Representative Seismic Line Generation for Efficient Database Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The petroleum industry faces challenges in efficiently processing and analyzing large seismic datasets, which are time-consuming and prone to missing representative seismic images, hindering timely decision-making and accurate geological interpretations.
Innovation Solution
A method and system that generate representative seismic lines by dividing seismic data into tiles, computing feature vectors, and clustering using eigenvalues and eigenvectors, allowing for efficient comparison and search for analogous geological regions within seismic databases, thereby accelerating data analysis and improving machine learning model training.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the entire seismic dataset is processed to ensure comprehensive analysis, then the accuracy of geological interpretations is improved, but the time required for processing increases significantly
Solution Approach 1:
The seismic dataset is divided into multiple seismic lines, which are further segmented into tiles. This segmentation allows the system to process manageable portions of data while maintaining comprehensive coverage. The method processes individual tiles and aggregates results to form complete seismic representations, balancing thoroughness with efficiency.
Solution Approach 2:
The system extracts representative seismic lines from the large dataset by computing feature vectors and applying unsupervised clustering. This extraction identifies and isolates the most informative seismic lines that capture the essential geological characteristics, allowing accurate interpretation without processing the entire dataset.
2Manufacturing precision
If more seismic lines are analyzed to ensure representative sampling, then the quality of training data improves, but the computational complexity increases
Solution Approach 1:
The system transforms seismic line data into feature vectors by computing statistical parameters such as mean, standard deviation, skewness, and kurtosis of amplitude values. This parameter transformation reduces the dimensionality and complexity of the data while preserving the essential characteristics needed for high-quality training data selection.
Solution Approach 2:
Feature vectors serve as an intermediary representation between the raw seismic data and the clustering algorithm. These compact feature vectors capture the essential characteristics of each seismic line, enabling efficient comparison and clustering without requiring direct processing of the full-resolution seismic data.
3Productivity
If the seismic dataset is reduced to accelerate processing, then the productivity increases, but the risk of missing representative seismic images increases
Solution Approach 1:
The system performs preliminary processing by computing feature vectors for all seismic lines before clustering. This preliminary action prepares the data in advance, allowing the clustering algorithm to quickly identify representative samples without missing critical information. The feature extraction is done once and reused throughout the analysis.
Solution Approach 2:
The unsupervised clustering algorithm provides feedback by identifying clusters of similar seismic lines and selecting representatives from each cluster. This feedback mechanism ensures that the selected subset comprehensively covers the diversity of the original dataset, maintaining reliability while reducing the data volume for final analysis.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A seismic dataset and a task to be performed with the seismic dataset may be received. A representative seismic line representative of the seismic dataset may be generated. The representative seismic line may include pixel data representative of the seismic dataset. Based on the representative seismic line, the task may be performed. The task may include at least finding an analogous geological region by searching for an analogous seismic dataset existing in a seismic database by comparing the representative seismic line with the analogous seismic dataset's representative seismic line.