Seismic Wavefield Deghosting via Orthogonal Matching Pursuit
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Marine seismic surveying is hindered by ghost data, which results in reduced accuracy in representing subsurface structures due to reflections from the air-water interface, causing amplitude and phase distortions and frequency elimination in measurement data.
Innovation Solution
The implementation of an iterative Orthogonal Generalized Matching Pursuits (OGMP) technique to determine a target wavefield, specifically an upgoing wavefield, by using dictionary elements that are products of a ghost operator and a complex exponential, allowing for deghosting and crossline interpolation, thereby removing ghost data and improving data accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If marine seismic surveying is performed with receivers positioned beneath the water surface, then the ability to collect subsurface seismic data is achieved, but ghost data is introduced due to reflections from the air-water interface
Solution Approach 1:
The patent extracts and removes the harmful ghost reflections from the measured seismic signal through iterative wavefield separation. By modeling the ghost operator and applying orthogonal matching pursuit, the algorithm separates the upgoing primary reflections from the downgoing ghost reflections, effectively removing the harmful component while preserving the useful signal.
Solution Approach 2:
The patent introduces an iterative wavefield separation algorithm as an intermediary processing step between data acquisition and subsurface imaging. This mediator uses dictionary learning and orthogonal matching pursuit to decompose the total wavefield into upgoing and downgoing components, enabling removal of ghost data without requiring physical modification of the survey setup.
2Measurement precision
If iterative wavefield separation using orthogonal matching pursuit is applied, then ghost data is removed and measurement precision is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary dictionary learning and atom selection before the main wavefield separation task. By pre-computing the ghost operator dictionary and selecting relevant atoms in advance, the algorithm reduces the computational burden during the iterative orthogonal matching pursuit phase, enabling efficient processing while maintaining high precision.
Solution Approach 2:
The patent applies partial action by limiting the number of iterations and selected atoms to the minimum necessary for effective ghost removal. Rather than performing exhaustive processing, the algorithm uses a controlled number of iterative steps with sparse atom selection, achieving sufficient precision while significantly reducing computational complexity compared to full-waveform inversion or other exhaustive methods.
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
The OGMP technique effectively removes ghost data, enhancing the accuracy of seismic data by reducing interference and increasing the spatial bandwidth for signal reconstruction, allowing for more precise imaging of subsurface structures.
Implementation Method 1
one or more seismic sources that are activated to produce seismic wavefields propagated into the subsurface structure
Implementation Method 2
A part of the seismic wavefields is reflected from the subsurface structure and detected by seismic receivers
Implementation Method 3
the seismic wavefield reflected from the subsurface structure continues to propagate upward past the receivers towards the air-water interface, where the seismic wavefield is reflected back downwardly. This reflected, generally downwardly traveling seismic wavefield from the air-water interface is detected by the seismic receivers as ghost data
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
Survey data corresponding to a subsurface region of interest is received. A wavefield is determined by iteratively performing the following until a specified condition is satisfied. For a current iteration, an element that includes a representation of at least one portion of the wavefield is selected based at least in part on a current residual representing an approximation error. For the current iteration, a respective data structure is computed from the selected element. The data structure is orthogonally projected onto a space spanned by a plurality of data structures including the computed data structure. The current residual is based at least in part on the orthogonal projection.