Seismic Imaging via Waveform Inversion by Relative Data Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional seismic imaging methods face challenges such as cycle skipping, noise inclusion, and difficulty in interpreting subsurface features due to reliance on ray-based algorithms and reflection full waveform inversion, which results in inaccurate earth models and poor seismic image quality.
Innovation Solution
The method of waveform inversion by relative data matching (WIRDM) uses extended reverse time migration and extended Born modeling to generate more accurate seismic images by cross-correlating wavefields, applying image processing, and subtracting multiple modeled datasets to create a data residual for iterative earth model updates, thereby improving focusing and noise attenuation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional seismic imaging methods using ray-based algorithms and reflection full waveform inversion are used, then the imaging process can be performed, but the earth model accuracy and seismic image quality deteriorate due to cycle skipping and noise inclusion
Solution Approach 1:
The patent applies inverse scattering theory to invert the scattering potential from high-frequency seismic data, reversing the conventional approach by starting from observed data and working backward to derive the earth model, thereby avoiding cycle skipping and improving both accuracy and reliability
Solution Approach 2:
The patent transforms the seismic imaging problem by changing from conventional reflection full waveform inversion to scattering potential inversion, modifying the mathematical parameters and approach to eliminate cycle skipping artifacts and improve image quality
2Productivity
If conventional seismic imaging methods are used, then the imaging process can be completed, but subsurface feature delineation deteriorates due to poor focusing and noise inclusion
Solution Approach 1:
By inverting the scattering potential rather than using conventional forward modeling and reflection inversion, the method achieves superior focusing of subsurface features while maintaining computational efficiency through direct inversion of high-frequency data
Solution Approach 2:
The patent extracts and removes noise components from the seismic data through the scattering potential inversion process, isolating the meaningful subsurface signals while eliminating harmful noise that degrades feature delineation
3Measurement precision
If high-frequency seismic data is used, then better resolution can be achieved, but noise inclusion and cycle skipping worsen
Solution Approach 1:
The patent converts the high-frequency noise and cycle skipping artifacts into beneficial information by using inverse scattering theory that specifically exploits high-frequency content to achieve better resolution while eliminating the harmful effects through proper mathematical inversion
Solution Approach 2:
By changing the inversion approach from conventional reflection methods to scattering potential inversion, the patent transforms the parameters and mathematical framework to handle high-frequency data effectively, converting potential harm into improved resolution
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 results in more accurate digital seismic images that better delineate subsurface features, including hydrocarbon deposits, enhancing exploration and production decisions by providing clearer boundaries and improved hydrocarbon reservoir identification.
Implementation Method 1
The sources generate seismic waves, which propagate into the geological medium creating pressure changes and vibrations
Implementation Method 2
cross-correlating wavefields, applying image processing, and subtracting multiple modeled datasets
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method is described for seismic imaging including generating extended image gathers by extended reverse time migration of a seismic dataset using an earth model; processing the extended image gathers to generate processed image gathers; performing extended modeling based on the processed image gathers to generate a modeled seismic dataset; enhancing the processed image gathers to generate an enhanced image; performing extended modeling based on the enhanced image gathers to generate a modeled enhanced dataset; differencing the modeled enhanced dataset and the modeled seismic dataset to determine a data residual; inverting the data residual to generate a model residual; updating the earth model based on the model residual to create an updated earth model; performing seismic imaging of the seismic dataset using the updated earth model to create an improved seismic image. The method may be executed by a computer system.