing all available data, the system maintains high classification precision while reducing computational complexity
Data Source
AI summary
A method to classify one or more seismic surfaces or surface patches based on measurements from seismic data, including: obtaining, by a computer, a training set including a plurality of previously obtained and labeled seismic surfaces or surface patches and one or more training seismic attributes measured or calculated at, above, and/or below the seismic surfaces; obtaining, by the computer, one or more unclassified seismic surfaces or surface patches and one or more seismic attributes measured or calculated at, above, and/or below the unclassified seismic surfaces; learning, by the computer, a classification model from the previously obtained and labeled seismic surfaces or surface patches and the one or more training seismic attributes; and classifying, by the computer, the unclassified seismic surfaces or surface patches based on a comparison between the classification model and the unclassified seismic surfaces or surface patches.