Seismic Fracture Identification via Hessian Matrix Eigenvalue Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current coherence algorithms are inadequate for identifying minor fractures in seismic data due to obscure breakpoints and multi-solutions, leading to reduced accuracy in fracture identification.
Innovation Solution
A method and device that determine three components of structure quantification for each data point in seismic data, construct a structure quantification matrix, and extract fracture attributes using a first-order derivative of the Gaussian function, allowing for improved identification of minor fractures through quantitative analysis of seismic data structure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If coherence algorithms are used for fracture identification, then the method is simple and widely applicable, but the accuracy of identifying minor fractures deteriorates due to obscure breakpoints and multi-solution problems
Solution Approach 1:
The patent transforms the fracture identification problem from direct coherence calculation to eigenvalue analysis of the Hessian matrix. By changing the mathematical parameters from simple correlation coefficients to second-derivative-based curvature measurements, the method achieves better sensitivity to minor fractures while maintaining computational feasibility through standardized matrix operations.
Solution Approach 2:
The patent replaces the traditional mechanical correlation-based coherence algorithm with a mathematical field approach using Hessian matrix eigenvalue analysis. This substitution transitions from direct trace-to-trace comparison to analyzing the curvature field derivatives, eliminating multi-solution problems while preserving ease of implementation through systematic matrix computation.
2Device complexity
If traditional coherence algorithms are applied, then computational complexity is low, but the ability to detect minor fractures with slight dislocations deteriorates
Solution Approach 1:
The patent changes the detection parameters from first-order coherence to second-order Hessian matrix eigenvalues. This parameter transformation increases detection sensitivity to minor fractures by capturing curvature changes, while the algorithmic complexity remains manageable through established linear algebra computational frameworks.
Solution Approach 2:
The patent adds a dimensional transformation by moving from 1D trace correlation to 2D spatial derivative analysis through the Hessian matrix. This dimensional elevation enables detection of subtle fracture patterns in multiple directions simultaneously, improving sensitivity without proportionally increasing computational burden.
3Productivity
If coherence algorithms are used, then processing speed is fast, but the reliability of fracture identification deteriorates due to multi-solution problems
Solution Approach 1:
The patent extracts the essential fracture information by isolating the dominant eigenvalue and its corresponding eigenvector from the Hessian matrix. This extraction eliminates multi-solution ambiguity by focusing on the primary curvature direction, while maintaining fast processing through efficient eigenvalue decomposition algorithms.
Solution Approach 2:
The patent transforms the identification parameter from multi-valued coherence coefficients to single-valued dominant eigenvalue metrics. This parameter change ensures unique solutions by capturing the primary fracture orientation and curvature, improving reliability while preserving computational efficiency through standardized eigenvalue solvers.
Data Source
AI summary
A method and device of identifying a fracture are provided in the embodiments of the present application. The method comprises: determining three components of structure quantification for each data point in a seismic data volume; constructing a structure quantification matrix of the data point according to the three components of structure quantification for each of the data points; determining feature value and feature vector of the structure quantification matrix of each of the data points; determining fracture attribute value of the data point according to the feature value and feature vector of the structure quantification matrix of each of the data points; constructing a data volume of the fracture attribute according to the fracture attribute values of respective data points; and performing a fracture extraction for the data volume of the fracture attribute according to the feature vectors of the structure quantification matrix of the respective data points. The embodiments of the present application can improve the accuracy of identifying a minor fracture, so as to realize an effective identification of the minor fracture.