Target-Oriented Reverse Time Migration for Prestack Depth Imaging
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Reverse-time migration (RTM) in geophysical prospecting is computationally intensive due to the high cost of solving two-way wavefield propagation, requiring significant storage and computation time, and existing methods to enhance efficiency are not sufficient to make it practical for widespread application.
Innovation Solution
A target-oriented reverse time migration method that focuses on specific subsurface areas by using Green's functions to relate receiver locations to imaging targets, reducing computational requirements by limiting wavefield propagation and storage to critical areas, and applying two-way scattered Green's functions directly for imaging without intermediate data datuming.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If reverse-time migration is used to achieve two-way wavefield propagation for imaging complex subsurface structures, then imaging accuracy is improved, but computational cost and storage requirements increase significantly
Solution Approach 1:
The patent divides the wavefield propagation process into forward and reverse components, and separates the imaging process into distinct stages: forward wavefield propagation, data back-extrapolation, and correlation with source wavefield. This segmentation allows for optimized computation at each stage, reducing overall computational complexity while maintaining imaging accuracy.
Solution Approach 2:
The patent implements target-oriented RTM that performs wavefield propagation and imaging only for specific target regions rather than the entire subsurface volume. By limiting the imaging domain to areas of interest and using adaptive grid refinement, the method reduces computational cost and storage requirements while maintaining full imaging accuracy for the targeted regions.
2Loss of information
If full 3D source wavefields are stored for reverse-time access, then imaging completeness is improved, but storage space requirements increase significantly
Solution Approach 1:
The patent extracts only the essential wavefield information needed for imaging from the full 3D source wavefield. By using one-way wave equations to propagate the source wavefield and selectively storing only the necessary components for correlation with back-extrapolated data, the method maintains imaging completeness while dramatically reducing storage requirements.
Solution Approach 2:
The patent uses numerical simulations and Green's function-based approaches to create simplified representations of the wavefield propagation. Instead of storing complete 3D wavefields, the method uses computed Green's functions and one-way wavefield solutions that replicate the essential imaging information with minimal storage requirements.
3Measurement precision
If two-way wavefield propagation is performed for RTM, then wave propagation accuracy in any direction is improved, but runtime cost increases to approximately twice that of forward full-wavefield modeling
Solution Approach 1:
The patent merges the forward and reverse wavefield propagation processes by using the same computational framework for both directions. The two-way wave equation is solved using a unified finite-difference scheme that efficiently handles both forward and backward propagation, reducing the total runtime compared to separate one-way propagations.
Solution Approach 2:
The patent employs time-domain sampling and frequency-domain transformations to efficiently compute wavefield propagation. By using Fourier transforms and frequency-wavenumber filtering, the method accelerates both forward and reverse propagation steps, reducing the overall runtime while maintaining wave propagation accuracy in all directions.
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 significantly reduces computational costs by limiting the scope of wavefield propagation and storage, allowing for efficient two-way imaging while maintaining the capabilities of RTM, and is applicable to both 2D and 3D data sets, enabling focused analysis and imaging of complex subsurface structures.
Implementation Method 1
Rigorous solutions of wave equation are highly accurate in simulating wave propagation through complex subsurface regions
Implementation Method 2
synthesizing wavefields along the reference surface to reconstruct Green's functions or other equivalent transfer functions which relate the receiver locations in the acquisition area to the target
Data Source
AI summary
A method and system for target-oriented reverse time migration for prestack depth imaging. One embodiment of the present invention includes determining an acquisition area within an earth model and also determining a reference surface near a target in the subsurface region of interest. The embodiment includes exciting wavefields from the reference surface and propagating the excited wavefields through the earth model. The embodiment additionally includes recording the wavefields at receiver locations in the acquisition area and at the target and synthesizing wavefields along the reference surface to reconstruct Green's functions which relate the receiver locations in the acquisition area to the target. The embodiment includes utilizing the Green's functions and prestack seismic data to determine subsurface characteristics of the subsurface region.