Seismic Streamer Deghosting via L1 Inversion and Radon Transform
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current receiver deghosting methods for seismic streamer data in geophysical prospecting face challenges such as distortions caused by the air-water interface, requiring dense wavefield sampling and often making assumptions that violate real data acquisition conditions, leading to interpolation errors and limited applicability in 3D scenarios.
Innovation Solution
The method employs a computer-implemented L1 inversion using a Redundant, Hybrid, Apex-Shifted Radon dictionary to reconstruct and deghost 3D seismic streamer data, allowing for simultaneous wavefield reconstruction and receiver deghosting, capable of handling sparse data acquisition and arbitrary streamer configurations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional receiver deghosting methods are used, then deghosting can be achieved, but dense wavefield sampling is required and assumptions are made that violate real data acquisition conditions
Solution Approach 1:
The patent changes the fundamental parameter of wavefield representation by using a Radon transform-based dictionary to sparsely represent the wavefield in terms of wavefront curvature and apex position, rather than requiring dense spatial sampling. This allows accurate deghosting with sparse data by transforming the problem from spatial domain to Radon domain where the wavefield exhibits sparsity.
Solution Approach 2:
The patent replaces the mechanical requirement of dense physical sampling with a computational approach using L1-norm inversion and Radon dictionary. Instead of physically acquiring dense wavefield samples, the method uses sparse sampling combined with computational reconstruction through dictionary matching and optimization algorithms.
2Measurement precision
If traditional deghosting methods with dense sampling are applied, then wavefield reconstruction is possible, but interpolation errors occur and applicability in 3D scenarios is limited
Solution Approach 1:
The patent creates a universal deghosting method that works across different acquisition geometries (2D and 3D, various streamer configurations) by using the Radon transform which can represent wavefields in arbitrary geometries. The same L1 inversion framework adapts to different dimensionalities and acquisition patterns without requiring geometry-specific assumptions.
Solution Approach 2:
The patent moves the problem from spatial domain to Radon domain by introducing new dimensions of wavefront curvature and apex position parameters. This dimensional transformation allows the method to handle sparse and irregular sampling patterns that would be problematic in traditional spatial domain approaches.
3Reliability
If L1 inversion with RHARD dictionary is used, then robust deghosting is achieved for sparse data, but computational complexity increases
Solution Approach 1:
The patent segments the large-scale L1 inversion problem into smaller sub-problems by dividing the Radon dictionary into manageable blocks and processing the inversion in stages. This segmentation allows the use of efficient sparse matrix operations and reduces the memory and computational burden of solving the full problem at once.
Solution Approach 2:
The patent introduces the Radon transform as an intermediary mathematical operation that bridges the sparse measured data and the reconstructed wavefield. This intermediary representation in Radon domain simplifies the inversion problem by exploiting the sparsity of wavefront parameters, making the computationally intensive L1 inversion more tractable.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This approach provides a robust and accurate deghosting solution that preserves bandwidth and resolution, effectively removing receiver ghost reflections and reconstructing wavefields at arbitrary positions, even in complex and sparse data scenarios.
Implementation Method 1
simultaneous wavefield reconstruction and receiver deghosting of three-dimensional (3D) seismic streamer data using an L1 inversion
Data Source
AI summary
Raw 3D seismic streamer wavefield data is received as a receiver-ghosted shot gather. The received receiver-ghosted shot gather shot gather is processed into a normalized form as normalized data. The normalized data is partitioned into a plurality of user-defined sub-gathers and processed to generate a complete receiver-deghosted shot gather. Output of the complete receiver-deghosted shot gather is initiated.