Source Wavelet Independent Seismic Inversion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seismic full waveform inversion methods depend on the accuracy of the source wavelet, which can lead to large discrepancies in the inverted model, especially at greater depths, making it challenging to achieve high-resolution geological models for complex subsurface structures.
Innovation Solution
The method modifies the misfit function to be independent of the source wavelet by using a matching filter and adjoint source calculations, allowing for the generation of inverted model parameters that are independent of the source wavelet choice.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional full waveform inversion methods are used, then the inverted model can be obtained, but the accuracy deteriorates at greater depths due to dependence on source wavelet
Solution Approach 1:
The patent extracts and removes the source wavelet component from the misfit function by using a matching filter approach. This allows the inversion to proceed without requiring accurate knowledge of the source wavelet, thereby eliminating the primary source of error that causes accuracy deterioration at greater depths.
Solution Approach 2:
The patent changes the parameterization of the misfit function from one that includes source wavelet convolution to one that uses a matching filter. This parameter change transforms the inversion problem to be independent of source wavelet characteristics, improving both accuracy and reliability throughout the depth range.
2Measurement precision
If source wavelet independent misfit function is used, then the inverted model accuracy is improved, but the device complexity increases due to matching filter and adjoint source calculations
Solution Approach 1:
The patent performs preliminary calculations of the matching filter and adjoint source before the main inversion loop. By pre-computing these components, the complexity is front-loaded and the main inversion iterations become simpler and more efficient, reducing the overall computational burden despite the added sophistication.
3Manufacturing precision
If high-resolution geological models are constructed, then the characterization of complex subsurface structures is improved, but the computational resources required increase
Solution Approach 1:
By extracting the source wavelet dependency from the inversion process, the method reduces computational errors that would otherwise require additional iterations and computational resources to correct. This leads to more efficient convergence to high-resolution models.
Solution Approach 2:
The patent uses adjoint source calculations that efficiently reuse forward modeling results through time-reversal, avoiding the need for separate forward simulations for each inversion iteration. This copying approach significantly reduces computational resources while maintaining high model resolution.
Data Source
AI summary
Method and apparatus for performing a seismic full waveform inversion independent of a source wavelet to generate a final velocity model of subsurface formations of a survey region are provided. The method includes positioning seismic data recording sensors in the survey region; blasting at points of incidence in the survey region to generate seismic waves; and sensing and recording the seismic waves using the seismic data recording sensor. The recorded seismic waves are seismic data. The method further includes transmitting the seismic data to a computer system including one or more memories and storing the seismic data in one or more memories; storing a source wavelet in the one or more memories; performing, by the computer system, a forward modeling operation using the source wavelet; and generating, by the computer system, the final velocity model for the seismic full waveform inversion using the forward modeling operation.