Seismic Fault Tracking via Gaussian Slope Decomposition
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current seismic signal processing techniques struggle with non-stationary seismic data, which are common in hydrocarbon geophysical prospecting, as they often assume stationarity, leading to inefficiencies and errors in higher dimensions, particularly in handling data with changing frequency content and curvature.
Innovation Solution
A method involving a cascade of filtering operations in the frequency-wavenumber domain using tiled Gaussian windows, followed by processing and adjoint operations with normalization, to decompose and reassemble seismic data, allowing for flexible handling of non-stationarity and incorporation of prior knowledge across multiple dimensions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If patch-based methods are used to handle non-stationary seismic data, then local processing efficiency is improved, but visible patch boundaries and merging errors occur
Solution Approach 1:
The patent applies segmentation by dividing the seismic data processing into local patches that are processed independently, then merged together. This allows efficient local processing while the merging step reconstructs the complete image, resolving the contradiction between processing efficiency and image continuity.
2Reliability
If non-stationary filters varying with position are used, then smooth variation without visible boundaries is achieved, but the problem becomes underdetermined requiring regularization
Solution Approach 1:
The patent applies local quality by using patches with localized support that are processed independently. Each patch captures local characteristics of non-stationary data, allowing smooth variation without visible boundaries while avoiding the underdetermined inverse problem through localized processing.
3Measurement precision
If the number of filter coefficients increases to capture non-stationarity, then filtering accuracy is improved, but computational cost scales poorly with dimensions
Solution Approach 1:
The patent segments the large-scale filtering problem into smaller local patches. Each patch requires fewer filter coefficients, maintaining filtering accuracy for local non-stationarity while dramatically reducing computational cost by avoiding the need to solve a single large underdetermined system.
Data Source
AI summary
Method for locating fault lines or surfaces in 2-D or 3-D seismic data based on the fact that fault discontinuities in the space domain span a wide range in a local slowness (slope) domain, whereas other dipping events in the space domain data, such as noise, tend to be coherent, and hence will appear focused in the slowness dimension. Therefore, the method comprises decomposing the seismic data (102) by a transformation to the local slowness domain, preferably using Gaussian slowness period packets as the local slowness or slope decomposition technique, thereby avoiding problems with the data stationary assumption. In the local slowness domain, faults may be identified (104) using the principle mentioned above, i.e. that faults are represented as a truncation in the space domain data, hence they will appear broadband in the slowness dimension.